Question asked on Linked in GCP group:Hi! For one of my studies, three months after the start of the recruitment, I discover the site has a competing protocol (same inclusion/exclusion criteria). What do you do in my place?
The question was commented as follows:
Stefania Bordin • I would suggest to check the Investigator availability in terms of time an recruitment potentials. Anyway, in order to avoid such risks during the fesibility phase and site choice the Investigator shoul have completed a questionnaire by indicating if he/ she has competitors' studies ongoing. In any case by signing the Protocol page he / she committs himself to conduct the trial according to Protocol which also includes all the time required to conduct the trial in the correct whay (beyond, for sure according to GCP ICH and all the regulations being applicable in this case). At the end, if you notice that the enrolment for your tiral is not good enough not even in terms of screening you can inform your head department and project manager in order to agree about the possibility to close the site.
Fatima EL GHAIB • Thanks Stefania.
During the feasibility phase, the Investigator completed the questionnaire indicating he has no competiting studies ongoing.
Just for information, how is it possible to use such a document legally?
Rebecca Georgevitch • He was already fabricating the truth prior to study acceptance. I don't believe this was a legal issue. At this point, I would be very cautious with this site and any Information given to me. He lied before he even began the trial. If they were open to enrollment and haven't enrolled in sometime, the sponsor/CRO may consider sending him a letter stating they will be shutting down sites that have less than x# of qualified subjects enrolled by a specific time. Try to close them down before the next IRB meeting/approval. This is costing them unnecessary money. I would also discuss future enrollment procedures with the PI and SC present.
Fatima EL GHAIB • Many thanks Rebecca :)
Shirley Isbill • There may be no need to do anything. It's not uncommon for study sites to run mulitple studies for several sponsors. The last study site I audited, had 30 active studies and they were recruiting for more. You can make an assessment of the study site staffing and comment to your study sponsor concerning the staff's ability to handle their clinical study workload. It would not be ethical for you to look at the details of the competing study; it might be a competitor's study of their drug for the same indication in your study.
Malaika Simmons • I agree with Shirley. Additionally, if there was a feasibility questionnaire completed that indicated that there were no competing studies, it may be a matter of opinion. For example, your study may be a fibroid study, and the site may have an ongoing endometriosis study. The PI may say that those are two different illnesses, therefore are noncompeting. However, we know that the inc/excl criteria will have similar if not nearly identical parameters.
That being said, you will still have to treat the site as you should all sites, and remind the PI /coordinator of the sites responsibility to recruit to the protocol specifications. It is a good idea to develop a recruitment plan (that includes methods as well as numbers) per site as part of the study initiation as a rule to mitigate under enrollment issues and reemphasizes roles and responsibilities (amazing that signing the 1572 does not immediately impart an overwhelming sense of responsibility! :-)).
Good luck
Fatima EL GHAIB • Many thanks to all of you! Your advices are very helpfull :)
In fact, when discussing with another site staff, it appears that they refused another study because very similar: same inclusion/exclusion criteria and same indication! It also appears that these 2 trials were both conducted by the same CRO (I was working for), that's why I had the possibility to read the competitive protocol.
The first site I was speaking about had been less realistic than the second one and accepted to conduct both studies, but finally wasn't able to reach his goal...
Kamakshi Sriram • Hi Fatima,
This is strange as while we are debating on the intentions of site one which has accepted to do competing studies at the same time, I have doubts on the intentions of the CRO now, who is going ahead and enrolling the same investigator for a competing study, the investigator would take the study up if CRO ( representing sponsor here as responsibilities are being delegated) is going ahead and even proposing a new study to the same site. The reason could be either the site is too talented..........or the recruitment period is over of the first study.............or sheer negligence on the part of the CRO.
Christoph Lohan (PgDip.) • Without having more details about the issue I would agree with Malaika. It may just be that the protocols are very similar but not wholly identical as they may target slightly different groups or conditions. In Oncology and Heamatology you will find it alot, e.g. breast cancer or AML. Woth looking at it in more detail.
Harpreet Kaur Anand • If Protocol is same, all Inclusion-Exclusion Criteria,Study Duration etc therefore its futile taking such studies.Its merely shows that the site is not interested in study, only concerned with the money they will receive from the it.
1) Patient recruitment will be less as site will be manage to enroll patient in both studies
2) This will be biased
If company is doing research on some molecule they must have targeted patients pool & site accordingly therefore if they will not get the sufficient data from sufficient number of patients then its useless conducting trial at such type of sites in near futute.
Rather than taking any strict action, the site should be warned about this & as per Miss Simmons PI should be reminded/educated of the sites responsibility to recruit to the protocol specifications.
Shirley Isbill • Fatima,
Re: It also appears that these 2 trials were both conducted by the same CRO (I was working for)
It's not uncommon for the left hand to not know what the right hand is doing; refer the common protocols to the CRO then forget about it.
Fatima EL GHAIB • I started in that CRO and on that study after the sites had been selected.
Kamakshi, I agree with you. As CRA I had to be informed by the CRO about the conduct of that competitive study so as to be more vigilant.
Thank you all for your active participation. Your comments are very constructive :)
20 days ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion0 . Lynda Cedar • It's common for a site to conduct similar studies (study design, same drug or simlar products), the same inc/exc criteria used.
- If a NDA is in place, the site is not supposed to disclose any information about its current studies, other than speaking about their experience with a specific disease or class of medication.
If a NDA is in place, allowing you to know about the other study, and even showing you the study protocol of the other sposnor, that is a violation of the NDA and confidential information provided to the site.
This disclosure is not ethical and might disavantage the other sponsor:if he does not know the same..
- Regarding the recruitment, no worries if a CTRA is in place and contains milestones for the recruitment, the site is going to deliver.
Maneesh Misra • You should consider Kamakshi, Rebecca & Malaika's comments seriously. The trial is not meant for only the Sponsors' drug & Study Design but also for effecasy & safety of patients. To fulfill the needs of NDA(Non Disclosure Agreements), you cannot compromise on the results of the trial which are more likely to be BIASED in this senarion mentioned by you.
It should be your moral responsibility to conduct the trial without any element of BIASED approach or ignorance as seen in this case. The PI may have taken the approach to enable the site to build more credibility & increase the list of achivements but as a professional we should take these issues seriously & act accordingly.
I am not in the correct position to advice you for the approach to be taken to resolve the violations, if any, as the details discussed is not sufficient for it. Ask your self & evaluate it agailst the GCP & IRB guidelines without any Biased thoughts & I am sure you will get the answer yourself.
All the participants made good approach & it is appreciable that these discussions will always help in learning more n more. I have also learned things from this discussion & everyones views & would like to Thank everybody.
Shannen Douglass • It is not uncommon for a site to have competing studies. As long as recruitment targets are being met, protocol is being followed and there is enough time for the staff to properly run both trials, you may not need to do anything at all.
Melvin paul • Hi Fatima,
Most often investigators do not disclose complete information as they feel additional project would help their profile. However in a situation like this it is good to speak to the investigator about your concern and let her/him know that there is an equal commitment that the PI should show to both the studies. You could suggest that he follows a fifty/fifty rule, where if he has two subjects screened on a day, then he should enroll one for one study and the other into the other study. If he could do this then he will ensure he has lived up to the commitments of both the studies. I quote this from a similar experience when i was a CRA. This startergy did work. However you will need to keep and eye on the recruitment time. If you see that the PI is not making any progress, it is better to either close the site or get permission to extend the recruitment time period.
Nancy Nahmias • Many times the site staff completes the feasibility, and the PI just signs it. It is up to the sponsor to do due diligence and review the feasibility and ask questions. Typically sponsors know what protocols are competing with theirs, and as a CRA I always ask this of the project lead.
Since the study has already started, I would put the PI's feet to the fire by ensuring you get regular enrollment logs, and having a lengthy discussion with the PI about enrollment at each visit. I would also mention that this competing protocol was not mentioned on the feasibility, but since they are participating ask the PI how he assigns subjects to a specific trial. If you get a lousy response to enrollment, I would tell the PI that you will close them if they have no enrollment after 3 months....
Bhugol Chandel, MD • Hi Fatima,
As long as the site and the investigator have time, patient population and the resources to enroll the patients, they need to be supported for their efforts to meet enrollment challenge.
Laura C. Collada Ali • What about the Ethics Committee? Is it possible that they do not point out such a case? Don't they have a register of ongoing studies?
Megan Mather • It is to be expected, especially when the PI is in a specialty, such as cardiac research, specific oncology, or neurology. Many moons ago I was a study coordinator working on migraines. I had as many as 7 studies all on migraines. As long as the site can provide a sufficient number of subjects and definite proof that appropriate washouts of drugs from trials are performed, there really shouldn't be a problem. The PI and coordinator should be interviewed on how they keep the trials separate and how they make sure they do not confuse the trials. All study documentation should be kept separate and away from other studies (no shared labs, etc.) and when information is addressed in a common patient file, if these are the PI's private patients as well, only the trial number should be listed in the patient chart. The one area I have special concerns over is during Phase I studies where they have "professional patients". We had to report a site in Florida where there were 2 Phase I units within one mile from the other and we found the patients were going from one unit to the other without washout and in some cases at the same time. Many of the studies were at risk, it was bad news!!
Stefania Bordin • I would suggest to check the Investigator availability in terms of time an recruitment potentials. Anyway, in order to avoid such risks during the fesibility phase and site choice the Investigator shoul have completed a questionnaire by indicating if he/ she has competitors' studies ongoing. In any case by signing the Protocol page he / she committs himself to conduct the trial according to Protocol which also includes all the time required to conduct the trial in the correct whay (beyond, for sure according to GCP ICH and all the regulations being applicable in this case). At the end, if you notice that the enrolment for your tiral is not good enough not even in terms of screening you can inform your head department and project manager in order to agree about the possibility to close the site.
Fatima EL GHAIB • Thanks Stefania.
During the feasibility phase, the Investigator completed the questionnaire indicating he has no competiting studies ongoing.
Just for information, how is it possible to use such a document legally?
Rebecca Georgevitch • He was already fabricating the truth prior to study acceptance. I don't believe this was a legal issue. At this point, I would be very cautious with this site and any Information given to me. He lied before he even began the trial. If they were open to enrollment and haven't enrolled in sometime, the sponsor/CRO may consider sending him a letter stating they will be shutting down sites that have less than x# of qualified subjects enrolled by a specific time. Try to close them down before the next IRB meeting/approval. This is costing them unnecessary money. I would also discuss future enrollment procedures with the PI and SC present.
Fatima EL GHAIB • Many thanks Rebecca :)
Shirley Isbill • There may be no need to do anything. It's not uncommon for study sites to run mulitple studies for several sponsors. The last study site I audited, had 30 active studies and they were recruiting for more. You can make an assessment of the study site staffing and comment to your study sponsor concerning the staff's ability to handle their clinical study workload. It would not be ethical for you to look at the details of the competing study; it might be a competitor's study of their drug for the same indication in your study.
Malaika Simmons • I agree with Shirley. Additionally, if there was a feasibility questionnaire completed that indicated that there were no competing studies, it may be a matter of opinion. For example, your study may be a fibroid study, and the site may have an ongoing endometriosis study. The PI may say that those are two different illnesses, therefore are noncompeting. However, we know that the inc/excl criteria will have similar if not nearly identical parameters.
That being said, you will still have to treat the site as you should all sites, and remind the PI /coordinator of the sites responsibility to recruit to the protocol specifications. It is a good idea to develop a recruitment plan (that includes methods as well as numbers) per site as part of the study initiation as a rule to mitigate under enrollment issues and reemphasizes roles and responsibilities (amazing that signing the 1572 does not immediately impart an overwhelming sense of responsibility! :-)).
Fatima EL GHAIB • Many thanks to all of you! Your advices are very helpfull :)
In fact, when discussing with another site staff, it appears that they refused another study because very similar: same inclusion/exclusion criteria and same indication! It also appears that these 2 trials were both conducted by the same CRO (I was working for), that's why I had the possibility to read the competitive protocol.
The first site I was speaking about had been less realistic than the second one and accepted to conduct both studies, but finally wasn't able to reach his goal...
Kamakshi Sriram • Hi Fatima,
This is strange as while we are debating on the intentions of site one which has accepted to do competing studies at the same time, I have doubts on the intentions of the CRO now, who is going ahead and enrolling the same investigator for a competing study, the investigator would take the study up if CRO ( representing sponsor here as responsibilities are being delegated) is going ahead and even proposing a new study to the same site. The reason could be either the site is too talented..........or the recruitment period is over of the first study.............or sheer negligence on the part of the CRO.
Christoph Lohan (PgDip.) • Without having more details about the issue I would agree with Malaika. It may just be that the protocols are very similar but not wholly identical as they may target slightly different groups or conditions. In Oncology and Heamatology you will find it alot, e.g. breast cancer or AML. Woth looking at it in more detail.
Harpreet Kaur Anand • If Protocol is same, all Inclusion-Exclusion Criteria,Study Duration etc therefore its futile taking such studies.Its merely shows that the site is not interested in study, only concerned with the money they will receive from the it.
1) Patient recruitment will be less as site will be manage to enroll patient in both studies
2) This will be biased
If company is doing research on some molecule they must have targeted patients pool & site accordingly therefore if they will not get the sufficient data from sufficient number of patients then its useless conducting trial at such type of sites in near futute.
Rather than taking any strict action, the site should be warned about this & as per Miss Simmons PI should be reminded/educated of the sites responsibility to recruit to the protocol specifications.
Shirley Isbill • Fatima,
Re: It also appears that these 2 trials were both conducted by the same CRO (I was working for)
It's not uncommon for the left hand to not know what the right hand is doing; refer the common protocols to the CRO then forget about it.
Fatima EL GHAIB • I started in that CRO and on that study after the sites had been selected.
Kamakshi, I agree with you. As CRA I had to be informed by the CRO about the conduct of that competitive study so as to be more vigilant.
Thank you all for your active participation. Your comments are very constructive :)
Lynda Cedar • It's common for a site to conduct similar studies (study design, same drug or simlar products), the same inc/exc criteria used.
- If a NDA is in place, the site is not supposed to disclose any information about its current studies, other than speaking about their experience with a specific disease or class of medication.
If a NDA is in place, allowing you to know about the other study, and even showing you the study protocol of the other sposnor, that is a violation of the NDA and confidential information provided to the site.
This disclosure is not ethical and might disavantage the other sponsor:if he does not know the same..
- Regarding the recruitment, no worries if a CTRA is in place and contains milestones for the recruitment, the site is going to deliver.
Maneesh Misra • You should consider Kamakshi, Rebecca & Malaika's comments seriously. The trial is not meant for only the Sponsors' drug & Study Design but also for effecasy & safety of patients. To fulfill the needs of NDA(Non Disclosure Agreements), you cannot compromise on the results of the trial which are more likely to be BIASED in this senarion mentioned by you.
It should be your moral responsibility to conduct the trial without any element of BIASED approach or ignorance as seen in this case. The PI may have taken the approach to enable the site to build more credibility & increase the list of achivements but as a professional we should take these issues seriously & act accordingly.
I am not in the correct position to advice you for the approach to be taken to resolve the violations, if any, as the details discussed is not sufficient for it. Ask your self & evaluate it agailst the GCP & IRB guidelines without any Biased thoughts & I am sure you will get the answer yourself.
All the participants made good approach & it is appreciable that these discussions will always help in learning more n more. I have also learned things from this discussion & everyones views & would like to Thank everybody.
Shannen Douglass • It is not uncommon for a site to have competing studies. As long as recruitment targets are being met, protocol is being followed and there is enough time for the staff to properly run both trials, you may not need to do anything at all.
Melvin paul • Hi Fatima,
Most often investigators do not disclose complete information as they feel additional project would help their profile. However in a situation like this it is good to speak to the investigator about your concern and let her/him know that there is an equal commitment that the PI should show to both the studies. You could suggest that he follows a fifty/fifty rule, where if he has two subjects screened on a day, then he should enroll one for one study and the other into the other study. If he could do this then he will ensure he has lived up to the commitments of both the studies. I quote this from a similar experience when i was a CRA. This startergy did work. However you will need to keep and eye on the recruitment time. If you see that the PI is not making any progress, it is better to either close the site or get permission to extend the recruitment time period.
Nancy Nahmias • Many times the site staff completes the feasibility, and the PI just signs it. It is up to the sponsor to do due diligence and review the feasibility and ask questions. Typically sponsors know what protocols are competing with theirs, and as a CRA I always ask this of the project lead.
Since the study has already started, I would put the PI's feet to the fire by ensuring you get regular enrollment logs, and having a lengthy discussion with the PI about enrollment at each visit. I would also mention that this competing protocol was not mentioned on the feasibility, but since they are participating ask the PI how he assigns subjects to a specific trial. If you get a lousy response to enrollment, I would tell the PI that you will close them if they have no enrollment after 3 months....
Bhugol Chandel, MD • Hi Fatima,
As long as the site and the investigator have time, patient population and the resources to enroll the patients, they need to be supported for their efforts to meet enrollment challenge.
Laura C. Collada Ali • What about the Ethics Committee? Is it possible that they do not point out such a case? Don't they have a register of ongoing studies?
Megan Mather • It is to be expected, especially when the PI is in a specialty, such as cardiac research, specific oncology, or neurology. Many moons ago I was a study coordinator working on migraines. I had as many as 7 studies all on migraines. As long as the site can provide a sufficient number of subjects and definite proof that appropriate washouts of drugs from trials are performed, there really shouldn't be a problem. The PI and coordinator should be interviewed on how they keep the trials separate and how they make sure they do not confuse the trials. All study documentation should be kept separate and away from other studies (no shared labs, etc.) and when information is addressed in a common patient file, if these are the PI's private patients as well, only the trial number should be listed in the patient chart. The one area I have special concerns over is during Phase I studies where they have "professional patients". We had to report a site in Florida where there were 2 Phase I units within one mile from the other and we found the patients were going from one unit to the other without washout and in some cases at the same time. Many of the studies were at risk, it was bad news!!
Tom Quegan • Very interesting discussion!
Fatima, there is another point that has not yet been discussed. Did/does the PI know that he has two competing studies? Does he have full oversight of all his studies, or does he delegate everything to study coordinators and sub-investigators?
As an auditor, I have been to sites where the PI could not tell me about critical parts of the protocol, so this may be something that you want to be aware of.
13 days ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion3 .
Molly Blake-Michaels • Hello Fatima,
The only comment I have to add to the discussion is that in cases where there are studies that compete for the same patients, it is always advisable to find out how the site will decide which patient will be enrolled into each study. For example, I have worked with sites in the past that have alternated studies; as eligible patients come in they are offered the study whose "turn" it is. If a procedure like that isn't in place, it is easy for an investigator's bias to result in different types of patients being enrolled in one study than the other (for example, if the investigator believes that heavier patients will do better on the drug in Study X, or older patients will do worse, it could easily affect their decision as to which study to present to the patient). The result of this bias can be that the full range of patients meeting the inclusion/exclusion criteria are not enrolled, at least at that site.
Anke Hoppe • If this constitutes a protocol violation, the chain of events starting with discovery of the issue and actions taken (e.g. re-training of PI, set up of recruitment plan, ensuring maintenance of confidentiality for each study etc.) need to be documented detailed in writing, signed and filed accordingly (ISF and TMF) and the site needs to inform their IRB.
Does the CTA address this kind of incidence?
Anupama Ramkumar • The objective of anyone who is assessing these sites(monitor,auditor etc) is to ensure that the data is credible and that the patients safety and well being is not compromised.
It is not uncommon for sites to be handling multiple trials at the same time and not only that but also be a highest recruiter!!
At all times if all the stakeholders can keep the objectives of data credibility and patient safety in mind ,there really are no restrictions.
Which means there has to be an audit trail which should include documentation to ensure that the inclusion/exclusion criteria are not being compromised and that the patient is not part of more than one trial.
Dennis McHugh • If inclusion/exclusion criteria and protocols are substantially equivalent suggest that they enroll in one trial on even days; the other on odd days or alternate weeks. This would be treating both programs equally and perhaps avoid bias. You do want to avoid investigator bias where they believe (consciously or unconsciously) that Study A might just have the superior intervention so they enroll the more challenging subjects there. That's an experimental model killer.
Patrick Martin • Fatima it appears that you are very conscientious and diligent in managing the site, which is great. Regarding your post: as I used to tell my team, when in doubt, ask.
There's nothing wrong in bringing those concerns to the PI in a professional and respectful manner. At the very least the site will know that you want open and honest communication. But the bottom line is do you honestly think that the site/PI has the ability to meet the obligations, including timely enrollment, for your study? If the answer is yes, then go forward. If no, then talk to your boss.
For what it's worth, if the site has expertise in a particular area then one would expect that other sponsors would be attracted to work w/ the site as well. If the site is large enough they may have the patient load to adequately enroll in your study and the other.
Best of luck with this.
Wednesday, January 18, 2012
Tuesday, January 17, 2012
The Site’s Side: Now You See Them, Now You Don’t - The Trend of Disappearing Study Sites
Discussion brought from Linked in GCP group: The Site’s Side: Now You See Them, Now You Don’t - The Trend of Disappearing Study Sites.
CenterWatch cautions that over the last 5 years, the number of study sites has decreased, with the loss of veteran PIs most sharply in decline. How will this trend affect sponsors, other stakeholders in the business of clinical research and the provision of new products to the public if it continues?
The Site’s Side: Now You See Them, Now You Don’t – The Trend of... blog.gobalto.com
By: Sherry Reuter A January. 5, 2012 report from CenterWatch cautions that over the last 5 years, the number of study sites has decreased, with the loss of veteran PIs most sharply in decline. Specifically: The proportion of...
Commented about it: Angela Rhodes • I agree, it is becoming more burdensome for a physician to conduct trials at his office. The documentation burden is high and in many cases monitors are less 'helpers' to the site and more 'police'. In many cases budgets 'piggyback' on top of standard of care and do not compensate. However, the documentation burden is high, even on these standard of care. Docs just want to 'do their job' and not complicate their lives/practices further. Sponsors should have a 'team' that helps set up a research endeavor at the practice, use their expertise to develop a flow and plan. Make it less burdensome for Docs. This should be separate from CRA/Monitors.
Natalie Borde • Sometime physician are also given the option to hire in extra help to lessen the burden. But one must try to ensure that valuable information is not lost along the way. Interesting point "sponsors should have a team that helps sent up a research endeavor........from CRA/monitors". The "team' would also need to be one which is unbiased. Is there a niche in the market for this?? With reference 'expertise to develop a flow and plan', do you meant how the PI and his team can efficiently incorporate trial activities within the daily work environment?
Angela Rhodes • Natalie, each individual office has its own ebb and flow. I envision a 'site set-up' geared more toward, how are you going to find the subjects? At what point in your daily practice. (most don't have searchable databases) What records are you going to need, who is going to push the button on the machine, write it down. etc. In many cases a 'coordinator' does this. This coordinator is usually only doing research part-time or as an add-on to regular duties. I have found that if staff have a "package' and are told to follow the flow, check the boxes, etc. It is easier on the staff and you get better results.
Sherry Reuter • Dear Natale and Angela,
Wonderful comments that sound like they would indeed help sites.
The term 'site support' has so many different meanings, but you describe real help. Partnering with sites to make sure they succeed helps all stakeholders to be successful! Kind regards, Sherry
Jose Manuel Masso • I agree with the comments so far. The burden of clinical trials is increasing as we are asking physicians for more patients and more quality, which are in fact necessary. This means that on one hand there is a tendency towards more documentation, assessments, visits at sites in current clinical trials and on the other hand monitors are more under the pressure of the sponsors SOPs, audits and co-monitorisations and, as Angela mentioned, less prone to help the investigators. My opinion is that in order to be successful stronger attention has to be put during the site selection visits, to make investigators aware of all these things (burden of some clinical trials) and also I think that the role of trial coordinators at site is gaining importance and trial coordinators are becoming a substantial part in the current clinical trials context.
Natalie Borde • So Angela, what you saying is trying to guide and stimulate the site to look in the right direction to look for potential subjects. Basically what one could propose is creating a manual for patient recruitment. But you mentioned the study co-ordinator, it is not really their role to identify the subjects, but the PI or SI.
Hey Jose Manuel, hope you are well. As you mentioned, and I agree entirely subject recruitment etc, this topic is address during the site selection. Here one identifies where the potential sources are for subjects.
Perhaps the introduction of a site coach is needed. An addition to the CRA. To support the sites, for additional trainings, the steer the sites in the right directions ie. recruitment. keep the momentum going...
Lynda Cedar • The Trend of Disappearing Study Sites !...... hum, yes! That’s a true; because of different factors and main them are the strict and demanding regulation that increase the study cost and the current devastating economy.
1. Some sites disappear from here but were transferred to elsewhere (emerging countries). Their manpower is cheap, the subjects are compensated 90% less, the regulation is less strict, the Ethics are different, and the quality is consequently less.
For the sites of here that continued to serve (because sponsors need some generic studies to be conducted here), some of them could not survive to the ''devastating" economy, since 2008 and it does not end yet!....
2. As explained above by the members, the regulation is very demanding (but it has to be as such), however this increases the study cost for the sponsors and the small CROs cannot survive..... At the end the small CROs disappear, and some empire CROs, very big disappear too... ''big room, big mistakes" and therefore penalized for the errors..... In Quebec, big CRO as Phoenix Int'l then became MDS pharma services, then closed..... Anapharm changed 2 - 3 times names and becomes very quiet or even shutting down.... Atlantic Life Sciences, a small CRO, is in restructuration..... other examples are for elsewhere in Canada, Europe, USA.....
3. For the sites operating in the field of complex studies, new medications, biosimilars, etc.. point 2 is even worst: the development takes time and the cost becomes excessive, just few CROs have the expertise and are just surviving....Sponsors should be aware of.
4. Laid-off and points 2-3 create a lot of resources available for consulting.... individual enterprises. Sponsors changed their strategy of employment.
5. For studies of late stage (Phase III), a lot of diseases (diabetes, BP/CV, Parkinson, Alzheimer, Cancer, HIV, ) are still needing discovery and improvement .... but again, because of the excessive cost, pharmaceuticals need to be selective when picking a medication to develop or a treatment to improve.
In this sector of clinical research, outside our countries, the patients do not have treatments available and are more willing to try the new medication in development (phase III), the recruitment looks easier, the cost looks less. I'm not sure that they are saving more money and time, as per some discussions on this forum and other LI groups, sometimes the preliminary of ICH or GCPs is not followed.... therefore, it looks less expensive (in appearance) but if they have to redo some studies, it costs them even more....
The cost is only 7% to 10% less when studies are performed in emerging countries at performing and compliant CROs with GCPs and FDA standards/regulations.
Ivalina Hristova • @ Lynda Cedar: "The cost is only 7% to 10% less when studies are performed in emerging countries at performing and compliant CROs with GCPs and FDA standards/regulations."
Do have reference for this?
Karen Levorson • Lynda, I do not agree with you that regulations in emerging countries are less strict. In my experience, studies performed in India for example, are scrutinized quite well by the authorities and also by Ethics committees.
And as for your comment about ethics being different, I totally disagree. There are researchers (and probably drug companies) who are proposing studies that are not ethical, that they could never perform in their own countries and they are going to the poor emerging countries, misrepresenting what they are doing - I have experienced that during due diligence audits!! But these are definitely not the "norm". There are investigators and researchers without ethics in every country!
In emerging countries there are conditions that are different that can really influence the subject to stay in a study, that they think that their doctor would never propose anything that could possibly harm them and then to not complain about any adverse event because they have no other recourse but the study to receive medication for their disease - even at the "risk" of receiving placebo!!
My fear for the future is that when studies are conducted under such conditions, we will have "surprises" when they come onto the market!!
Otherwise, "the times they are a changing" for everyone in clinical research, it is truly a very "interesting" time with exciting challenges!
Lynda Cedar • Karen, I understand and respect your opinion but did I mention India? If the other emerging countries conduct the studies as India (some CROs of India) does, probably we would see more CROs shutting down here....!!!
That has been said, I work with some CROs in India when my clients pick them or I do for bioanalytical work for example, the quality is good, the studies are conducted under GCPs and FDA standards. However, these CROs are not that cheap as we largely assume, they are just a bit competitive (7 to 10% less than we (at my previous CRO) were charging.
Natalie Borde • hey all, I think everyone will agree with me, that all companies wanting to bring their product to market will try to pick a route of least resistance. But one must also keep in mind the patient population. Using the example of India, a company cannot do a complete trial in India and no where else. As the population of India does not reflect all ethnic groups. In addition to the choosing emerging countries, is that the patients here are medication naive.
13 hours ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion1 . Lynda Cedar • I agree with you Nathalie about mainly the resistance that sponsors may face. In comparing to here (I mean either, Europe, Canada or USA), the process and recruitment/enrolement look faster in the emerging countries ..... and that's is one of the most important things for sponsors.
In my previous comment, after reading those of others, I wanted to bring the economic issue that is contributing to the trend of disappearing of study sites.
Natalie Borde • recruiting faster means can mean time won in the long run.. more revenue for the sponsor.
Economics is certainly effecting the industry. I think if one looks carefully, i.e. organon was taken over by sharing and within 2 years was taken over by Merck. The economics is then filtering in down to the trial conduct. Sponsors may not be as patient as they use to be. Recruiting less sites, but more recruitment is what they want. Often sponsors hire CROs to do the work, but the figures provided by the CROs are not always realistic. Sponsor what to find a balance.
Lynda Cedar: @Natalie: Yes, I agree with you, I came across many times to CROs that promise unrealistic things (timing and/or prices), they have strong marketing/sales people, and sponsors should not fall in the trap.
On the other hand, when sponsors outsource the work to those who know what they are doing, the work is achieved right from the first time, no time nor money is wasted. Knowledge and experience enable CROs to gauge how trial protocols will manifest in the real life, the sponsoring companies can consequently plan and budget based on real-world projections.
CRO and Sponsors, both want to do more with less but that can be achieved only when experienced people drive the projects on both sides (Sponsor and CROs), they determine realistic milestones and budget, and that how they can hit the study objectives and the deadlines.
CenterWatch cautions that over the last 5 years, the number of study sites has decreased, with the loss of veteran PIs most sharply in decline. How will this trend affect sponsors, other stakeholders in the business of clinical research and the provision of new products to the public if it continues?
The Site’s Side: Now You See Them, Now You Don’t – The Trend of... blog.gobalto.com
By: Sherry Reuter A January. 5, 2012 report from CenterWatch cautions that over the last 5 years, the number of study sites has decreased, with the loss of veteran PIs most sharply in decline. Specifically: The proportion of...
Commented about it: Angela Rhodes • I agree, it is becoming more burdensome for a physician to conduct trials at his office. The documentation burden is high and in many cases monitors are less 'helpers' to the site and more 'police'. In many cases budgets 'piggyback' on top of standard of care and do not compensate. However, the documentation burden is high, even on these standard of care. Docs just want to 'do their job' and not complicate their lives/practices further. Sponsors should have a 'team' that helps set up a research endeavor at the practice, use their expertise to develop a flow and plan. Make it less burdensome for Docs. This should be separate from CRA/Monitors.
Natalie Borde • Sometime physician are also given the option to hire in extra help to lessen the burden. But one must try to ensure that valuable information is not lost along the way. Interesting point "sponsors should have a team that helps sent up a research endeavor........from CRA/monitors". The "team' would also need to be one which is unbiased. Is there a niche in the market for this?? With reference 'expertise to develop a flow and plan', do you meant how the PI and his team can efficiently incorporate trial activities within the daily work environment?
Angela Rhodes • Natalie, each individual office has its own ebb and flow. I envision a 'site set-up' geared more toward, how are you going to find the subjects? At what point in your daily practice. (most don't have searchable databases) What records are you going to need, who is going to push the button on the machine, write it down. etc. In many cases a 'coordinator' does this. This coordinator is usually only doing research part-time or as an add-on to regular duties. I have found that if staff have a "package' and are told to follow the flow, check the boxes, etc. It is easier on the staff and you get better results.
Sherry Reuter • Dear Natale and Angela,
Wonderful comments that sound like they would indeed help sites.
The term 'site support' has so many different meanings, but you describe real help. Partnering with sites to make sure they succeed helps all stakeholders to be successful! Kind regards, Sherry
Jose Manuel Masso • I agree with the comments so far. The burden of clinical trials is increasing as we are asking physicians for more patients and more quality, which are in fact necessary. This means that on one hand there is a tendency towards more documentation, assessments, visits at sites in current clinical trials and on the other hand monitors are more under the pressure of the sponsors SOPs, audits and co-monitorisations and, as Angela mentioned, less prone to help the investigators. My opinion is that in order to be successful stronger attention has to be put during the site selection visits, to make investigators aware of all these things (burden of some clinical trials) and also I think that the role of trial coordinators at site is gaining importance and trial coordinators are becoming a substantial part in the current clinical trials context.
Natalie Borde • So Angela, what you saying is trying to guide and stimulate the site to look in the right direction to look for potential subjects. Basically what one could propose is creating a manual for patient recruitment. But you mentioned the study co-ordinator, it is not really their role to identify the subjects, but the PI or SI.
Hey Jose Manuel, hope you are well. As you mentioned, and I agree entirely subject recruitment etc, this topic is address during the site selection. Here one identifies where the potential sources are for subjects.
Perhaps the introduction of a site coach is needed. An addition to the CRA. To support the sites, for additional trainings, the steer the sites in the right directions ie. recruitment. keep the momentum going...
Lynda Cedar • The Trend of Disappearing Study Sites !...... hum, yes! That’s a true; because of different factors and main them are the strict and demanding regulation that increase the study cost and the current devastating economy.
1. Some sites disappear from here but were transferred to elsewhere (emerging countries). Their manpower is cheap, the subjects are compensated 90% less, the regulation is less strict, the Ethics are different, and the quality is consequently less.
For the sites of here that continued to serve (because sponsors need some generic studies to be conducted here), some of them could not survive to the ''devastating" economy, since 2008 and it does not end yet!....
2. As explained above by the members, the regulation is very demanding (but it has to be as such), however this increases the study cost for the sponsors and the small CROs cannot survive..... At the end the small CROs disappear, and some empire CROs, very big disappear too... ''big room, big mistakes" and therefore penalized for the errors..... In Quebec, big CRO as Phoenix Int'l then became MDS pharma services, then closed..... Anapharm changed 2 - 3 times names and becomes very quiet or even shutting down.... Atlantic Life Sciences, a small CRO, is in restructuration..... other examples are for elsewhere in Canada, Europe, USA.....
3. For the sites operating in the field of complex studies, new medications, biosimilars, etc.. point 2 is even worst: the development takes time and the cost becomes excessive, just few CROs have the expertise and are just surviving....Sponsors should be aware of.
4. Laid-off and points 2-3 create a lot of resources available for consulting.... individual enterprises. Sponsors changed their strategy of employment.
5. For studies of late stage (Phase III), a lot of diseases (diabetes, BP/CV, Parkinson, Alzheimer, Cancer, HIV, ) are still needing discovery and improvement .... but again, because of the excessive cost, pharmaceuticals need to be selective when picking a medication to develop or a treatment to improve.
In this sector of clinical research, outside our countries, the patients do not have treatments available and are more willing to try the new medication in development (phase III), the recruitment looks easier, the cost looks less. I'm not sure that they are saving more money and time, as per some discussions on this forum and other LI groups, sometimes the preliminary of ICH or GCPs is not followed.... therefore, it looks less expensive (in appearance) but if they have to redo some studies, it costs them even more....
The cost is only 7% to 10% less when studies are performed in emerging countries at performing and compliant CROs with GCPs and FDA standards/regulations.
Ivalina Hristova • @ Lynda Cedar: "The cost is only 7% to 10% less when studies are performed in emerging countries at performing and compliant CROs with GCPs and FDA standards/regulations."
Do have reference for this?
Karen Levorson • Lynda, I do not agree with you that regulations in emerging countries are less strict. In my experience, studies performed in India for example, are scrutinized quite well by the authorities and also by Ethics committees.
And as for your comment about ethics being different, I totally disagree. There are researchers (and probably drug companies) who are proposing studies that are not ethical, that they could never perform in their own countries and they are going to the poor emerging countries, misrepresenting what they are doing - I have experienced that during due diligence audits!! But these are definitely not the "norm". There are investigators and researchers without ethics in every country!
In emerging countries there are conditions that are different that can really influence the subject to stay in a study, that they think that their doctor would never propose anything that could possibly harm them and then to not complain about any adverse event because they have no other recourse but the study to receive medication for their disease - even at the "risk" of receiving placebo!!
My fear for the future is that when studies are conducted under such conditions, we will have "surprises" when they come onto the market!!
Otherwise, "the times they are a changing" for everyone in clinical research, it is truly a very "interesting" time with exciting challenges!
Lynda Cedar • Karen, I understand and respect your opinion but did I mention India? If the other emerging countries conduct the studies as India (some CROs of India) does, probably we would see more CROs shutting down here....!!!
That has been said, I work with some CROs in India when my clients pick them or I do for bioanalytical work for example, the quality is good, the studies are conducted under GCPs and FDA standards. However, these CROs are not that cheap as we largely assume, they are just a bit competitive (7 to 10% less than we (at my previous CRO) were charging.
Natalie Borde • hey all, I think everyone will agree with me, that all companies wanting to bring their product to market will try to pick a route of least resistance. But one must also keep in mind the patient population. Using the example of India, a company cannot do a complete trial in India and no where else. As the population of India does not reflect all ethnic groups. In addition to the choosing emerging countries, is that the patients here are medication naive.
13 hours ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion1 . Lynda Cedar • I agree with you Nathalie about mainly the resistance that sponsors may face. In comparing to here (I mean either, Europe, Canada or USA), the process and recruitment/enrolement look faster in the emerging countries ..... and that's is one of the most important things for sponsors.
In my previous comment, after reading those of others, I wanted to bring the economic issue that is contributing to the trend of disappearing of study sites.
Natalie Borde • recruiting faster means can mean time won in the long run.. more revenue for the sponsor.
Economics is certainly effecting the industry. I think if one looks carefully, i.e. organon was taken over by sharing and within 2 years was taken over by Merck. The economics is then filtering in down to the trial conduct. Sponsors may not be as patient as they use to be. Recruiting less sites, but more recruitment is what they want. Often sponsors hire CROs to do the work, but the figures provided by the CROs are not always realistic. Sponsor what to find a balance.
Lynda Cedar: @Natalie: Yes, I agree with you, I came across many times to CROs that promise unrealistic things (timing and/or prices), they have strong marketing/sales people, and sponsors should not fall in the trap.
On the other hand, when sponsors outsource the work to those who know what they are doing, the work is achieved right from the first time, no time nor money is wasted. Knowledge and experience enable CROs to gauge how trial protocols will manifest in the real life, the sponsoring companies can consequently plan and budget based on real-world projections.
CRO and Sponsors, both want to do more with less but that can be achieved only when experienced people drive the projects on both sides (Sponsor and CROs), they determine realistic milestones and budget, and that how they can hit the study objectives and the deadlines.
Saturday, January 14, 2012
Localisation of signed Informed Consent - Where exactly do I have to store a signed informed consent and why?
The following question was asked on Linked in, GCP group:
Localisation of signed Informed Consent - Where exactly do I have to store a signed informed consent and why? As I am in a discussion with a monitor I would like to ask you.
Please understand I think I do know my GCP, and think I have a normal/good common sense ;-). As we, the monitor and I, differ in this discussion about the exact localisation (as in detail - which dossier/file) and why, I am curious as what you will have to say, without telling you what/how I think about it. So you will have an open mind about it. I am working on site/hospital in hemato-oncology as a CCRC. Thank you so much for your attention.
The question was commented as follows:
Linda James • ICHGCP E6, 8.2.3 references that the blank models to be filed in ISF and Sponsor File.
8.3.12 reference states that signed forms should only be filed in the ISF. The protects data privacy of the patient too.
Jim Sheets • ICH says they should be "located in the files of the investigator/institution." It doesn't matter if they're all kept together in one reguatory binder at the site, separately in each patients' case history, or with individual patient case report form binders. As long as the access is limited and controlled, you are free to choose the system that works best for your site.
Lynda Cedar • The following procedure works for both, Investigator and sponsor sites. It is just one of the ways of handling study components.
Regarding the ICF part, the objective of the way is to prevent the confidentiality during the process and also in knowing where to find them (as put in the same place), it makes them easily accessible at any time as needed.
- At the pre-study period: at this step, the study has just started, it is new to the staff, it's important that all documents of recruitment stay at one place, with two persons (in case, one is absent). In having them at one place, it prevents to loose any important document at this step.
The confidentiality of subjects is protected better as only the staff involved with the recruitment of this study would access the subjects’ chart.
Either with EDC or manual recruitment, the status of the subjects enrolled can be known at a glance at any time
The ICFs are inserted in the subjects' chart.
- At the on-study period: when the ICFs have been signed, checked and considered final, the QA signs the form of transferring them/it into the reg. binder(s). This make the ICFs safe, and if needed (for any reason), they are easily accessible all the time.
- At the post-study: when all the study binders are considered complete, checked and final (after the study close-out), they are transferred to the archives.
@Dominique: I occupied similar position at the antitumor center, Institute Jules Bordet (Brussels, Belgium), for the patients of our hospital included in the clinical studies; I managed the files as explained above (little adjustment). That has been few years ago but if you want I can provide you with more details (privately).
Shirley Isbill • Re: - At the post-study: when all the study binders are considered complete, checked and final (after the study close-out), they are transferred to the archives.
Sometimes it is difficult to retrieve the ICFs and other records from an archive, especially a hospital record archive. The ICFs and other records need to be available to regulatory auditors.
Lynda Cedar • Yes, that is true. However, as the sponsor is responsible of archiving the study components, the originals are kept with him. If the study conduct was delegated to a CRO, a copy is also archived by the CRO (internally or externally).
As for the hospitals, as a monitor, I provide them with the visits at the study start and I recall and confirm the visit a long time before I show up; and most of the time I don't find the charts when I arrive...... Therefore, it would be better to make the info available to the auditors by the sponsor or the CRO and let them to decide whether they go to the hospital or not.
Dominique S. • Thank you all for your answers! I was so curious about the reactions. Thank you.
On our site the process is as follows (on study i.e. active recruiting enrolment period)
1. To file the original and signed informed consent in the patient medical file. It took me a while to ensure and convince some of the PI/SI to be very careful with a signed IC – now they all do it correctly! I don’t feel to change this embedded procedure. And this way medical personnel will know about the IC as well.
2. In ISF there is a ‘Note to File’ where to find the signed IC and the Inclusion and Enrolment Log Form. Inclusion Log also reveals date of IC.
3. If a monitor needs to view a medical file the only one who will provide him with one is my college or myself.
4. All monitors will have to sign a confidentiallity agreement about patient files and patient digital data on site.
This monitor states – “I cannot open a medical file if I have not seen a signed IC. If an IC is in the patient dossier I cannot verify it.” He wants to have IC in the ISF. I am totally content if he makes a copy and files it in the ISF. He does not want to do the copying.
As Linda and Jim refer to GCP chapter 8, I totally agree – the handy list. And indeed it only states “located in the files of the investigator/institution.” it does not mention the name of the binder or file. And the funny thing is that I just had a re-read of 8.3.13 which states [… To include original documents related to the trial, …] again “located in the files of the investigator/institution” – is this my final answer ;-)?
Thank you Shirley for the archive tip. Though sometimes a patient file is still easier to find than boxes with trial archives of trials conducted by retired physicians and no body knows…
And in my case a sponsor will probably never be in charge of archiving trial components of our site (as trial is closed). And as IC shows the true name of the patient, and not an anonymous trial number – I think this cannot be in the sponsor archives.
Thank you again for your reaction!
Shirley Isbill • If you intend to comply with ICH, sections 8.3.12 and 8.3.13 require that the source documents and signed ICF documents be filed in the investigator’s files, not the sponsor’s files (see essential document filing locations). Generally, the sponsor has no need to know the identity of the subjects which would be included in the ICFs (originals or copies). Additionally, 21 CFR 312.62 requires the Investigator (not the study sponsor) retain the drug study records including the ICFs. There is no provision within 21 CFR 312 for the transfer of storage of the study records from the Investigator to the sponsor; 21 CFR 812.140 does provide for this transfer, but for device studies only. As an FDA auditor I would cite an Investigator if he did not have the original drug study records and they were not available for review at the time of my audit. Having the sponsor supply Investigator records to an FDA auditor is not required or recommended anywhere, it impedes the FDA audit of the Investigator’s records and provides an opportunity for the sponsor to limit record access. The original ICFs should be filed in the Investigator’s study records.
Shirley Isbill • Re: This monitor states – “I cannot open a medical file if I have not seen a signed IC. If an IC is in the patient dossier I cannot verify it.” He wants to have IC in the ISF. I am totally content if he makes a copy and files it in the ISF. He does not want to do the copying.
You need to refer this matter to the sponsor; the monitor is creating an unnecessary problem and is lazy.
Dominique S. • Thank you Shirley. Indeed all our documents are filed on site with proper labeling. And yes all site required documents will be stored/archived on site according to GCP Chapter 8. In another comment it was mentioned as archiving could be done at sponsor, not in my case/on my site.
And thank you for the smile I had a about the lazy monitor ;-)
Lynda Cedar • @Dominique: I just want to clarify that at the time of archiving the study components by the sponsor, the ICFs are put in envelops that are closed with a legal stamp (red circle), and a note is printed on the envelops (can be opened by an authorized person only and in following SOP_____). The Sponsors and CROs have SOPs: how to disclose the subject ID all the time when needed, or how to disclose the randomization code when needed,......etc.....
On the other hand, the sponsor is responsible for archiving the study for 25 years Iin Canada), he has the rights to ask for the originals and copies are kept by the site (hospital) and the CRO, for their own interests.
When a study has been finished, completed, closed and archived, why a sponsor would access it if not needed? if so, the log-book of the archives is signed by the person who needs to access the study files, and the following information is collected: Date(s), time (in and out), title and position, and the reason.
The study files cannot be taken out of the archives, unless for a big big reason (as an inspection for few days) and it requests the sponsor permission.
Dominique S. • Thank yo Lynda for your explanation. I am not familiar with archiving at sponsor. In the Netherlands its 15 years - both sponsor and investigator are responsible as in chapter 8.
1 day ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as
Jim Sheets • Informed consents containing signatures should remain at the investigator's facility.
Lynda Cedar • The investigator keep a copy for his own concern, as the patients were enrolled and treated by him, they are his patients….. they per say ‘‘belong” to him (sorry for the English), I’m not sure, but I think (regarding the ICFs) he can decide to keep the originals and provide the sponsor with a copy.
The ICFs can remain with the investigator and kept with the medical chart, reg. binder or as by the site's procedure, however, the sponsor has the rights to decide to archive all the study components including the ICFs, mainly for some kind of medication (new medication that is not well known or the case of some biosimilars),
The sponsor is the first concerned by any happening with the medication, for many years after a study has been closed-out. In Canada, since 2001, the archiving period is for 25 years. In my experience, when sponsors wanted to keep the ICFs as well as the other components of the study, the attorney advised to put such document (ICF) in an envelop that is sealed legally , and a warning note written on it, plus a SOP of handling it. The regulatory authorities were consulted and were comfortable with this way of ensuring the confidentiality of the ICFs during their archiving for 25 years.
Sponsors are advised to archive the study components at 2 different places, to ensure that at least one copy stays safe.
Martin Robinson • As long as the signed copy is readily available (for audit, inspection or monitoring)at the investigator site it doesn't matter where you keep it as long as it is safe. Some investigators keep them all together in the investigator site file, others keep the forms with the patient notes
21 hours ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion1 .
Glenda Guest • And among all this debate, it is proper for the monitor to verify informed consent prior to looking into a subject's confidential records. Where that document is filed and how it is accesed are certainly flexible as this discussion has shown.
Localisation of signed Informed Consent - Where exactly do I have to store a signed informed consent and why? As I am in a discussion with a monitor I would like to ask you.
Please understand I think I do know my GCP, and think I have a normal/good common sense ;-). As we, the monitor and I, differ in this discussion about the exact localisation (as in detail - which dossier/file) and why, I am curious as what you will have to say, without telling you what/how I think about it. So you will have an open mind about it. I am working on site/hospital in hemato-oncology as a CCRC. Thank you so much for your attention.
The question was commented as follows:
Linda James • ICHGCP E6, 8.2.3 references that the blank models to be filed in ISF and Sponsor File.
8.3.12 reference states that signed forms should only be filed in the ISF. The protects data privacy of the patient too.
Jim Sheets • ICH says they should be "located in the files of the investigator/institution." It doesn't matter if they're all kept together in one reguatory binder at the site, separately in each patients' case history, or with individual patient case report form binders. As long as the access is limited and controlled, you are free to choose the system that works best for your site.
Lynda Cedar • The following procedure works for both, Investigator and sponsor sites. It is just one of the ways of handling study components.
Regarding the ICF part, the objective of the way is to prevent the confidentiality during the process and also in knowing where to find them (as put in the same place), it makes them easily accessible at any time as needed.
- At the pre-study period: at this step, the study has just started, it is new to the staff, it's important that all documents of recruitment stay at one place, with two persons (in case, one is absent). In having them at one place, it prevents to loose any important document at this step.
The confidentiality of subjects is protected better as only the staff involved with the recruitment of this study would access the subjects’ chart.
Either with EDC or manual recruitment, the status of the subjects enrolled can be known at a glance at any time
The ICFs are inserted in the subjects' chart.
- At the on-study period: when the ICFs have been signed, checked and considered final, the QA signs the form of transferring them/it into the reg. binder(s). This make the ICFs safe, and if needed (for any reason), they are easily accessible all the time.
- At the post-study: when all the study binders are considered complete, checked and final (after the study close-out), they are transferred to the archives.
@Dominique: I occupied similar position at the antitumor center, Institute Jules Bordet (Brussels, Belgium), for the patients of our hospital included in the clinical studies; I managed the files as explained above (little adjustment). That has been few years ago but if you want I can provide you with more details (privately).
Shirley Isbill • Re: - At the post-study: when all the study binders are considered complete, checked and final (after the study close-out), they are transferred to the archives.
Sometimes it is difficult to retrieve the ICFs and other records from an archive, especially a hospital record archive. The ICFs and other records need to be available to regulatory auditors.
Lynda Cedar • Yes, that is true. However, as the sponsor is responsible of archiving the study components, the originals are kept with him. If the study conduct was delegated to a CRO, a copy is also archived by the CRO (internally or externally).
As for the hospitals, as a monitor, I provide them with the visits at the study start and I recall and confirm the visit a long time before I show up; and most of the time I don't find the charts when I arrive...... Therefore, it would be better to make the info available to the auditors by the sponsor or the CRO and let them to decide whether they go to the hospital or not.
Dominique S. • Thank you all for your answers! I was so curious about the reactions. Thank you.
On our site the process is as follows (on study i.e. active recruiting enrolment period)
1. To file the original and signed informed consent in the patient medical file. It took me a while to ensure and convince some of the PI/SI to be very careful with a signed IC – now they all do it correctly! I don’t feel to change this embedded procedure. And this way medical personnel will know about the IC as well.
2. In ISF there is a ‘Note to File’ where to find the signed IC and the Inclusion and Enrolment Log Form. Inclusion Log also reveals date of IC.
3. If a monitor needs to view a medical file the only one who will provide him with one is my college or myself.
4. All monitors will have to sign a confidentiallity agreement about patient files and patient digital data on site.
This monitor states – “I cannot open a medical file if I have not seen a signed IC. If an IC is in the patient dossier I cannot verify it.” He wants to have IC in the ISF. I am totally content if he makes a copy and files it in the ISF. He does not want to do the copying.
As Linda and Jim refer to GCP chapter 8, I totally agree – the handy list. And indeed it only states “located in the files of the investigator/institution.” it does not mention the name of the binder or file. And the funny thing is that I just had a re-read of 8.3.13 which states [… To include original documents related to the trial, …] again “located in the files of the investigator/institution” – is this my final answer ;-)?
Thank you Shirley for the archive tip. Though sometimes a patient file is still easier to find than boxes with trial archives of trials conducted by retired physicians and no body knows…
And in my case a sponsor will probably never be in charge of archiving trial components of our site (as trial is closed). And as IC shows the true name of the patient, and not an anonymous trial number – I think this cannot be in the sponsor archives.
Thank you again for your reaction!
Shirley Isbill • If you intend to comply with ICH, sections 8.3.12 and 8.3.13 require that the source documents and signed ICF documents be filed in the investigator’s files, not the sponsor’s files (see essential document filing locations). Generally, the sponsor has no need to know the identity of the subjects which would be included in the ICFs (originals or copies). Additionally, 21 CFR 312.62 requires the Investigator (not the study sponsor) retain the drug study records including the ICFs. There is no provision within 21 CFR 312 for the transfer of storage of the study records from the Investigator to the sponsor; 21 CFR 812.140 does provide for this transfer, but for device studies only. As an FDA auditor I would cite an Investigator if he did not have the original drug study records and they were not available for review at the time of my audit. Having the sponsor supply Investigator records to an FDA auditor is not required or recommended anywhere, it impedes the FDA audit of the Investigator’s records and provides an opportunity for the sponsor to limit record access. The original ICFs should be filed in the Investigator’s study records.
Shirley Isbill • Re: This monitor states – “I cannot open a medical file if I have not seen a signed IC. If an IC is in the patient dossier I cannot verify it.” He wants to have IC in the ISF. I am totally content if he makes a copy and files it in the ISF. He does not want to do the copying.
You need to refer this matter to the sponsor; the monitor is creating an unnecessary problem and is lazy.
Dominique S. • Thank you Shirley. Indeed all our documents are filed on site with proper labeling. And yes all site required documents will be stored/archived on site according to GCP Chapter 8. In another comment it was mentioned as archiving could be done at sponsor, not in my case/on my site.
And thank you for the smile I had a about the lazy monitor ;-)
Lynda Cedar • @Dominique: I just want to clarify that at the time of archiving the study components by the sponsor, the ICFs are put in envelops that are closed with a legal stamp (red circle), and a note is printed on the envelops (can be opened by an authorized person only and in following SOP_____). The Sponsors and CROs have SOPs: how to disclose the subject ID all the time when needed, or how to disclose the randomization code when needed,......etc.....
On the other hand, the sponsor is responsible for archiving the study for 25 years Iin Canada), he has the rights to ask for the originals and copies are kept by the site (hospital) and the CRO, for their own interests.
When a study has been finished, completed, closed and archived, why a sponsor would access it if not needed? if so, the log-book of the archives is signed by the person who needs to access the study files, and the following information is collected: Date(s), time (in and out), title and position, and the reason.
The study files cannot be taken out of the archives, unless for a big big reason (as an inspection for few days) and it requests the sponsor permission.
Dominique S. • Thank yo Lynda for your explanation. I am not familiar with archiving at sponsor. In the Netherlands its 15 years - both sponsor and investigator are responsible as in chapter 8.
1 day ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as
Jim Sheets • Informed consents containing signatures should remain at the investigator's facility.
Lynda Cedar • The investigator keep a copy for his own concern, as the patients were enrolled and treated by him, they are his patients….. they per say ‘‘belong” to him (sorry for the English), I’m not sure, but I think (regarding the ICFs) he can decide to keep the originals and provide the sponsor with a copy.
The ICFs can remain with the investigator and kept with the medical chart, reg. binder or as by the site's procedure, however, the sponsor has the rights to decide to archive all the study components including the ICFs, mainly for some kind of medication (new medication that is not well known or the case of some biosimilars),
The sponsor is the first concerned by any happening with the medication, for many years after a study has been closed-out. In Canada, since 2001, the archiving period is for 25 years. In my experience, when sponsors wanted to keep the ICFs as well as the other components of the study, the attorney advised to put such document (ICF) in an envelop that is sealed legally , and a warning note written on it, plus a SOP of handling it. The regulatory authorities were consulted and were comfortable with this way of ensuring the confidentiality of the ICFs during their archiving for 25 years.
Sponsors are advised to archive the study components at 2 different places, to ensure that at least one copy stays safe.
Martin Robinson • As long as the signed copy is readily available (for audit, inspection or monitoring)at the investigator site it doesn't matter where you keep it as long as it is safe. Some investigators keep them all together in the investigator site file, others keep the forms with the patient notes
21 hours ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion1 .
Glenda Guest • And among all this debate, it is proper for the monitor to verify informed consent prior to looking into a subject's confidential records. Where that document is filed and how it is accesed are certainly flexible as this discussion has shown.
Sunday, January 8, 2012
If a site has already disposed of the remaining drug supplies without completing the final inventory of what has been used/lost. What do you do?
"In a close out visit, you find the site has already disposed of the remaining drug supplies and a final inventory of what had been used/lost/broken or returned was not completed. What can you do? "
The question was asked on Linkedin and commented as follows:
Parag Desai • Protocol violation has to be notified to EC. Ideally this situation should not happen if the same monitor is there, as the site should be trained at SIV and also reminded during monitoring to retain all records and ensure IP is returned. If its the CRA has changed then he should check all monitoring reports of past and in-house site file, also before going for close out tell coordinator list of things they should keep ready for close out. If this has happened at close out then there is gross error. There should be a co-monitoring by the PM or CTL for any study for quality checks. Ideally IP should go for destruction at close out.
Fiona Waddell • Agree with Parag but would add that as much evidence as possible should be collected to support investigator's claim of destruction activities.
Lynda Cedar • What kind of medication is it?
Anyone involved in clinical studies knows about the drug accountability and its importance in the study conduct. Its is the sponsor responsibility to select and pick a qualified site.
It is not a violation of protocol only, it is a violation of ICH statement.
What can you do other than reporting the violation immediately, document as much as possible: since the time the medication was received by the site until the last record ....
make sure that the medication was detected in the patients' body (either via PK or PD parameters measurement).
Jorge Garagorri • I agree with you Lynda. The drug accountability is one of the most important activities that a CRA should perform while a site. Personally, I consider this activity should be clearly defined in all monitoring activities to remind the site that the CRA is the only one in giving the "green light" to the site in order that they can dispose the IP medication.
Lynda Cedar • The medication is the sponsor's property therefore he determines how the medication not used is going to be handled either returned to him or to be disposed by a site. In any case , SOPs are followed to track and record exactly the handling of the medication from the time it quits the sponsor facility until its outcome as directed by the sposnor (return, destruction or re-use when possible).
That is determined and agreed on before the study starts, it cannot be missed as there is some cost for the medication destruction and even its return.
In any case, suppose that the sponsor did not specify anything about it at the time of study preparation, the site must be qualified enough to follow the preliminary of the GCPs and handle the medication use appropriately.
I'm wondering whether things like this still happens nowdays.
Fatima EL GHAIB • See 21 CFR 312.59 and 62(a)
Liping Zhou • Agree with the proposed actions. In some countries/regions, there is clear law/regulation to stipute the handling of investigational product, e.g no destruction should be performed prior to the check from authroity officer. If this is the case, it would be more serious. Additionally, except for the inventory log, any way is available to identify the study drug adminsitration to subjects? All of these need to be discussed with your study management team and QA department to evaluate the impact to the data, besides reporting to the EC.
Carla Duarte • I agree with all the opinions. Most of the site's should have a apropriate local, closed and with the appropaty conditions to kept the medication. This local must be closed and should have a responsable person and a pharmacy to have acess the medication, is important to have a back-up. The "Drug Accontability", should be fill in every time you have a patient in the site. If a patient lost any medication or if any of the medication was lost at the site, this should be comunicate to the monitor and describe what happen in the clinical file of the patient. We should perform according to the ICH / GCP guidelines, the use medication sloud be kept in the center for the monitor verify every time he perform a visit to the site. Most of the Investigator's meeting that I have been present, many persons site don't have any aknologe of this or other mather's. It is important, to help them to have information about this matter and another matter's concerning the GCP/ICH. But is important kept all the medication used and unsed in the site, according to the Sponsor indication. If the medication is in capsules and the visits in your center are to long, telephone to the patient betewen visit's to ask how they are doing and remenber to bring the medication on the next visit.
It is import for you the investigators if they want to perform clinical trials, they must have a dedicate study coordinator and person available to dedicate time to the study.
Christoph Lohan (PgDip.) • Leaving the violation statements aside! I am not sure how it is done in your part of the world, but in the UK you have a good chance of tracing back most of the missing information. Most IMPs are handled via a pharmacy, hence they should have detailed information about to whom what was dispensed. If the accountability logs haven't been filled in at receptition of closure of trial you ought to have the accountability of the IMP whilst the trial was open. On the other hand you have the research team who usually sees the patient at study visits and they most often document the study drug compliance. Finally you do have delivery notes from your end to match batch numbers and quanitity. Thus you can gather quite a bit of information around the missing IMPs.
Otherwise I agree that accountability and destruction procedure needs to be clear with start of study.
Michael Raucci • First check your reg binder for Invoices of all medication, these should of been filed asap once you recieved drug. If by chance you do not have all your drug invoices contact the sponsor and have them track all the medication that was sent to your site, including rescue medication. Their vendor will have copies of the inventory sent to your specific site. Once you have established how much medication was sent to your site you will then refer back to each and every source document of all subjects for that study. Reveiw each chart in detail. A good site will document any missed or lossed drug in the progress notes after each visit. My staff documents the amount used not used, rescue med taken, study med number etc... in the source at each visit. You should be able to come up with a very accurate drug accountablilty log. Of course you will document your errors in a memo to file notiy the IRB and make sure your PI is involed in the whole process. Progress notes are so key, make it your practice standard to document most details to create a paper trail incase of a Audit. Laslty, learn from you mistakes which I am sure you have.
Lynda Cedar • I suggest to read the SOPs to see how the medication was supposed to be handled by the site / hospital, then follow the record (if done) in looking at each place (physic, shelves, fridge, documents) where the medication passed through.
If needed, for example, if the data (when measurable assessment is possible: PK or PD) are questionable, ask the IRB permission and provide the patients with a questionnaire about the medication adminsitration (as much as they could remember). The IRB is able to designate someone to interview the patients (if a small size).
The QA has the duty to make sure that the clinical staff uses and accomplishes their tasks in respect of SOPs in place regarding: the log book of the Pharmacy access (by who, when, why and therefore all the information might be trachked if recorded). The reg binder is also a good tool (see Michael comments about it), the CRF completed by the person who dispensed the medication and by the person who administered the medication ( the date, the time and the dose received are reported).
If the patients are dosed at the hospital, go and get information from the pharmacist of the hospital.
Carla Duarte • I agree with all of you, but my experience as a study coordinator this process must be done every time the patient is in the center to perform a visit. If you do that or advice the sites to do that you will avoid many mistakes.
I'm from Portugal, and myself and the pharmacist perform according to what I have descrive.
What I'm try to transmite is that a center can be a god recretument center, but has dificulties with the guidelines of ICH/GCP.
If we want to make good clinical pratice, we must input to the sites so formations in ICH/GCP.
This is my opinion as a study coordinator.
Francois Peterlongo • The main purpose of the certificate of disposal is to prove that the remaining drug was not used "illegally", e.g. administered to patients outside of the protocol, or in "hidden study" (I have seen that occasionaly in the past!)
In the above situation, this would be a important concern of authorithies in case of inspection. So I would suggest (in addition to the above advises) to collect convincing evidence that all the drug not delivered to study subjects was actually disposed of.
harun njago • I do agree with all of you but this issue takes us back to the fundamentals i.e proper site and PI (and study team)selection, working closely with site for proper site management and preparation of the site for close out. It would be important to map this as a risk area during study start and closely monitor it to prevent this occurrence.
6 days ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion0 .
Megan
Megan Mather • The first thing I would want are reports from the coordinator and the PI as to exactly what happened, how the medications were disposed of and all corroborating documentation that is available. What is done is done, but as a QA auditor, I would want everything possible fully documented and I would want to see that the Investigator Brochure spelled out what was to be done with the drug at the end of the study, how the PI and coordinator were trained, were they at the Investigator Meeting, if there was one, did their contract include that they fully understood their duties and responsibilites, etc. A full paper trail of training, study expectations, and sponsor information should be gathered so that a corrective and preventive action (CAPA) plan could be immediately compiled. If there was something that indicated that the sites were not aware of study responsibilities, then the other sites need to be made immediately aware of the drug accountability procedures. There should also be a CRA report from the site initiation visit.
Many times problems arise when the coordinator is new to the study, perhaps hired after the Investigator Meeting or taking over for another coordinator or the coordinator has too many studies. Sometimes the PI hands a study over to a coordinator and doesn't oversee the study as they should, and also there are times when the sponsor is new to clinical trials (small pharma company) and doesn't make sure that all aspects of a clinical trial are addressed in the Investigator Brochure/meeting or even in the contract duties. I went to a site that actually gave away the study drug because the area was poor and patients needed the drug.
About all that can be done in these cases is to get as much of the drug back as possible or document how it was destroyed and then again, document all that happened preferably separate statements from the PI and the coordinator, and then set the CAPA plan to show that all other sites were made aware of the error, that they are informed of the proper procedures, and that there is proof that policies will be put into place to ensure that the error will not be repeated. CRA training should also be required and documented.
Dominique Chesnais • The problem you have exposed is serious and needs a full investigation of its root causes. The CRA should have already collated some information of what the site said about that event and should be shared with CQA to investigate further.
As a simple reason, it could be a lack of experience in clinical research. However, any researcher knows that nothing can be disposed of or destroyed, without the sponsor's agreement or its representatives. It raises an issue that the drug accountability form was not clear enough for the site to see that the CRA had to review the whole drug accountability and counter-sign the form. Full and detailed drug accountability is essential at all stages, verified by a CRA.
If no drug accountability was performed by the CRA during the whole study, or very little, it is impossible to know about each subject's drug accountability. Therefore, the data at that site should be excluded from the efficacy analyses.
It is important to know where the drugs are gone. They must be returned at the sponsor's office or be really destroyed. No hidden and secondary use with IMPs.
CQA should help in defining the event as an incident or a case of serious malpractice, not only for drugs but also for the whole clinical trial at that site.
Lynda Cedar • I fully agree with you Dominique, the data will be questionnable and might be rejected if there is no record confirming that they were obtained after the investigational drug administration.
The Clozapine generic was developed before the patent expiration in one of the countries where patents are not protected as not recognized, therfore if no record can be found about the investigational drug outcome, the intellectual property of the sponsor might be compromized.
The question was asked on Linkedin and commented as follows:
Parag Desai • Protocol violation has to be notified to EC. Ideally this situation should not happen if the same monitor is there, as the site should be trained at SIV and also reminded during monitoring to retain all records and ensure IP is returned. If its the CRA has changed then he should check all monitoring reports of past and in-house site file, also before going for close out tell coordinator list of things they should keep ready for close out. If this has happened at close out then there is gross error. There should be a co-monitoring by the PM or CTL for any study for quality checks. Ideally IP should go for destruction at close out.
Fiona Waddell • Agree with Parag but would add that as much evidence as possible should be collected to support investigator's claim of destruction activities.
Lynda Cedar • What kind of medication is it?
Anyone involved in clinical studies knows about the drug accountability and its importance in the study conduct. Its is the sponsor responsibility to select and pick a qualified site.
It is not a violation of protocol only, it is a violation of ICH statement.
What can you do other than reporting the violation immediately, document as much as possible: since the time the medication was received by the site until the last record ....
make sure that the medication was detected in the patients' body (either via PK or PD parameters measurement).
Jorge Garagorri • I agree with you Lynda. The drug accountability is one of the most important activities that a CRA should perform while a site. Personally, I consider this activity should be clearly defined in all monitoring activities to remind the site that the CRA is the only one in giving the "green light" to the site in order that they can dispose the IP medication.
Lynda Cedar • The medication is the sponsor's property therefore he determines how the medication not used is going to be handled either returned to him or to be disposed by a site. In any case , SOPs are followed to track and record exactly the handling of the medication from the time it quits the sponsor facility until its outcome as directed by the sposnor (return, destruction or re-use when possible).
That is determined and agreed on before the study starts, it cannot be missed as there is some cost for the medication destruction and even its return.
In any case, suppose that the sponsor did not specify anything about it at the time of study preparation, the site must be qualified enough to follow the preliminary of the GCPs and handle the medication use appropriately.
I'm wondering whether things like this still happens nowdays.
Fatima EL GHAIB • See 21 CFR 312.59 and 62(a)
Liping Zhou • Agree with the proposed actions. In some countries/regions, there is clear law/regulation to stipute the handling of investigational product, e.g no destruction should be performed prior to the check from authroity officer. If this is the case, it would be more serious. Additionally, except for the inventory log, any way is available to identify the study drug adminsitration to subjects? All of these need to be discussed with your study management team and QA department to evaluate the impact to the data, besides reporting to the EC.
Carla Duarte • I agree with all the opinions. Most of the site's should have a apropriate local, closed and with the appropaty conditions to kept the medication. This local must be closed and should have a responsable person and a pharmacy to have acess the medication, is important to have a back-up. The "Drug Accontability", should be fill in every time you have a patient in the site. If a patient lost any medication or if any of the medication was lost at the site, this should be comunicate to the monitor and describe what happen in the clinical file of the patient. We should perform according to the ICH / GCP guidelines, the use medication sloud be kept in the center for the monitor verify every time he perform a visit to the site. Most of the Investigator's meeting that I have been present, many persons site don't have any aknologe of this or other mather's. It is important, to help them to have information about this matter and another matter's concerning the GCP/ICH. But is important kept all the medication used and unsed in the site, according to the Sponsor indication. If the medication is in capsules and the visits in your center are to long, telephone to the patient betewen visit's to ask how they are doing and remenber to bring the medication on the next visit.
It is import for you the investigators if they want to perform clinical trials, they must have a dedicate study coordinator and person available to dedicate time to the study.
Christoph Lohan (PgDip.) • Leaving the violation statements aside! I am not sure how it is done in your part of the world, but in the UK you have a good chance of tracing back most of the missing information. Most IMPs are handled via a pharmacy, hence they should have detailed information about to whom what was dispensed. If the accountability logs haven't been filled in at receptition of closure of trial you ought to have the accountability of the IMP whilst the trial was open. On the other hand you have the research team who usually sees the patient at study visits and they most often document the study drug compliance. Finally you do have delivery notes from your end to match batch numbers and quanitity. Thus you can gather quite a bit of information around the missing IMPs.
Otherwise I agree that accountability and destruction procedure needs to be clear with start of study.
Michael Raucci • First check your reg binder for Invoices of all medication, these should of been filed asap once you recieved drug. If by chance you do not have all your drug invoices contact the sponsor and have them track all the medication that was sent to your site, including rescue medication. Their vendor will have copies of the inventory sent to your specific site. Once you have established how much medication was sent to your site you will then refer back to each and every source document of all subjects for that study. Reveiw each chart in detail. A good site will document any missed or lossed drug in the progress notes after each visit. My staff documents the amount used not used, rescue med taken, study med number etc... in the source at each visit. You should be able to come up with a very accurate drug accountablilty log. Of course you will document your errors in a memo to file notiy the IRB and make sure your PI is involed in the whole process. Progress notes are so key, make it your practice standard to document most details to create a paper trail incase of a Audit. Laslty, learn from you mistakes which I am sure you have.
Lynda Cedar • I suggest to read the SOPs to see how the medication was supposed to be handled by the site / hospital, then follow the record (if done) in looking at each place (physic, shelves, fridge, documents) where the medication passed through.
If needed, for example, if the data (when measurable assessment is possible: PK or PD) are questionable, ask the IRB permission and provide the patients with a questionnaire about the medication adminsitration (as much as they could remember). The IRB is able to designate someone to interview the patients (if a small size).
The QA has the duty to make sure that the clinical staff uses and accomplishes their tasks in respect of SOPs in place regarding: the log book of the Pharmacy access (by who, when, why and therefore all the information might be trachked if recorded). The reg binder is also a good tool (see Michael comments about it), the CRF completed by the person who dispensed the medication and by the person who administered the medication ( the date, the time and the dose received are reported).
If the patients are dosed at the hospital, go and get information from the pharmacist of the hospital.
Carla Duarte • I agree with all of you, but my experience as a study coordinator this process must be done every time the patient is in the center to perform a visit. If you do that or advice the sites to do that you will avoid many mistakes.
I'm from Portugal, and myself and the pharmacist perform according to what I have descrive.
What I'm try to transmite is that a center can be a god recretument center, but has dificulties with the guidelines of ICH/GCP.
If we want to make good clinical pratice, we must input to the sites so formations in ICH/GCP.
This is my opinion as a study coordinator.
Francois Peterlongo • The main purpose of the certificate of disposal is to prove that the remaining drug was not used "illegally", e.g. administered to patients outside of the protocol, or in "hidden study" (I have seen that occasionaly in the past!)
In the above situation, this would be a important concern of authorithies in case of inspection. So I would suggest (in addition to the above advises) to collect convincing evidence that all the drug not delivered to study subjects was actually disposed of.
harun njago • I do agree with all of you but this issue takes us back to the fundamentals i.e proper site and PI (and study team)selection, working closely with site for proper site management and preparation of the site for close out. It would be important to map this as a risk area during study start and closely monitor it to prevent this occurrence.
6 days ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion0 .
Megan
Megan Mather • The first thing I would want are reports from the coordinator and the PI as to exactly what happened, how the medications were disposed of and all corroborating documentation that is available. What is done is done, but as a QA auditor, I would want everything possible fully documented and I would want to see that the Investigator Brochure spelled out what was to be done with the drug at the end of the study, how the PI and coordinator were trained, were they at the Investigator Meeting, if there was one, did their contract include that they fully understood their duties and responsibilites, etc. A full paper trail of training, study expectations, and sponsor information should be gathered so that a corrective and preventive action (CAPA) plan could be immediately compiled. If there was something that indicated that the sites were not aware of study responsibilities, then the other sites need to be made immediately aware of the drug accountability procedures. There should also be a CRA report from the site initiation visit.
Many times problems arise when the coordinator is new to the study, perhaps hired after the Investigator Meeting or taking over for another coordinator or the coordinator has too many studies. Sometimes the PI hands a study over to a coordinator and doesn't oversee the study as they should, and also there are times when the sponsor is new to clinical trials (small pharma company) and doesn't make sure that all aspects of a clinical trial are addressed in the Investigator Brochure/meeting or even in the contract duties. I went to a site that actually gave away the study drug because the area was poor and patients needed the drug.
About all that can be done in these cases is to get as much of the drug back as possible or document how it was destroyed and then again, document all that happened preferably separate statements from the PI and the coordinator, and then set the CAPA plan to show that all other sites were made aware of the error, that they are informed of the proper procedures, and that there is proof that policies will be put into place to ensure that the error will not be repeated. CRA training should also be required and documented.
Dominique Chesnais • The problem you have exposed is serious and needs a full investigation of its root causes. The CRA should have already collated some information of what the site said about that event and should be shared with CQA to investigate further.
As a simple reason, it could be a lack of experience in clinical research. However, any researcher knows that nothing can be disposed of or destroyed, without the sponsor's agreement or its representatives. It raises an issue that the drug accountability form was not clear enough for the site to see that the CRA had to review the whole drug accountability and counter-sign the form. Full and detailed drug accountability is essential at all stages, verified by a CRA.
If no drug accountability was performed by the CRA during the whole study, or very little, it is impossible to know about each subject's drug accountability. Therefore, the data at that site should be excluded from the efficacy analyses.
It is important to know where the drugs are gone. They must be returned at the sponsor's office or be really destroyed. No hidden and secondary use with IMPs.
CQA should help in defining the event as an incident or a case of serious malpractice, not only for drugs but also for the whole clinical trial at that site.
Lynda Cedar • I fully agree with you Dominique, the data will be questionnable and might be rejected if there is no record confirming that they were obtained after the investigational drug administration.
The Clozapine generic was developed before the patent expiration in one of the countries where patents are not protected as not recognized, therfore if no record can be found about the investigational drug outcome, the intellectual property of the sponsor might be compromized.
Thursday, December 22, 2011
Does data generated by conducting clinical trial in one region will be applicable/useful for conducting a clinical trial in a another region?
Does data generated by conducting a clinical trial in one region will be applicable or useful for conducting a clinical trial in another region having different environmental condition?
Marie-Christine Reymond • Your question should not find a correct answer without knowing the therapeutic area you deal with. For example, it is well known that cardiovascular risk factors are different from US and European subjects, so that your conclusive findings can not directly be derived between both. Hope it helps.
Satish Champaneria • Environmental conditions are didderent from region to region..and so does conditions and various factors..extrapolation of clinical trials will depend on segment of trials..dosage and other factors will have to be born in mind..
Lynda Cedar • @Kunal: In any case all the clinical data obtained must be included in the study file at the submission, mainly when the action (see the question) was not planned at the beginning of the study....and it would be helpful if you tell us the reason to change the region.
In general it is certainly helpful to have clinical (pharmacology) data and explore, design next studies. Examples: don't we go into first-in-man studies based on the data obtained in animals? Are not the data of international studies (late stage) pooled to complete the study?
Pooling the data obtained by different sites or in different regions, that cannot be generalised, further information is requested, clinical research is not white or black, most of the time grey zones come up, and therefore clinical and medical judgements apply.
What is the different environmental condition: is it genetic? Is it the life style; is it the culture of clinical research? Etc,
What are regions A and B?
What is the medication (therapeutic area)?
What is the stage of the clinical study?
Etc.,
Etc.,
At least for economic reasons: If not a large population is needed, try to run a pilot study (a sample) in the second region and then compare to the data obtained initially.
Dennis McHugh • Many examples including analgesia, infectious diseases, gastroenterology, anti-coagulation, cardiovascular medicine support the fact that populations respond differently. Genetics, environment, standard of care, and even subject expectations all conspire to effect outcomes. Even in the US where everyone came from somewhere else at some point and you might anticipate a degree of homogenization, there are regional differences. Multi-center trials combined with sufficient torturing of the data 'till it yields the "right" answer is the only way to go.
Lynda Cedar • Dennis, my understanding of the question is that data were obtained at site(s) in one region and the sponsor wants to continue the study in another region. You are right, Phase III studies involve different populations (that is even a FDA requiremnent: 18y and up, different gender, different race, style of life, etc.) and mainly when rare diseases are concerned, however, when changes happen in clinical conduct at the middle or at the end of study, there is no general answer, it should be taken case per case, andappropriate recommendations might be provided.
Kunal Desale • Thanks to all of you,
for all of your reference I want to repeat my question with example,
I wanted to know that, does data generated by conducting a Dermatological trial in one region ( e.g. European region) will be useful for conducting a clinical trial in another region ( e.g. Asian region) and if it is useful or applicable then, should sponsor have to conduct only BABE study or whole clinical research of same molecule in another region...???
hope you all get my point.
Dr. Akhilesh Sharma • Clinical Data generated from a region for dermatology study would be suggestive of its efficacy and safety in the concerned population. It would also provide inputs about the dugs efficacy and safety. However, it may not be plausible to extrapolate this data for another region(unless the other regions are part f multi-centric study and the differences in response outcome are included while powering the study and calculating sample size). Clinical situations and factors for a dermatology disease can vary e.g. for acne or psoriasis in Europe exacerbations are seen twice a year mostly during winter & when sunshine is inadequate. The seasons in Europe are different and vary in intensiy, for example as compared to Asian countries like India and China, so disease prognosis/outcome/remissions differs. Similarly, other factors like food habits, skin-cleansing etc also contributes to the prognosis of a disease in skin condition. Nevertheless, data generated in a region is good indicator if not confirmatory of what can be expected from a drug in terms of its efficacy & safety. On the other hand certain skin conditions which have a heavy immune and genetic involvement, the outcome becomes more predictive about the response of a drug in a captive genetic population rather than regional, though careful interpretations are required because genetic variations exist between regions and ethnic groups as we all know e.g. outcome of Gefitinib in Head & Neck cancer in US & Europe as compared to caucasians.
So as stated by Lynda multi-centric, multi-national studies are sufficiently powered enough to look at these differences. In case a study has been completed and results analysed for a region and if another region was not a part of the study but data is required in this 'another' region a proper statistical powering of the study is required based on endpoints to be evaluated and therefore the number of patients is based on the expected outcome in that population/group. This may be a full fledged phase of a clinical trial, however the sample size calculation becomes more easy (and sample requirement 'maybe' less) from the response seen in clinical trial conducted in another region. These are indicators more towards phase IIb/ III studies, I am not sure as to which phase of the clinical trial is being referred to here.
Lynda Cedar • Since it is about a bioequivalence study, the efficacy was demonstrated at the time of developing the original. Depending on how the comparison is assessed, via PK data, PD data or visual scale. If PK data or measurable PD data cannot be obtained, therefore it is about therapeutic bioequivalence. In fact, there is a difference (physiology - ADME) between Caucasian, Asian and black, and a population might be more or less receptive/sensitive to the treatment consequently.
I recommend a pilot exploratory clinical study and then do the adjustment as/if needed. Also, check about the local regulations, whether foreign clinical data can be used or not.
Anders Fuglsang • In a BE trial the primary endpoint is the T/R ratio. While regional differences in phenotype might affect drug response or PK for any given molecule, the T/R ratio or the variability associated with it might not be affected. This is why BE done on population X is often acceptable for a claim in population Y.
It needs to be said though that a good design in population X might need to be tweaked somewhat if the trial for some reason is to be repeated in population Y. Example: Blood sampling regimens in BE trials often need to be planned so that AUCt is 80% or more of AUCinf. Therefore, if population Y is on average (or median) more slowly metabolizing than population X, the timing of blood samples should reflect this.
A similar consideration might apply to equivalence in terms of PD.
Best regards,
A.
Lynda Cedar • Anders, that true if PK or PD data are possible to be measured. This is a dermatology study; the data might be observational, example: heeling, less lesions, decreasing of redness or size, itching, wrinkles, spots, etc.
Francois Peterlongo • The general question was thoroughly debated at the beginning of the ICH process. It was even a serious obstacle to the harmonization between the 3 ICH regions. I remember that Japan authorities were especially reluctant to accept that clinical trials results observed in western populations could be adequate to support product registration for Japanese population.
Fortunately, a consensus was finally found on scientific basis, ans it is the content of the guideline ICH E5 "Ethnic Factors in the Acceptability of Foreign Clinical Data". As far as I know, it is still the reference text about this question.
Lynda Cedar • Yes François, I was at school by the time of process of mutual recognition.
Sponsor still have some difficulties I think when importing data obtained outside the 3 ICH regions.
Marie-Christine Reymond • Your question should not find a correct answer without knowing the therapeutic area you deal with. For example, it is well known that cardiovascular risk factors are different from US and European subjects, so that your conclusive findings can not directly be derived between both. Hope it helps.
Satish Champaneria • Environmental conditions are didderent from region to region..and so does conditions and various factors..extrapolation of clinical trials will depend on segment of trials..dosage and other factors will have to be born in mind..
Lynda Cedar • @Kunal: In any case all the clinical data obtained must be included in the study file at the submission, mainly when the action (see the question) was not planned at the beginning of the study....and it would be helpful if you tell us the reason to change the region.
In general it is certainly helpful to have clinical (pharmacology) data and explore, design next studies. Examples: don't we go into first-in-man studies based on the data obtained in animals? Are not the data of international studies (late stage) pooled to complete the study?
Pooling the data obtained by different sites or in different regions, that cannot be generalised, further information is requested, clinical research is not white or black, most of the time grey zones come up, and therefore clinical and medical judgements apply.
What is the different environmental condition: is it genetic? Is it the life style; is it the culture of clinical research? Etc,
What are regions A and B?
What is the medication (therapeutic area)?
What is the stage of the clinical study?
Etc.,
Etc.,
At least for economic reasons: If not a large population is needed, try to run a pilot study (a sample) in the second region and then compare to the data obtained initially.
Dennis McHugh • Many examples including analgesia, infectious diseases, gastroenterology, anti-coagulation, cardiovascular medicine support the fact that populations respond differently. Genetics, environment, standard of care, and even subject expectations all conspire to effect outcomes. Even in the US where everyone came from somewhere else at some point and you might anticipate a degree of homogenization, there are regional differences. Multi-center trials combined with sufficient torturing of the data 'till it yields the "right" answer is the only way to go.
Lynda Cedar • Dennis, my understanding of the question is that data were obtained at site(s) in one region and the sponsor wants to continue the study in another region. You are right, Phase III studies involve different populations (that is even a FDA requiremnent: 18y and up, different gender, different race, style of life, etc.) and mainly when rare diseases are concerned, however, when changes happen in clinical conduct at the middle or at the end of study, there is no general answer, it should be taken case per case, andappropriate recommendations might be provided.
Kunal Desale • Thanks to all of you,
for all of your reference I want to repeat my question with example,
I wanted to know that, does data generated by conducting a Dermatological trial in one region ( e.g. European region) will be useful for conducting a clinical trial in another region ( e.g. Asian region) and if it is useful or applicable then, should sponsor have to conduct only BABE study or whole clinical research of same molecule in another region...???
hope you all get my point.
Dr. Akhilesh Sharma • Clinical Data generated from a region for dermatology study would be suggestive of its efficacy and safety in the concerned population. It would also provide inputs about the dugs efficacy and safety. However, it may not be plausible to extrapolate this data for another region(unless the other regions are part f multi-centric study and the differences in response outcome are included while powering the study and calculating sample size). Clinical situations and factors for a dermatology disease can vary e.g. for acne or psoriasis in Europe exacerbations are seen twice a year mostly during winter & when sunshine is inadequate. The seasons in Europe are different and vary in intensiy, for example as compared to Asian countries like India and China, so disease prognosis/outcome/remissions differs. Similarly, other factors like food habits, skin-cleansing etc also contributes to the prognosis of a disease in skin condition. Nevertheless, data generated in a region is good indicator if not confirmatory of what can be expected from a drug in terms of its efficacy & safety. On the other hand certain skin conditions which have a heavy immune and genetic involvement, the outcome becomes more predictive about the response of a drug in a captive genetic population rather than regional, though careful interpretations are required because genetic variations exist between regions and ethnic groups as we all know e.g. outcome of Gefitinib in Head & Neck cancer in US & Europe as compared to caucasians.
So as stated by Lynda multi-centric, multi-national studies are sufficiently powered enough to look at these differences. In case a study has been completed and results analysed for a region and if another region was not a part of the study but data is required in this 'another' region a proper statistical powering of the study is required based on endpoints to be evaluated and therefore the number of patients is based on the expected outcome in that population/group. This may be a full fledged phase of a clinical trial, however the sample size calculation becomes more easy (and sample requirement 'maybe' less) from the response seen in clinical trial conducted in another region. These are indicators more towards phase IIb/ III studies, I am not sure as to which phase of the clinical trial is being referred to here.
Lynda Cedar • Since it is about a bioequivalence study, the efficacy was demonstrated at the time of developing the original. Depending on how the comparison is assessed, via PK data, PD data or visual scale. If PK data or measurable PD data cannot be obtained, therefore it is about therapeutic bioequivalence. In fact, there is a difference (physiology - ADME) between Caucasian, Asian and black, and a population might be more or less receptive/sensitive to the treatment consequently.
I recommend a pilot exploratory clinical study and then do the adjustment as/if needed. Also, check about the local regulations, whether foreign clinical data can be used or not.
Anders Fuglsang • In a BE trial the primary endpoint is the T/R ratio. While regional differences in phenotype might affect drug response or PK for any given molecule, the T/R ratio or the variability associated with it might not be affected. This is why BE done on population X is often acceptable for a claim in population Y.
It needs to be said though that a good design in population X might need to be tweaked somewhat if the trial for some reason is to be repeated in population Y. Example: Blood sampling regimens in BE trials often need to be planned so that AUCt is 80% or more of AUCinf. Therefore, if population Y is on average (or median) more slowly metabolizing than population X, the timing of blood samples should reflect this.
A similar consideration might apply to equivalence in terms of PD.
Best regards,
A.
Lynda Cedar • Anders, that true if PK or PD data are possible to be measured. This is a dermatology study; the data might be observational, example: heeling, less lesions, decreasing of redness or size, itching, wrinkles, spots, etc.
Francois Peterlongo • The general question was thoroughly debated at the beginning of the ICH process. It was even a serious obstacle to the harmonization between the 3 ICH regions. I remember that Japan authorities were especially reluctant to accept that clinical trials results observed in western populations could be adequate to support product registration for Japanese population.
Fortunately, a consensus was finally found on scientific basis, ans it is the content of the guideline ICH E5 "Ethnic Factors in the Acceptability of Foreign Clinical Data". As far as I know, it is still the reference text about this question.
Lynda Cedar • Yes François, I was at school by the time of process of mutual recognition.
Sponsor still have some difficulties I think when importing data obtained outside the 3 ICH regions.
CRA found missing data in the source records. How he/she should deal with that?
The question was posted and commented on a Linked in group:
Myrna Loughrey • Can the CRA still enter the data into the database? If the database is closed, the CRA must notify the project PM, inform the sponsor, and inform the PI. A note to file may be written to document the inadvertent error or directly write the error on the source and or the CRFs. This inadvertent error must be documented and appropriate folks must be notified. The documentation will help during regulatory or sponsor audits.
Bakhtiyar Valiyev • Hi Myrna, thank you for your response!
Yes, the database is open. I wonder how the fact of missing data should be considered? Should we consider these missing data a protocol and/or GCP deviation? If yes, what are the criteria?
Let's say some evaluation was done but was not recorded in the source records at the visit (e.g. urine pregnancy test). By the time of the monitoring visit, the CRA found that there is no result in the patient chart and may be some vague recollection of site staff that the test was done and it was negative.
What steps can be recommended the CRA to make (with regards to site staff, PI, trip report, follow up)? Is it enough to make an action item, e.g.: "the result of pregnancy test is missing in the patient chart; the site staff should enter missing data"?
As I understand in this case you propose 1) to write a note to file to document the issue or 2) add comments in the source records as well as to notify the appropriate folks (which ones in this case?). Any other options or the mentioned above will suffise?
Jim Sheets • You could just have the site staff record a late entry in the patient's medical chart (dated today), and then update your open study database with the new information. Issue a query and have the PI/subinvestigator sign.
Lynda Cedar • Jim, late entries are data (like vital signs for example) that were performed but the staff did not write it down or entered it in the data base for x, y reasons.
Regarding the pregnancy, since there is no proof that the pregnancy has been performed, no late entries can be considered. I came across to a similar situation, during the morning monitoring before the dosing (it was about a Phase I study), the data were not entered and the clinical staff ensured that the pregnancy tests were performed, I put a note to file, to document the event and for follow-up. Later, the site was able to demonstrate (via the log book of purchase) the number of units (pregnancy test) available before and after the day where tests were supposed to be performed. This was acceptable to the IRB/IEC.
Jim Sheets • Bakhtiyar says: "By the time of the monitoring visit, the CRA found that there is no result in the patient chart and may be some vague recollection of site staff that the test was done and it was negative". If the site can retrospectively remember such information, I don't see anything wrong with documenting something like "pregnancy test was performed on (date), and was negative" - PI sign/date with today's date. "Negative" is entered in the database. A pregnancy test result is easier to remember than a vital sign.
I don't believe proof is required in order to remember something and retrospectively document. Obviously, the ability to recall information like that has to be considered on a case-by-case basis.
2 days ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion0 . Lynda Cedar • Exactly, it is case-by-case, some sites do not perform the test to save money and declare that the test was done. How can we ensure that?
If I'm not convinced about it, to protect the sponsor and the rights of participating female, I reporte it as missing data (therfore a protocole deviation) and put a note to file to document it for track record and follow-up. This way, we are on a safe side, if a female was pregnant or becomes pregnant while she was under medication effect.
Myrna Loughrey • Bakhtiyar I inform my staff and the sites that I visit, if its not documented it never happend. The site can recollect all they want; however, if the study procedure is missed and not documented it never occured. Is it a protocol deviation? Yes of course it is! Study procedures written in the study protocol must be performed. There is no way around that. It's part of the study protocol!
What the CRA must do to move forward to ensure? It is an action item that should be included in the monitoring report, sit down with the site site staff and the PI and inform the site missing data and/or miss procedure is not acceptable, inform PM and discuss the occurence during CRA conference call (it might be occuring at other sites).
No other options but document..
Lynda Cedar • @Myrna: your comments are pertinent, if the study is multi-center, you should not inform the other sites but monitor them to see whether they are performing the pregnancy test or not. If a BE study, the process is much simple, I just report it as missing date (protocol deviation in the study report) and document it for follow-up.
Jim Sheets • It is documented...if you document it now, retrospectively. Sure, it does not meet GCP criteria for good documentation practices in that it is not "contemporaneous"; therefore, (if we want to use cliches) are we saying if it isnt documented contemporaneously it "didn't happen." Not sure I agree that is always the case.
Lynda Cedar • Jim, I respect all opinions, mine is just one of them: for me if a test was not done and discovered by a monitor or a CRA, it was not done (missing data and protocol deviation).
Let's give the site a chance, the usual good practice, if the staff forgets to report pregnancy test results, they would do report it quite rapidly, it cannot wait until the CRA dicovers it and then it is reported (as late entry).
Thanks for sharing and best regards.
Lynda.
Shirley Isbill • If a pregnancy test was done, there should be a record somewhere. If the test was not done (the conclusion in the absence of a record) then a NTF that it was done and was negative would be a false medical record entry.
Ann Baumann • From an ex-inspector point of view, I can tell you that this is definetely at least a major finding . If something is not documented, it does not exist. What irritates me, is the fact that "the site staff only vaguely remembers having performed a pregnancy test", which is a crucial test in a clinical trial.
I must agree with Shirley, a NTF stating that the test was negative is definetely a false statement. So you should document in the NTF what really happened.
1 day ago •
Myrna Loughrey • @ Ann: Thank you for your comment! " If something is not documented, it does not exits" This statement is what I've heard form all the regulatory inspectors I've encountered.
Lynda Cedar • I agree and that what Jim explained: it is not GCP reporting a test done while there is no record. In this case, regulatory inspector and strict monitors (like you and I) would state, if not done, it is not done. The only thing the site can do is a NTF documenting it: when the deviation was discovered and what were the actions undertaken. .
John T. Davidson • In my experience this is sadly more common that it should be. Having said this, the key to avoid this is for the Sponsor staff to develop a relationship with the Investigator and investigator staff that is transparent and where information flows rapidly and freely. This includes “bad” news.
I know the adage, “If it isn’t documented, it didn’t happen.” While I appreciate the sediment of the saying, the biggest mistake is to not document the incident at all since “any documentation is better than no documentation.” Of course I am not implying that anyone should make the facts up but if the staff only “remembers” doing the test, state this clearly and bluntly, stating that the documentation is retrospective.
To Ann’s point, if an inspector finds the incident, you are looking at a major, but in my experience if you have been careful to accurately document the issue before the inspection, you will at minimum gain the respect of the inspector and it will improve your working relationship with the Agency and at best the inspector could give you a break if it is a one off mistake.
So, document, document, document, but if you didn’t do it prospectively, document, document, document honestly, bluntly and transparently retrospectively since this may be the best you can do.
Bakhtiyar Valiyev • Myrna, Jim, Lynda, Shirley, Ann, John, thank you very much for your feedback! Your ideas and suggestions are very useful and give great hints as to how to deal with such cases in everyday CRA work.
1 day ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion
Batya Kornfeld • I think when coming across events of missing information either in the source documents or in the CRF it is essential to sit down with the site personnel discuss the issue and retrain the site to provide the re-occurrence of such issues. If a test was not done this is obviously a protocol deviation which needs to be documented and submitted to the EC and reported to the PM/Sponsor. If it was not entered into the CRF this should be done at once.
10 hours ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion.
Jorge Garagorri • Dear All,
A personal opinion is to notify this omission data as protocol violation inmediatly, request to the site to perform the pregnancy test asap and inform to medical monitor about this, also. In addition, request to the site an action and preventive plan to avoid this happen again and follow up with the site closely its implementation.
Since we cannot assure if this pregnancy test was done, late entry is not valid because ALCOA criteria is not being followed.
4 hours ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion
John T. Davidson • Good discussion. George, I thought about the PG testing, but by the time the oversight of the PG test is found, it is likely the patient already knows if she is pregnant or not. This does make me think that the IRB and your Drug Safety group would have to be notified and of course legal would have to be brought in to the loop in the event the patient is indeed pregnant or worse experienced a miscarriage during the treatment phase of the study.
Lynda Cedar: Jorge and others (including myself) would report the event as a protocol deviation, in fact, is it a deviation or a violation of the protocol?
When a test is done but not as directed by the protocol or SOP, it is a deviation; but when not done, mainly dicovered by the CRA before any documentation or follow up was undertaken by the site, that is a violation involving the consequences brought by the members above.
Myrna Loughrey • Can the CRA still enter the data into the database? If the database is closed, the CRA must notify the project PM, inform the sponsor, and inform the PI. A note to file may be written to document the inadvertent error or directly write the error on the source and or the CRFs. This inadvertent error must be documented and appropriate folks must be notified. The documentation will help during regulatory or sponsor audits.
Bakhtiyar Valiyev • Hi Myrna, thank you for your response!
Yes, the database is open. I wonder how the fact of missing data should be considered? Should we consider these missing data a protocol and/or GCP deviation? If yes, what are the criteria?
Let's say some evaluation was done but was not recorded in the source records at the visit (e.g. urine pregnancy test). By the time of the monitoring visit, the CRA found that there is no result in the patient chart and may be some vague recollection of site staff that the test was done and it was negative.
What steps can be recommended the CRA to make (with regards to site staff, PI, trip report, follow up)? Is it enough to make an action item, e.g.: "the result of pregnancy test is missing in the patient chart; the site staff should enter missing data"?
As I understand in this case you propose 1) to write a note to file to document the issue or 2) add comments in the source records as well as to notify the appropriate folks (which ones in this case?). Any other options or the mentioned above will suffise?
Jim Sheets • You could just have the site staff record a late entry in the patient's medical chart (dated today), and then update your open study database with the new information. Issue a query and have the PI/subinvestigator sign.
Lynda Cedar • Jim, late entries are data (like vital signs for example) that were performed but the staff did not write it down or entered it in the data base for x, y reasons.
Regarding the pregnancy, since there is no proof that the pregnancy has been performed, no late entries can be considered. I came across to a similar situation, during the morning monitoring before the dosing (it was about a Phase I study), the data were not entered and the clinical staff ensured that the pregnancy tests were performed, I put a note to file, to document the event and for follow-up. Later, the site was able to demonstrate (via the log book of purchase) the number of units (pregnancy test) available before and after the day where tests were supposed to be performed. This was acceptable to the IRB/IEC.
Jim Sheets • Bakhtiyar says: "By the time of the monitoring visit, the CRA found that there is no result in the patient chart and may be some vague recollection of site staff that the test was done and it was negative". If the site can retrospectively remember such information, I don't see anything wrong with documenting something like "pregnancy test was performed on (date), and was negative" - PI sign/date with today's date. "Negative" is entered in the database. A pregnancy test result is easier to remember than a vital sign.
I don't believe proof is required in order to remember something and retrospectively document. Obviously, the ability to recall information like that has to be considered on a case-by-case basis.
2 days ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion0 . Lynda Cedar • Exactly, it is case-by-case, some sites do not perform the test to save money and declare that the test was done. How can we ensure that?
If I'm not convinced about it, to protect the sponsor and the rights of participating female, I reporte it as missing data (therfore a protocole deviation) and put a note to file to document it for track record and follow-up. This way, we are on a safe side, if a female was pregnant or becomes pregnant while she was under medication effect.
Myrna Loughrey • Bakhtiyar I inform my staff and the sites that I visit, if its not documented it never happend. The site can recollect all they want; however, if the study procedure is missed and not documented it never occured. Is it a protocol deviation? Yes of course it is! Study procedures written in the study protocol must be performed. There is no way around that. It's part of the study protocol!
What the CRA must do to move forward to ensure? It is an action item that should be included in the monitoring report, sit down with the site site staff and the PI and inform the site missing data and/or miss procedure is not acceptable, inform PM and discuss the occurence during CRA conference call (it might be occuring at other sites).
No other options but document..
Lynda Cedar • @Myrna: your comments are pertinent, if the study is multi-center, you should not inform the other sites but monitor them to see whether they are performing the pregnancy test or not. If a BE study, the process is much simple, I just report it as missing date (protocol deviation in the study report) and document it for follow-up.
Jim Sheets • It is documented...if you document it now, retrospectively. Sure, it does not meet GCP criteria for good documentation practices in that it is not "contemporaneous"; therefore, (if we want to use cliches) are we saying if it isnt documented contemporaneously it "didn't happen." Not sure I agree that is always the case.
Lynda Cedar • Jim, I respect all opinions, mine is just one of them: for me if a test was not done and discovered by a monitor or a CRA, it was not done (missing data and protocol deviation).
Let's give the site a chance, the usual good practice, if the staff forgets to report pregnancy test results, they would do report it quite rapidly, it cannot wait until the CRA dicovers it and then it is reported (as late entry).
Thanks for sharing and best regards.
Lynda.
Shirley Isbill • If a pregnancy test was done, there should be a record somewhere. If the test was not done (the conclusion in the absence of a record) then a NTF that it was done and was negative would be a false medical record entry.
Ann Baumann • From an ex-inspector point of view, I can tell you that this is definetely at least a major finding . If something is not documented, it does not exist. What irritates me, is the fact that "the site staff only vaguely remembers having performed a pregnancy test", which is a crucial test in a clinical trial.
I must agree with Shirley, a NTF stating that the test was negative is definetely a false statement. So you should document in the NTF what really happened.
1 day ago •
Myrna Loughrey • @ Ann: Thank you for your comment! " If something is not documented, it does not exits" This statement is what I've heard form all the regulatory inspectors I've encountered.
Lynda Cedar • I agree and that what Jim explained: it is not GCP reporting a test done while there is no record. In this case, regulatory inspector and strict monitors (like you and I) would state, if not done, it is not done. The only thing the site can do is a NTF documenting it: when the deviation was discovered and what were the actions undertaken. .
John T. Davidson • In my experience this is sadly more common that it should be. Having said this, the key to avoid this is for the Sponsor staff to develop a relationship with the Investigator and investigator staff that is transparent and where information flows rapidly and freely. This includes “bad” news.
I know the adage, “If it isn’t documented, it didn’t happen.” While I appreciate the sediment of the saying, the biggest mistake is to not document the incident at all since “any documentation is better than no documentation.” Of course I am not implying that anyone should make the facts up but if the staff only “remembers” doing the test, state this clearly and bluntly, stating that the documentation is retrospective.
To Ann’s point, if an inspector finds the incident, you are looking at a major, but in my experience if you have been careful to accurately document the issue before the inspection, you will at minimum gain the respect of the inspector and it will improve your working relationship with the Agency and at best the inspector could give you a break if it is a one off mistake.
So, document, document, document, but if you didn’t do it prospectively, document, document, document honestly, bluntly and transparently retrospectively since this may be the best you can do.
Bakhtiyar Valiyev • Myrna, Jim, Lynda, Shirley, Ann, John, thank you very much for your feedback! Your ideas and suggestions are very useful and give great hints as to how to deal with such cases in everyday CRA work.
1 day ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion
Batya Kornfeld • I think when coming across events of missing information either in the source documents or in the CRF it is essential to sit down with the site personnel discuss the issue and retrain the site to provide the re-occurrence of such issues. If a test was not done this is obviously a protocol deviation which needs to be documented and submitted to the EC and reported to the PM/Sponsor. If it was not entered into the CRF this should be done at once.
10 hours ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion.
Jorge Garagorri • Dear All,
A personal opinion is to notify this omission data as protocol violation inmediatly, request to the site to perform the pregnancy test asap and inform to medical monitor about this, also. In addition, request to the site an action and preventive plan to avoid this happen again and follow up with the site closely its implementation.
Since we cannot assure if this pregnancy test was done, late entry is not valid because ALCOA criteria is not being followed.
4 hours ago • Unlike • Like • Reply privately• Flag as inappropriate • Flag as promotion
John T. Davidson • Good discussion. George, I thought about the PG testing, but by the time the oversight of the PG test is found, it is likely the patient already knows if she is pregnant or not. This does make me think that the IRB and your Drug Safety group would have to be notified and of course legal would have to be brought in to the loop in the event the patient is indeed pregnant or worse experienced a miscarriage during the treatment phase of the study.
Lynda Cedar: Jorge and others (including myself) would report the event as a protocol deviation, in fact, is it a deviation or a violation of the protocol?
When a test is done but not as directed by the protocol or SOP, it is a deviation; but when not done, mainly dicovered by the CRA before any documentation or follow up was undertaken by the site, that is a violation involving the consequences brought by the members above.
Thursday, December 15, 2011
Should an abnormal lab value automatically be considered an AE if the value was normal at baseline?
The question was posted on a Linked in group: Should an abnormal lab value automatically be considered an AE if the value was normal at baseline? Obviously the investigator would classify as CS or NCS, but would it automatically be an AE?
It was commented 200 times as follows:
pallavi Borse • It is depend on the medication ,IF it is realated to the IP product then it is definately AE
Dr. A. D. Paul • Strictly speaking, an abnormal laboratory finding is to be regarded as an adverse event (AE), irrespective of its clinical significance. Whether such an abnormal lab. value is clinically significant or not, is its interpretation. If the abnormal lab. value is due to an investigational product, it constitutes the causality of the AE.
Sarah McLeod • The answer is yes unless otherwise specified by the protocol.
Biswajeet Dasgupta • ICH E6 ------ 1.2 Adverse Event (AE)
Any untoward medical occurrence in a patient or clinical investigation subject
administered a pharmaceutical product and which does not necessarily have a causal
relationship with this treatment. An adverse event (AE) can therefore be any
unfavourable and unintended sign (including an abnormal laboratory finding),
symptom, or disease temporally associated with the use of a medicinal
(investigational) product, whether or not related to the medicinal (investigational)
product.
Valera Bussell, CCRC, CCRA • Thank you to all of you for your input. As you may have noted; there are varied opinions regarding the issue. Years ago, when I was a coordinator; I was taught that any change in lab values from normal at baseline would be considered an AE, but apparently many companies now view that differently and only consider Clinically Significant abnormal lab values to be an AE; the clinical significance being decided by the investigator. I would like to see the FDA's opinion on the issue. Thanks again for commenting.
Lakshmi Guduri • Completely agree with Valera. Even, I was taught that an abnormal laboratory finding is to be regarded as an adverse event (AE), irrespective of its clinical significance.
Dr.Avani Badani • I do not think an abnormal lab value should always be reported as an AE. There are scenarios where even 0.1 increase or decrease of parameter happens,and that isnt a AE by anycase. It should always be reported as an AE depending on PI discretion or Sponsor's reporting requirements.
Shahid Ashfaque • The official and most accepted definitions in most countries are based on ICH E2A guidelines and are as follows:
Adverse Event defined by ICH :
Any untoward occurrence in a patient or clinical investigation subject administered a pharmaceutical product and and which does not necessarily have to have a causal relationship with this treatment. Any unfavorable and unintended sign( including an abnormal laboratory finding, that is sign or symptom, or disease temporally associated with the use of any dose of a medicinal product, wether or not considered related to the medicinal product.
Having stated the above it is an Investigator's perogative as what should be reported as an adverse event (it is his/her responsibility to assess the safety of the subject). You cannot ask the investigator wether to report any lab value as an AE or not to report it. You query or advise but ultimately it is investigator's judgement based on his expertise and medical judgement
The GCP states as follows for the investigator:
4.11.2 Adverse events and/or laboratory abnormalities identified in the protocol as critical to safety evaluations should be reported to the sponsor according to the reporting requirements and within the time periods specified by the sponsor in the protocol.
Dan Rossignol • Small deltas from within normal limits to just outside the normal range are meaningless to patients and should not be used to define AEs (in fact, lab values can move from normal to abnormal just by being converted to different units and rounded!). Sponsors should devise standards that define a minimum delta that determines when a change becomes an AE, either as a discrete value change or a percentage change. These can be used in combination with claims of clinical significance by the site investigator.
In addition, when not dealing with potentially-overtly toxic drugs (eg cytotoxic chemo therapies) we had a derived list of treatment-emergent abnormal values (TEAVs) that were based on amount of change from baseline. By this analysis a change from high but normal to just outside of normal range might not be a TEAV as the percent change was small, however a dramatic change within a normal range would be flagged as a TEAV. This (as well as shift tables) was helpful to look at possible lab changes in populations that would go otherwise unnoticed as even though there could be dramatic changes, they would not be defined as adverse events.
Shahid Ashfaque • I totally agree with Dan Rossignol
Valera Bussell, CCRC, CCRA • And I totally agree with both Dan and Shahid. Seems like there could be a standard to go by, rather than relying on the investigator to decide whether to report as an AE or not.
Marie Gabrielle Laguna MD • I agree: it is investigator's judgement based on his or her expertise and medical judgement.
Shilpa Garg Agrawal • It is the investigator's judgement which would finally decide if it is AE or not. Also, it would depend on the protocol. For eg. if there is a change in the lab value, which shows exacerbation of the underlying disease, if the protocol states that exacerbation of underlying disease is not to be reported as AE, then, this will not be an AE. However, if the protocol states that exacerbation is to be reported as AE, then, it may have to be reported as AE. Also, although investigator's judgement is final, but the CRA should be vigilant to identify any such abnormal lab values which the investigator may state as NCS, If the CRA is concerned about it, the best thing is to discuss it out with investigator, and if still not convinced, the CRA should discuss the matter with medical monitor, and take it forward..
Dan Rossignol • I agree, and this all makes sense for the use of local labs, especially
when a lab finding leads to an intervention. For an international study in
very sick subjects, our group actually defined an AE as a finding that
necessitated medical intervention. That reduced the number of minor
observations including labs and physical exam finds that counted as AEs.
But another point on lab findings is what to do with central lab results.
Variability due to judgment can be reduced, by use of a good central lab,
especially when a large study is done on a very sick population. In this
case, the site investigator's opinion is not included in the results and
the use of guidelines to interpret the data (changes in values) can be very
important.
Marie Gabrielle Laguna MD • I like these type of discussions rather than job posts.. more....more.....more please! Let us exercise our brains!
Sara Lampinen • I agree with Marie, this discussion it great! This issue comes up so frequently, it's nice to see all the different opinions.
Dan Rossignol • LOL.. It seems that this issue comes up with every study! Where would there be a "master guidance" (besides the regs of course) on things like this so we do not have to reinvent the wheel each time?
These discussions are good, but for some us (eg. the currently in transition types), the job postings may be of use!
But I have another question on how AEs can be standardized for severity in non- cancer trials. maybe start another string.
Konstantinos Kakos MD • Dan Rossignol-
A central lab will always be a better solution than local labs for a multicenter trial,especially an international one. Collecting data will be much easier and problems in the trial set up such as different test methods, different validated methods, different reference ranges etc. are eliminated but the central lab would never replace the Investigator in his role of the judging if a lab finding is CS or not.
Steve Mcateer • I agree more or less with Dan, but would emphasise that the biggest issue with using the reference range as a tool is that there is a danger of just scanning the labs looking for flags, wheras the biggest problems can be found when results are still well within the reference range. As a lab professional, we tend to pretty well ignore reference ranges and concentrate on delta change within an individual. Any decent computer system can track these changes and flag accordingly (ask your lab to do it!). Just on a technical point, reference ranges are calculated from 95% confidence limits of a 'normal' population, so 5% of results MUST be outside of the quoted range if the range is valid - thats an awful lot of AE's to be notified!
Finally, re the role of the investigator - their responsibility is for their own patient, and they are well judged to comment on clinical significance, but unless they have full access to all other patient results, they cannot see trends in other centres, which coupled with their own NCS, might tell a whole different story - the CRA is usually the only person to see this (apart from the lab of course). My overall message is talk to the lab, get them involved in setting criteria at the protocol stage, but that mantra's an old one!
Dan Rossignol • I agree. And the turn-around time for a central lab takes it out of the
"timely information" loop as well, but central lab results are great for
analyzing final data in a clean, consistent way though.
Ana Isabel Alvarez Retuerto • I believe that while the ultimate decision is the investigator's, this is at the end a team effort to get the possibly most accurate data while keeping the patient/participant's safety as a priority and keeping compliance with protocols, so if something out of the ordinary comes up it should certainly be brought up by the team members, talked and discussed by the relevant investigator/s and depending on the study/protocol, the "unusual finding" and having the patient/participant's safety at the top priority, make a decision that stays contained within the protocol.
Shahid Ashfaque • Finally, re the role of the investigator - their responsibility is for their own patient, and they are well judged to comment on clinical significance, but unless they have full access to all other patient results, they cannot see trends in other centres, which coupled with their own NCS, might tell a whole different story - the CRA is usually the only person to see this (apart from the lab of course). My overall message is talk to the lab, get them involved in setting criteria at the protocol stage, but that mantra's an old one!
Steve,My comments on above paragraph:
I see your point but we are talking two different things. We should not confuse with the responsibility of Investigator with the responsibility of Sponsor.Although both are responsible for the safety of the patients Investigator is responsible at the level of his site and patients, while Sponsor is responsible for the aggregate analysis, while the study is ongoing it is blinded, and the safety plays the dtective through signal detection. All DIL letters during the trial to investigator is the method to inform EC and Investigator of any serious adverse events occurrence. The non serious adverse event trending can be better done by sponsor since they have the panoramic review. At the end the investgator brochure is the reference they can wisely use.
We are drifting from our main question what is an adverse event? is any abnormal lab value is adverse event. The answer is NO. Use your judgement by utilizing the wholistic approach to the compound under study and considering other variables of therapeutic are of study, type of patient population, and what is the end point
Patricia Hollis, MBA • it really depends on the protocol and who you work for.
Patricia Hollis, MBA • I would also like to add that many people depend on the job posts so please do not stop them.
Essack Mitha • I agree with Steve, that changes in lab values are as important as once off out of range values.
Thanks to those that quoted ICH GCP. I regard only CS abnormal values as AE's. The reason is simply that there are normal fluctuations in lab values that sometimes may go into abnormal range. HOwever, these could be due to variety of factors, including time of day. Once i am of the opinion that a lab value is CS, i report it as an AE.
I have sometimes encountered changes in lab value over a few visits, where the value is close to lower end of normal, and gradually elevates till it is close to upper end normal. Although these are still within normal range, i consider these as AE's as well, as the change in lab value could be significant, while still being within normal range.
S K. • Strictly speaking, Its automatically an AE
manorama Mpanwar9 • I agree with Essack Mitha. If the change in lab value is CS then it should be an AE but if it is NCS then it is not necessary to take it as an AE until it is the requirement of the protocol. We can take a simple example - if lab test of a same person is performed at 2 different labs on same day we can easily see the variations in the results. As, the variation can occur so there is no need of taking all changes as AE except for those values that are CS.
S K. • I do agree with Manorama, Consider this example: Suppose PI denotes the lab value as NCS in all visit 1, 2 3, & 4 for same subject. Finally in visit 5 too he got the value as NCS then PI may decide to consider it as CS and will report it as AE.
Konstantinos Kakos MD • I think that there's a greater confusion on the issue. Clinical trials is
not just following guidelines, there's logic involved. An Investigator as a
physician will take a number of factors into consideration before reporting
an abnormal lab value as an AE. Simple fluctuations in values from baseline
do not necessarily mean AE. Of course if the protocol strictly requires
reporting as AEs every abnormal value no matter what the Investigator will
have to do it.
Remember that lab tests are a tool allowing the physician to get a general
picture on the patient's condition, they shouldn't be looked at as simple
numbers of a scale.
Brad Brush • An AE must be reported, irrespective of the Investigator's opinion of causality or individual opinion of clinical significance. ICH guidelines aside, what if we all did not report an increase in a laboratory parameter (out of the reference range) and this occurred in a large number of subjects? We need this data!
Valera Bussell, CCRC, CCRA • Hi Brad,
As I am sure you have seen within the numerous comments; not everyone agrees that all abnormal labs should be reported as AEs. Something that I find interesting since the FDA guidelines includes abnormal lab values within their definition of an AE. It would be so nice if we could standardize some of these issues across the industry!
Dr. Darpan Gangil • It should be an AE.
Shalini Dayananda • I agree with Steve regarding the reference ranges being calculated from 95% confidence limits of a 'normal' population, so 5% of results MUST be outside of the quoted range if the range is valid -and the PI land up reporting a number of AEs which have actually not occurred. However, Protocols usually define the how much deviation from the reference range - e.g twice/ thrice of UNL to be reported as lab AE. Accordingly, the AE should be reported.
Coming to Investigator's prerogative to decide upon CS or NCS - in either case AE should be reported with an additional info stating NCS as per the Investigator. This will definitely ensure that the data does not go unreported.And, this will be irrespective of whether it is related or not to the IP.
Cristina Gonzalez • I would say it depends on the situation. If the lab value is a symptom of an underlying disease, the investigator will not consider it as an adverse event unless it indicates something different as expected. For instance, for liver metastases is expected the ALT and AST values will be higher than normal, however are not an AE themsleves if the liver metastases are confirmed and no other cause is found.
Otherwise we could fall in overeporting Adverse Events.
nithiyarajan nambirajan • hi, i am a DENTIST, now pursuing clinical research course, by seeing all this comment i got good information regarding AE, so thanks for all ur comments.....
Hamdi Akan • İf you are doing a hematology/oncology trial, this issue can a bigggg problem. The best way to overcome this, is specifying limits for expected abnormal lab. results related to the underlying disease and leave it to the investigator but by definition all of these are AE's
Valera Bussell, CCRC, CCRA • Again, thank you for the lively debate and all of the input; it should prove quite helpful; especially for those just getting into clinical research. Our SOP is to report all abnormal labs after baseline (assuming they were normal at baseline) as AEs, unless the issue is clearly addressed in the protocol. I.E. if some changes are to be anticipated based on the indication being studied and the protocol clearly states that these changes will not be considered AEs or will be assessed by the investigator, then they will not be reported as such. I would encourage all sites to be written SOPs in place at their site. Please let me know if you don't have written SOPs, but would like assistance in putting them in place. Perhaps if we could all address more of these "gray areas" in the forums; we could finally start moving toward real standardized processes within clinical research. Thanks again! Valera
Shilpa Garg Agrawal • Thanks Valera, I really liked the idea of having a written SOP for this specific issue.
Adrian Johnson • Unless it is specified otherwise in the protocol, an abnormal lab value outside baseline regardless of relationship to IP is considered an AE.
Valera Bussell, CCRC, CCRA • HI Adrian,
That was my opinion as well; however it appears that some companies (CROs) may have a different opinion on that and think that is should not be captured as an AE if the abnormal value is slight and not clinically significant in the opinion of the PI. I personally think that leads to confusion, especially on the part of the study coordinator who may be assisting in reporting AEs.
Dana Austin • Also, its dependent upon other variables, such as was it an "expected or unexpected" event? Those will help determine if it should be captured as an AE until resolution or not.
Dr. Datta Pawar • Anything new finding during study should automatically be reported as AE. Over-reporting is always beneficial than under-reporting. Though it may not be related to Investigational product.
Oussama OUESLATI • if the abnormal result is observed for the first time after medication this result must be registered as AE, In any way, the CRA have not to say the significance degree: it is the physician role!
The CRA have to notify this observation as AE.
Pat Sabatini • Adrian - I agree and it should never be left to the Investigator.
Valera Bussell, CCRC, CCRA • I would agree and as previously stated, it is our SOP that abnormal labs that were normal at baseline should be reported as an AE, unless it is explicitly addressed within the protocol. I lean toward the over reporting rather than take a chance on not reporting an AE that should have been reported. It is unfortunate that some of the CROs have decided to insist that the PI decide whether to report as an AE or not to report. I would prefer a directive from the FDA before we change our SOP. Thanks to all for your opinions.
Shamshad Ali • I totally agree that based on safety aspects any abnormal value should be reported as an AE
Essack Mitha • I can see that there will not be a strict guideline on what should be reported as an AE, as Investigators have their own opinions. However, the medical monitor and the sponsor get results of every subject on the study, so even if an abnormal result is not reported as an AE, the capturing of the lab data will show changes in lab values. It is this that is also taken into account when looking at the effects of IP, not only what the investigators report.
Dan Rossignol • I agree with Essack Mitha as eventually all changes, even if they occur
within normal ranges (eg platelet counts that can be high normal at
baseline and low normal post treatment) need to be evaluated (and compared
to placebo where possible) to see what is going on in the treated
population. While we talk about AEs or lack thereof, these are simply
"cut-offs" and should be understood to be "clinically significant" or "non
clinically significant" but in the end, there is usually more info across
all subjects over time (with post-treatment recoveries) that are evaluated
in the clinical study report. Signals seen at low dose IP may indicate
potential targets for close observation as you move to higher doses,
extended treatment, or treatment of more impaired patients.
Emmy Merkley • I think I have always followed the following criteria,. If the investigator deems it is clinically significant after repeat testing of an abnormal value to confirm the 1st abnormal value and it may require some kind of treatment, follow-up testing or obeservation, then it does become an adverse event.
Dan Rossignol • That seems like an ideal definition. Simple and clear. The requirement for
an intervention makes a case for significance as well.
YJ Bi • This is a great discussion! Just want to add my humble opinion:
During the collecting of data (investigational site/investigator and site monitor), in general, an out of range lab value should always be captured as an AE on the CRF. Once it reached datamanagement, usually involve medical monitoring and data review/coding/cleaning which is also done under FDA regulations/SOPs/study data specifications, Such AE can be kicked back via data query to remove or just confirm the AE on case by case basis (e.g. lower WBC while the patient had a flu/other viral infections, transient transaminase increase after a dozen beers previous night at a birthday party, etc).
As all AEs, the clinical significance and severity is rated by the investigator (while serious is by FDA definition). It not uncommen that an AE of out of range lab value, to be queried.
However, before CRFs is pulled out, an out of range lab value should be reported as an AE. But this can still generate an query. Such kind of query is a part of the datamanagement process, even site/investigator and monitor are doing the right thing to capture such AE.
Thus, out of range lab value is to be reported as an AE unless study specific documentation indicated otherwise on the CRF. It may generat a data query, then it may be handled differently depending on the data convention and clinical scenario.
Carlos Estrada, MD • I agree with Biswajeet Dasgupta, It is considered an AE, as an Investigator, I am able to suggest if it is related or not to the investigational product, but it is definitely an AE
Sebastian Antonelli • This discussion seems closed by now, but anyway, my 2 cents: in general, report an abnormal lab as AE when it triggers a medical intervention (thats why it is called "medically significant"), and such medical intervention could range from a medical advice (i.e. "you should really reduce the number of MacCombos per week"), a drug prescription, or med change, or dose adjustment or discontinuing of an ongoing med, or additional testings that confirm a diagnosis. If there are symptoms accompanying the abnormal lab, please don't duplicate entries, AEs should be diagnosis if possible. And every AE should have an action taken.
Sundarkia Hill • As a former CRA, I always required changes from baseline to be reported as adverse events. Even if it is exacerbation of a pre-existing medical condition, the exacerbation itself is an adverse event. Whether the reference range is set by the lab or by the investigator's judgment, any lab result deemed abnormal is an adverse event, regardless of clinical significance. Determining reporting relevance at the site level can lead to obscuring trends across the study which could give further insight into how the drug works. In my opinion, it is the duty of data management and the data analysis committee to adjudicate these events once they review data from all sites, from all countries, for the entire study. There will always be room for debate between sites as to what is relevant, even when a master guidance is given. Therefore, I believe the only necessary master guidance is to report all abnormalities and let those individuals privy to all study data determine relevance.
Carlos Estrada, MD • I agree 100% with Sundarkia, as a PI and Medical Monitor, the abnormal changes in labs need to be reported as AE's, and it is the discretion of the PI and/or Medical monitor ( or Data Safety Monitoring Board) to adjudicate the AE to a particular Drug/procedure/device or not.
Jacqueline Gough • Fascinating discussion that truly does seem to occur on a regular basis.As others have stated, the reference ranges are only useful for understanding where the patient's result lands with respect to the "normal" distribution of results. This is usually the reason that the investigator's opinion of clinical significance has been included in the determination of whether an abnormal lab value is considered an adverse event. What might be abnormal according to the reference ranges may be completely normal for a particular patient and only the investigator, with their more complete experience with that patient, can assess this. This is completely distinct from relatedness.Additionally, although it is conceptually more conservative to include all abnormal lab values as adverse events, there is a risk that any safety signal from that information could get lost in the noise of so many adverse events that are not clinically significant.Instead, perhaps adverse events should be reserved for what the investigator assesses as clinically significant and the actual lab values can be analysed for trends or items that the investigator has failed to notice.The purpose of collecting this data is to identify any safety risks associated with whatever intervention being studied. Patient management should, in my opinion, be left to the individual investigator.
Sundarkia Hill • While I agree with the group's interpretation of a reference range, I must disagree with the interpretation of clinical significance and its role in clinical studies. Clinical significance has no bearing on determining if an abnormal lab is an adverse event. The ICH definition of an AE shows an abnormal lab in itself constitutes an AE. If you are consistently not documenting abnormal labs as AEs and you are audited by a CRO, FDA, or IRB you will likely be dinged for not properly reporting AEs.
Clinical significance is followed for the purpose of monitoring the status of the event with the subject. For example, if I am a CRA monitoring a subject’s chart and found a lab value was 4 times the upper limit of normal, I expect to see active intervention by the investigator to bring this value back to normal (change in meds, retesting, etc). I would also make it an action item in my report to follow up on this abnormality to see if the intervention had an effect, resolving the AE.
Regarding AEs in general, sometimes the side effects of drugs are hidden in the small changes and I think it would do a disservice to the industry to ignore them. Although reporting these events can lead to queries, this is actually a good thing. The queries I often saw as a monitor would say something along the lines of “elevated ABC plus decreased DEF codes to condition XYZ. Please confirm if the subject has been evaluated for XYZ.” This could be from a small increase over baseline, just barely putting the result out of range. Something that would likely be marked as not clinically significant when looked at separately can become significant when looked at collectively for a subject and accross subjects at different sites. Such queries come from AE reports, not from the lab. This also shows the level of analysis of these events and can sometimes make the site staff dig deeper into something that may have been accidentally overlooked, simply because the changes were not significant in the opinion of the investigator. We should not rely soley on investigator opinion because every investigatior's opinion is subjective, no matter how experienced a physician one is. My advice is to report all abnormal labs and allow the data monitoring committee to adjudicate.
Dan Rossignol • Beautiful. Put writing of this sort in the protocol and you should have
everythingyou need (between this and the analysis of lab changes) And I
like your sense of humor. But I would consult regulatory to see if you have
to list the BK and Wendy combos as well in the example.
Valera Bussell, CCRC, CCRA • Thank you again for all of your opinions; I am especially happy to see that many of you agree with me that a change in lab value from normal at baseline to abnormal after starting the study drug should be considered an AE based on the FDA's definition of an AE and should not be based on the evaluation of how far outside normal range it is in the opinion of the investigator. I will continue to direct my research staff to capture these change as AEs, with the investigator merely evaluating the seriousness of the abnormality (clinically significant or not) etc.
Gulben S. • If the investigator classify it as CS then it would be an AE. Except 'special cases' if it is NCS it would not be an AE.
Carlos Estrada, MD • Again, as the question said, if the laboratory value is ABNORMAL, it should be reported as an AE, regardless of causality, if the laboratory value is DIFFERENT from baseline, it needs to be assessed first as CS or NCS, before being reported as an AE.
During an FDA audit this is something the inspectors will look at and you might get a 483 for not reporting as an AE.
It is the responsibility of the PI, Medical Monitor and or the DSMB to adjudicate the AE to a Drug/device/procedure.
Erika Tokaji • To ensure consistent reporting leading to meaningful safety profile and help both the sites and CRAs with SDV, the protocol should provide guidance what is to be reported as an AE. One commonly applied approach is to specify CTC severity grades as from which the result should be reported as an AE (usually grade 3, 4).
Artie Jhappan • My experience working in Phase 1 trials - only abnormal lab results that are deemed CS by investigator were reported as an AE.
Artie Jhappan • Often the regulations are broad and non specific.I think this is something that needs to be defined clearly in the protocol.
Artie Jhappan • Joyce -Excacerbation of an existing condition should be considered an AE.
Cynthia J. Robinson • If the lab value is significantly high from base line then it is the judgment of the investigator to determine if the value is an AE or not.
Pat Sabatini • If it were to either be written in the protocol or left up to the investigator AEs would be all over the map.
Valera Bussell, CCRC, CCRA • Pat,
I agree, but there are some CROs that don't want all abnormal labs captured as AEs and have specified that it should be assessed by the PI and then a decision made whether to capture it as an AE or not. Historically, all of the sites that I have worked with have reported abnormal values as an AE, even if not deemed CS. I see it the same as you apparently; where do we draw the line if we leave it up to a PI? Can they consider other issues as not worthy of capturing as an AE as well if that is the case etc.?
Abhijit Sharma • Abnormal lab value + Clinical significance = AE
Abnormal lab value - Clinical significance = Not an AE
Clinical significance= standardiztion in Protocol/PI judgement
In case of CNS abnormal lab value ratinale can be documented.
Pat Sabatini • Hi Valera - I'm not suggesting that all abnormal labs be captured as AEs but that CS must. The definition of CS has always been if it requries intervention. The same goes for the severity of the AE, this is not defined by the PI, there is a clear definition of this and every PI must be given the same yard stick by which to measure this. It is up to the monitor to verify that they are all using the same yard stick.
Meenakshi Manerikar • yes it is AE
Karolien Timmermans • Hi,If you look strictly to the definitions in ISO or GCP, an abnormal lab value should be documented as an AE. However, as already pointed out before, you also have to look at the clinical significance of the abnormal value.
Also, if no treatment is initiated to correct the lab value, i would not consider it to be an AE.
Nevertheless, I would obtain confirmation from the PI in the center and make a note in the monioring report.
Cynthia J. Robinson • Shilpa Garg, I agree it is the principal investigator’s (PI) judgment that makes the final Adverse Event (AE) decision. However, it also depends on the protocol and whether this is an exacerbation from a chronic illness. Prior to final decision, PI should repeat labs to confirm significantly high value. If CRA is still concern whether this entails an AE then CRA should collaborate with the Medical Monitor.
Muhammad Ather Latif • Abnormal value which has / could have relation with administration of investigational product and was not recorded in baseline or history of the patient/subject will be considered and regarded as AE. Its significance may be analyzed by PI later.
Priya Reshma • its very good kind of discussion....getting more information and also helping us to update our knowledge..........thanks to all
Jhappan • a stimulating discussion.... I like it
John R. Waldron, CCRP • This is definitely a case-by-case situation with no general rule or regulation to follow. Lab abnormalities that constitute worsening of the disease under study would not be an AE. For example, elevated liver enzymes in patients with known progressive liver metastasis. In addition, lab values slightly above or below ref range could be deemed normal for that patient, interpreted as NCS and not an AE.
On the contrary, lab abnormalities that constitute worsening of pre-existing conditions found in medical history would be an AE. All lab abnormalities that are study drug related should be interpreted as CS and listed as AE.
Samuel W Boellner, MD • depending on the study I would recommend a repeat
Luigi Nardacchione • We cannot consider the sponsors' or investigators' opinion (or interpretations) but just the official guidelines. So no doubt about it as per ICH E6 and related FDA/EMA requirements: it is sponsor and investigator responsibility to strictly follow them in order to protect patients and also not to jeopardize study results.
Helen Russo • While everyone continues to quote the untoward regulation, I think the boat has been missed...not all abnormal labs constitute an untoward event, particularly if the PI considers them NCS. Also, an abnormal lab value, in and of itself, would not be the AE, but rather the underlying condition, even if it's the probable/possible effect of the IP on liver fx, e.g. In other words, "elevated ALT" is not the AE.
As a monitor reviewing signed labs, I always tell my PIs that I reserve the right to ask them "why" they feel an abnormal lab is NCS if I disagree. It may be that they have reviewed that particular report away from others and not noticed a trend that I have, or they may have just missed it...it happens, but the bottom line is that it's the PI's responsibility to make these decisions, absent specific language in the protocol, and it's the monitor's responsibility to ensure the PI is doing so.
There are also instances where I was taught that an abnormality in particular assay is always CS and must be repeated to ensure it was not a lab error: minute elevations in serum calcium, e.g., a 0.1 mg/dL elevation is likely a lab error, but must be repeated to ensure that fact as even that small an elevation can indicate pituitary tumors; should the value return to WNL, there would be no AE. I had a PI overlook an elevation like this until the subject started presenting other symptoms and he ended up with a very serious problem on his hands!
Bottom line, if all lab abnormalities were automatically listed as AEs we'd be so completely overwhelmed we never be able to do our jobs.
Dianne Edwards • I agree with Joyce. If not part of the patient's MH; it is an AE unless otherwise specified in the protocol for the intented investigation under study and the change is expected and documented in the IB or PI.
However, that said, if the change in the value is clinically significant (outside of patient's "normal" normal value); increases in severity, frequency or duration - your must document this as an AE according to ICH E6 and GCP guidelines,
Dan Rossignol • Helen,
Excellent points here. It seems that the reporting of AEs as well as
careful analysis of trends in groups based on hard evaluations of changes
in lab values (not just changes that move from normal to abnormal) need to
be part of a final study evaluation. It will confuse things to have minor
changes reported as AEs. As has been stated, the protocol needs to be
clear on this.
Despite any protocol discussion however, the PI does have final say and
should be able to explain the AE claim. One problem that I have seen
however, is that investigators may be overly conservative in claiming Aes
in fear of an agency audit that picks up abnormal values and asks why they
were not claimed as AEs. This again is where the protocol language
(presumably approved by the agency) as well as careful monitoring protects
them.
Dr. Ashish Indani • I agree to the consideration that altered laboratory profile should be considered as Adverese event. These findings are, in fact, more important than clinical findings in many cases, especially when investigator is specialist of one perticular decipline. this is because, these investigations are performed for this very reason. Let's review a few facts
1) The investigator is always a medical professional and just like routine, the investigations may have been be left to his discreetion, if the designer would not want to note the sub-clinical (NCS) changes
2) The basis of calling for a perticular set of investigation by a Study designer is based upon the preclinical, previous phase clinical and Theory information of investigational product and the substance of disease, whatever is available.
3) the adverse event reported for a perticular drug must not always be bad. Infact we must keep example of Sildenaphil, Aspirin in consideration, that these drugs are highly sucessful in more than one ailments in different dosage. This whole phenomenon was based upon the evaluation of adverse events reported (Clinical in Sildenaphil and laboratory findings in Aspirin)
Considering the fact that in a clinical trial, we design the study with certain statistical qualifications as posted by Steve, hence frequency of such reporting will be automatically taken as significant or non-significant at the time of evaluation. There is also a reference data from previous phases and control of the study available to estimate and evaluate whether the changes in investigations are pertaining to the investigational product. For this analysis to go impartial and unbiased, Reporting of any any every change in the subject's overall status, including physician's assessment, any scores which are designed, any laboratory investigations etc. must be reported in time.
Michelle Perkins • Site for dtp jobs, information about graphic and computer jobs. http://lnkd.in/7hH7e3http://lnkd.in/Zfqbhj
Ramune Kanapieniene • I suggest reading the Journal of Best Clinical Practices "Documenting Clinically Significant Lab Values" by S. Eric Ceh http://firstclinical.com/journal/2009/0901_CS_Documentation.pdf
Vindu Seal • All abnormal lab values need not be documented as AEs. They have to be classified as CS or NCS and all NCS findings need to be documented as an AE. However if the abnormal lab value is falling within the defined alert parameter in the protocol then it has to be documented as an AE.
Vinette Zinkand • There are certain medication that we expect to see a change in lab results. In a study situation you would be blinded to those result unless they fall into a critical range that would necessitate clinical intervention. Then the lab result would be an AE or an SAE.
marita mcdonough • In most instances this would qualify as an AE unless specified otherwise in the protocol.
Dr. Dipti Kale • I think if should be classified into 2 categories clinically significantly abnormal and non-significant abnormal lab value. if it is clinically significantly abnormal value then it should be recorded as a AE.
Dr. Dipti Kale • many a times it happens any of the subject's lab value may be abnormal before taking IP and it may continue as the same after taking IP also and sometimes the subject may be nutritionally deficient . so it becomes necessary to track abnormal lab value right from screening visit to subsequent visit and it also becomes necessary to find out the causative factor. if it is done mandatory then it can increase work load for the site and central lab also. the abnormal lab value can be of significance and it can not also.
Lynda Cedar, Ph.D. • Change from the baseline for medical diagnostic values happen often in the clinical studies.
The changeis reported and interpreted by the medical staff as CS or NCS. If found unfavourable it is therefor reported as an AE and treated as such.
Sometimes, a healthy subject is accepted to participate while the values were within the range but bordeline. At the end of the study, such values might become out of the range of acceptable values. In this case, based on the medical judgement, the study medical doctor may decide to refer the subject to a family physician for more exploration and further diagnostic. At my site we document that the subject was informed and advised to consult his family doctor.
Anand Kawade • An abnormal lab value ( lab doing investigation has to state normal ranges for them) is always an AE, even if found incidently. but PI has to decide their clinical significance, intensity, and relatedness
Brown Kim • Great thread. I would love to see a lot more like this one. Thank you everyone for your well thought out replies to the question. I learned a thing or two.
Liljana Stevceva • FDA toxicity tables define the range that should be considered an AE as well as the grade of the AE.
Dr Faisal Khan • Yes, it should be considered as an AE and grading to that has to be agreed upon with the investigator
Binay Thakur • Not all abnormal laboratory findings are often associated with untoward medical occurences and therefore should not be recorded as AE up front. Further, for any particular parameter unless associated with any sign or symptoms such subjects would more resemble a healthy population (having similar degree of variations in the lab value) than a subject population. It would therefore be more prudent to know if there is any sign or symptoms associated with a lab abnormality before considering it as an "AE" unless there is sponsor/protocol specified guidelines such as "CTCAE or similar guidelines would be followed for recording AEs regardless of their clinical significance". Also parameters where laboratory error or false positives are anticipated, if deemed neccessary by the investigator re-tests should be done to reconfirm the abnormality and only then an AE should be recorded.
Objectivity in capturing AE data is good however at the end of the study to make a sense out of it would be very difficult, as each subject with the same degree of abnormality presents to the investigators with totally different case histories, this is where subjectivity is important and neccessary and therefore an investigator's judgment must be honoured (or argued with medical monitors if need be) while recording an AE.
Raji Varghese • AE reporting for abnormal lab values are mostly PI depended some values are CS and some may be NCS,it is better to leave at PI's discretion bcoz they know patients condition better than anyone else,to report all abnormal lab values means lot of reporting to be done.
Sergiy Gryshyn • I think that any lab abnormality is AE. Another question is, should we report it to Sponsor?.Any AE should be documented in SD with clarification is it clinical significant or not. Then we have to follow to protocol requirements. According to my experience, we have to be very careful with interpretation of AEs and have to understand what does it mean clinical significant or not. Leucopenia grade1 as per CTC during chemotherapy, is it significant or predictable and not significant? And doctor treats it.
Rosemarie Harris • It also depends on the indication (i.e. oncology) and how sick the patient population starts out at baseline.
June Rosas • Depends if at baseline if the value is normal for that patient. Also, depends on if the sponsor/ protocol.. wants this reported or not. But definelty should be documented.
Liljana Stevceva • This issues are defined in the protocol. Shouldn't the PI grade the lab abnormalities per protocol? I think yes.
Raji Varghese • Most of you are missing a point that investigator is the right person to mark a abnormal lab value as AE or not until and unless the protocol specifies so.Only the investigator is in contact with the patient it is just not the lab factor that is taken in to consideration there are other factors which is taken in to consideration as Konstantinos Kakos mentioned it is tool that is used to get the picture of patient's condition
Helen Russo • Agree with Raji, and as I noted before an abnormal lab value is a diagnostic tool, not a diagnosis, and in and of itself may/may not be "untoward." as with any other AE, one does not list the symptom (i.e., runny nose, swollen finger), but the diagnosis. An abnormal lab may or may not be an indicator of a problem or it may be a normal variant in the subject's life, but the lab, itself, would not be the AE...
Sergiy Gryshyn • According to ICH GCP, definition of AE doesn't content any Information about PI's judgment. Of course, it's theoretically, on practice, we can meet a lot of different decisions, etc. But in this case, who should be responsible for such decisions?
Traian Andrei Manu • I want to raise some issues with you in regards to what should or shouldn't be reported as an AE for an abnormal lab finding. First of all the majority of studies have a central lab involved and the results are pretty much standardized so you cannot go around saying if you change the units for a lab value you would have an AE.
In regards to what is abnormal you should keep in mind that the subjects involved in phase 2-4 studies are sick. It is very possible that their lab results wpuld come abnormal for some markers. Let's say for a RA study : of course you will have a PCR value for most of the subjects above normal. It is clinical significant because it shows that the subject has an inflamatory response to the disease.
According to GCP every abnormal lab finding can be considered an AE but ultimately the PI will decide if that is an AE or not. The role of the monitoring team is to make sure the lab results are not overlooked or not taken seriously.
Please keep in mind that nowadays central labs offer services for evaluating for soem alb results the difference from baseline to any other visit in the study so you will be able to see a trend for that subject, site , country and IP safety. Do remember that you also have services for a central lab aside the site's , CRA's role to verify the lab results, in regards to Hy's law and that lab values for the liver function if it's above Hy's law standards it will automatically be a SAE.
Lucian Popa • Andrei, regarding your post, the responsability of the investigator is to assess if the AE is CS or NCS, not to define what is an AE, for AE we have a standard definition in ICH-GCP and most of the times the protocol define what AE should be reported or not. The problem is that sometimes the investigators don`t want to record AE that aren`t related with IP or study procedures, in that cases is CRA duty is to identify and make sure that they are reported.
Adel M. Medhkour, MD • abnormal lab values occurring during a clinical investigation are to be regarded as AEs, unless (1) specifically highlighted in the study protocol to be a natural occurence without clinical relevance, or (2) such values are known historically to fluctuate during disease progression with or without the introduction of the investigational product, or (3) such change in lab value is a contributory to patient improvement or a pure sign of efficacy of the investigational product.
I don't think it would a good idea to report an efficacy variable or endpoint as an AE!
Rahul Dawkhar • Lab values: if abnormal, may be associated by many factors and to have a correct judgment study physicians can review the AE’s to give solid evidence, justifying the abnormality as an AE. Instead of jumping the gun, this practice would always help. All the data reported for LAB as abnormal values are extracted in a listing and reviewed on ongoing basis for the study by physicians.
Also protocol does specify some abnormalities that can be classified as an AE but only after considering other factors which a study physician can bring into the picture.
sumit saxena • Dr Sumit Saxena : As per the ICH GCP Guidelines and most acceptable by the industry
Adverse Event defined by ICH :
Any untoward occurrence in a patient or clinical investigation subject administered a pharmaceutical product and and which does not necessarily have to have a causal relationship with this treatment. Any unfavorable and unintended sign( including an abnormal laboratory finding, that is sign or symptom, or disease temporally associated with the use of any dose of a medicinal product, whether or not considered related to the medicinal product.
Still we have to correlate with baseline values and GCP said about abnormal laboratory finding it could be CS or NCS as per investigator discretion.
Jon Ruckle • "Normal" ranges are statistical definitions, and even healthy people can vary a little above or below those ranges simply because that is the way the body responds to multiple factors, even apart from a clinical trial. My opinion: as a clinician, and a PI with extensive Phase I experience in healthy/normal individuals, lab values outside the "normal range" should be assessed by the Investigator case by case, and should not automatically be classified as AEs.
Charles H Pierce MD, PhD, CPI • Hi, I am a bit late in this discussion but when I saw the question "Should an abnormal lab value automatically be considered an AE if the value was normal at baseline? Obviously the investigator would classify as CS or NCS, but would it automatically be an AE?" I had to answer as I had just given a Webinar on "Adverse Events".
The answer is an emphatic "No", a value of a laboratory variable above or below the reference range is not (NOT) necessarily an AE. In fact, it would only be an AE if, in the opinion (Medical judgement) of the PI backed up in the Protocol that the value was affected by the Investigational Medicinal Product (IMP) and indicated an effect of the IMP. I agree with Shilpa above on this clearly
For example, if the CK value was 2 or 3 times the ULR (Upper Limit of the Reference range) and the Protocol stated that above 5 times the ULR was an AE, one would not question that it was not. If the Protocol was silent in this circumstance but the patient admitted to painting his dining room the day before, I would also not think that this was an AE as an elevation of CK following muscular exertion wold be expected and not unfavorable.
Incidentally, the term we must all use is the "Reference Range" which is correct for the mathematical and statistical lab ranges we all use, which imply, correctly, that one could be "normal" either above or below the reference range. Basically, the term "Normal" range or "Normal" Value (for lab data) has no place in clinical research. For this reason alone one would never (NEVER) state that any value above or below the Reference Range would be, by definition, an AE.
Also agree with Biswajeet as far as he goes but it remains for the Protocol and/or the PI to decide whether or not the elevated or depressed lab value is an unintended or unfavorable sign. Charles if you wish clarification
Harold Doshan, Ph.D. • I have to agree with Dr. Pierce on all counts. One or more isolated changes in lab parameters that fall outside the reference range are not inherently AEs unless so stipulated per protocol. They are, like "sniffles" or "sore throat," possibly signs and symptoms of an AE (a "cold"), related or unrelated to study treatment, and either CS or NCS, as judged by the PI. It is normally the PI's call to report such occurrences as AEs, based in his assessment of the subject's condition.
Further analysis, including identification of progressive changes in the same or related lab results in the same or other study subjects can and should be evaluated longitudinally to shed additional light on what may, at first, be perceived as an isolated occurrence. Indeed, even temporal trends (so-called "shift analysis") that remain within the reference range may be suggestive of treatment-related effects. For this reason alone, all trends, both within and across study subjects, are evaluated as part of the study analysis.
The study staff has an obligation to report and annotate as "out of range" any lab result that fits that criterion, but to deem it automatically an AE is generally unwarranted.
Lynda Cedar, Ph.D. • I agree with Charles too. IThe lab vaues of baseline are used to see whether a significant change happens to the subject at the end of study or during its monitoring. The significance is left to the PI (medical judgement) which prevails on the study protocol. Additional tests even not scheduled by the protocol might be requested by the PI/medical doctor in charge of the study.
Example: if the CPK are significantly high, the subject might be not allowed to participate in a study. However if they become significantly high during the study or at the end, after a physical examination, medical history, the medical doctor might ask for a CPK-MB test, an ECG, ultrasound or any other test.
When speaking of borderline values, the same value might be considered normal for a patient but abnormal for another one. The results of blood and urine analysis are not considered individually. Each group of values reflects the function of a system: cardiovascular, kidney, liver, nervous, etc.
Andrea Bianchi • I agree with the mayority of the answers posted here, however, in every Regulatory Agency Inspection I have attended, the issue is always adressed in the same manner: if the protocol states a maximum lab value after baseline determinations, it should be at least consulted with the sponsor, before making a call on an AE; this electronic and paper trail, can make the difference between an observation in the final inspection report, or a protocol deviation.
The PI' s judment is always respected, as long as it has documented grounds, and means of demonstrating that there was at least a query regarding the finding.
Subhash Amrutham • I absolutely agree with Charles. Lab Abnormal value not necessarily be an AE; its purely Principal Investigator discretion.
Dadahayath Shaik• I agree with Dr.Charles
Ming Q. Lu, MD, PhD • An abnormal lab value will only be considered an AE if it is clinically significant. If the value was abnormal at baseline and it remains basically the same during the trial, it is not an AE.
Ming Q. Lu, MD, PhD • CIOMS VI group report is a good reference for detailed PVG guidance.
Diane Bartusiak • I agree with Harold: "One or more isolated changes in lab parameters that fall outside the reference range are not inherently AEs unless so stipulated per protocol." Populations like Oncology are evaluated using a wider range of reference values and these Oncology lab ranges are used when the sponsor is reviewing lab tables. Any laboratory trends will be captured by Stats or the study team. Therefore, it's up to the PI to classify, per protocol, whether any patient has an Adverse Event when a lab is slightly abnormal. It's helpful if PI can research a diagnosis/illness asssociated with lab change. Source documentation is crucial so the monitor can confirm that the site's AE reporting is within guidelines/protocol.
Carlos Estrada, MD • HI Guys, I believe we are loosing the original point, the question was: "Should an ABNORMAL lab value automatically be considered an AE if the value was NORMAL at baseline?" So, referring to this particular question, my answer as a Physician and Auditor is YES, it should be reported as an AE, regardless of it is Clinical or not Clinical Significant. If a Subject starts the trial with normal values at baseline and now it is abnormal, it is an AE.
ICH E6 States: "An adverse event (AE) can therefore be any unfavourable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medicinal (investigational) product, whether or not related to the medicinal (investigational) product"
Ming Q. Lu, MD, PhD • In the real clinical world, this will generate tremendous AEs which are not meaningful. e.g. normal range for diastolic BP is 90 mmHg, if the subject has 89 mmHg at baseline and 91 mmHg occasionally during the trial, it is just a common fluctuation.
Therefore, the answer is NO for the question "Should an ABNORMAL lab value automatically be considered an AE if the value was NORMAL at baseline?"
Helen Russo • Dr. Estrada, I think you just negated your own point with your ICH quote. An abnormal lab value, even if normal at baseline, is NOT necessarily unfavorable or unintended. As many have pointed out, labs fluctuate under completely normal circumstances. Reference ranges are just that, reference ranges, and it is up to the clinician to determine if a value outside those ranges constitute a clinically significant abnormality. If the answer to the question is yes, then the lab should be used as diagnostic criteria for an AE, if the answer is no, then it is NOT considered diagnostically relevant, and is noted as such by a physician on the report/subject's chart.
As a monitor and auditor, I've seen (and questioned) labs on both sides...an elevated fasting glucose in a non-diabetic noted as NCS, e.g., I've also requested repeat labs for things like elevated serum calcium which is only elevated under two (to my knowledge) circumstances: lab error or pituitary tumor.
Also, unless the protocol specifies differently, an abnormal lab value itself is not generally the AE, and many sponsors specifically ask that "symptoms" NOT be listed as AEs, which is what an abnormal lab value would be. Many times I have instructed sites to remove multiple symptoms and request the physician to use one diagnosis, this would include abnormal labs, i.e., elevatd WBCs, stuffy nose, earache, etc., in favor of a diagnosis of URI. I have also seen instances of wildly abnormal CK values, e.g., and cme to find out the subject was exercising extensively. Usually in a case like is, and absent of any other untoward signs/symptoms, the subjects are asked to discontinue their training and the lab is repeated until it return to normal. These situations are not generally AEs, or reported only once, and ended when the value returns to WNL.
The bottom line is that a blanket response of "yes" to the original question of "should abnormal labs be listed as AEs if normal at baseline" cannot be made.
Tim Pratt • We recently had a lengthy discussion on this topic prompted by "anaemia" as an AE.
Labs vary in what they consider normal, so a finding below an absolute value won't work (in this case it was 10). More appropriate, perhaps, is a relative, so a decline of X% from baseline. That, however leads to other problems - decline in haemoglobin during menses for example is normal.
We proposed to our DSMB, and they accepted a 2 part criteria for this particular value to be an AE - 1) a shift of X% from baseline (the relative method) AND 2) for which medical intervention/treatment was provided. the latter could be as innocuous as giving Fe supplements or as dramatic as transfusions. Our DSMB thought these criteria, in tandem, were very appropriate. This model is along the lines of that proposed by Dan R above.
Theresa Dunaway RN,BSN,MBA,CCRP • I think we are getting off topic here. I don't think there is one and only one answer for this question. Yes the protocol needs to state the labs that are significant to the study. The reference range for the lab being used for the study (if not a central lab) needs to be filed with the sponsor. Honestly it is the PI, and only the PI that makes this determination. If a woman's hgb is significantly decreased from baseline, there may be a problem. It should go down but not to significantly, if she is on a drug study you do want to acknowledge this. Also Dr Lu I don't think the example you gave is that accurate...we are speaking of ranges but not those that would be 'normal' for a particular patient.
I must say that for any lab out of range you do need to consider the baseline and relevance to a single individual. I do strongly believe in re-tests, I do believe there are errors and fluctuations, this is also for the PI to determine. But I agree with Dr Estrada, If the protocol doesn't state that small out of range results are acceptable then the PI has to consider the this as an AE.
David Radin • By definition "Normal" means the value is within the 95% confidence range. In other words you would expect out of 1000 subjects 25 of them would be "high" and 25 of them would be "low." If the protocol calls for reporting of abnormal labs int he absence of clinical manifestations then it would be an AE by protocol specific definitions. If, however, the difference is within what the clinician would consider an expected variation in a patient without an underlying disease burden it is simply "normal" and not an AE in and of itself.
Paula Singleton • For concern for the patient's safety, as well as the overall review of the study, I think any change from baseline should be listed as an AE. A case that comes to mind is a patient who had an infected injury and was prescribed an antibiotic which caused a marked change in one of the lab ranges, a factor that was a) not unexpected, and b) would have definitely been considered an AE had it not been for the antibiotic that was prescribed. In this case, the change in the lab value was not because of the study drug, it was because of the antibiotic. Regardless, I feel the change should be marked as an AE and then noted as a response to an antibiotic. For the patient's safety, any anomalies while on the study drug should be documented, and if not in the AE log then where?
David Radin • Which makes MY point -- unless the antibiotic makes the lab change ALWAYS you have a change due to a medication (an AE) which cannot be 100% attributed to the con med. A further examples are GGT and ferritin -- these vary so much without clinical manifestations that Medicare specifically excluded them from blood chemistries and will only pay for them if there is a specific supporting diagnosis/clinical syndrome. Abnormal labs without clinical manifestations can always be captured by Data Management and will not be "lost."
Paula Singleton • Good point. For my example above, if it creates an expected out of range value for any and all patients who take the con med, then it would NOT be an AE. but should it only create out of range values for some patients, it should be listed as an AE for those patients.
For the original question, Using the reasoning above, if the reason for the change is specific to the patient, isn't defined as expected for any reason, whether in the protocol or other documentation, I believe it should be an AE.
Andras Koser MD, MBA,CPI, FHM • I am not sure if any you who responded are PI's. I am a PI, and it is my duty and responsibility to determine what is an AE and what is not, unless I have specific instructions in the protocol. To do that the PI need to be involved, know medicine well and know their patients and their history.
We can speculate long on specific parameters we measure, there cannot be one rule for all.
Theresa Dunaway RN,BSN,MBA,CCRP • Thank you Dr Koser. I think we all need to look at Dr Koser's comment as the correct answer.
Charles H Pierce MD, PhD, CPI • Amen to what Andras has said. Also, I would work with Theresa any time as she understands this issue. I am and have been a PI and have >20 years in the Clinical Research industry. I suspect that any and all who answered this question as you and I did - that it is the PI who decides if and when (and in keeping with the Protocol) that a lab result outside of the Reference limits (ULR) is an AE or not. I have been astounded by the responses who suggest that any result above of below the Limits (and they sometimes wrongly use the term "Normal Limits") is an AE. Have just given Webinars on "PI Responsibility" and "Adverse Event Reporting" and nowhere in the Regs does it state otherwise. Period
Helen Russo • I think that last few responses have pretty much hit the nail on the head with regard that you CANNOT make a global statement that all abnormal labs s/b treated as AEs, and at the risk of repeating myself, in every initiation visit I have ever done, my "schtick" is that you are the physician and it's up to you to determine what is/is not an AE, however, I reserve the right, as a monitor, to ask you WHY you have made such a determination if it is unclear to me, and to put the reason in writing either on the lab orin a chart note, because if it's unclear to me, it's going to be to someone else such as an auditor.
I have also, frequently, requested physicians to differentiate between WNL and abnormal/NCS, because they absolutely do have different meanings, and as Dr. Pearce pointed out many incorrectly use WNL.
Theresa Dunaway RN,BSN,MBA,CCRP • Thank you the for the compliment Dr Pierce, I only have 11 years but it still feels good! Helen I agree there is a difference between abnormal/AE and NCS, but you as the monitor still has no responsibility here, your responsibility as well as a coordinator's is to review the documentation. Each of us involved (Coordinator, sub-Investigators, Monitors, sponsors and FDA) need to ask the PI what his rationale is. I have even seen Investigators debate on AE's vs. baseline and many other issues, but it is still their responsibility to make this determination and support it.
Lynda Cedar, Ph.D. • Any event or change that happens during the study is reported,documented and interpreted. Speaking of lab. values change, deviating from the baseline, the change is captured as an AE but its clinical signififance belongs to the PI (medical doctor) whether it is NCS or CS.
Helen Russo • Sorry Theresa, but I disagree, though I never have said that I have the responsibility. It most definitely is part of my responsibility to 1) make sure the data being submitted is accurate, and 2) that the physician reviewing the data is clear to the reviewer. If I do not understand a physicuan's rationale for for a decision, and ask him/her about it, then I am not doing my job properly and it will likely be questioned by someone later. As long as it is properly documented and does not go against something specifically in the protocol, I personally don't care if a physician determines soothing is CS or NCS, however if I see no support for that or that support isn't clear I have not done my job to the best of my ability.
Lynda, I'm sorry, but an abnormal lab, or rather one that is ouside of the reference ranges, is not globally an AE. For one thing, a lab value is a diagnostic tool not a diagnosis. I couldn't count the number of data queries I've seen over the years on exactly that: "elevated (insert lab value of choice) is not a diagnosis. Please use the correct medical diagnosis." forget the fact that if every out of range lab value were listed, each study subject would have likely have pages and pages of of AEs when combined with the usual AEs such as the flu, URIs, UTIs, sprains, cut, bruises, etc., that
This is precisely why most central lab reports have places for physicians to note CS/NCS and comments. In almost 20 years, I have never once seen an NCS out of range lab value listed as an AE even if it was normal at baseline.
Helen Russo • Oops, meant do not ask...
Theresa Dunaway RN,BSN,MBA,CCRP • Helen you are right. The accuracy and clarity of the documentation does lean on the shoulders of the coordinator and monitor. It is still the PI's responsibility to determine the significance, the support for that rationale (whether CS or NCS and/or AE) is only the PI's.
Andras Koser MD, MBA,CPI, FHM • We have a lot of action around this question. I am still standing by my previous statement. It is PI 's responsibility to determine what is CS, NCS, what is AE and what is not unless the protocol specify.
Of course if the monitor or the sponsor, of the FDA asks about one or another lab result, the PI's better have a good explanation why did not report this one way or another. Also any decision that could be later questioned should be well documented.
To be able to do this the PI must know EVERYTHING about the subject, the study, and has to be a well trained physician. You have a problem when you have a ghost PI, or an ungifted physician trying to be a "scientist".
Would be great if the sponsors grade us, P Investigators by the quality of work we do.
David Radin • As a PI with 18 years in clinical trials I whole heartedly agree with Andras
Charles H Pierce MD, PhD, CPI • Good point Andras and I also agree. Let me point out that the ACRP/APPI have a certification process for Investigators that would answer you main point. The CPI will be increasingly important down the path and, presently, one major Pharmaceutical firm lets Certified PI's pass on the Investigator training sessions so many have. Certification is not a substitute for experience and common sense but at least one knows GCP and why the Protocol is what is followed.
From what I have seen, those who know agree that a lab result above or below the "Reference Range" is not an AE unless the PI so states and that statement is in the Protocol. Actually, if the Protocol stated that "All lab results above or below the 'Reference Range' is an AE", I would not agree or sign it until this was removed.
Theresa Dunaway RN,BSN,MBA,CCRP • I agree as well...This is the point I was trying to make. The PI is responsible for this determination of significance and/or if it is an AE. With that responsibility he must be able to document and support it in a way it is relevant to the study to sponsors, investigators, regulators and colleagues.
Helen Russo • Dr. Koser, sponsors do "grade" PIs and sites for the quality of their work, though they may not always share that information. You can also usually tell, if you have a good monitor who knows what he/she is doing, by what your monitor is telling you. Sadly, there are monitors and sites who are conducting clinical trials without a good understanding of basic research techniques and when you combine an inexperienced monitor and site, and I'm trying to be polite when I use the term "inexperienced" as I've worked with both experienced monitors and sites for whom that term can only be loosely applied, you get a poor end result with things like potentially significant AEs being missed.
It has always been my standing that a physician is responsible for determining and documenting the CS of any abnormality noted during a clinical trial, this is not, unfortunately, always the reality.
Standard source documents are partly to blame as they tend to discourage any kind of written discourse in favor of "fill in the blanks and check box" responses. I strongly encourage my sites to add chart note pages to standard sources, and that "more information written in a source/chartis better;" and not all of the information is needed for transcription to the data forms. It is a pleasure for me to work with physicians who actually KNOW the study subjects to which I'm referring when I have a question, but again, that is not always the case
Lynda Cedar, Ph.D. • What do you call this event: when a subject is injured at home, transported to the hospital, got medications and surgery, and his participation is interrupted consequently... ?
By definition, an adverse event (AE) is any adverse change in health or side effect that occurs in a person who participates in a clinical trial while the patient is receiving the treatment (study medication, application of the study device, etc.) or within a previously specified period of time after the treatment has been completed.
Helen Russo • In my experience, this example is both unintended and untoward and would be an AE, and most physicians I've worked with would agree. In a double-blind study there is no real way to know if the test article caused some dizziness, verrtigo, decrease of special senses, whatever that caused the accident which caused the injury.
That is a completely different issue from an out of range lab, 1) which may fluctuate "normally," and 2) may or may or may not be of CS by any definition, and I think most physicians would agree. I've seen labs at panic levels which turned out to be NCS when evaluated by a physician, however they are generally followed closely along with other diagnostic parameters reviewed by those physicians before coming to that determination.
The bottom line is that one cannot make a global statement that ALL out of range lab values are AEs and that is what Drs. Pearce and Koser and others have pointed out. An abnormal lab is only part of the diagnostic puzzle and what may be diagnostically significant for one subject may not be for another.
If it is a case of a particular lab ending up out of range, regardless of significance, across the board for the study, the statisticians will pick that up when looking at the global study data.
Helen Russo • Also, a change in a lab may or may not constitute a change in health...
Lynda Cedar, Ph.D. • Helen, I agree with you on the first paragraph, in interpreting the happening of the event as an AE, it was even classified as SAE by the FDA.
Regarding the change of lab values, no one is saying that all out of range values are AEs or constitute a change in health..... Regulations, standards and medical judgement apply.
Modern CROs are using the Adverse Event Reporting System (AERS which is in compliance with the international safety reporting guidance (ICH E2B) issued by ICH. Adverse events in AERS are coded to terms in the Medical Dictionary for Regulatory Activities terminology (MedDRA).
Assessment of Lab values is as follows:
- If the lab value changes but it is still fluctuating within the range of normal values, whether it decreases or increases, it is not reported as an AE. However when reaching the borderline values or exceeding them a bit, the PI attention is needed, he decides about the follow-up.
- If a value was within the normal range (or acceptable in the PI judgement) and its change is major at any time of the study life, the change is reported as an AE and its interpretation as CS (clinically significant) or NCS (not clinically significant) belongs to the PI’s judgement. The follow-up will continue until the value is found similar to its baseline.
Theresa Dunaway RN,BSN,MBA,CCRP • I agree with Lynda, it is still the PI's call...significance in lab changes, AE vs SAE. We have regulations and guidelines to construct a framework and guidance in addition to the protocol, after that it is all up to the PI.
Lynda Cedar, Ph.D. • Totally agree with you Theresa and others about the PI's freedom in interpreting the significance of an adverse event, that could be lab value, a symptom or any medical issue. In fact, the subject's safety is under his responsibility and the sponsor.
Regarding what is called AE and what is not, there are many sources explaining it and guiding clinical staff about reporting it.
SOMNATH MONDAL • As per my understanding of definition of ICH GCP, an abnormal lab value must be considered an AE whether it is clinically significant or NCS. As because, a meticulous approach is essential to achieve a bright out come.
Hence, this must be documented in scientific manner ..............
Charles H Pierce MD, PhD, CPI • Please, Everyone, Read the entries above by several experienced Investigators (and others like Lynda, Theresa, et all). Lab values are reported with an attached "Reference range" which implies by definition that one (the patient or the value) could be "normal" and be above or below the reference range limits. So, a lab level above or below is not (NOT) an Adverse Event (AE) unless the PI deems it to be an AE.
The PI would, of course, review all lab values and then make the decision as to whether or not the lab value is CS or NCS. If it is deemed CS, then it would most likely be an AE. Nowhere in the ICH GCP Guidelines or the FDA Regs does it state that lab values above or below the "Reference Ranges" is an AE.(Period)
Binay Thakur • I have gone through the comments posted by Dr. koser, Lynda, Theresa and
Helen and find it very informative and interesting. I agree with the
comments posted by Helen that although it is a sole responsibility of PI to
determine what is an AE and what is not, it is always the duty of the CRA
to cross verify the decisions of the PI (unless it is very obvious) and
discuss with the investigator about what prompted him to mark a particular
event as AE. The outcome may still be the same however the CRA doubts, the
discussion and the conclusion must be documented clearly in the source of
the subject and in the monitor's notes for future
references/audits/inspections.
I am not as experienced as most of you ( as have only 5.5 years of
experience) but quite a time PI have changed their decisions (turning an
event from AE to non-AE) after discussion with the CRA. It happens
sometimes when your investigators are not well trained on AE recording
requirements.
So, the decision to make an event AE always rests with PI however it is the
duty of the CRA to find out if any AE is created just out of
misunderstanding or is not in compliance with the protocol lines or if you
are following a certain guidelines such as CTCAE for toxicity grading and
specific Sponsor requirements.
Theresa Dunaway RN,BSN,MBA,CCRP • Billy, remember the question, 'If a lab value is out of range and a change from baseline, whether CS or NCS is it an AE'. The CRA does need to inquire following review of the records, so do the investigators and coordinators. If there are many inquires, the site may need retraining. Let's not forget a good site with a good coordinator should be the first line of inquiry and these questions and support should be documented clear enough that the CRA understands the rationale. The CRA should be the last line of review and inquiry, not the first, but then not all sites, coordinators and CRA's are equal are they....
Liljana Stevceva • The key to the answer is the protocol. If the event is AE according to the protocol toxicity tables, than it is an AE. Usually, the toxicity table would specify the grading of the event as well.
SOMNATH MONDAL • I do agree with Linjana, obviously protocol is the ultimatum. But if the same event recur subsequently for a number of subjects then the question of reconsideration about the protocol arise.
Whatever the event CS or NCS one need to document it for future consideration and discussion have a decision among the team members first, if discussion found rationale , PI must have to convey the same to the Sponsor...for their judgement....and final decision.
Helen Russo • I think we're, again, losing site of the original question. Most protocols, particularly in recent years, address the issue of hepatic and renal toxicities, and in some cases, issues pertaining to bleeding events, platelet aggregation, etc. These are almost universal these days.
As Dr. Pearce stated earlier, there is nothing in either FDA regs or ICH guidelines which state, categorically, all lab abnormalities (or changes from baseline, for that matter) are AEs. It would be ludicrous to say otherwise, not to mention labor intensive for both the sites and the monitors. We would be completely overwhelmed recording and monitoring events which are, in most cases, normal variances.
I depend on my physicians to review labs reports and make a clinical decision as to their significance and part of my job is to ensure they are doing this in a timely manner. The reality is that sites are busy, and physicians are busy therefore, things get missed or looked at in isolation (one visit's labs apart from a subject's entire chart).
If this didn't happen, I would be put of a job, and quite frankly, I love my work! I appreciate physicians like Drs. Pearce and Koser who know and understand not only the laws governing clinical research, but their responsibilities to the study subjects and the study itself. I have had many discussions with physicians regarding lab reports, and when we've disagreed about their clinical decision (rare though that may be) I defer to the medical monitor. MOST cases, my site physicians/PIs are able to justify their determination quite easily, and I merely request them to put it in writing...
Helen Russo • Sorry, not liking touch pad typing right now, meant out of a job...
Theresa Dunaway RN,BSN,MBA,CCRP • Helen I think you are just restating the point we are all making. The PI calls it depending on this medical and study knowledge. A research site should have most of this documented for you, although we are all human and have oversights (or the FDA auditors would be out of a job!). And I think your job is safe my friend, the sponsor still needs someone to review the data on site for them and find these PI rationales and oversights. Good job my friend!
Harold Doshan, Ph.D. • Just a follow-up to Steve Mcateer's comment: It is true that reference ranges for lab tests are established as 95% confidence limits on data for each clinical laboratory. These ranges are recalculated periodically, based on the most recent set (often tens of thousands of lab tests) of data available to that institution. However, please note that these ranges are NOT a reflection of a purely "normal, healthy" population. Lab tests are most frequently ordered by physicians and hospitals for patients with medical complaints or undergoing treatment. Many of these patients may have abnormal labs associated with the conditions for which the labs were ordered. Other labs, of course, are ordered as part of routine physical exams and prove to be well within the normal range.
Taken together, the reference range arrived at is a composite of all of these results and if the sample size is large enough and represents an adequate mix of healthy and variably ill patients, the range for each test should generally reflect, but is likely to be broader than, that for a purely healthy population. For large labs with many, many samples, the fit is good. Differences in reference rangess for smaller, local labs can be explained by their smaller sample sizes and the possibility of skewing due to a larger percentage of ill patients (as well as different responses for their instruments, test reagents, etc).
All the more reason to tread carefully when calling a shift in a lab finding (by the same lab) from within to just outside the reference range an AE. Many protocols (especial in oncology) stipulate the extent of the deviation from "normal" that must be considered an AE. Lacking these guidelines, the PI must decide, but he/she should also consider whether the out-of-range value is, in itself, an AE, or whether taken together with other signs and symptoms points to a specific conditions that is an AE, whether related to study treatment or not.
Dominique Chesnais • Lab values are well investigated in clinical trial analyses and reported in the clinical study reports. Refer to http://www.fda.gov/downloads/regulatoryinformation/guidances/ucm129456.pdf
In the FDA Structure and Content of Clinical Study Reports, lab data are reviewed in sections 12.4 and 14.3 about the following topics:
Clinical Laboratory Evaluation
Clinical Laboratory Evaluation of Individual Laboratory Measurements by Patient (Appendix 16.2.8) and Each Abnormal Laboratory Value (see section 14.3.4)
Evaluation of Each Laboratory Parameter
Laboratory Values Over Time
Individual Patient Changes
Individual Clinically Significant Abnormalities
Safety Data Summary figures and tables
Abnormal Laboratory Value Listing (each patient)
To report abnormal lab data, beyond CS or NCS, as AEs will be duplicating the information, unless the abnormal lab value is also part of a wider clinical pathology problem.
Babu R R • an abnormal value to be repeated to rule out the lab error. further if it reveals higher/lower than the normal range, if it is CS then it is AE. if it is NCS then it couldn't be an AE. correct me if am wrong
Lynda Cedar, Ph.D. • Hello Dominique. The comments above show 2 opinions regarding an AE statement, what is your definition of an AE: only when the change is found CS by the PI or at anytime when an out-of-range value is captured, or otherwise. Please explain.
In fact, the ICH integrated clinical report (or the link you posted above), the lab values are assessed as: it lists the range, the values at the baseline, at each study visit and at the end of study. Descriptive statistics are sometimes done.
The central lab provides an interactive file that is recognized by the computerized operating data base, programmed to capture the changes automatically. The programming is done by the IT, the biostatistician and the clinician to determine which change is determined as an AE. At this stage a change can be captured as an AE regardless to its relevance as CS or NCS in the medical judgement.
Thanks Dominique to share your practice with us. Lynda.
Dominique Chesnais • Hi Lynda, The first issue is that AEs are frequently under-reported by investigators during clinical trials. One needs to look at disparity between sites in reporting AEs, whatever their type. What benefits will it bring for an investigator to report an increase of one or two unrelated abnormal lab. values as AEs, as they are already noted in the lab. report as out-of-range and CS?
All study lab data are specified as CS or NCS after review by the investigator. Furthermore, they will be thoroughly analysed and reported, as required in the ICH E3guideline (Structure and Content of Clinical Study Reports). All out-of-range lab values will be highlighted in graphs, tables, listing and even discussion; same with the changes over time and values reported as CS. All that information strongly highlight some potentially pathological problems that Sponsor personnel should careful investigate.
However, when there are some increase of AST, ALT, GGT, alk. phosphatase, etc. the investigator should report a liver pathology; same about abnormal blood tests looking for blood disorders; too for the renal function markers. Drugs are reported as having liver, renal, blood toxicities, based on their clinical and biological signs.
In conclusion and in my opinion, it is these biological pathological problems that investigators should concentrate and not miss to report rather than some single, unrelated, out-of-range, CS laboratory values.
Lynda Cedar, Ph.D. • Hi Dominique. Thank you for adding your comments.
The liver enzymes and CPK are good examples to illustrate the discussion. They may respectively increase if a subject took alcohol or did exercise within 48-24 hours before blood testing. They are captured by the system of the central lab and the CRO computerized database as AEs. At this stage, the PI will look at them and decide about the outcome,.
I agree, lab values as per their name, medical diagnosis values, reflect the function of a system (liver, kidney, heart, etc.), a single value might be not relevant. Example, if Na+ is high, it does not mean blood pressure or kidney but if K+ if high the meaning is different.
When selecting a central lab, I ask them to provide the curve of calibration of the day, plus, to insert controls (small, medium and high) within the chain of tubes of analysis.
Thank you for sharing. Lynda.
Christoph Lohan • Based on the way the question is asked one near enough has to say that the "abnormal" lab result has to be classified as an AE. Rightly quoted by Mr. Dasgupta in its defenition.
Yet I would like to add that most often protocols are more diverse and more specific altogether if that makes sense. It is not only limited to AE as such. You will find other classifications such as "expected" adverse reaction and "unexpected" adverse reaction, of which both can be serious or not. Therefore a abnormal lab result could well be expected and therefore will not classed by companies as an AE per se. If it is tends to be unexpected, especially the SUSARs, then it will require flagging up to sponsor and regulator.
Thus I also agree with Ms McLeod "yes it's an AE unless stated otherwise in the protocol" :-) Happy discussing
Lynda Cedar, Ph.D. • Hi Christoph. To the best of my knowledge when an event is expected, it is reported as assuch but per the difinition of an AE that event is not an AE. If you can provide a specific example, it might illustrate the case.
Dominique Chesnais • >>>>>> SURVEY <<<<<<
I have initiated a basic survey (using LinkedIn Polls facilities with their limitations) based on a current discussion on reporting abnormal lab data as AE. I would be pleased if you could complete it at http://linkd.in/v0g4YI
Lynda Cedar, Ph.D. • Hi Dominique. It is a good idea of doing the survey. If I'm not mistaken, you asked the question differently.
The discussions have shown divergence: when an abnormal lab value that has changed comparing to the baseline is considered as an AE:
1) when captured by a computerized bata base as an AE,
2) only when found/interpreted as CS by a PI.
Sylvie JOUVE • In you protocol you also need to anticipate abnormal diseases' related values such as CRP when infection, AST/ALT (usually > x3 or 4 times is acceptables in some diseases) you also need to use age & gender related NR for at then having PI to apply his evaluation accordingly to these and medical practice
In the study report, you have to present lab data with predefined criteria for PCI or non PCI;
Declaring as a AE is as per PI judgement and usually we ask PI to declare rather than 'cancel" an AE.
So this is why a good knowledge of the disease and related abnormal vlaues can mak esome differences.
Hope this helps
Evy Moreno, BSHS, CCRP • I agree, this is a good topic to discuss because it happens all the time during clinical trials. So the combination of FDA regulations, the PI judgments and the input from the sponsor or protocol takes a role when deciding on A/Es
Nishat Hedayetullah • Going back to the original question (Should an abnormal lab value automatically be considered an AE if the value was normal at baseline? Obviously the investigator would classify as CS or NCS, but would it automatically be an AE) and seeing the recent exchanges, the clinical significance of the lab result should be considered in the context of the patient's overall health and in consideration of ongoing medical conditions and concomitant medications that are being taken. The treating physicain (not necessarily the PI) is best qualified to answer what the clinical significance is. Since the patient is on a clinical study, the treating physician/study investigator can discuss with the study sponsor whether the lab results have a causal relationship to the study drug.
Dipali joshi • It was really great discussion but some times we have to follow protocols and few protocols specify that "grade 2 or 3" abnormal values should be captures as AE , or some times baseline lab values will be considered.As I am handling Oncology trials where AEs are more frequent so capturing AEs differes from protocol to protocol and also based on the indication of trials.
Lori Jacobi • Again, the answer is yes, unless the protocol states otherwise. Generally speaking, anything that is a change from baseline is an AE regardless of it's clinical significance.
Lynda Cedar • Lori: indeed that it is, if located out of the normal range. The PI decides about its outcome.
Helen Russo • Lori, that's not correct. If every NCS lab were reported as an AE, sites, monitors, and data management would melt down. It is simply not feasible, nor does it meet the definition of an AE which is an "untoward and/or unexpected" event.
As many have mentioned, lab values WILL fluctuate outside of reference range values in a normal, healthy patient with no diagnostic significance. Others have pointed out, and many protocols include, specific ranges for analytes like LFTs, because drugs are metabolized in the liver and therefore can effect liver function. For example, historically, LFTs, depending on the specific analyte, are considered clinically significant if >2-2.5x the ULN, some sponsors alter this guideline up or down depending on the drug or use CTAE toxicity ranges.
The bottom line, is that I've never worked with a physician who 1) treats lab values in isolation, but as a diagnostic tool to determine a diagnosis, and 2) would agree that an NCS out of range lab would be an AE, especially absent of any clinical symptoms.
As for, global, study wide abnormalities or delta changes across a specific panel, that was statisticians are looking at. Once all the clinical data are gathered, changes like these are analyzed for statistical significance.
Prasad Vaishampayan • Great commnet Helen,,,an eyeopener !!
Andras Koser MD, MBA,CPI, FHM • 176 comments so far, read my comment somewhere at the beginning of this lengthy thread and you will find that that is is the right answer to the question. Many agreed so far. It is time to talk about something else. :)
Olusegun Adewusi • Adverse events after normal baseline records might signify the onset of traumatic episodes. So it should be a worthwhile 'in-vitro' investigation protocol afterall.
Lynda Cedar • Agree with Andras. It was a great question and great answers/discussions, great idea for surveys, now it's time to come up with another topic of discussions. Many thanks. L.C.
Charles H Pierce MD, PhD, CPI • Who would have believed that this one question would illicit such a wide range of responses. Some clearly wrong has been the largest surprise. I second my colleague Andras view that it is time to move on. Helen, above, and many other have it right that lab results are taken in context and / or according to the protocol and / or it is essential to remember that a patient or subject may be above or below the "Reference range" and be completely normal (by definition). 'nuff said. CHP
Andras Koser MD, MBA,CPI, FHM • If you have a great meal this thanksgiving but the turkey looks and taste different from the years before, or god forbid from your moms turkey it does not mean that it is bad. If you as a PI like it it is not an AE :) Happy Thanksgiving to all who participated.
MARIA KORKOVILI • We all already know what an AE is. Having at the same time laboratory experience I would say that an abnormal value should not be automatically regarded as an AE. Many times biochemical analyzers and haematology analyzers go through internal errors. Thus what I would recommend is run the test once more and then decide whether to report it as an AE or not. In addition, the nature and the sensitivity of the test should be taken into consideration.
Kind regards to all!
Anna Zimmermann • I think a laboratory value is something very specific. It is something real and possible to prove. Thus a change may indeed be automatically considered an AE.
Mickey O'Brien • Wow, that's a loaded question, for which there is not a black and white answer. Much depends upon the analyte, the amount of change from the baseline value, and the health status of the subject/patient. Effect upon related analytes would need to be assessed as well. Rule of thumb is to investigate ANY abnormal lab if the baseline lab was in normal range, but it is not necessarily an AE. Sometimes it is just lab error, but having worked for numerous labs, that can be an overused reason for not investigating further.
Paula Singleton • One thing I have learned in my 3 whole years in Data Management, is that if you assume you have the right answer, there is usually going to be someone, or some document, that can refute it, and refute it well. Or, if anything, bring into light new considerations you hadn't realized before-hand. This question, whether an abnormal lab value should be an AE, is completely and totally up to the Investigator/Physician. When I first answered this question, I looked at it from an outside perspective, as if it were a test question. If I were an investigator, what would I consider.
All things questionable, regarding labs and adverse events and so many other things, fall into the investigator's realm of responsibility. We have electronic edit checks that flag all out of range values, but not all of those out of range values show up on an AE log. I have learned to realize, in this industry, if you think you have the answer, there may be a chance you haven't looked broad enough into the question.
Cathryn Evans • THese things are generally quite clearly defined in the individual study Protocol and SAP, as well as GCRP guidelines. Not advisable to make an absolute statement without the context of the first two items.
Theresa Dunaway RN,BSN,MBA,CCRP • I agree with Cathryn, there is no absolute statement due to the variability in protocols and endpoints for each individual trial.
Nicholas McWilliams • Though I understand the desire/concern to not miss AEs, I think that people need to remember that the PIs are judging clinical significance for these labs. Just because out of range labs can be AEs doesn't mean all are required to be.
Nicholas McWilliams • Why have the PI assess significance then? The PIs have the ability to see the labs in a greater clinical context. If you require NCS labs to be reported as AE then you end up potentially creating a lot of noise in the safety data and end up risking creating a skewed and inaccurate label for your drug.
Nicholas McWilliams • If there are labs of interest then the protocol should state how they are addressed, including any threshold for abnormal values vs AE. Shift tables from baseline has been mentioned previously as a valuable tool in the context of detecting lab oriented signals, and I agree highly with that approach (sorry this had to get split into 3 comments)
Sateesh Kumar • It should be consider as AE once the Investigator indicates that it is clinically significant. All lab parameters should check by Investigator in each visit by judging clinically and also should compare with the baseline incase of significant increase or decrease in the values it should be captured as AE though it is not mentioned in the protocol. All these changes should be recorded in source as well as on the reports.It is a good practice to record abnormal values significance in source. It is an ideal procedure for Investigator who follows strictly ICH -GCP
Shweta Gupta • Here are my 2 cents.
It should be PI decision of CS or NCS. If it is CS than yes it is an AE. For e.g. For Hemoglobin the normal range for Adult males: 14-18 gm/dl. If the range on test report is 18.2 gm/dl and indication for IP is pain and PI thinks it is NCS it is not an AE.
Thanks!
JATIN SHARMA • As per my experience until or unless investigator has reported a abnormal value as Clinically significant it cannot be reported as AE this is the guidelines maximum sponsor follows and i am agree with the way its reported. To report any abnormal value as CS or NCS is investigators call.
Pharma JobLinks.com LLC • Yes, an AE is an AE. This should be included in the PI especially to track if this happens more often. Anything else seems fishy & like you're hiding something...
Ajay Reddy Jhampa • As per the E2A guidelines, AE is an
1. Any abnormality ( Untoward medical occurance)
2. After the first dose of Pharmaceutical administration ( After the first dose of study
drug)
3. It may or may not have casual relatiomship (Relatedness).
Therefore any abnormality that occurs after the first dose should be considered as an AE whether it is clinically significant or not. It is not an Investigators discretion.
FDA accepts data in SDTM standards and we submitted all AE's in AE dataset. If we get the information of clinically significance from the CRF, we submit this information of clinically significanvr in SUPPAE( Supplimentary Qualifier dataset).
Charles H Pierce MD, PhD, CPI • Please, folks, go back and read the many comments up to 4 months ago when this was started. It should be crystal clear by now that whether or not to call a lab result that is above or below the reference range limits an AE is a PI decision. A lab result above or below the reference range is not (NOT) an Adverse Event unless the PI (after review) states that it is an AE. To do or think otherwise is incorrect.
Agree with the ICH E2A definition but notice that it does not say that a lab result above or below the reference range is an AE by definition. I am sure you know that by definition one could be entirely "Normal" and be above or below the reference range limits. See earlier comments making this point that the terms "Normal limits" or "Normal values" has no (NO) place in clinical research.
I will be giving Webinar on this subject in January. Will let this group know of the date when set.
Sunee Thiravanitkul • I agree with Dr. Charles H Pierce that it depends on the clinical stage of the patient. Some out of normal lab value is just very much normal for some particular patients due to the underlining disease, e.g. low hemoglubin value in anemia patients.
Cathryn Evans • Thank you, Charles. To add to this comment, since all abnormal lab values must by regulation be tabulated and summarized in the CSR under the Safety Section, these abnormalities never go unreported. The clinical significance of the abnormality is what leads the investigator to include it as an AE — thus sometimes abnormal lab values are designated as "AEs"; other times they are retained in the Safety/Laboratory Test section of the report.
Helen Russo • @Dr. Pearce: This does seem to be the discussion that never ends!
@Cathryn: You pretty much said what I and a number of lab people have said: seemingly NCS trends get caught in the statistics.
Lynda Cedar • This question generated 199 comments, I wanted to bring it to 200.
In fact the statistics use the data captured by the computerised database which is programmed to capture any change in comparison to the baseline (called AE). After the medical interpretation some AEs are classified as NCS and others as CS associated with a number from 1 to 5 as selected by the PI.
In capturing all the changes versus the baseline is the only way to ensure that all the AEs were explored and interpreted by the PI .
It was commented 200 times as follows:
pallavi Borse • It is depend on the medication ,IF it is realated to the IP product then it is definately AE
Dr. A. D. Paul • Strictly speaking, an abnormal laboratory finding is to be regarded as an adverse event (AE), irrespective of its clinical significance. Whether such an abnormal lab. value is clinically significant or not, is its interpretation. If the abnormal lab. value is due to an investigational product, it constitutes the causality of the AE.
Sarah McLeod • The answer is yes unless otherwise specified by the protocol.
Biswajeet Dasgupta • ICH E6 ------ 1.2 Adverse Event (AE)
Any untoward medical occurrence in a patient or clinical investigation subject
administered a pharmaceutical product and which does not necessarily have a causal
relationship with this treatment. An adverse event (AE) can therefore be any
unfavourable and unintended sign (including an abnormal laboratory finding),
symptom, or disease temporally associated with the use of a medicinal
(investigational) product, whether or not related to the medicinal (investigational)
product.
Valera Bussell, CCRC, CCRA • Thank you to all of you for your input. As you may have noted; there are varied opinions regarding the issue. Years ago, when I was a coordinator; I was taught that any change in lab values from normal at baseline would be considered an AE, but apparently many companies now view that differently and only consider Clinically Significant abnormal lab values to be an AE; the clinical significance being decided by the investigator. I would like to see the FDA's opinion on the issue. Thanks again for commenting.
Lakshmi Guduri • Completely agree with Valera. Even, I was taught that an abnormal laboratory finding is to be regarded as an adverse event (AE), irrespective of its clinical significance.
Dr.Avani Badani • I do not think an abnormal lab value should always be reported as an AE. There are scenarios where even 0.1 increase or decrease of parameter happens,and that isnt a AE by anycase. It should always be reported as an AE depending on PI discretion or Sponsor's reporting requirements.
Shahid Ashfaque • The official and most accepted definitions in most countries are based on ICH E2A guidelines and are as follows:
Adverse Event defined by ICH :
Any untoward occurrence in a patient or clinical investigation subject administered a pharmaceutical product and and which does not necessarily have to have a causal relationship with this treatment. Any unfavorable and unintended sign( including an abnormal laboratory finding, that is sign or symptom, or disease temporally associated with the use of any dose of a medicinal product, wether or not considered related to the medicinal product.
Having stated the above it is an Investigator's perogative as what should be reported as an adverse event (it is his/her responsibility to assess the safety of the subject). You cannot ask the investigator wether to report any lab value as an AE or not to report it. You query or advise but ultimately it is investigator's judgement based on his expertise and medical judgement
The GCP states as follows for the investigator:
4.11.2 Adverse events and/or laboratory abnormalities identified in the protocol as critical to safety evaluations should be reported to the sponsor according to the reporting requirements and within the time periods specified by the sponsor in the protocol.
Dan Rossignol • Small deltas from within normal limits to just outside the normal range are meaningless to patients and should not be used to define AEs (in fact, lab values can move from normal to abnormal just by being converted to different units and rounded!). Sponsors should devise standards that define a minimum delta that determines when a change becomes an AE, either as a discrete value change or a percentage change. These can be used in combination with claims of clinical significance by the site investigator.
In addition, when not dealing with potentially-overtly toxic drugs (eg cytotoxic chemo therapies) we had a derived list of treatment-emergent abnormal values (TEAVs) that were based on amount of change from baseline. By this analysis a change from high but normal to just outside of normal range might not be a TEAV as the percent change was small, however a dramatic change within a normal range would be flagged as a TEAV. This (as well as shift tables) was helpful to look at possible lab changes in populations that would go otherwise unnoticed as even though there could be dramatic changes, they would not be defined as adverse events.
Shahid Ashfaque • I totally agree with Dan Rossignol
Valera Bussell, CCRC, CCRA • And I totally agree with both Dan and Shahid. Seems like there could be a standard to go by, rather than relying on the investigator to decide whether to report as an AE or not.
Marie Gabrielle Laguna MD • I agree: it is investigator's judgement based on his or her expertise and medical judgement.
Shilpa Garg Agrawal • It is the investigator's judgement which would finally decide if it is AE or not. Also, it would depend on the protocol. For eg. if there is a change in the lab value, which shows exacerbation of the underlying disease, if the protocol states that exacerbation of underlying disease is not to be reported as AE, then, this will not be an AE. However, if the protocol states that exacerbation is to be reported as AE, then, it may have to be reported as AE. Also, although investigator's judgement is final, but the CRA should be vigilant to identify any such abnormal lab values which the investigator may state as NCS, If the CRA is concerned about it, the best thing is to discuss it out with investigator, and if still not convinced, the CRA should discuss the matter with medical monitor, and take it forward..
Dan Rossignol • I agree, and this all makes sense for the use of local labs, especially
when a lab finding leads to an intervention. For an international study in
very sick subjects, our group actually defined an AE as a finding that
necessitated medical intervention. That reduced the number of minor
observations including labs and physical exam finds that counted as AEs.
But another point on lab findings is what to do with central lab results.
Variability due to judgment can be reduced, by use of a good central lab,
especially when a large study is done on a very sick population. In this
case, the site investigator's opinion is not included in the results and
the use of guidelines to interpret the data (changes in values) can be very
important.
Marie Gabrielle Laguna MD • I like these type of discussions rather than job posts.. more....more.....more please! Let us exercise our brains!
Sara Lampinen • I agree with Marie, this discussion it great! This issue comes up so frequently, it's nice to see all the different opinions.
Dan Rossignol • LOL.. It seems that this issue comes up with every study! Where would there be a "master guidance" (besides the regs of course) on things like this so we do not have to reinvent the wheel each time?
These discussions are good, but for some us (eg. the currently in transition types), the job postings may be of use!
But I have another question on how AEs can be standardized for severity in non- cancer trials. maybe start another string.
Konstantinos Kakos MD • Dan Rossignol-
A central lab will always be a better solution than local labs for a multicenter trial,especially an international one. Collecting data will be much easier and problems in the trial set up such as different test methods, different validated methods, different reference ranges etc. are eliminated but the central lab would never replace the Investigator in his role of the judging if a lab finding is CS or not.
Steve Mcateer • I agree more or less with Dan, but would emphasise that the biggest issue with using the reference range as a tool is that there is a danger of just scanning the labs looking for flags, wheras the biggest problems can be found when results are still well within the reference range. As a lab professional, we tend to pretty well ignore reference ranges and concentrate on delta change within an individual. Any decent computer system can track these changes and flag accordingly (ask your lab to do it!). Just on a technical point, reference ranges are calculated from 95% confidence limits of a 'normal' population, so 5% of results MUST be outside of the quoted range if the range is valid - thats an awful lot of AE's to be notified!
Finally, re the role of the investigator - their responsibility is for their own patient, and they are well judged to comment on clinical significance, but unless they have full access to all other patient results, they cannot see trends in other centres, which coupled with their own NCS, might tell a whole different story - the CRA is usually the only person to see this (apart from the lab of course). My overall message is talk to the lab, get them involved in setting criteria at the protocol stage, but that mantra's an old one!
Dan Rossignol • I agree. And the turn-around time for a central lab takes it out of the
"timely information" loop as well, but central lab results are great for
analyzing final data in a clean, consistent way though.
Ana Isabel Alvarez Retuerto • I believe that while the ultimate decision is the investigator's, this is at the end a team effort to get the possibly most accurate data while keeping the patient/participant's safety as a priority and keeping compliance with protocols, so if something out of the ordinary comes up it should certainly be brought up by the team members, talked and discussed by the relevant investigator/s and depending on the study/protocol, the "unusual finding" and having the patient/participant's safety at the top priority, make a decision that stays contained within the protocol.
Shahid Ashfaque • Finally, re the role of the investigator - their responsibility is for their own patient, and they are well judged to comment on clinical significance, but unless they have full access to all other patient results, they cannot see trends in other centres, which coupled with their own NCS, might tell a whole different story - the CRA is usually the only person to see this (apart from the lab of course). My overall message is talk to the lab, get them involved in setting criteria at the protocol stage, but that mantra's an old one!
Steve,My comments on above paragraph:
I see your point but we are talking two different things. We should not confuse with the responsibility of Investigator with the responsibility of Sponsor.Although both are responsible for the safety of the patients Investigator is responsible at the level of his site and patients, while Sponsor is responsible for the aggregate analysis, while the study is ongoing it is blinded, and the safety plays the dtective through signal detection. All DIL letters during the trial to investigator is the method to inform EC and Investigator of any serious adverse events occurrence. The non serious adverse event trending can be better done by sponsor since they have the panoramic review. At the end the investgator brochure is the reference they can wisely use.
We are drifting from our main question what is an adverse event? is any abnormal lab value is adverse event. The answer is NO. Use your judgement by utilizing the wholistic approach to the compound under study and considering other variables of therapeutic are of study, type of patient population, and what is the end point
Patricia Hollis, MBA • it really depends on the protocol and who you work for.
Patricia Hollis, MBA • I would also like to add that many people depend on the job posts so please do not stop them.
Essack Mitha • I agree with Steve, that changes in lab values are as important as once off out of range values.
Thanks to those that quoted ICH GCP. I regard only CS abnormal values as AE's. The reason is simply that there are normal fluctuations in lab values that sometimes may go into abnormal range. HOwever, these could be due to variety of factors, including time of day. Once i am of the opinion that a lab value is CS, i report it as an AE.
I have sometimes encountered changes in lab value over a few visits, where the value is close to lower end of normal, and gradually elevates till it is close to upper end normal. Although these are still within normal range, i consider these as AE's as well, as the change in lab value could be significant, while still being within normal range.
S K. • Strictly speaking, Its automatically an AE
manorama Mpanwar9 • I agree with Essack Mitha. If the change in lab value is CS then it should be an AE but if it is NCS then it is not necessary to take it as an AE until it is the requirement of the protocol. We can take a simple example - if lab test of a same person is performed at 2 different labs on same day we can easily see the variations in the results. As, the variation can occur so there is no need of taking all changes as AE except for those values that are CS.
S K. • I do agree with Manorama, Consider this example: Suppose PI denotes the lab value as NCS in all visit 1, 2 3, & 4 for same subject. Finally in visit 5 too he got the value as NCS then PI may decide to consider it as CS and will report it as AE.
Konstantinos Kakos MD • I think that there's a greater confusion on the issue. Clinical trials is
not just following guidelines, there's logic involved. An Investigator as a
physician will take a number of factors into consideration before reporting
an abnormal lab value as an AE. Simple fluctuations in values from baseline
do not necessarily mean AE. Of course if the protocol strictly requires
reporting as AEs every abnormal value no matter what the Investigator will
have to do it.
Remember that lab tests are a tool allowing the physician to get a general
picture on the patient's condition, they shouldn't be looked at as simple
numbers of a scale.
Brad Brush • An AE must be reported, irrespective of the Investigator's opinion of causality or individual opinion of clinical significance. ICH guidelines aside, what if we all did not report an increase in a laboratory parameter (out of the reference range) and this occurred in a large number of subjects? We need this data!
Valera Bussell, CCRC, CCRA • Hi Brad,
As I am sure you have seen within the numerous comments; not everyone agrees that all abnormal labs should be reported as AEs. Something that I find interesting since the FDA guidelines includes abnormal lab values within their definition of an AE. It would be so nice if we could standardize some of these issues across the industry!
Dr. Darpan Gangil • It should be an AE.
Shalini Dayananda • I agree with Steve regarding the reference ranges being calculated from 95% confidence limits of a 'normal' population, so 5% of results MUST be outside of the quoted range if the range is valid -and the PI land up reporting a number of AEs which have actually not occurred. However, Protocols usually define the how much deviation from the reference range - e.g twice/ thrice of UNL to be reported as lab AE. Accordingly, the AE should be reported.
Coming to Investigator's prerogative to decide upon CS or NCS - in either case AE should be reported with an additional info stating NCS as per the Investigator. This will definitely ensure that the data does not go unreported.And, this will be irrespective of whether it is related or not to the IP.
Cristina Gonzalez • I would say it depends on the situation. If the lab value is a symptom of an underlying disease, the investigator will not consider it as an adverse event unless it indicates something different as expected. For instance, for liver metastases is expected the ALT and AST values will be higher than normal, however are not an AE themsleves if the liver metastases are confirmed and no other cause is found.
Otherwise we could fall in overeporting Adverse Events.
nithiyarajan nambirajan • hi, i am a DENTIST, now pursuing clinical research course, by seeing all this comment i got good information regarding AE, so thanks for all ur comments.....
Hamdi Akan • İf you are doing a hematology/oncology trial, this issue can a bigggg problem. The best way to overcome this, is specifying limits for expected abnormal lab. results related to the underlying disease and leave it to the investigator but by definition all of these are AE's
Valera Bussell, CCRC, CCRA • Again, thank you for the lively debate and all of the input; it should prove quite helpful; especially for those just getting into clinical research. Our SOP is to report all abnormal labs after baseline (assuming they were normal at baseline) as AEs, unless the issue is clearly addressed in the protocol. I.E. if some changes are to be anticipated based on the indication being studied and the protocol clearly states that these changes will not be considered AEs or will be assessed by the investigator, then they will not be reported as such. I would encourage all sites to be written SOPs in place at their site. Please let me know if you don't have written SOPs, but would like assistance in putting them in place. Perhaps if we could all address more of these "gray areas" in the forums; we could finally start moving toward real standardized processes within clinical research. Thanks again! Valera
Shilpa Garg Agrawal • Thanks Valera, I really liked the idea of having a written SOP for this specific issue.
Adrian Johnson • Unless it is specified otherwise in the protocol, an abnormal lab value outside baseline regardless of relationship to IP is considered an AE.
Valera Bussell, CCRC, CCRA • HI Adrian,
That was my opinion as well; however it appears that some companies (CROs) may have a different opinion on that and think that is should not be captured as an AE if the abnormal value is slight and not clinically significant in the opinion of the PI. I personally think that leads to confusion, especially on the part of the study coordinator who may be assisting in reporting AEs.
Dana Austin • Also, its dependent upon other variables, such as was it an "expected or unexpected" event? Those will help determine if it should be captured as an AE until resolution or not.
Dr. Datta Pawar • Anything new finding during study should automatically be reported as AE. Over-reporting is always beneficial than under-reporting. Though it may not be related to Investigational product.
Oussama OUESLATI • if the abnormal result is observed for the first time after medication this result must be registered as AE, In any way, the CRA have not to say the significance degree: it is the physician role!
The CRA have to notify this observation as AE.
Pat Sabatini • Adrian - I agree and it should never be left to the Investigator.
Valera Bussell, CCRC, CCRA • I would agree and as previously stated, it is our SOP that abnormal labs that were normal at baseline should be reported as an AE, unless it is explicitly addressed within the protocol. I lean toward the over reporting rather than take a chance on not reporting an AE that should have been reported. It is unfortunate that some of the CROs have decided to insist that the PI decide whether to report as an AE or not to report. I would prefer a directive from the FDA before we change our SOP. Thanks to all for your opinions.
Shamshad Ali • I totally agree that based on safety aspects any abnormal value should be reported as an AE
Essack Mitha • I can see that there will not be a strict guideline on what should be reported as an AE, as Investigators have their own opinions. However, the medical monitor and the sponsor get results of every subject on the study, so even if an abnormal result is not reported as an AE, the capturing of the lab data will show changes in lab values. It is this that is also taken into account when looking at the effects of IP, not only what the investigators report.
Dan Rossignol • I agree with Essack Mitha as eventually all changes, even if they occur
within normal ranges (eg platelet counts that can be high normal at
baseline and low normal post treatment) need to be evaluated (and compared
to placebo where possible) to see what is going on in the treated
population. While we talk about AEs or lack thereof, these are simply
"cut-offs" and should be understood to be "clinically significant" or "non
clinically significant" but in the end, there is usually more info across
all subjects over time (with post-treatment recoveries) that are evaluated
in the clinical study report. Signals seen at low dose IP may indicate
potential targets for close observation as you move to higher doses,
extended treatment, or treatment of more impaired patients.
Emmy Merkley • I think I have always followed the following criteria,. If the investigator deems it is clinically significant after repeat testing of an abnormal value to confirm the 1st abnormal value and it may require some kind of treatment, follow-up testing or obeservation, then it does become an adverse event.
Dan Rossignol • That seems like an ideal definition. Simple and clear. The requirement for
an intervention makes a case for significance as well.
YJ Bi • This is a great discussion! Just want to add my humble opinion:
During the collecting of data (investigational site/investigator and site monitor), in general, an out of range lab value should always be captured as an AE on the CRF. Once it reached datamanagement, usually involve medical monitoring and data review/coding/cleaning which is also done under FDA regulations/SOPs/study data specifications, Such AE can be kicked back via data query to remove or just confirm the AE on case by case basis (e.g. lower WBC while the patient had a flu/other viral infections, transient transaminase increase after a dozen beers previous night at a birthday party, etc).
As all AEs, the clinical significance and severity is rated by the investigator (while serious is by FDA definition). It not uncommen that an AE of out of range lab value, to be queried.
However, before CRFs is pulled out, an out of range lab value should be reported as an AE. But this can still generate an query. Such kind of query is a part of the datamanagement process, even site/investigator and monitor are doing the right thing to capture such AE.
Thus, out of range lab value is to be reported as an AE unless study specific documentation indicated otherwise on the CRF. It may generat a data query, then it may be handled differently depending on the data convention and clinical scenario.
Carlos Estrada, MD • I agree with Biswajeet Dasgupta, It is considered an AE, as an Investigator, I am able to suggest if it is related or not to the investigational product, but it is definitely an AE
Sebastian Antonelli • This discussion seems closed by now, but anyway, my 2 cents: in general, report an abnormal lab as AE when it triggers a medical intervention (thats why it is called "medically significant"), and such medical intervention could range from a medical advice (i.e. "you should really reduce the number of MacCombos per week"), a drug prescription, or med change, or dose adjustment or discontinuing of an ongoing med, or additional testings that confirm a diagnosis. If there are symptoms accompanying the abnormal lab, please don't duplicate entries, AEs should be diagnosis if possible. And every AE should have an action taken.
Sundarkia Hill • As a former CRA, I always required changes from baseline to be reported as adverse events. Even if it is exacerbation of a pre-existing medical condition, the exacerbation itself is an adverse event. Whether the reference range is set by the lab or by the investigator's judgment, any lab result deemed abnormal is an adverse event, regardless of clinical significance. Determining reporting relevance at the site level can lead to obscuring trends across the study which could give further insight into how the drug works. In my opinion, it is the duty of data management and the data analysis committee to adjudicate these events once they review data from all sites, from all countries, for the entire study. There will always be room for debate between sites as to what is relevant, even when a master guidance is given. Therefore, I believe the only necessary master guidance is to report all abnormalities and let those individuals privy to all study data determine relevance.
Carlos Estrada, MD • I agree 100% with Sundarkia, as a PI and Medical Monitor, the abnormal changes in labs need to be reported as AE's, and it is the discretion of the PI and/or Medical monitor ( or Data Safety Monitoring Board) to adjudicate the AE to a particular Drug/procedure/device or not.
Jacqueline Gough • Fascinating discussion that truly does seem to occur on a regular basis.As others have stated, the reference ranges are only useful for understanding where the patient's result lands with respect to the "normal" distribution of results. This is usually the reason that the investigator's opinion of clinical significance has been included in the determination of whether an abnormal lab value is considered an adverse event. What might be abnormal according to the reference ranges may be completely normal for a particular patient and only the investigator, with their more complete experience with that patient, can assess this. This is completely distinct from relatedness.Additionally, although it is conceptually more conservative to include all abnormal lab values as adverse events, there is a risk that any safety signal from that information could get lost in the noise of so many adverse events that are not clinically significant.Instead, perhaps adverse events should be reserved for what the investigator assesses as clinically significant and the actual lab values can be analysed for trends or items that the investigator has failed to notice.The purpose of collecting this data is to identify any safety risks associated with whatever intervention being studied. Patient management should, in my opinion, be left to the individual investigator.
Sundarkia Hill • While I agree with the group's interpretation of a reference range, I must disagree with the interpretation of clinical significance and its role in clinical studies. Clinical significance has no bearing on determining if an abnormal lab is an adverse event. The ICH definition of an AE shows an abnormal lab in itself constitutes an AE. If you are consistently not documenting abnormal labs as AEs and you are audited by a CRO, FDA, or IRB you will likely be dinged for not properly reporting AEs.
Clinical significance is followed for the purpose of monitoring the status of the event with the subject. For example, if I am a CRA monitoring a subject’s chart and found a lab value was 4 times the upper limit of normal, I expect to see active intervention by the investigator to bring this value back to normal (change in meds, retesting, etc). I would also make it an action item in my report to follow up on this abnormality to see if the intervention had an effect, resolving the AE.
Regarding AEs in general, sometimes the side effects of drugs are hidden in the small changes and I think it would do a disservice to the industry to ignore them. Although reporting these events can lead to queries, this is actually a good thing. The queries I often saw as a monitor would say something along the lines of “elevated ABC plus decreased DEF codes to condition XYZ. Please confirm if the subject has been evaluated for XYZ.” This could be from a small increase over baseline, just barely putting the result out of range. Something that would likely be marked as not clinically significant when looked at separately can become significant when looked at collectively for a subject and accross subjects at different sites. Such queries come from AE reports, not from the lab. This also shows the level of analysis of these events and can sometimes make the site staff dig deeper into something that may have been accidentally overlooked, simply because the changes were not significant in the opinion of the investigator. We should not rely soley on investigator opinion because every investigatior's opinion is subjective, no matter how experienced a physician one is. My advice is to report all abnormal labs and allow the data monitoring committee to adjudicate.
Dan Rossignol • Beautiful. Put writing of this sort in the protocol and you should have
everythingyou need (between this and the analysis of lab changes) And I
like your sense of humor. But I would consult regulatory to see if you have
to list the BK and Wendy combos as well in the example.
Valera Bussell, CCRC, CCRA • Thank you again for all of your opinions; I am especially happy to see that many of you agree with me that a change in lab value from normal at baseline to abnormal after starting the study drug should be considered an AE based on the FDA's definition of an AE and should not be based on the evaluation of how far outside normal range it is in the opinion of the investigator. I will continue to direct my research staff to capture these change as AEs, with the investigator merely evaluating the seriousness of the abnormality (clinically significant or not) etc.
Gulben S. • If the investigator classify it as CS then it would be an AE. Except 'special cases' if it is NCS it would not be an AE.
Carlos Estrada, MD • Again, as the question said, if the laboratory value is ABNORMAL, it should be reported as an AE, regardless of causality, if the laboratory value is DIFFERENT from baseline, it needs to be assessed first as CS or NCS, before being reported as an AE.
During an FDA audit this is something the inspectors will look at and you might get a 483 for not reporting as an AE.
It is the responsibility of the PI, Medical Monitor and or the DSMB to adjudicate the AE to a Drug/device/procedure.
Erika Tokaji • To ensure consistent reporting leading to meaningful safety profile and help both the sites and CRAs with SDV, the protocol should provide guidance what is to be reported as an AE. One commonly applied approach is to specify CTC severity grades as from which the result should be reported as an AE (usually grade 3, 4).
Artie Jhappan • My experience working in Phase 1 trials - only abnormal lab results that are deemed CS by investigator were reported as an AE.
Artie Jhappan • Often the regulations are broad and non specific.I think this is something that needs to be defined clearly in the protocol.
Artie Jhappan • Joyce -Excacerbation of an existing condition should be considered an AE.
Cynthia J. Robinson • If the lab value is significantly high from base line then it is the judgment of the investigator to determine if the value is an AE or not.
Pat Sabatini • If it were to either be written in the protocol or left up to the investigator AEs would be all over the map.
Valera Bussell, CCRC, CCRA • Pat,
I agree, but there are some CROs that don't want all abnormal labs captured as AEs and have specified that it should be assessed by the PI and then a decision made whether to capture it as an AE or not. Historically, all of the sites that I have worked with have reported abnormal values as an AE, even if not deemed CS. I see it the same as you apparently; where do we draw the line if we leave it up to a PI? Can they consider other issues as not worthy of capturing as an AE as well if that is the case etc.?
Abhijit Sharma • Abnormal lab value + Clinical significance = AE
Abnormal lab value - Clinical significance = Not an AE
Clinical significance= standardiztion in Protocol/PI judgement
In case of CNS abnormal lab value ratinale can be documented.
Pat Sabatini • Hi Valera - I'm not suggesting that all abnormal labs be captured as AEs but that CS must. The definition of CS has always been if it requries intervention. The same goes for the severity of the AE, this is not defined by the PI, there is a clear definition of this and every PI must be given the same yard stick by which to measure this. It is up to the monitor to verify that they are all using the same yard stick.
Meenakshi Manerikar • yes it is AE
Karolien Timmermans • Hi,If you look strictly to the definitions in ISO or GCP, an abnormal lab value should be documented as an AE. However, as already pointed out before, you also have to look at the clinical significance of the abnormal value.
Also, if no treatment is initiated to correct the lab value, i would not consider it to be an AE.
Nevertheless, I would obtain confirmation from the PI in the center and make a note in the monioring report.
Cynthia J. Robinson • Shilpa Garg, I agree it is the principal investigator’s (PI) judgment that makes the final Adverse Event (AE) decision. However, it also depends on the protocol and whether this is an exacerbation from a chronic illness. Prior to final decision, PI should repeat labs to confirm significantly high value. If CRA is still concern whether this entails an AE then CRA should collaborate with the Medical Monitor.
Muhammad Ather Latif • Abnormal value which has / could have relation with administration of investigational product and was not recorded in baseline or history of the patient/subject will be considered and regarded as AE. Its significance may be analyzed by PI later.
Priya Reshma • its very good kind of discussion....getting more information and also helping us to update our knowledge..........thanks to all
Jhappan • a stimulating discussion.... I like it
John R. Waldron, CCRP • This is definitely a case-by-case situation with no general rule or regulation to follow. Lab abnormalities that constitute worsening of the disease under study would not be an AE. For example, elevated liver enzymes in patients with known progressive liver metastasis. In addition, lab values slightly above or below ref range could be deemed normal for that patient, interpreted as NCS and not an AE.
On the contrary, lab abnormalities that constitute worsening of pre-existing conditions found in medical history would be an AE. All lab abnormalities that are study drug related should be interpreted as CS and listed as AE.
Samuel W Boellner, MD • depending on the study I would recommend a repeat
Luigi Nardacchione • We cannot consider the sponsors' or investigators' opinion (or interpretations) but just the official guidelines. So no doubt about it as per ICH E6 and related FDA/EMA requirements: it is sponsor and investigator responsibility to strictly follow them in order to protect patients and also not to jeopardize study results.
Helen Russo • While everyone continues to quote the untoward regulation, I think the boat has been missed...not all abnormal labs constitute an untoward event, particularly if the PI considers them NCS. Also, an abnormal lab value, in and of itself, would not be the AE, but rather the underlying condition, even if it's the probable/possible effect of the IP on liver fx, e.g. In other words, "elevated ALT" is not the AE.
As a monitor reviewing signed labs, I always tell my PIs that I reserve the right to ask them "why" they feel an abnormal lab is NCS if I disagree. It may be that they have reviewed that particular report away from others and not noticed a trend that I have, or they may have just missed it...it happens, but the bottom line is that it's the PI's responsibility to make these decisions, absent specific language in the protocol, and it's the monitor's responsibility to ensure the PI is doing so.
There are also instances where I was taught that an abnormality in particular assay is always CS and must be repeated to ensure it was not a lab error: minute elevations in serum calcium, e.g., a 0.1 mg/dL elevation is likely a lab error, but must be repeated to ensure that fact as even that small an elevation can indicate pituitary tumors; should the value return to WNL, there would be no AE. I had a PI overlook an elevation like this until the subject started presenting other symptoms and he ended up with a very serious problem on his hands!
Bottom line, if all lab abnormalities were automatically listed as AEs we'd be so completely overwhelmed we never be able to do our jobs.
Dianne Edwards • I agree with Joyce. If not part of the patient's MH; it is an AE unless otherwise specified in the protocol for the intented investigation under study and the change is expected and documented in the IB or PI.
However, that said, if the change in the value is clinically significant (outside of patient's "normal" normal value); increases in severity, frequency or duration - your must document this as an AE according to ICH E6 and GCP guidelines,
Dan Rossignol • Helen,
Excellent points here. It seems that the reporting of AEs as well as
careful analysis of trends in groups based on hard evaluations of changes
in lab values (not just changes that move from normal to abnormal) need to
be part of a final study evaluation. It will confuse things to have minor
changes reported as AEs. As has been stated, the protocol needs to be
clear on this.
Despite any protocol discussion however, the PI does have final say and
should be able to explain the AE claim. One problem that I have seen
however, is that investigators may be overly conservative in claiming Aes
in fear of an agency audit that picks up abnormal values and asks why they
were not claimed as AEs. This again is where the protocol language
(presumably approved by the agency) as well as careful monitoring protects
them.
Dr. Ashish Indani • I agree to the consideration that altered laboratory profile should be considered as Adverese event. These findings are, in fact, more important than clinical findings in many cases, especially when investigator is specialist of one perticular decipline. this is because, these investigations are performed for this very reason. Let's review a few facts
1) The investigator is always a medical professional and just like routine, the investigations may have been be left to his discreetion, if the designer would not want to note the sub-clinical (NCS) changes
2) The basis of calling for a perticular set of investigation by a Study designer is based upon the preclinical, previous phase clinical and Theory information of investigational product and the substance of disease, whatever is available.
3) the adverse event reported for a perticular drug must not always be bad. Infact we must keep example of Sildenaphil, Aspirin in consideration, that these drugs are highly sucessful in more than one ailments in different dosage. This whole phenomenon was based upon the evaluation of adverse events reported (Clinical in Sildenaphil and laboratory findings in Aspirin)
Considering the fact that in a clinical trial, we design the study with certain statistical qualifications as posted by Steve, hence frequency of such reporting will be automatically taken as significant or non-significant at the time of evaluation. There is also a reference data from previous phases and control of the study available to estimate and evaluate whether the changes in investigations are pertaining to the investigational product. For this analysis to go impartial and unbiased, Reporting of any any every change in the subject's overall status, including physician's assessment, any scores which are designed, any laboratory investigations etc. must be reported in time.
Michelle Perkins • Site for dtp jobs, information about graphic and computer jobs. http://lnkd.in/7hH7e3http://lnkd.in/Zfqbhj
Ramune Kanapieniene • I suggest reading the Journal of Best Clinical Practices "Documenting Clinically Significant Lab Values" by S. Eric Ceh http://firstclinical.com/journal/2009/0901_CS_Documentation.pdf
Vindu Seal • All abnormal lab values need not be documented as AEs. They have to be classified as CS or NCS and all NCS findings need to be documented as an AE. However if the abnormal lab value is falling within the defined alert parameter in the protocol then it has to be documented as an AE.
Vinette Zinkand • There are certain medication that we expect to see a change in lab results. In a study situation you would be blinded to those result unless they fall into a critical range that would necessitate clinical intervention. Then the lab result would be an AE or an SAE.
marita mcdonough • In most instances this would qualify as an AE unless specified otherwise in the protocol.
Dr. Dipti Kale • I think if should be classified into 2 categories clinically significantly abnormal and non-significant abnormal lab value. if it is clinically significantly abnormal value then it should be recorded as a AE.
Dr. Dipti Kale • many a times it happens any of the subject's lab value may be abnormal before taking IP and it may continue as the same after taking IP also and sometimes the subject may be nutritionally deficient . so it becomes necessary to track abnormal lab value right from screening visit to subsequent visit and it also becomes necessary to find out the causative factor. if it is done mandatory then it can increase work load for the site and central lab also. the abnormal lab value can be of significance and it can not also.
Lynda Cedar, Ph.D. • Change from the baseline for medical diagnostic values happen often in the clinical studies.
The changeis reported and interpreted by the medical staff as CS or NCS. If found unfavourable it is therefor reported as an AE and treated as such.
Sometimes, a healthy subject is accepted to participate while the values were within the range but bordeline. At the end of the study, such values might become out of the range of acceptable values. In this case, based on the medical judgement, the study medical doctor may decide to refer the subject to a family physician for more exploration and further diagnostic. At my site we document that the subject was informed and advised to consult his family doctor.
Anand Kawade • An abnormal lab value ( lab doing investigation has to state normal ranges for them) is always an AE, even if found incidently. but PI has to decide their clinical significance, intensity, and relatedness
Brown Kim • Great thread. I would love to see a lot more like this one. Thank you everyone for your well thought out replies to the question. I learned a thing or two.
Liljana Stevceva • FDA toxicity tables define the range that should be considered an AE as well as the grade of the AE.
Dr Faisal Khan • Yes, it should be considered as an AE and grading to that has to be agreed upon with the investigator
Binay Thakur • Not all abnormal laboratory findings are often associated with untoward medical occurences and therefore should not be recorded as AE up front. Further, for any particular parameter unless associated with any sign or symptoms such subjects would more resemble a healthy population (having similar degree of variations in the lab value) than a subject population. It would therefore be more prudent to know if there is any sign or symptoms associated with a lab abnormality before considering it as an "AE" unless there is sponsor/protocol specified guidelines such as "CTCAE or similar guidelines would be followed for recording AEs regardless of their clinical significance". Also parameters where laboratory error or false positives are anticipated, if deemed neccessary by the investigator re-tests should be done to reconfirm the abnormality and only then an AE should be recorded.
Objectivity in capturing AE data is good however at the end of the study to make a sense out of it would be very difficult, as each subject with the same degree of abnormality presents to the investigators with totally different case histories, this is where subjectivity is important and neccessary and therefore an investigator's judgment must be honoured (or argued with medical monitors if need be) while recording an AE.
Raji Varghese • AE reporting for abnormal lab values are mostly PI depended some values are CS and some may be NCS,it is better to leave at PI's discretion bcoz they know patients condition better than anyone else,to report all abnormal lab values means lot of reporting to be done.
Sergiy Gryshyn • I think that any lab abnormality is AE. Another question is, should we report it to Sponsor?.Any AE should be documented in SD with clarification is it clinical significant or not. Then we have to follow to protocol requirements. According to my experience, we have to be very careful with interpretation of AEs and have to understand what does it mean clinical significant or not. Leucopenia grade1 as per CTC during chemotherapy, is it significant or predictable and not significant? And doctor treats it.
Rosemarie Harris • It also depends on the indication (i.e. oncology) and how sick the patient population starts out at baseline.
June Rosas • Depends if at baseline if the value is normal for that patient. Also, depends on if the sponsor/ protocol.. wants this reported or not. But definelty should be documented.
Liljana Stevceva • This issues are defined in the protocol. Shouldn't the PI grade the lab abnormalities per protocol? I think yes.
Raji Varghese • Most of you are missing a point that investigator is the right person to mark a abnormal lab value as AE or not until and unless the protocol specifies so.Only the investigator is in contact with the patient it is just not the lab factor that is taken in to consideration there are other factors which is taken in to consideration as Konstantinos Kakos mentioned it is tool that is used to get the picture of patient's condition
Helen Russo • Agree with Raji, and as I noted before an abnormal lab value is a diagnostic tool, not a diagnosis, and in and of itself may/may not be "untoward." as with any other AE, one does not list the symptom (i.e., runny nose, swollen finger), but the diagnosis. An abnormal lab may or may not be an indicator of a problem or it may be a normal variant in the subject's life, but the lab, itself, would not be the AE...
Sergiy Gryshyn • According to ICH GCP, definition of AE doesn't content any Information about PI's judgment. Of course, it's theoretically, on practice, we can meet a lot of different decisions, etc. But in this case, who should be responsible for such decisions?
Traian Andrei Manu • I want to raise some issues with you in regards to what should or shouldn't be reported as an AE for an abnormal lab finding. First of all the majority of studies have a central lab involved and the results are pretty much standardized so you cannot go around saying if you change the units for a lab value you would have an AE.
In regards to what is abnormal you should keep in mind that the subjects involved in phase 2-4 studies are sick. It is very possible that their lab results wpuld come abnormal for some markers. Let's say for a RA study : of course you will have a PCR value for most of the subjects above normal. It is clinical significant because it shows that the subject has an inflamatory response to the disease.
According to GCP every abnormal lab finding can be considered an AE but ultimately the PI will decide if that is an AE or not. The role of the monitoring team is to make sure the lab results are not overlooked or not taken seriously.
Please keep in mind that nowadays central labs offer services for evaluating for soem alb results the difference from baseline to any other visit in the study so you will be able to see a trend for that subject, site , country and IP safety. Do remember that you also have services for a central lab aside the site's , CRA's role to verify the lab results, in regards to Hy's law and that lab values for the liver function if it's above Hy's law standards it will automatically be a SAE.
Lucian Popa • Andrei, regarding your post, the responsability of the investigator is to assess if the AE is CS or NCS, not to define what is an AE, for AE we have a standard definition in ICH-GCP and most of the times the protocol define what AE should be reported or not. The problem is that sometimes the investigators don`t want to record AE that aren`t related with IP or study procedures, in that cases is CRA duty is to identify and make sure that they are reported.
Adel M. Medhkour, MD • abnormal lab values occurring during a clinical investigation are to be regarded as AEs, unless (1) specifically highlighted in the study protocol to be a natural occurence without clinical relevance, or (2) such values are known historically to fluctuate during disease progression with or without the introduction of the investigational product, or (3) such change in lab value is a contributory to patient improvement or a pure sign of efficacy of the investigational product.
I don't think it would a good idea to report an efficacy variable or endpoint as an AE!
Rahul Dawkhar • Lab values: if abnormal, may be associated by many factors and to have a correct judgment study physicians can review the AE’s to give solid evidence, justifying the abnormality as an AE. Instead of jumping the gun, this practice would always help. All the data reported for LAB as abnormal values are extracted in a listing and reviewed on ongoing basis for the study by physicians.
Also protocol does specify some abnormalities that can be classified as an AE but only after considering other factors which a study physician can bring into the picture.
sumit saxena • Dr Sumit Saxena : As per the ICH GCP Guidelines and most acceptable by the industry
Adverse Event defined by ICH :
Any untoward occurrence in a patient or clinical investigation subject administered a pharmaceutical product and and which does not necessarily have to have a causal relationship with this treatment. Any unfavorable and unintended sign( including an abnormal laboratory finding, that is sign or symptom, or disease temporally associated with the use of any dose of a medicinal product, whether or not considered related to the medicinal product.
Still we have to correlate with baseline values and GCP said about abnormal laboratory finding it could be CS or NCS as per investigator discretion.
Jon Ruckle • "Normal" ranges are statistical definitions, and even healthy people can vary a little above or below those ranges simply because that is the way the body responds to multiple factors, even apart from a clinical trial. My opinion: as a clinician, and a PI with extensive Phase I experience in healthy/normal individuals, lab values outside the "normal range" should be assessed by the Investigator case by case, and should not automatically be classified as AEs.
Charles H Pierce MD, PhD, CPI • Hi, I am a bit late in this discussion but when I saw the question "Should an abnormal lab value automatically be considered an AE if the value was normal at baseline? Obviously the investigator would classify as CS or NCS, but would it automatically be an AE?" I had to answer as I had just given a Webinar on "Adverse Events".
The answer is an emphatic "No", a value of a laboratory variable above or below the reference range is not (NOT) necessarily an AE. In fact, it would only be an AE if, in the opinion (Medical judgement) of the PI backed up in the Protocol that the value was affected by the Investigational Medicinal Product (IMP) and indicated an effect of the IMP. I agree with Shilpa above on this clearly
For example, if the CK value was 2 or 3 times the ULR (Upper Limit of the Reference range) and the Protocol stated that above 5 times the ULR was an AE, one would not question that it was not. If the Protocol was silent in this circumstance but the patient admitted to painting his dining room the day before, I would also not think that this was an AE as an elevation of CK following muscular exertion wold be expected and not unfavorable.
Incidentally, the term we must all use is the "Reference Range" which is correct for the mathematical and statistical lab ranges we all use, which imply, correctly, that one could be "normal" either above or below the reference range. Basically, the term "Normal" range or "Normal" Value (for lab data) has no place in clinical research. For this reason alone one would never (NEVER) state that any value above or below the Reference Range would be, by definition, an AE.
Also agree with Biswajeet as far as he goes but it remains for the Protocol and/or the PI to decide whether or not the elevated or depressed lab value is an unintended or unfavorable sign. Charles
Harold Doshan, Ph.D. • I have to agree with Dr. Pierce on all counts. One or more isolated changes in lab parameters that fall outside the reference range are not inherently AEs unless so stipulated per protocol. They are, like "sniffles" or "sore throat," possibly signs and symptoms of an AE (a "cold"), related or unrelated to study treatment, and either CS or NCS, as judged by the PI. It is normally the PI's call to report such occurrences as AEs, based in his assessment of the subject's condition.
Further analysis, including identification of progressive changes in the same or related lab results in the same or other study subjects can and should be evaluated longitudinally to shed additional light on what may, at first, be perceived as an isolated occurrence. Indeed, even temporal trends (so-called "shift analysis") that remain within the reference range may be suggestive of treatment-related effects. For this reason alone, all trends, both within and across study subjects, are evaluated as part of the study analysis.
The study staff has an obligation to report and annotate as "out of range" any lab result that fits that criterion, but to deem it automatically an AE is generally unwarranted.
Lynda Cedar, Ph.D. • I agree with Charles too. IThe lab vaues of baseline are used to see whether a significant change happens to the subject at the end of study or during its monitoring. The significance is left to the PI (medical judgement) which prevails on the study protocol. Additional tests even not scheduled by the protocol might be requested by the PI/medical doctor in charge of the study.
Example: if the CPK are significantly high, the subject might be not allowed to participate in a study. However if they become significantly high during the study or at the end, after a physical examination, medical history, the medical doctor might ask for a CPK-MB test, an ECG, ultrasound or any other test.
When speaking of borderline values, the same value might be considered normal for a patient but abnormal for another one. The results of blood and urine analysis are not considered individually. Each group of values reflects the function of a system: cardiovascular, kidney, liver, nervous, etc.
Andrea Bianchi • I agree with the mayority of the answers posted here, however, in every Regulatory Agency Inspection I have attended, the issue is always adressed in the same manner: if the protocol states a maximum lab value after baseline determinations, it should be at least consulted with the sponsor, before making a call on an AE; this electronic and paper trail, can make the difference between an observation in the final inspection report, or a protocol deviation.
The PI' s judment is always respected, as long as it has documented grounds, and means of demonstrating that there was at least a query regarding the finding.
Subhash Amrutham • I absolutely agree with Charles. Lab Abnormal value not necessarily be an AE; its purely Principal Investigator discretion.
Dadahayath Shaik• I agree with Dr.Charles
Ming Q. Lu, MD, PhD • An abnormal lab value will only be considered an AE if it is clinically significant. If the value was abnormal at baseline and it remains basically the same during the trial, it is not an AE.
Ming Q. Lu, MD, PhD • CIOMS VI group report is a good reference for detailed PVG guidance.
Diane Bartusiak • I agree with Harold: "One or more isolated changes in lab parameters that fall outside the reference range are not inherently AEs unless so stipulated per protocol." Populations like Oncology are evaluated using a wider range of reference values and these Oncology lab ranges are used when the sponsor is reviewing lab tables. Any laboratory trends will be captured by Stats or the study team. Therefore, it's up to the PI to classify, per protocol, whether any patient has an Adverse Event when a lab is slightly abnormal. It's helpful if PI can research a diagnosis/illness asssociated with lab change. Source documentation is crucial so the monitor can confirm that the site's AE reporting is within guidelines/protocol.
Carlos Estrada, MD • HI Guys, I believe we are loosing the original point, the question was: "Should an ABNORMAL lab value automatically be considered an AE if the value was NORMAL at baseline?" So, referring to this particular question, my answer as a Physician and Auditor is YES, it should be reported as an AE, regardless of it is Clinical or not Clinical Significant. If a Subject starts the trial with normal values at baseline and now it is abnormal, it is an AE.
ICH E6 States: "An adverse event (AE) can therefore be any unfavourable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medicinal (investigational) product, whether or not related to the medicinal (investigational) product"
Ming Q. Lu, MD, PhD • In the real clinical world, this will generate tremendous AEs which are not meaningful. e.g. normal range for diastolic BP is 90 mmHg, if the subject has 89 mmHg at baseline and 91 mmHg occasionally during the trial, it is just a common fluctuation.
Therefore, the answer is NO for the question "Should an ABNORMAL lab value automatically be considered an AE if the value was NORMAL at baseline?"
Helen Russo • Dr. Estrada, I think you just negated your own point with your ICH quote. An abnormal lab value, even if normal at baseline, is NOT necessarily unfavorable or unintended. As many have pointed out, labs fluctuate under completely normal circumstances. Reference ranges are just that, reference ranges, and it is up to the clinician to determine if a value outside those ranges constitute a clinically significant abnormality. If the answer to the question is yes, then the lab should be used as diagnostic criteria for an AE, if the answer is no, then it is NOT considered diagnostically relevant, and is noted as such by a physician on the report/subject's chart.
As a monitor and auditor, I've seen (and questioned) labs on both sides...an elevated fasting glucose in a non-diabetic noted as NCS, e.g., I've also requested repeat labs for things like elevated serum calcium which is only elevated under two (to my knowledge) circumstances: lab error or pituitary tumor.
Also, unless the protocol specifies differently, an abnormal lab value itself is not generally the AE, and many sponsors specifically ask that "symptoms" NOT be listed as AEs, which is what an abnormal lab value would be. Many times I have instructed sites to remove multiple symptoms and request the physician to use one diagnosis, this would include abnormal labs, i.e., elevatd WBCs, stuffy nose, earache, etc., in favor of a diagnosis of URI. I have also seen instances of wildly abnormal CK values, e.g., and cme to find out the subject was exercising extensively. Usually in a case like is, and absent of any other untoward signs/symptoms, the subjects are asked to discontinue their training and the lab is repeated until it return to normal. These situations are not generally AEs, or reported only once, and ended when the value returns to WNL.
The bottom line is that a blanket response of "yes" to the original question of "should abnormal labs be listed as AEs if normal at baseline" cannot be made.
Tim Pratt • We recently had a lengthy discussion on this topic prompted by "anaemia" as an AE.
Labs vary in what they consider normal, so a finding below an absolute value won't work (in this case it was 10). More appropriate, perhaps, is a relative, so a decline of X% from baseline. That, however leads to other problems - decline in haemoglobin during menses for example is normal.
We proposed to our DSMB, and they accepted a 2 part criteria for this particular value to be an AE - 1) a shift of X% from baseline (the relative method) AND 2) for which medical intervention/treatment was provided. the latter could be as innocuous as giving Fe supplements or as dramatic as transfusions. Our DSMB thought these criteria, in tandem, were very appropriate. This model is along the lines of that proposed by Dan R above.
Theresa Dunaway RN,BSN,MBA,CCRP • I think we are getting off topic here. I don't think there is one and only one answer for this question. Yes the protocol needs to state the labs that are significant to the study. The reference range for the lab being used for the study (if not a central lab) needs to be filed with the sponsor. Honestly it is the PI, and only the PI that makes this determination. If a woman's hgb is significantly decreased from baseline, there may be a problem. It should go down but not to significantly, if she is on a drug study you do want to acknowledge this. Also Dr Lu I don't think the example you gave is that accurate...we are speaking of ranges but not those that would be 'normal' for a particular patient.
I must say that for any lab out of range you do need to consider the baseline and relevance to a single individual. I do strongly believe in re-tests, I do believe there are errors and fluctuations, this is also for the PI to determine. But I agree with Dr Estrada, If the protocol doesn't state that small out of range results are acceptable then the PI has to consider the this as an AE.
David Radin • By definition "Normal" means the value is within the 95% confidence range. In other words you would expect out of 1000 subjects 25 of them would be "high" and 25 of them would be "low." If the protocol calls for reporting of abnormal labs int he absence of clinical manifestations then it would be an AE by protocol specific definitions. If, however, the difference is within what the clinician would consider an expected variation in a patient without an underlying disease burden it is simply "normal" and not an AE in and of itself.
Paula Singleton • For concern for the patient's safety, as well as the overall review of the study, I think any change from baseline should be listed as an AE. A case that comes to mind is a patient who had an infected injury and was prescribed an antibiotic which caused a marked change in one of the lab ranges, a factor that was a) not unexpected, and b) would have definitely been considered an AE had it not been for the antibiotic that was prescribed. In this case, the change in the lab value was not because of the study drug, it was because of the antibiotic. Regardless, I feel the change should be marked as an AE and then noted as a response to an antibiotic. For the patient's safety, any anomalies while on the study drug should be documented, and if not in the AE log then where?
David Radin • Which makes MY point -- unless the antibiotic makes the lab change ALWAYS you have a change due to a medication (an AE) which cannot be 100% attributed to the con med. A further examples are GGT and ferritin -- these vary so much without clinical manifestations that Medicare specifically excluded them from blood chemistries and will only pay for them if there is a specific supporting diagnosis/clinical syndrome. Abnormal labs without clinical manifestations can always be captured by Data Management and will not be "lost."
Paula Singleton • Good point. For my example above, if it creates an expected out of range value for any and all patients who take the con med, then it would NOT be an AE. but should it only create out of range values for some patients, it should be listed as an AE for those patients.
For the original question, Using the reasoning above, if the reason for the change is specific to the patient, isn't defined as expected for any reason, whether in the protocol or other documentation, I believe it should be an AE.
Andras Koser MD, MBA,CPI, FHM • I am not sure if any you who responded are PI's. I am a PI, and it is my duty and responsibility to determine what is an AE and what is not, unless I have specific instructions in the protocol. To do that the PI need to be involved, know medicine well and know their patients and their history.
We can speculate long on specific parameters we measure, there cannot be one rule for all.
Theresa Dunaway RN,BSN,MBA,CCRP • Thank you Dr Koser. I think we all need to look at Dr Koser's comment as the correct answer.
Charles H Pierce MD, PhD, CPI • Amen to what Andras has said. Also, I would work with Theresa any time as she understands this issue. I am and have been a PI and have >20 years in the Clinical Research industry. I suspect that any and all who answered this question as you and I did - that it is the PI who decides if and when (and in keeping with the Protocol) that a lab result outside of the Reference limits (ULR) is an AE or not. I have been astounded by the responses who suggest that any result above of below the Limits (and they sometimes wrongly use the term "Normal Limits") is an AE. Have just given Webinars on "PI Responsibility" and "Adverse Event Reporting" and nowhere in the Regs does it state otherwise. Period
Helen Russo • I think that last few responses have pretty much hit the nail on the head with regard that you CANNOT make a global statement that all abnormal labs s/b treated as AEs, and at the risk of repeating myself, in every initiation visit I have ever done, my "schtick" is that you are the physician and it's up to you to determine what is/is not an AE, however, I reserve the right, as a monitor, to ask you WHY you have made such a determination if it is unclear to me, and to put the reason in writing either on the lab orin a chart note, because if it's unclear to me, it's going to be to someone else such as an auditor.
I have also, frequently, requested physicians to differentiate between WNL and abnormal/NCS, because they absolutely do have different meanings, and as Dr. Pearce pointed out many incorrectly use WNL.
Theresa Dunaway RN,BSN,MBA,CCRP • Thank you the for the compliment Dr Pierce, I only have 11 years but it still feels good! Helen I agree there is a difference between abnormal/AE and NCS, but you as the monitor still has no responsibility here, your responsibility as well as a coordinator's is to review the documentation. Each of us involved (Coordinator, sub-Investigators, Monitors, sponsors and FDA) need to ask the PI what his rationale is. I have even seen Investigators debate on AE's vs. baseline and many other issues, but it is still their responsibility to make this determination and support it.
Lynda Cedar, Ph.D. • Any event or change that happens during the study is reported,documented and interpreted. Speaking of lab. values change, deviating from the baseline, the change is captured as an AE but its clinical signififance belongs to the PI (medical doctor) whether it is NCS or CS.
Helen Russo • Sorry Theresa, but I disagree, though I never have said that I have the responsibility. It most definitely is part of my responsibility to 1) make sure the data being submitted is accurate, and 2) that the physician reviewing the data is clear to the reviewer. If I do not understand a physicuan's rationale for for a decision, and ask him/her about it, then I am not doing my job properly and it will likely be questioned by someone later. As long as it is properly documented and does not go against something specifically in the protocol, I personally don't care if a physician determines soothing is CS or NCS, however if I see no support for that or that support isn't clear I have not done my job to the best of my ability.
Lynda, I'm sorry, but an abnormal lab, or rather one that is ouside of the reference ranges, is not globally an AE. For one thing, a lab value is a diagnostic tool not a diagnosis. I couldn't count the number of data queries I've seen over the years on exactly that: "elevated (insert lab value of choice) is not a diagnosis. Please use the correct medical diagnosis." forget the fact that if every out of range lab value were listed, each study subject would have likely have pages and pages of of AEs when combined with the usual AEs such as the flu, URIs, UTIs, sprains, cut, bruises, etc., that
This is precisely why most central lab reports have places for physicians to note CS/NCS and comments. In almost 20 years, I have never once seen an NCS out of range lab value listed as an AE even if it was normal at baseline.
Helen Russo • Oops, meant do not ask...
Theresa Dunaway RN,BSN,MBA,CCRP • Helen you are right. The accuracy and clarity of the documentation does lean on the shoulders of the coordinator and monitor. It is still the PI's responsibility to determine the significance, the support for that rationale (whether CS or NCS and/or AE) is only the PI's.
Andras Koser MD, MBA,CPI, FHM • We have a lot of action around this question. I am still standing by my previous statement. It is PI 's responsibility to determine what is CS, NCS, what is AE and what is not unless the protocol specify.
Of course if the monitor or the sponsor, of the FDA asks about one or another lab result, the PI's better have a good explanation why did not report this one way or another. Also any decision that could be later questioned should be well documented.
To be able to do this the PI must know EVERYTHING about the subject, the study, and has to be a well trained physician. You have a problem when you have a ghost PI, or an ungifted physician trying to be a "scientist".
Would be great if the sponsors grade us, P Investigators by the quality of work we do.
David Radin • As a PI with 18 years in clinical trials I whole heartedly agree with Andras
Charles H Pierce MD, PhD, CPI • Good point Andras and I also agree. Let me point out that the ACRP/APPI have a certification process for Investigators that would answer you main point. The CPI will be increasingly important down the path and, presently, one major Pharmaceutical firm lets Certified PI's pass on the Investigator training sessions so many have. Certification is not a substitute for experience and common sense but at least one knows GCP and why the Protocol is what is followed.
From what I have seen, those who know agree that a lab result above or below the "Reference Range" is not an AE unless the PI so states and that statement is in the Protocol. Actually, if the Protocol stated that "All lab results above or below the 'Reference Range' is an AE", I would not agree or sign it until this was removed.
Theresa Dunaway RN,BSN,MBA,CCRP • I agree as well...This is the point I was trying to make. The PI is responsible for this determination of significance and/or if it is an AE. With that responsibility he must be able to document and support it in a way it is relevant to the study to sponsors, investigators, regulators and colleagues.
Helen Russo • Dr. Koser, sponsors do "grade" PIs and sites for the quality of their work, though they may not always share that information. You can also usually tell, if you have a good monitor who knows what he/she is doing, by what your monitor is telling you. Sadly, there are monitors and sites who are conducting clinical trials without a good understanding of basic research techniques and when you combine an inexperienced monitor and site, and I'm trying to be polite when I use the term "inexperienced" as I've worked with both experienced monitors and sites for whom that term can only be loosely applied, you get a poor end result with things like potentially significant AEs being missed.
It has always been my standing that a physician is responsible for determining and documenting the CS of any abnormality noted during a clinical trial, this is not, unfortunately, always the reality.
Standard source documents are partly to blame as they tend to discourage any kind of written discourse in favor of "fill in the blanks and check box" responses. I strongly encourage my sites to add chart note pages to standard sources, and that "more information written in a source/chartis better;" and not all of the information is needed for transcription to the data forms. It is a pleasure for me to work with physicians who actually KNOW the study subjects to which I'm referring when I have a question, but again, that is not always the case
Lynda Cedar, Ph.D. • What do you call this event: when a subject is injured at home, transported to the hospital, got medications and surgery, and his participation is interrupted consequently... ?
By definition, an adverse event (AE) is any adverse change in health or side effect that occurs in a person who participates in a clinical trial while the patient is receiving the treatment (study medication, application of the study device, etc.) or within a previously specified period of time after the treatment has been completed.
Helen Russo • In my experience, this example is both unintended and untoward and would be an AE, and most physicians I've worked with would agree. In a double-blind study there is no real way to know if the test article caused some dizziness, verrtigo, decrease of special senses, whatever that caused the accident which caused the injury.
That is a completely different issue from an out of range lab, 1) which may fluctuate "normally," and 2) may or may or may not be of CS by any definition, and I think most physicians would agree. I've seen labs at panic levels which turned out to be NCS when evaluated by a physician, however they are generally followed closely along with other diagnostic parameters reviewed by those physicians before coming to that determination.
The bottom line is that one cannot make a global statement that ALL out of range lab values are AEs and that is what Drs. Pearce and Koser and others have pointed out. An abnormal lab is only part of the diagnostic puzzle and what may be diagnostically significant for one subject may not be for another.
If it is a case of a particular lab ending up out of range, regardless of significance, across the board for the study, the statisticians will pick that up when looking at the global study data.
Helen Russo • Also, a change in a lab may or may not constitute a change in health...
Lynda Cedar, Ph.D. • Helen, I agree with you on the first paragraph, in interpreting the happening of the event as an AE, it was even classified as SAE by the FDA.
Regarding the change of lab values, no one is saying that all out of range values are AEs or constitute a change in health..... Regulations, standards and medical judgement apply.
Modern CROs are using the Adverse Event Reporting System (AERS which is in compliance with the international safety reporting guidance (ICH E2B) issued by ICH. Adverse events in AERS are coded to terms in the Medical Dictionary for Regulatory Activities terminology (MedDRA).
Assessment of Lab values is as follows:
- If the lab value changes but it is still fluctuating within the range of normal values, whether it decreases or increases, it is not reported as an AE. However when reaching the borderline values or exceeding them a bit, the PI attention is needed, he decides about the follow-up.
- If a value was within the normal range (or acceptable in the PI judgement) and its change is major at any time of the study life, the change is reported as an AE and its interpretation as CS (clinically significant) or NCS (not clinically significant) belongs to the PI’s judgement. The follow-up will continue until the value is found similar to its baseline.
Theresa Dunaway RN,BSN,MBA,CCRP • I agree with Lynda, it is still the PI's call...significance in lab changes, AE vs SAE. We have regulations and guidelines to construct a framework and guidance in addition to the protocol, after that it is all up to the PI.
Lynda Cedar, Ph.D. • Totally agree with you Theresa and others about the PI's freedom in interpreting the significance of an adverse event, that could be lab value, a symptom or any medical issue. In fact, the subject's safety is under his responsibility and the sponsor.
Regarding what is called AE and what is not, there are many sources explaining it and guiding clinical staff about reporting it.
SOMNATH MONDAL • As per my understanding of definition of ICH GCP, an abnormal lab value must be considered an AE whether it is clinically significant or NCS. As because, a meticulous approach is essential to achieve a bright out come.
Hence, this must be documented in scientific manner ..............
Charles H Pierce MD, PhD, CPI • Please, Everyone, Read the entries above by several experienced Investigators (and others like Lynda, Theresa, et all). Lab values are reported with an attached "Reference range" which implies by definition that one (the patient or the value) could be "normal" and be above or below the reference range limits. So, a lab level above or below is not (NOT) an Adverse Event (AE) unless the PI deems it to be an AE.
The PI would, of course, review all lab values and then make the decision as to whether or not the lab value is CS or NCS. If it is deemed CS, then it would most likely be an AE. Nowhere in the ICH GCP Guidelines or the FDA Regs does it state that lab values above or below the "Reference Ranges" is an AE.(Period)
Binay Thakur • I have gone through the comments posted by Dr. koser, Lynda, Theresa and
Helen and find it very informative and interesting. I agree with the
comments posted by Helen that although it is a sole responsibility of PI to
determine what is an AE and what is not, it is always the duty of the CRA
to cross verify the decisions of the PI (unless it is very obvious) and
discuss with the investigator about what prompted him to mark a particular
event as AE. The outcome may still be the same however the CRA doubts, the
discussion and the conclusion must be documented clearly in the source of
the subject and in the monitor's notes for future
references/audits/inspections.
I am not as experienced as most of you ( as have only 5.5 years of
experience) but quite a time PI have changed their decisions (turning an
event from AE to non-AE) after discussion with the CRA. It happens
sometimes when your investigators are not well trained on AE recording
requirements.
So, the decision to make an event AE always rests with PI however it is the
duty of the CRA to find out if any AE is created just out of
misunderstanding or is not in compliance with the protocol lines or if you
are following a certain guidelines such as CTCAE for toxicity grading and
specific Sponsor requirements.
Theresa Dunaway RN,BSN,MBA,CCRP • Billy, remember the question, 'If a lab value is out of range and a change from baseline, whether CS or NCS is it an AE'. The CRA does need to inquire following review of the records, so do the investigators and coordinators. If there are many inquires, the site may need retraining. Let's not forget a good site with a good coordinator should be the first line of inquiry and these questions and support should be documented clear enough that the CRA understands the rationale. The CRA should be the last line of review and inquiry, not the first, but then not all sites, coordinators and CRA's are equal are they....
Liljana Stevceva • The key to the answer is the protocol. If the event is AE according to the protocol toxicity tables, than it is an AE. Usually, the toxicity table would specify the grading of the event as well.
SOMNATH MONDAL • I do agree with Linjana, obviously protocol is the ultimatum. But if the same event recur subsequently for a number of subjects then the question of reconsideration about the protocol arise.
Whatever the event CS or NCS one need to document it for future consideration and discussion have a decision among the team members first, if discussion found rationale , PI must have to convey the same to the Sponsor...for their judgement....and final decision.
Helen Russo • I think we're, again, losing site of the original question. Most protocols, particularly in recent years, address the issue of hepatic and renal toxicities, and in some cases, issues pertaining to bleeding events, platelet aggregation, etc. These are almost universal these days.
As Dr. Pearce stated earlier, there is nothing in either FDA regs or ICH guidelines which state, categorically, all lab abnormalities (or changes from baseline, for that matter) are AEs. It would be ludicrous to say otherwise, not to mention labor intensive for both the sites and the monitors. We would be completely overwhelmed recording and monitoring events which are, in most cases, normal variances.
I depend on my physicians to review labs reports and make a clinical decision as to their significance and part of my job is to ensure they are doing this in a timely manner. The reality is that sites are busy, and physicians are busy therefore, things get missed or looked at in isolation (one visit's labs apart from a subject's entire chart).
If this didn't happen, I would be put of a job, and quite frankly, I love my work! I appreciate physicians like Drs. Pearce and Koser who know and understand not only the laws governing clinical research, but their responsibilities to the study subjects and the study itself. I have had many discussions with physicians regarding lab reports, and when we've disagreed about their clinical decision (rare though that may be) I defer to the medical monitor. MOST cases, my site physicians/PIs are able to justify their determination quite easily, and I merely request them to put it in writing...
Helen Russo • Sorry, not liking touch pad typing right now, meant out of a job...
Theresa Dunaway RN,BSN,MBA,CCRP • Helen I think you are just restating the point we are all making. The PI calls it depending on this medical and study knowledge. A research site should have most of this documented for you, although we are all human and have oversights (or the FDA auditors would be out of a job!). And I think your job is safe my friend, the sponsor still needs someone to review the data on site for them and find these PI rationales and oversights. Good job my friend!
Harold Doshan, Ph.D. • Just a follow-up to Steve Mcateer's comment: It is true that reference ranges for lab tests are established as 95% confidence limits on data for each clinical laboratory. These ranges are recalculated periodically, based on the most recent set (often tens of thousands of lab tests) of data available to that institution. However, please note that these ranges are NOT a reflection of a purely "normal, healthy" population. Lab tests are most frequently ordered by physicians and hospitals for patients with medical complaints or undergoing treatment. Many of these patients may have abnormal labs associated with the conditions for which the labs were ordered. Other labs, of course, are ordered as part of routine physical exams and prove to be well within the normal range.
Taken together, the reference range arrived at is a composite of all of these results and if the sample size is large enough and represents an adequate mix of healthy and variably ill patients, the range for each test should generally reflect, but is likely to be broader than, that for a purely healthy population. For large labs with many, many samples, the fit is good. Differences in reference rangess for smaller, local labs can be explained by their smaller sample sizes and the possibility of skewing due to a larger percentage of ill patients (as well as different responses for their instruments, test reagents, etc).
All the more reason to tread carefully when calling a shift in a lab finding (by the same lab) from within to just outside the reference range an AE. Many protocols (especial in oncology) stipulate the extent of the deviation from "normal" that must be considered an AE. Lacking these guidelines, the PI must decide, but he/she should also consider whether the out-of-range value is, in itself, an AE, or whether taken together with other signs and symptoms points to a specific conditions that is an AE, whether related to study treatment or not.
Dominique Chesnais • Lab values are well investigated in clinical trial analyses and reported in the clinical study reports. Refer to http://www.fda.gov/downloads/regulatoryinformation/guidances/ucm129456.pdf
In the FDA Structure and Content of Clinical Study Reports, lab data are reviewed in sections 12.4 and 14.3 about the following topics:
Clinical Laboratory Evaluation
Clinical Laboratory Evaluation of Individual Laboratory Measurements by Patient (Appendix 16.2.8) and Each Abnormal Laboratory Value (see section 14.3.4)
Evaluation of Each Laboratory Parameter
Laboratory Values Over Time
Individual Patient Changes
Individual Clinically Significant Abnormalities
Safety Data Summary figures and tables
Abnormal Laboratory Value Listing (each patient)
To report abnormal lab data, beyond CS or NCS, as AEs will be duplicating the information, unless the abnormal lab value is also part of a wider clinical pathology problem.
Babu R R • an abnormal value to be repeated to rule out the lab error. further if it reveals higher/lower than the normal range, if it is CS then it is AE. if it is NCS then it couldn't be an AE. correct me if am wrong
Lynda Cedar, Ph.D. • Hello Dominique. The comments above show 2 opinions regarding an AE statement, what is your definition of an AE: only when the change is found CS by the PI or at anytime when an out-of-range value is captured, or otherwise. Please explain.
In fact, the ICH integrated clinical report (or the link you posted above), the lab values are assessed as: it lists the range, the values at the baseline, at each study visit and at the end of study. Descriptive statistics are sometimes done.
The central lab provides an interactive file that is recognized by the computerized operating data base, programmed to capture the changes automatically. The programming is done by the IT, the biostatistician and the clinician to determine which change is determined as an AE. At this stage a change can be captured as an AE regardless to its relevance as CS or NCS in the medical judgement.
Thanks Dominique to share your practice with us. Lynda.
Dominique Chesnais • Hi Lynda, The first issue is that AEs are frequently under-reported by investigators during clinical trials. One needs to look at disparity between sites in reporting AEs, whatever their type. What benefits will it bring for an investigator to report an increase of one or two unrelated abnormal lab. values as AEs, as they are already noted in the lab. report as out-of-range and CS?
All study lab data are specified as CS or NCS after review by the investigator. Furthermore, they will be thoroughly analysed and reported, as required in the ICH E3guideline (Structure and Content of Clinical Study Reports). All out-of-range lab values will be highlighted in graphs, tables, listing and even discussion; same with the changes over time and values reported as CS. All that information strongly highlight some potentially pathological problems that Sponsor personnel should careful investigate.
However, when there are some increase of AST, ALT, GGT, alk. phosphatase, etc. the investigator should report a liver pathology; same about abnormal blood tests looking for blood disorders; too for the renal function markers. Drugs are reported as having liver, renal, blood toxicities, based on their clinical and biological signs.
In conclusion and in my opinion, it is these biological pathological problems that investigators should concentrate and not miss to report rather than some single, unrelated, out-of-range, CS laboratory values.
Lynda Cedar, Ph.D. • Hi Dominique. Thank you for adding your comments.
The liver enzymes and CPK are good examples to illustrate the discussion. They may respectively increase if a subject took alcohol or did exercise within 48-24 hours before blood testing. They are captured by the system of the central lab and the CRO computerized database as AEs. At this stage, the PI will look at them and decide about the outcome,.
I agree, lab values as per their name, medical diagnosis values, reflect the function of a system (liver, kidney, heart, etc.), a single value might be not relevant. Example, if Na+ is high, it does not mean blood pressure or kidney but if K+ if high the meaning is different.
When selecting a central lab, I ask them to provide the curve of calibration of the day, plus, to insert controls (small, medium and high) within the chain of tubes of analysis.
Thank you for sharing. Lynda.
Christoph Lohan • Based on the way the question is asked one near enough has to say that the "abnormal" lab result has to be classified as an AE. Rightly quoted by Mr. Dasgupta in its defenition.
Yet I would like to add that most often protocols are more diverse and more specific altogether if that makes sense. It is not only limited to AE as such. You will find other classifications such as "expected" adverse reaction and "unexpected" adverse reaction, of which both can be serious or not. Therefore a abnormal lab result could well be expected and therefore will not classed by companies as an AE per se. If it is tends to be unexpected, especially the SUSARs, then it will require flagging up to sponsor and regulator.
Thus I also agree with Ms McLeod "yes it's an AE unless stated otherwise in the protocol" :-) Happy discussing
Lynda Cedar, Ph.D. • Hi Christoph. To the best of my knowledge when an event is expected, it is reported as assuch but per the difinition of an AE that event is not an AE. If you can provide a specific example, it might illustrate the case.
Dominique Chesnais • >>>>>> SURVEY <<<<<<
I have initiated a basic survey (using LinkedIn Polls facilities with their limitations) based on a current discussion on reporting abnormal lab data as AE. I would be pleased if you could complete it at http://linkd.in/v0g4YI
Lynda Cedar, Ph.D. • Hi Dominique. It is a good idea of doing the survey. If I'm not mistaken, you asked the question differently.
The discussions have shown divergence: when an abnormal lab value that has changed comparing to the baseline is considered as an AE:
1) when captured by a computerized bata base as an AE,
2) only when found/interpreted as CS by a PI.
Sylvie JOUVE • In you protocol you also need to anticipate abnormal diseases' related values such as CRP when infection, AST/ALT (usually > x3 or 4 times is acceptables in some diseases) you also need to use age & gender related NR for at then having PI to apply his evaluation accordingly to these and medical practice
In the study report, you have to present lab data with predefined criteria for PCI or non PCI;
Declaring as a AE is as per PI judgement and usually we ask PI to declare rather than 'cancel" an AE.
So this is why a good knowledge of the disease and related abnormal vlaues can mak esome differences.
Hope this helps
Evy Moreno, BSHS, CCRP • I agree, this is a good topic to discuss because it happens all the time during clinical trials. So the combination of FDA regulations, the PI judgments and the input from the sponsor or protocol takes a role when deciding on A/Es
Nishat Hedayetullah • Going back to the original question (Should an abnormal lab value automatically be considered an AE if the value was normal at baseline? Obviously the investigator would classify as CS or NCS, but would it automatically be an AE) and seeing the recent exchanges, the clinical significance of the lab result should be considered in the context of the patient's overall health and in consideration of ongoing medical conditions and concomitant medications that are being taken. The treating physicain (not necessarily the PI) is best qualified to answer what the clinical significance is. Since the patient is on a clinical study, the treating physician/study investigator can discuss with the study sponsor whether the lab results have a causal relationship to the study drug.
Dipali joshi • It was really great discussion but some times we have to follow protocols and few protocols specify that "grade 2 or 3" abnormal values should be captures as AE , or some times baseline lab values will be considered.As I am handling Oncology trials where AEs are more frequent so capturing AEs differes from protocol to protocol and also based on the indication of trials.
Lori Jacobi • Again, the answer is yes, unless the protocol states otherwise. Generally speaking, anything that is a change from baseline is an AE regardless of it's clinical significance.
Lynda Cedar • Lori: indeed that it is, if located out of the normal range. The PI decides about its outcome.
Helen Russo • Lori, that's not correct. If every NCS lab were reported as an AE, sites, monitors, and data management would melt down. It is simply not feasible, nor does it meet the definition of an AE which is an "untoward and/or unexpected" event.
As many have mentioned, lab values WILL fluctuate outside of reference range values in a normal, healthy patient with no diagnostic significance. Others have pointed out, and many protocols include, specific ranges for analytes like LFTs, because drugs are metabolized in the liver and therefore can effect liver function. For example, historically, LFTs, depending on the specific analyte, are considered clinically significant if >2-2.5x the ULN, some sponsors alter this guideline up or down depending on the drug or use CTAE toxicity ranges.
The bottom line, is that I've never worked with a physician who 1) treats lab values in isolation, but as a diagnostic tool to determine a diagnosis, and 2) would agree that an NCS out of range lab would be an AE, especially absent of any clinical symptoms.
As for, global, study wide abnormalities or delta changes across a specific panel, that was statisticians are looking at. Once all the clinical data are gathered, changes like these are analyzed for statistical significance.
Prasad Vaishampayan • Great commnet Helen,,,an eyeopener !!
Andras Koser MD, MBA,CPI, FHM • 176 comments so far, read my comment somewhere at the beginning of this lengthy thread and you will find that that is is the right answer to the question. Many agreed so far. It is time to talk about something else. :)
Olusegun Adewusi • Adverse events after normal baseline records might signify the onset of traumatic episodes. So it should be a worthwhile 'in-vitro' investigation protocol afterall.
Lynda Cedar • Agree with Andras. It was a great question and great answers/discussions, great idea for surveys, now it's time to come up with another topic of discussions. Many thanks. L.C.
Charles H Pierce MD, PhD, CPI • Who would have believed that this one question would illicit such a wide range of responses. Some clearly wrong has been the largest surprise. I second my colleague Andras view that it is time to move on. Helen, above, and many other have it right that lab results are taken in context and / or according to the protocol and / or it is essential to remember that a patient or subject may be above or below the "Reference range" and be completely normal (by definition). 'nuff said. CHP
Andras Koser MD, MBA,CPI, FHM • If you have a great meal this thanksgiving but the turkey looks and taste different from the years before, or god forbid from your moms turkey it does not mean that it is bad. If you as a PI like it it is not an AE :) Happy Thanksgiving to all who participated.
MARIA KORKOVILI • We all already know what an AE is. Having at the same time laboratory experience I would say that an abnormal value should not be automatically regarded as an AE. Many times biochemical analyzers and haematology analyzers go through internal errors. Thus what I would recommend is run the test once more and then decide whether to report it as an AE or not. In addition, the nature and the sensitivity of the test should be taken into consideration.
Kind regards to all!
Anna Zimmermann • I think a laboratory value is something very specific. It is something real and possible to prove. Thus a change may indeed be automatically considered an AE.
Mickey O'Brien • Wow, that's a loaded question, for which there is not a black and white answer. Much depends upon the analyte, the amount of change from the baseline value, and the health status of the subject/patient. Effect upon related analytes would need to be assessed as well. Rule of thumb is to investigate ANY abnormal lab if the baseline lab was in normal range, but it is not necessarily an AE. Sometimes it is just lab error, but having worked for numerous labs, that can be an overused reason for not investigating further.
Paula Singleton • One thing I have learned in my 3 whole years in Data Management, is that if you assume you have the right answer, there is usually going to be someone, or some document, that can refute it, and refute it well. Or, if anything, bring into light new considerations you hadn't realized before-hand. This question, whether an abnormal lab value should be an AE, is completely and totally up to the Investigator/Physician. When I first answered this question, I looked at it from an outside perspective, as if it were a test question. If I were an investigator, what would I consider.
All things questionable, regarding labs and adverse events and so many other things, fall into the investigator's realm of responsibility. We have electronic edit checks that flag all out of range values, but not all of those out of range values show up on an AE log. I have learned to realize, in this industry, if you think you have the answer, there may be a chance you haven't looked broad enough into the question.
Cathryn Evans • THese things are generally quite clearly defined in the individual study Protocol and SAP, as well as GCRP guidelines. Not advisable to make an absolute statement without the context of the first two items.
Theresa Dunaway RN,BSN,MBA,CCRP • I agree with Cathryn, there is no absolute statement due to the variability in protocols and endpoints for each individual trial.
Nicholas McWilliams • Though I understand the desire/concern to not miss AEs, I think that people need to remember that the PIs are judging clinical significance for these labs. Just because out of range labs can be AEs doesn't mean all are required to be.
Nicholas McWilliams • Why have the PI assess significance then? The PIs have the ability to see the labs in a greater clinical context. If you require NCS labs to be reported as AE then you end up potentially creating a lot of noise in the safety data and end up risking creating a skewed and inaccurate label for your drug.
Nicholas McWilliams • If there are labs of interest then the protocol should state how they are addressed, including any threshold for abnormal values vs AE. Shift tables from baseline has been mentioned previously as a valuable tool in the context of detecting lab oriented signals, and I agree highly with that approach (sorry this had to get split into 3 comments)
Sateesh Kumar • It should be consider as AE once the Investigator indicates that it is clinically significant. All lab parameters should check by Investigator in each visit by judging clinically and also should compare with the baseline incase of significant increase or decrease in the values it should be captured as AE though it is not mentioned in the protocol. All these changes should be recorded in source as well as on the reports.It is a good practice to record abnormal values significance in source. It is an ideal procedure for Investigator who follows strictly ICH -GCP
Shweta Gupta • Here are my 2 cents.
It should be PI decision of CS or NCS. If it is CS than yes it is an AE. For e.g. For Hemoglobin the normal range for Adult males: 14-18 gm/dl. If the range on test report is 18.2 gm/dl and indication for IP is pain and PI thinks it is NCS it is not an AE.
Thanks!
JATIN SHARMA • As per my experience until or unless investigator has reported a abnormal value as Clinically significant it cannot be reported as AE this is the guidelines maximum sponsor follows and i am agree with the way its reported. To report any abnormal value as CS or NCS is investigators call.
Pharma JobLinks.com LLC • Yes, an AE is an AE. This should be included in the PI especially to track if this happens more often. Anything else seems fishy & like you're hiding something...
Ajay Reddy Jhampa • As per the E2A guidelines, AE is an
1. Any abnormality ( Untoward medical occurance)
2. After the first dose of Pharmaceutical administration ( After the first dose of study
drug)
3. It may or may not have casual relatiomship (Relatedness).
Therefore any abnormality that occurs after the first dose should be considered as an AE whether it is clinically significant or not. It is not an Investigators discretion.
FDA accepts data in SDTM standards and we submitted all AE's in AE dataset. If we get the information of clinically significance from the CRF, we submit this information of clinically significanvr in SUPPAE( Supplimentary Qualifier dataset).
Charles H Pierce MD, PhD, CPI • Please, folks, go back and read the many comments up to 4 months ago when this was started. It should be crystal clear by now that whether or not to call a lab result that is above or below the reference range limits an AE is a PI decision. A lab result above or below the reference range is not (NOT) an Adverse Event unless the PI (after review) states that it is an AE. To do or think otherwise is incorrect.
Agree with the ICH E2A definition but notice that it does not say that a lab result above or below the reference range is an AE by definition. I am sure you know that by definition one could be entirely "Normal" and be above or below the reference range limits. See earlier comments making this point that the terms "Normal limits" or "Normal values" has no (NO) place in clinical research.
I will be giving Webinar on this subject in January. Will let this group know of the date when set.
Sunee Thiravanitkul • I agree with Dr. Charles H Pierce that it depends on the clinical stage of the patient. Some out of normal lab value is just very much normal for some particular patients due to the underlining disease, e.g. low hemoglubin value in anemia patients.
Cathryn Evans • Thank you, Charles. To add to this comment, since all abnormal lab values must by regulation be tabulated and summarized in the CSR under the Safety Section, these abnormalities never go unreported. The clinical significance of the abnormality is what leads the investigator to include it as an AE — thus sometimes abnormal lab values are designated as "AEs"; other times they are retained in the Safety/Laboratory Test section of the report.
Helen Russo • @Dr. Pearce: This does seem to be the discussion that never ends!
@Cathryn: You pretty much said what I and a number of lab people have said: seemingly NCS trends get caught in the statistics.
Lynda Cedar • This question generated 199 comments, I wanted to bring it to 200.
In fact the statistics use the data captured by the computerised database which is programmed to capture any change in comparison to the baseline (called AE). After the medical interpretation some AEs are classified as NCS and others as CS associated with a number from 1 to 5 as selected by the PI.
In capturing all the changes versus the baseline is the only way to ensure that all the AEs were explored and interpreted by the PI .
Subscribe to:
Posts (Atom)