A couple of years ago Applied Clinical Trials magazine ran a survey on the most troublesome aspect of conducting clinical trials. There were about a dozen different categories listed including government regulations. However, the number one choice, of nearly half the respondents (clinical trial professionals), was the difficulty in recruiting subjects/participants to the studies. Now the New York Times has connected the problem with recruitment of clinical trial volunteers to the war against cancer (see Blogroll: 123 NY Times…).
There is one very big mistake in the article. The claim is made that each “adverse reaction” needs to be reported to the institutional review board (IRB) for evaluation and that each participant needs to be notified, regardless if the event is related to the study drug or is the progression of the disease. This just isn’t true. And the misunderstanding about reporting adverse drug or device events causes a lot of problems for people involved with conducting clinical trials.
If you read the FDA regulations in 21 CFR 312.66 it says that the researcher, “…will promptly report to the IRB all changes in the research activity and all unanticipated problems involving risk to human subjects or others…”
This is very different than reporting the progression of the disease, or stubbing one’s toe. There is a lot of confusion about defining and reporting adverse events to IRBs. Some IRBs will refuse to accept anything unless it meets the standard of “unanticipated problems involving risk to human subjects.” Other IRBs, those without much experience, demand all adverse events or all serious adverse events (SAEs as defined by FDA regulations) be reported. This isn’t helpful and it really isn’t helpful that many clinical trial sponsors or contract research organizations require over-reporting “just to be safe.” The whole purpose of expedited reporting of an unanticipated problem is to bring it to the attention of the IRB. If the IRB is inundated with with unrelated events, then the real problems, that require the IRB’s attention, get lost in the flood of paperwork.
FDA regulations and guidance documents are in serious need of updating. How to defne and report adverse events would be a very good place to start.
UPDATE: I did actually try to get a letter published by the Times. But you know how that goes. Its one of the reasons I have a blog. Here it is for those who are interested:
“Lack of Study Participants Said To Hobble Fight Against Cancer” repeats a common misunderstanding about reporting adverse events to an institutional review board (IRB). The article states that every time a patient has an adverse reaction that participating medical centers needed to inform each study participant and respond to the IRB, even if the reaction is unrelated to the drug and due to disease progression. Nothing could be further than the truth. FDA regulations only require reporting “all unanticipated problems involving risk.”*
Disease progression in a cancer clinical trial can hardly be described as an unanticipated problem. The over-reporting of adverse events to IRBs is a major problem clogging research activities. Reporting each adverse event defeats the purpose of expedited reporting when there really is a problem. Actual unanticipated problems are lost in the mountain of routine reports. There are several reasons for this.
The first is that many IRBs and sponsors of clinical trials incorrectly believe that all adverse events need to be reported. FDA has not had the resources to effectively communicate requirements for safety reporting. Next, there are confusing, multiple definitions for adverse event, adverse experience, and adverse reaction. Finally, FDA’s clinical trial regulations were written in the 1970s, are outdated, and sometimes conflict with the regulations of the U.S. Office of Human Research Protections.
One obvious solution is to give FDA the resources and support necessary to update their regulations and guidance documents, coordinate them with other agencies, and then consistently apply them during inspections.