Wednesday, March 05, 2008

The FDA Needs Your Input on Diabetes Medicines!

On Tuesday, I wrote about the opportunity for people with diabetes to attend the NIH/NIDDK's jointly-sponsored (along with JDRF) "Artificial Pancreas Workshop" in July. But I also mentioned that it was merely the first opportunity for you as patients to contribute something by representing real-life patient needs to our Government-funded agencies.

Another opportunity you should be aware of is that the FDA has just issued a document with Draft Guidance for Industry for Diabetes Mellitus: Developing Drugs and Therapeutic Biologics for Treatment and Prevention, which provides recommendations to industry regarding the development of drugs and more specifically, therapeutic biologics regulated within the Center for Drug Evaluation and Research (CDER) in the Food and Drug Administration (FDA) for new medicines intended to treat this disease.

Biologics aren't simple chemical pills, but more complex proteins that are "grown" in yeast or bacteria cells, rather than simply combining chemicals. Insulin is perhaps the most visible example of a biologic used to treat diabetes, but some newer treatments such as Byetta used in the treatment of type 2 diabetes also falls within this category. The FDA is now seeking public comments on these proposed guidelines.

Why is it that they do such a poor job of soliciting public opinion on these things? I cannot comment, but lets just say that I am communicating the issue to everyone in the Diabetes OC to encourage their participation in the process.

Comments and suggestions regarding this draft document should be submitted within 60 days of publication noted in the Federal Register of the notice announcing the availability of the draft guidance, and most of the details on how to participate are contained there. For questions regarding this draft document contact Ilan Irony at (301) 796-2290.

My Own Initial Observation on the Guidance

One area I am somewhat troubled by is a comment under type 1 diabetes which discusses blinding (or lack thereof) of studies and clinical trials. They note that unblinded, controlled trials may be appropriate in some circumstances, particularly for trials incorporating clearly objective endpoints. Really? I'm not necessarily convinced, and think the FDA needs to provide some more clarity on this issue.

Consider the following, real-life example:

Insulin is typically evaluated based on its ability to reduce blood sugar levels -- therefore, reduced glycemia is the "clearly objective endpoint" (probably measured by hemoglobin A1C) in this case. But why shouldn't a study involving insulin be blinded? The FDA suggests that avoiding hypoglycemia is the reason in its draft guidance. But the fact is that regular blood testing should address this issue, and why would anyone who enters a clinical trial not be testing when they do not know what their blood glucose is going to do?

As I wrote last year, Germany's esteemed Institute for Quality and Efficiency in Health Care (IQWiG) conducted a comprehensive meta-analysis of clinical trials done for insulin analogs and determined that there was no, reliable scientific evidence available of the superiority of rapid-acting insulin analogs over regular insulin in the treatment of adult patients with type 1 diabetes mellitus.

The chief criticism of the research they reviewed was that none of the studies included in their widespread review was blinded, meaning both the patients and the physicians knew which type of insulin was being injected. Think about what this means from a practical standpoint. Without blinding, there is a very real risk that patients, knowing their type of insulin, could behave differently (such as testing more frequently, for example) than they would normally, which would subsequently lead to a bias in the results of the studies. One could make a compelling argument that if a patient tests more often, they are likely to have better A1c's regardless of what type of insulin they use -- therefore rendering those results inapplicable to the diabetes population at large. In other words, the behavior change is responsible for the reduction, not necessarily the drug (in the IQWiG example, the insulin analog) being tested.

The FDA does note that "On the other hand, unblinding can severely limit the interpretability of subjective endpoints (i.e., patient-reported outcomes) that might be incorporated as secondary assessments of efficacy" but it seems to me that the guidance from the FDA needs to be more specific, and they need to note under what circumstances is an unblinded study acceptable vs. when should it be blinded.

Obviously, there are a variety of other elements which also need to be considered, but as people with diabetes who are likely to be directly influenced by this guidance, I would urge you to review it and give the FDA your comments.

Remember, the FDA notes that as a regulatory agency, the FDA publishes rules that establish or modify the way it regulates foods, drugs, biologics, cosmetics, radiation-emitting electronic products, and medical devices -- commodities close to the daily lives of all Americans. The FDA rules have considerable impact on the nation's health, industries and economy. These rules are not created arbitrarily or in a vacuum (or at least, they shouldn't be). They are formed with the public's help, and that doesn't just mean drug company representatives.

By law, anyone can participate in the rule-making process by commenting in writing on rules FDA proposes. The FDA routinely allows sufficient time for public input and carefully considers these comments when it draws up its final rule(s), so make your voices heard!!

No comments: