The FDA should overhaul its approval process for new antidiabetics, argues Robert Misbin, from the FDA’s Division of Endocrinology and Metabolism and the medical officer who initially reviewed Avandia’s (rosiglitazone) application. Misbin’s commentary in the December issue of Diabetes Care envisages a new paradigm for the clinical development of drugs for type 2 diabetes.

Misbin draws several lessons from the ongoing controversy surrounding rosiglitazone and an earlier controversy surrounding the related drug muraglitazar. For example, just because a drug lowers glucose levels more effectively than placebo that does not necessarily mean it will provide long-term benefit for patients with type 2 diabetes. Furthermore, the American Diabetes Association recommends that patients with type 2 diabetes should generally receive metformin at diagnosis, making long-term placebo monotherapy difficult and unethical.

The article also notes that the FDA approved Rezulin (troglitazone), which was subsequently withdrawn because of a high risk of liver failure as monotherapy and in combination with insulin or sulfonylureas, but not metformin. Rosiglitazone’s initial approval was add-on therapy to metformin. Misbin, who stresses that the views in the article are personal and do not necessarily reflect FDA policy, suggests that a desire to fulfil a market niche seems to have been a more important factor driving rosiglitazone’s development than differences in pharmacology.

Agency 'partly responsible' for pharma seeking marketing niche
The article suggests that the FDA is partly responsible for companies seeking to fill a marketing niche. Currently, companies need to submit clinical trials for each situation – monotherapy and each combination. However, in the paper Misbin points out that there are no examples of approved drugs for diabetes that are effective as monotherapy or in combination with metformin, but that are ineffective when combined with other medications. So efficacy trials for each combination may be unnecessary. However, adverse reactions can emerge in some situations but not others. For example, congestive heart failure emerged in the study of rosiglitazone with insulin, but not as monotherapy. Therefore, combination studies with insulin should focus on safety.

Misbin’s paradigm envisages, firstly, a placebo controlled comparison lasting between six and 12 weeks, the maximum he recommends given the issues mentioned above. Secondly, a monotherapy trial, lasting between 6 and 12 months, that compares the new drug with metformin. The third trial should compare add-in therapy with new drug against placebo in patients whose hyperglycaemia is not adequately controlled with metformin. Again this study should last between 6 and 12 months.

The final study, also lasting 6 to 12 months, compares the new drug with placebo in patients receiving insulin with or without oral agents. Researchers should adjust the insulin dose so that there is no significant difference in the change in glucose levels. Any safety issues that emerge are, therefore, attributable to the new drug and not glycaemic control. In addition, pre-marketing safety trials should not exclude patients at high-risk of developing adverse events.

Extensions to these four studies should continue while the FDA reviews the NDA. The company should submit safety updates to the FDA before approval. As a condition of approval, post-marketing surveillance should last at least two years and include extensions to the four studies and, where appropriate, additional trials to examine safety issues that emerged during review.

In the article, Misbin suggests approval if the new drug is more effective than placebo and non-inferior to metformin. In additional, the new drug’s profile of serious adverse events should be non-inferior to standard therapy and the novel antidiabetic should not show other serious safety problems. The paper suggests that the FDA may approve a new drug that is less effective than metformin, particularly if the antidiabetic offers a novel mechanism of action and is free of safety concerns.

“New drugs should continue to be developed to reduce the global burden of diabetes and its complications,” the paper concludes. “Regulators should set high standards but should also be pragmatic. Fear of uncertainty should not be allowed to stifle innovation”. Mark Greener