FDA briefing holds sharp criticisms about Avandia review

The FDA released the agenda and background material for its controversial two-day meeting to readjudicate the diabetes drug rosiglitazone. The briefing materials, with criticisms from and about FDA officials and others, offered a preview to what may become a contentious event.

The FDA said it assembled the joint meeting of its Endocrinologic and Metabolic Drugs Advisory Committee and the Drug Safety and Risk Management Advisory Committee to review results of RECORD (Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycaemia in Diabetes), a study designed to compare rosiglitazone (Avandia) with other drugs. The reviewers are to consider if regulatory decisions made in 2010 that tightly restricted use of the drug should be modified.

The agenda will address methodology; statistical analyses; the FDA’s inspection of the Duke Clinical Research Institute (DCRI), the contractor who performed the adjudication; new epidemiological data; a summary of a meta-analysis; drug utilization data; and a summary of the Risk Evaluation and Mitigation Strategy program for rosiglitazone and products containing rosiglitazone. The agency acknowledged that “parties inside and outside the FDA have strongly held views on the cardiovascular safety of rosiglitazone.”

Briefing materials include charges from Thomas A. Marciniak, MD, medical team leader of the division of cardiovascular and renal products at the FDA. At the panel meeting in 2010, he described RECORD as “a very complex and confusing study” with missing endpoints, missing information and potential biases. “If consulted in advance, I would have rejected this study design as completely inappropriate and biased,” he said in 2010.

The briefing also detailed reasoning for the readjudication, which is described by other FDA officials as acceptable and not marred by bias.    

In a memorandum dated April 23, Marciniak lists five issues he has with the RECORD adjudication. He claims the DCRI has “poor track records of detecting or reporting problems” and dealing with fraud; that re-adjudication was not independent; that new information is minimal; it fails to provide reassurance on safety; and the study remains flawed.

One section prepared by Karen Murry Mahoney, MD, lead medical officer of the diabetes team in the division of metabolism and endocrinology products, states that Marciniak’s assessment of the trial design is “useful and objective, ” but she takes umbrage at claims of trial misconduct and readjudication misconduct. “Regrettably, the case reviews in his original consult were not blinded; although he states that some type of blinding occurred, it appears that he himself did the redactions, then followed with his own review. This is not an acceptable review procedure if one wants to put forth one’s review work as blinded.

“Furthermore, from the earliest meetings in the original RECORD review process, Dr. Marciniak was openly discussing specific patients, including their patient identification numbers and treatment assignments. Therefore, from the earliest stages, his review was unblinded. Nevertheless, his allegations during the original review cycle were taken quite seriously, and extensive inspections focused heavily on cases provided by Dr. Marciniak as examples of possible trial misconduct. The inspection team did not find evidence of misconduct or fraud.”

Others found fault with what they described as unprofessional language that disparages FDA staff and review units and questions the competence of DCRI and Avandia maker GlaxoSmithKline.

The meeting will be held June 5 and 6 in Silver Spring, Md.

The briefing and attachments are available here. Other meeting materials can be found here.