After an FDA external advisory panel last week provided mixed recommendations regarding Avandia (rosiglitazone), the American Association of Clinical Endocrinologists, American Diabetes Association, and The Endocrine Society issued a Joint Statement for Health Care Providers.
The 33-member panel was composed of the Endocrinologic and Metabolic Drugs Advisory Committee and the Drugs Safety and Risk Management Advisory Committee. Although the advisory panel was dominated by experts in statistics and epidemiology, it included only a few MD’s and fewer practicing endocrinologists.
After a two-day hearing, in which the advisory panel heard 18 presentations with over 500 slides from both external and FDA subject-matter experts and reviewed over 1000 pages of documents submitted in advance, the panel voted 20 to 12 (with one abstention) not to remove Avandia from the market.
According to the joint statement, the key debate revolved around the potential cardiovascular side effects of rosiglitazone. On this issue, the advisory panel seemed divided because of conflicting data on the drug based on the existing studies, which encompassed a variety of different patient populations and study designs.
For example, Nissen and Wolski’s 2007 meta-analysis of 42 studies, which was updated in a recently published study that added an additional 14 studies, showed that Avandia increased risk for MI (but not cardiovascular death). In addition, a retrospective study of the Medicare database by Graham et al. showed an increased risk of death (but not MI) for those over the age of 65 with Avandia compared with Actos (pioglitazone).
The RECORD trial, which was an open-label, prospective randomized control, non-inferiority trial sponsored by GlaxoSmithKilne, showed that Avandia did not have such a risk compared to Active Comparators. This study, which is the longest trial of any drug in diabetes, was scrutinized in detail by the FDA because of design and execution and the fact that it was conducted in a relatively low risk population.
On the other hand, the meta-analyses of studies using data from Active Comparator trials rather than placebo-controlled trials showed no signal of risk from Avandia. In addition, post-hoc analysis of the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) study, a prospective randomized controlled trial in a high risk population and a similar analysis of the Veterans Administration Diabetes Trial (VADT) also showed no increase in cardiovascular risk.
Given this information, “the advisory panel wrestled with the limitations and strengths of the retrospective data analyses that constituted the bulk of the current information, given that there are many precedents in which such retrospective studies were not confirmed in prospective, randomized control trials.” Consequently, the panel showed “considerable difference of opinion about the appropriate interpretation of the data and their implications for patient safety.”
The advisory panel also had a significant discussion comparing Avandia with Actos, since there are no prospective head-to-head trials comparing the two drugs and the comparisons are based on meta-analyses using different patient populations and with different designs. The outcome of this discussion yielded a 19 to 11 vote (with 2 abstentions and 1 non-voting) to allow the FDA-ordered TZD Intervention with Vitamin D Evaluation (TIDE) trial to continue, although it is unclear whether the study will be continued at this time.
Ultimately, the joint statement noted that “the final decision of whether to pull the drug from the market remains with the FDA leadership. Since the outcome remains uncertain, AACE, ADA, and the Endocrine Society told doctors that based on information from the hearings to inform patients:
– Do not stop taking your medication without first discussing it with your endocrinologist or health care provider;
– There are alternative medications for diabetes if needed; and
– Retain good glucose control to avoid short and long term complications of diabetes