Evaluating the Safety of Diabetes Drugs
Perspective of a Food and Drug Administration insider
- Robert I. Misbin, MD
- From the Division of Endocrine and Metabolic Drug Products, Food and Drug Administration, Rockville, Maryland
- Address correspondence and reprint requests to Robert I. Misbin, MD, 5727 Crawford Dr., Rockville, MD 20851. E-mail: misbinr{at}cder.fda.gov
- ALT, alanine aminotransferase
- CHF, congestive heart failure
- FDA, Food and Drug Administration
- TZD, thiazolidinedione
- ULN, upper limit of normal
Problems had been festering at the Food and Drug Administration (FDA) for years (1,2). In 2004, the abscess finally broke. First came the recognition that widely used antidepressants were associated with suicide in young patients, and then came the withdrawal of Vioxx. Drugs in diabetes and metabolism have not been spared. The obesity drug Meridia and the lipid-lowering agent Crestor are two of the five drugs whose safety was challenged by FDA epidemiologist Dr. David Graham in testimony before a Senate committee (3).
Unique among drug withdrawals is the diabetes drug phenformin, which was declared an imminent hazard by Secretary Califano and was removed from the market in 1977 because of lactic acidosis (4). As was the case with Vioxx, more recent withdrawals of diabetes and metabolism drugs were done by the manufacturers. The obesity drug Redux was removed in 1997 because of heart valvulopathy (5). The diabetes drug Rezulin was removed in 2000 because of liver failure and the lipid-lowering drug Baycol in 2001 because of rhabdomyolysis (6).
But the news for diabetes drugs has not been all bad. The concern about lactic acidosis that accompanied the approval of Glucophage has been found to be greatly exaggerated (7). Despite reports of liver injury that appeared shortly after the approval of Avandia, there is little, if any, evidence that either Avandia or the related drug Actos increase the risk of liver injury over background (8).
One lesson that should be learned from these examples is that it is hazardous to make firm conclusions about the safety of a new drug. The approval of a new drug is based on safety data from a few thousand patients treated under the intense medical supervision of a clinical trial. The very patients who are likely to have side …
