Study: Avandia linked to higher death risk

More bad news for GlaxoSmithKline's Avandia. A new study shows that 65-and-older patients taking the diabetes med are more likely to die than are patients using Avandia rival Actos.

The retrospective trial used Medicare records to compare 14,000 patients on Avandia to 14,000 using Actos, made by Japan's Takeda Pharmaceuticals. The Avandia patients had a 15 percent higher rate of death from all causes, a difference that Derek Lowe at In the Pipeline called "startlingly high." The patients also were more likely to develop congestive heart failure; that risk was about 13 percent higher for Avandia patients.

The one bit of good news for GSK in this study was that Avandia users didn't appear to suffer more heart attacks or strokes than the patients using Actos did. One of the researchers, however, theorized that the higher death rates among Avandia patients could stem from underlying cardiovascular disease. "In much older adults, it is possible if they do have a stroke or myocardial infarction, they might actually die immediately and never make it to the hospital for a diagnosis," Dr. Wolfgang Winkelmeyer, lead author, told the New York Times, "so the excess cardiac events might show up as deaths."

Consumer advocate Sidney Wolfe of Public Citizen said he hoped that this study would be "the last nail in the coffin of this drug." But GSK questioned the study, saying this data doesn't match its own internal research into Avandia's risks. Safety questions about the drug continue to mount, so stay tuned.

- read the NYT story

Suggested Articles

CEPI, which started to help prepare the world for new outbreaks, has awarded Inovio and Moderna money for vaccine work against the new coronavirus.

The real estate impresario that built a chain of upscale drug recovery facilities is now building a gene and cell therapy CDMO near Philadelphia.

The seven-year Astellas venture served as a model for Amgen's recent $2.7 billion tie-up with BeiGene in China—and now it's amping up there, too.