September 16, 2010
Thank you for the opportunity to testify as president of the National Research Center for Women & Families.
Our Center is dedicated to improving the health and safety of adults and children, and we do that by scrutinizing medical and scientific research to determine what is known and not known about specific treatments and prevention strategies. We do not accept contributions from companies that make medical products, so I have no conflicts of interest.
In addition, I am a fellow at the University of Pennsylvania Center for Bioethics, and a board member for two nonprofit organizations that work to improve resources for the FDA: the Alliance for a Stronger FDA, and the Reagan Udall Foundation.
My perspective is as a researcher. I was on the faculty at Yale and Vassar, was trained in epidemiology at Yale Medical School, conducted research at Harvard, and I have worked on FDA policy issues in the U.S. Congress, the Institute of Medicine, and for nonprofit organizations.
The big question facing you is: has this drug been proven safe and effective, and if not, what studies are needed to determine if the FDA should approve it in the future? The FDA has suggested a long list of possible risks, and asked you to determine if more research is needed to examine those risks.
The results for Lorcaserin, as presented in the information provided by the sponsor and the FDA, clearly do not merit approval. Additional research might change that conclusion, but it should be pre-market research, definitely not post-market research.
Lorcaserin was administered in conjunction with behavior modification, and as a result, the placebo group – which had behavior modification but no drug – does quite well in losing weight. The addition of Lorcaserin helps, but not that much. Patients taking the drug averaged only about a 3% difference in weight loss compared to placebo and therefore did not meet the FDA’s criteria of at least a 5% difference in weight loss.
Remember that about half the patients in the study dropped out, whether they took placebo or lorcaserin. Patients who lost less weight tended to drop out, so the weight loss statistics for patients at the one-year mark are considerably higher than the actual average weight loss for all patients. The patients who dropped out of the study lost less weight, and then often gained it back by the one-year mark.
One in four in the placebo group – which had behavior modification only – lost at least 5% of baseline body weight after one year in the study. That’s quite impressive. That percentage didn’t quite double for the Lorcaserin BID group. FDA concludes that it barely met the categorical efficacy criteria that were set in an FDA draft guidance for weight management products.
I want to praise Arena Pharmaceuticals for including subgroup analyses of African-American and Latino patients. We don’t see that as often as we should. It is especially important in obesity studies because of racial and ethnic differences in obesity, and the diseases caused by obesity. Unfortunately, minority patients were less likely to lose weight and less likely to stay in the study for 12 months, compared to White patients.
It is important that the company continued one study into a second year. The placebo group remained on placebo for the second year, and half the lorcaserin patients continued the drug while the other half were switched to placebo. The focus of the analysis was on the patients who had lost at least 5% of their body weight during the first year, to see if they maintained that weight loss for the second year. Two-thirds of the drug group kept their 5% loss in the second year compared to only half the patients who were switched from the drug group to the placebo group. That’s disappointing because so many of the patients stopped taking the drug during the first year of the study or stayed on it but didn’t lose at least 5% during the first year. In other words, even the small and especially successful group of weight loss patients from the first year couldn’t keep their weight off for two years.
In contrast to the modest benefits of lorcaserin compared to behavior modification alone, especially for people of color, the health risks were quite impressive. Most frightening was the increase in valvular heart disease, which is the condition that Fen-phen caused and why it was removed from the market. Although the increase in VHD was small during the first year, it is such a dangerous condition that more research is clearly needed. It is certainly possible that more people stopped taking the drug early because of adverse reactions than would be the case in the real world. That would bias the results.
Neuro-psychiatric and cognitive adverse events were reported by 21% of patients taking lorcaserin compared to only 12% taking placebo. These symptoms included dizziness, fatigue, parethesias and abnormal dreams. Patients taking the drug were twice as likely to report adverse events categorized broadly as depression. And, although rather rare, memory impairment, disturbance in attention, amnesia, and other cognitive problems were three times as likely with patients taking the drug, compared to placebo. These are all debilitating symptoms.
Malignant tumors were found in rats taking lorcaserin at a dose that was within 7 fold of the proposed clinical dose of 10 mg BID. It’s hard to know how to interpret those animal findings, but more research is clearly needed. Any study of cancer in humans would need to be much longer-term than either the BLOSSOM or BLOOM trial.
In summary, the studies are an important reminder that behavior modification is a safe and effective alternative to diet pills. Lorcaserin does not appear to be effective for most patients, and especially not for most African-American or Latino patients. More extensive research is needed on the risks. Until research is conducted to determine who might be most likely to benefit from lorcaserin and who is least likely to be harmed, and until we are able to understand those risks and benefits more conclusively, this drug should not be approved by the FDA.