Yesterday's FDA advisory committee meeting on Vivus' weight loss pill proved to us, once again, how incredibly difficult it is to successfully develop and register new drugs for obesity. Mountain View, CA-based Vivus
The good news is that the combination -- tested by Vivus at three dose levels -- was effective, showing a clear dose response that resulted in a median weight loss (at the high dose) of about 15 percent for patients who completed 56 weeks of treatment. That's a serious amount of weight, considering that your average obese patients weigh around 220 pounds or more. Furthermore, the Vivus drug demonstrated some very impressive salutary effects on other endpoints that travel with obesity -- including blood glucose, blood pressure, HDL and LDL cholesterol, sleep apnea, and inflammation markers.
There were some compelling speeches from patient advocates, too. I have to confess, I choked up a bit when one young and striking patient who had been transformed with tremendous weight loss during her trial experience on the Vivus drug tearfully pleaded to the panel to support approval, saying, "We need help. I ... need help." She, like others who were quoted by Kelly L. Close, who edits a journal called diaTribe and spoke passionately for approval -- view obesity as "a nightmare I cannot wake up from."
It's true -- obesity is a lifelong struggle. She and the rest of us need to have good tools in the hands of our physicians to keep weight in check, just like we do for hypertension, high cholesterol, and other chronic diseases. It's very touching stuff, and sobering, too. Especially when you consider that 30 percent or more of U.S. adults are obese, and that the strategies being taken to stop the problem through diet and exercise are about as effective against established obesity as a sponge mop on the oil-soaked beaches of Louisiana.
So what prompted the advisory committee to vote against approval by a resounding 10-6 majority? It's a bit of a story in itself, one that started earlier in the same room, where the fate of the diabetes drug rosiglitazone (Avandia) was discussed and vigorously debated for two days. Vivus' Qnexa was the closing act to follow a bitterly contested decision of whether to keep GlaxoSmithKline's Avandia on the market -- a matter that was in balance due to the question of whether the drug increases risk of cardiovascular events -- events that were foreseen by some and contested by others, but which took tens of thousands of patients to assess. It is painful for everyone to go back to ask if approving a drug was the right decision. Hindsight is a mean master, and the master had just taught the field a serious lesson.
What lesson? I think if one were to characterize the problem with a phrase, it would be that timeless warning made to physicians in training -- "Primum non nocere," or "First, do no harm." Another way to say it is -- be careful, and make sure you don't make a decision you're going to regret. Don't end up in the same tense room 10 years later trying to figure out what, if anything, went wrong.
Yes, there were signals and other concerns with Vivus' treatment, falling into a few main categories inherent to the two molecules packaged together in the drug. First, despite improving blood pressure and lipids, Qnexa increased heart rate by a small amount (on average, 1.6 beats per minute over 56 weeks of treatment) and increased the percentage of people experiencing rapid heart rates. This is a known function of phentermine. There was no escaping that observation, despite very positive effects observed with Qnexa treatment on blood pressure, lipids, and inflammatory markers. The committee expressed clearly that they would prefer seeing a clean profile in patient populations that would be at increased risk of cardiovascular problems. That means more studies, or perhaps a first approval of Qnexa's lower (but still effective) doses.
Second, there is an issue with topirimate itself. The higher doses used to treat epilepsy increase rates of cleft palate, a birth defect, in topirimate-exposed pregnancies. The Qnexa protocols counseled women to avoid pregnancy by using multiple forms of birth control. Well, a few women did get pregnant during the trials and delivered babies -- thankfully without any birth defects -- making it pretty clear that avoiding pregnancy in clinical trials is easier planned than accomplished. To compound the matter, altered attention span and memory loss are additional side effects of topirimate. This might make it harder for women on the drug to remember to take birth control pills, or to adhere to other contraceptive procedures. More work is needed, in the opinion of the panel.
Third, another topirimate-related matter was unresolved. Topirimate inhibits an enzyme called carbonic anhydrase, which is involved in keeping our blood balanced with respect to negatively charged components like chloride, carbonate, and lactic acid. Too big of a "gap" in anion-cation balance leads to bone loss and kidney stones. This needs more work, too, in the opinion of the panel.
These problems, partially studied in the Vivus trials, are of real concern to the panel members and could lead to dramatic and unwanted consequences if a really effective agent were released into the U.S. obesity market -- a market just seething with unmet, pent up demand for therapies. The next months will determine whether Vivus can adequately address these concerns, adjust their proposed label and risk management plan, and win approval.
So are there take-home lessons here?
I've been going to as many of the FDA events as I can lately, to try to learn what matters in drug approval so that plans for our own obesity treatment can be as good as possible. What is the committee looking for? What can we do with our clinical program at Zafgen to give it the best possible chance at success at gaining approval? I'm glad to say I'm encouraged. There are some steps we can take to improve our chances here.
First, we need to stop treating obesity as a single, homogeneous problem. There are important differences between different groups of obese patients. Severe obese patients with sky-high cardiovascular risk profiles may not be the same as metabolically fit obese patients. Asian patients will have different body mass index (BMI) profiles worthy of treatment than Caucasians or Hispanics or African-Americans. Women may respond differently from men. Careful profiling -- and registration in a manner that targets the groups who will benefit the most from a given treatment -- makes enormous sense. Patient registries may be essential to allow the community to collect the kinds of data that can sculpt use of products once they are available. Our medical records in the U.S. are behind the times, relative to other more advanced health-care systems such as Sweden's. We can't readily study treatment outcomes post-market without formal registries. It's time to fix that.
Second, we may need to carefully limit and monitor drug access in this country in a way that ensures that the correct and well-studied patient population gets the drug without leaking across to people who simply think they should have it, or who want it to their own detriment. Obesity is an area ripe for abuse. Having carefully managed access will likely need to be a part of any registration path for new drugs that would be acceptable in the right population but potentially devastating in the wrong people. Companies filing for registration of new molecules should consider these aspects early, and seek counsel of the agencies and leaders in the field.
This advisory meeting was chapter one of a three-part miniseries. There are two more advisory committee meetings planned for new obesity products this year -- for Arena Pharmaceuticals'
More from Xconomy.com:
- Qualcomm Looks to Re-Energize the Brew Faithful at Inaugural Uplinq Conference
- Tesla Rocks the Nasdaq, But Analysts Skeptical Electric Car Company Can Go The Distance
- HP Buys Melodeo, RumbleFish Partners With YouTube, Motricity Declines & More Seattle-Area Deals News
Tom Hughes is CEO of Cambridge, MA-based Zafgen.
