European Regulators to Antigenics: Take Your Subset and Shove It

(MoneyWatch)  News that a European advisory committee won't recommend approval of Antigenics' kidney cancer vaccine Oncophage (vitespen) provides yet another reminder -- in case anyone didn't catch it the first few hundred times -- that regulators don't like retrospective subset data.

Oncophage failed to improve progression-free survival or overall survival in a Phase III kidney cancer trial nearly four years ago. But Antigenics presented a retrospective analysis at ASCO this year showing that a subset of "intermediate risk" patients from the failed study had a 46 percent lower risk of death (p = 0.036).
Subset analyses are fine and dandy if you're still trying to figure out which patients your drug works best in and you've got plans to conduct confirmatory clinical trials. But most regulators won't touch you with a 10-foot pole until those prospective confirmatory trials come up clean. Otherwise, as FDA cancer drug czar Richard Pazdur famously said, you're basically shooting an arrow into the wall and then drawing a bull's-eye around it.

Antigenics knows this -- it didn't submit Oncophage to the FDA. But apparently the company thought the Europeans might be a bit more open-minded, especially since the vaccine is approved in Russia.

Investors apparently thought so too -- they pushed the stock up some 300 percent over the summer. But it came crashing down last week when the Committee for Medicinal Products for Human Use told the company to expect a negative opinion on its approval bid.

What now? Antigenics says it may appeal, given that "there are no approved drugs in Europe or the United States for the post-surgical treatment of adjuvant kidney cancer."

Well, yeah, that's true -- but there are approved kidney cancer drugs like Nexavar (sorafenib, Onyx Pharmaceuticals and Bayer) and Sutent (sunitinib, Pfizer), so Antigenics may have a tough time playing the "unmet need" card.

What's sad is that the drug might actually work in the right patients, but finding out for sure is going to take a 10-year, $500 million confirmatory trial. Ironically though, if Antigenics had started that trial back in early 2006 when the kidney cancer trial failed, they'd be pretty close to half-way done by now.

Target photo by Flickr user gothick matt, CC.