Wednesday, February 16, 2011

Going Early Into Humans Doesn't Always Work

It all looked so good in theory. When UK biotech Renovo started out in 1999, hunting for treatments to prevent scarring, the idea was to test the company's drugs in humans very early on, allowing more informed decisions as to whether to proceed with expensive Phase III trials.

Back then, this wasn't called translational medicine, it was called experimental medicine. And this, Renovo's founder and CEO Prof. Mark Ferguson told START-UP back then, was Renovo's hallmark.

Pity it didn't pay off: Renovo on Feb. 11 announced that Phase III trials of its lead scar reduction treatment, Juvista (human TGF-beta3) had definitively failed. Failed on its primary endpoint (assessment of the scars by independent experts) and on its secondary endpoint (assessment by the patients themselves).

It's more bad news for UK biotech, as if we needed any, after Antisoma's latest crash-and-burn. Juvista is dead, so is the company, says one analyst. Bye-bye to the £100 million odd that Renovo had raised since inception (a figure that excludes the $75 million in up-front cash that Shire paid in 2007 for marketing rights to Juvista). Let's just hope e-therapeutics, with its contrasting in silico approach to quantifying the likelihood of successful drug discovery, lives up to e-xpectations.

U.K. biotech lamenting aside, there's more so say about Juvista. This was a compound that, in Phase II 'within-patient' trials (that's to say, among healthy volunteers who had two cuts made, one on each arm, with one receiving placebo treatment and the other Juvista) had shown outstanding results. Thus apart from being "shocking", both to Ferguson and the small posse of perplexed analysts following the beleaguered stock, the Phase III failure in scar revision surgery calls into question Renovo's entire portfolio. "The other products they have for scar revision are all being developed using this control system," says Samir Devani at Nomura Code in London. The problem is that the Phase II model doesn't appear to be representative of clinical situations.

That's hardly groundbreaking news on a sector level: if good Phase IIIs always followed good Phase IIs, Big Pharma wouldn't be buying toothpaste and cough-medicine businesses and biotech wouldn't have lost most of its investors. But in the case of a treatment like Juvista, applied intra-dermally to scars, one might have expected Phase III results to more closely replicate Phase II -- particularly since some of the outstanding Phase II results were in the exact same indication as the Phase III studies, so it wasn't just about the Phase III scars being larger, and not a result of a planned incision.

What went wrong, then? Several analysts on the call announcing the news sounded desperate to identify some data collection mishap that would explain the results so as to make their bullish predictions look less...well, wrong. But Ferguson didn't mince his words: "This was not a near miss, it was a clear failure." He's adamant that there were no technical or executional problems with the trials, either, which involved 350 patients worldwide.

The company now is trying figure out why in some patients, the half of the scar treated with placebo seemed to heal better than the Juvista-treated half. Ferguson doesn't think the treatment's making it worse (how could it?!) but that there's unexplained variation in the scar "where we wouldn't expect any."

That's science for you. Unpredictable. As for Renovo's future: perhaps predictably short. The company is 'considering all options', including winding the company up and returning (some of) the £44 million cash to shareholders.

No comments: