Data Look Good, Approval Unknown

Curis (Nasdaq: CRIS  ) announced in March that the clinical trial testing vismodegib in advanced basal cell carcinoma was successful, but its partner Roche held the data for a medical meeting. With the data released yesterday, we know how successful, but the risk hasn't been totally negated.

In the phase 2 trial, 43% of patients with localized basal cell carcinoma saw their tumors shrink or their skin cancer lesions healed. In patients with cancer that had spread to other parts of the body, the response rate was 30%.

How did patients who didn't receive vismodegib fare? There wasn't a control group to compare them to. Of course, it's not like spontaneous shrinkage happens all that often in cancer, so we can assume it would be close to zero if there was a control group.

The data should give Curis' investors added confidence that vismodegib works, but whether it can get approved with this limited data remains to be seen.

Roche plans to file for an accelerated approval for vismodegib using the uncontrolled phase 2 data from 104 patients. That's not exactly a strong data package.

Working in vismodegib's favor is that these basal cell carcinoma patients have limited additional treatment options. Basal cell carcinoma is a different type of skin cancer than melanoma, which has seen recent advances from Bristol-Myers Squibb (NYSE: BMY  ) and Roche.

Working against Curis is that Roche's track record for accelerated approvals hasn't been that great lately. The FDA refused to accept the marketing application for Roche and ImmunoGen's (Nasdaq: IMGN  ) trastuzumab-DM1. And the FDA pulled back its accelerated approval for Roche's Avastin in breast cancer after trials failed to confirm the drug worked.

Those weren't necessarily Roche's fault, but they are indicative of the FDA's current conservative attitude toward accelerated approvals.

The good news for Curis: Even in the worst-case scenario, where Roche isn't able to get an accelerated approval, it's pretty clear that vismodegib should show an effect in a phase 3 trial and possibly other cancer types.