Biotech Values

[AlpineBV_Miller]

Dr. Scher talked about this tangentially and noted there need to be standards set. That said, there are two distinct issues.

The first issue is drug activity. With no standard for assessment of the bone scans, a company could take advantage and oversell the data.

With EXEL, it's a little hard to oversell these data. The scans are remarkable and every doc I've talked to with experience using XL184 says the benefits of the drug are obvious, happen quickly, and disappear quickly when the drug is withdrawn (RDT has a randomized component that deprives non- soft-tissue progressors of drug). Anecdotal information suggests crossover after placebo in the RDT results in obvious, quick drug effect.

Osteoclastic (CTx) and osteoblastic (ALP/ALK) bone markers are largely in the same direction as the bone scans. Lessening of anemia and pain are also clinically-relevant associates of bone disease.

Finally, there is no experience in PCa of spontaneous bone disease reduction. These patients don't get better without intervention. Furthermore, there is no intervention that has done this previously -- which is one reason why nobody has bothered to create a standard for bone scans.

So from a "is mgmt lying/overstating" view of the issue of no standardized definition of response, I think this isn't an issue.

The second issue is regulatory path. This is a problem. Bone scan progression has been the death of many PCa drugs. Reading the Clinical Trials Working Group recommendations on this is instructive. They recommend ignoring initial bone scan progressions when adjudicating response. It's not until two and sometimes three scans later under these guidelines that you confirm bone scan progression (and then backdate progression to the initial date, incidentally).

There are many reasons for this, but the simple explanation is scintographs target osteoblastic bone activity -- which can be tumors or natural bone remodeling. Newer scanning tech (PET) will help here, but are new enough that there are also no standards.

Coming into the FDA with endpoints that create standards is risky business. This is what EXEL will be doing with XL184. IMO, this is where the risk is -- i.e. regulatory, not efficacy. Regulatory risk is nothing to sneeze at when entering Pazdur's house.

So while I might not agree completely with how Dew got there, I agree caution is needed.