I don't think it is likely they will accept bone scan results as the primary endpoint of the study. It's more likely they would use an endpoint like skeletal-related events, which is typically used in bisphosphonate studies. Ultimately, bone scan results are simply a surrogate - if they aren't accompanied by improved survival or quality of life, then the clinical benefit is questionable, IMO.
If Exelixis believes that there is considerable activity against the primary tumor or soft tissue metastases, then I don't see why the FDA would not expect typical efficacy measures as the primary endpoint (which for hormone-refractory prostate cancer, would be overall survival).