· Data presented at ESMO 2012 for ifosfamide (a pro-drug for pali) further reinforces our confidence. We were encouraged by data presented at the European Society for Medical Oncology (ESMO) 2012 annual meeting in Austria, which showed that ifosfamide + doxorubicin significantly increased PFS vs. doxorubicin alone by almost 3 months (7.4 months vs. 4.6 months; p=0.002). As a reminder, pali is the stabilized active metabolite of ifosfamide but is potentially far less toxic. Importantly, PFS in the control arm of this study (4.6 mos) was very close to the 4.3 month median PFS assumption for the doxorubicin arm of PICASSO 3, which underpins the trial’s powering assumptions. We also note that this trial enrolled younger STS patients in general vs. PICASSO 3, which could have helped boost the PFS time as younger patients tend to take a little longer to progress vs. older patients.
· We view PFS as an approvable endpoint for STS. Originally the protocol for PICASSO 3 stated that PFS was the primary endpoint for accelerated approval, with OS the primary endpoint for full approval. Recently, the protocol was amended to have PFS as the primary endpoint for full approval (with OS as a key secondary endpoint). There has been skepticism among some investors surrounding the risk of gaining approval based on PFS, but we believe this protocol amendment should help alleviate those concerns, although the FDA's decision will ultimately depend on the strength of the data itself. We also note that the FDA approved GSK’s Votrient for STS (in the 2L+ setting) in April last year based on a ~3-month PFS benefit (without a significant OS benefit) in a single Phase 3 trial. We believe a similar PFS benefit for pali could be sufficient for approval (assuming no meaningful safety concerns) and be viewed as clinically meaningful. In addition, we note that four products were approved based on PFS as a primary endpoint in 2011 (Sutent and everolimus for primitive neuroectodermal tumors, vandeti