For patients
randomized to bevacizumab 7.5mg/kg, the unstratified hazard ratio for
progression-free survival was 0.79, p=0.0318. Median progression-free
survival was 8.0 months in the placebo arm and 8.7 months in this bevacizumab
arm.

The unstratified
hazard ratio for progression-free survival in the bevacizumab 15mg/kg trial
arm versus placebo was 0.72, p=0.0099. Median progression-free survival was
8.8 months in this bevacizumab arm.

Here, median PFS is
not entirely reflective of treatment effect as it is based on only one point
on the curve, whereas hazard ratio is representative of the whole study
period.

In the analysis for
FDA purposes, which is the stratified analysis with data censored for
non-protocol therapy before disease progression (patients with NPT before PD:
3% in placebo arm, 4% in 7.5mg/kg bevacizumab arm and 3% in 15mg/kg
bevacizumab arm), the hazard ratios for PFS were 0.69 for the lower and 0.61
for the higher bevacizumab dose.