what could go wrong?, page-94

What is your question?

It's abit late now to be worrying about P values from the Phase 2.

The Phase 2 study was considered too small a sample size to show statistically significant improvements in efficiency, and was also designed to measure safety not efficiency ( so the PFS result was very encouraging). The original premise was that HA delivery would decrease toxicities associated with Irinotecan.

As we all know the phase 3 has a primary end point of PFS, and is powered to show a statistical significance from smaller differences. If differences in OS remain similar to the Phase 2 study, then from my understanding that end point could potentially be achieved aswell.

Since PFS is the primary end point, that's all we're really concerned about at the moment( we also won't hear about OS until the final data is released later in the year, or perhaps early in 2015). OS is a tougher end point to meet, however it's a secondary end point in this study, so a statistically significant improvement in OS doesn't need to be achieved in order to apply for FDA approval- I assume the same can be said about the decease control rate...Someone please correct me if I've said anything wrong.