In contrast to AF's prior statements, that the responder analysis could "produce the appearance of positive results", as is more often the case it has done just the opposite.
The higher bar of the responder analysis in the statistical analysis plan for the US did not reach. But the prespecified European SAP which called for continuous variable on the same data set did in 3 of 4 primary endpoints (Power 1 met on both stair climb power and lean body mass, but Power 2 only on LBM). Still not enough on its own, given prior guidance was the FDA and EMA sought SCP, but such are the facts.
It’s worth noting the massive amount of anemia seen in the Power 2 (predominate Gemzar) trial, which was reported as 45-50%. In Power 1 anemia was reported at 24%. This result alone could account for reduced SCP overall, making any responder delta related to enobosarm an even greater challenge to statistically
verify. My issue on this trial remains the inclusion of 31% ECOG 0 pts. I remain convinced that the right setting for testing SARMs in cacx is in ECOG 1&2, given too many ECOG 0 pts may never develop cachexia.
One note on survival. The reported 20%+ hazard ratio benefit is not something one can just ignore. Yes, it is early. Yes it was just an exploratory endpoint that is not yet significantly different.
But here's the thing. Recent studies in cachexia continue to build the cases that LBM is a powerful and important independent predictor of survival. (Tsai et al. "Importance of lean body mass in the oncologic patient." Nutr Clin Pract. 2012 Oct;27(5):593-8. Also see Martin "Cancer cachexia in the age of obesity: skeletal muscle depletion is a powerful prognostic factor, independent of body mass index." J Clin Oncol. 2013 Apr 20;31(12):1539-47) Today's Ph3 results of GTx are in concordance with the emerging science that says “muscle mass matters”.
Yes, regulators continue to favor functional assessments over LBM. But this is a reaction to poor results on older agents that for example primarily bulk muscle via water weight. Such is not the effect with SARMs.
Cancer cachexia remains a very challenging setting to conduct trials given all the confounding effects of tumor burden and chemo. I continue to believe that enobosarm is an agent with efficacy on LBM and has a higher therapeutic window than just 3 mg. Of course, there are much more potent SARMs out there, including Ligand’s (LGND) LGD-4033.
When you look at the literature on LBM and cachexia in cancer you are mostly looking at the natural consequence of a devastating disorder (cachexia) under unusually harsh (active cancer) conditions. Few, if any, studies have shown that specific improvements in LBM lead to improved outcomes. Loss of muscle mass matters, regaining mass is ?????
You are asking for a major leap of fail by suggesting that increased LBM and outcomes are co-dependent, when all the intervening steps are NS.
AF is entitled to his jig. He guessed right on the direction of the stock. He would have won the honey pot if he was able to own shares or options.
Ironically, when he made his opinion, AF admits he had no clue the FDA sought a responder analysis, as he was thinking both the FDA & EMEA sought the more traditional continuous variable SAP. And there, LBM was a success in both, and SCP in 1 of 2. As such, enobosarm Ph3 is a fail under US sought responders, the same Ph3 results also reject AF’s alt hypothesis that "enobosarm is a placebo".
The test of character continues. If he was intellectually honest, he’d take the high road and admit his prior work was based on errors of math and fact on the Dobs Ph2b data. But I suspect he'll sweep this saga under the carpet in hope all forgive and forget.