That is problematical. The trial randomized 232 to -L+SOC and 99 to SOC. The 233 did not live longer than the 99. That in itself is such a huge red flag that it stuns me that anybody cannot see it. But let us drill down for the details.
Yes, they blame crossover for the lack of shown efficacy, But making excuses for failure is, well making excuses for failure.
Then we have the formal problem that they did not change the analysis plan until after they knew the predefined primary PFS and secondary OS endpoints failed. That is a huge issue called multiplicity. But whatever,
Then they use the comparison to an ECA that is summary data scraper off published papers. Both FDA and EMA laugh at this for good reason.
And last. Even against the ECA there is a huge issue with patient selection. The JAMA paper notes that patients in the ECA with near complete resections live long, just as long as the -L patients with near complete resections. And the -L trial required intent for total resection.
Market Data 
Markets 
AI
