CheckMate-227:
Agreed completely. If much higher percentage of randomized patients - let’s say >90% - were tested for TMB, I’d tend to give the benefit of doubt of combo result. As of right now, TMB is an exploratory marker that needs to be confirmed in a prespecified trial. Look at failed CheckMate-026, BMY post hoc TMB from that trial got nivolumab vs chemo PFS HR=0.62 in just over 100 patients, CheckMate-227 post-randomization for the same TMB got nivolumab vs chemo PFS HR=0.95 with more patients!
It is amazing to me how size of company like BMY get to this point from the already messy CheckMate-227 I posted before. BMY highlighted nivolumab combo with chemo in CheckMate-227 at last year ASCO. If they don’t release nivolumab vs chemo in PD-L1+ and nivolumab + chemo vs chemo in PD-L1- results at some point, I’d assume results are indeed inferior to Keynote-042 and Keynote-189. This doesn’t bode well for 2nd part of CheckMate-227 nivolumab + chemo vs chemo in unselected patients. Why does this matter this late for 1st line NSCLC? Because it will CONFIRM nivolumab is indeed inferior to pembrolizumab in 1st line NSCLC, and it is hard to believe this will be the only indication nivolumab is inferior to pembrolizumab. We will have many more readouts from 1st line ph3 trials in next 18 months from both nivolumab and pembrolizumab. I don’t think the market is pricing nivolumab is indeed inferior to pembrolizumab in 1st line NSCLC and more.
One minor correction to your post, OS secondary endpoints for nivolumab + ipilimumab vs chemo is in TMB>=10 mutations/Mb population, only nivolumab vs chemo is in PD-L1+ and TMB>=13 mutations/Mb population.
