InvestorsHub Logo
Followers 10
Posts 162
Boards Moderated 0
Alias Born 01/20/2018

Re: Doc logic post# 260743

Wednesday, 01/22/2020 12:00:58 PM

Wednesday, January 22, 2020 12:00:58 PM

Post# of 701320
Hi Doc,

I’ve been doing some reading on the different trial designs, specifically with crossover built in. Are you familiar with the Keynote-024 trial?

Essentially Merck designed this trial for patients with NSCLC who progressed to have the option to crossover. Patients were assigned to either The treatment group (Keytruda) or control group (Chemo).

In order to account for crossover confoundment they used three crossover adjustment analysis methods:
1) simplified two stage method 2) rank preserving structural failure time and 3) inverse probability of censoring weighting

Long story short, after using these statistical methods they were able to adjust the control arm OS KM curve.

The result was a movement towards what is typically seen with OS for control, moving the whole curve towards the left and the median OS from 14.2 to 8.7 months. Treatment saw a median OS of 30 months.

Anyway, I was curious if you or anyone on this board was aware of this crossover adjustment in this trial. The updated analysis was published January 2019 in the journal of clinical oncology .

TIA!
Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent NWBO News