3 Comments

Good paper. Two questions:

1. Is there any evidence that there should have been some survival data in the papers which didn’t report it?

2. Are there any studies which appear to have been “buried” without publication?

Obviously my overarching point here is: are you sure you haven’t performed (in essence) a meta-analysis of a dataset curated to bias the drugs by a pharma-captured publication system?

Expand full comment

Great work! Been looking for something like this. Looking forward to digging into the paper. A bit of confirmation bias here but those are the results I would have expected

Expand full comment

BTW, looking at table 2 in the supplement - it would be interesting to add the OS improvement vs control arm (and whether stat sig)... would be able to see at a glance the IO landscape in one chart... just a thought

Expand full comment