Ann: OncoSil AGM CEO Presentation, page-45

tinyteddy-completely agree, BSI are the arbiters. However, when the company slides state they consider modern combinations like folfirinox and gemcitabine + abraxane to be "state of the art" (and proceed to use them in their own study!), and yet those are mysteriously absent from a supposedly "systematic" review and meta-analysis ( which instead almost exclusively includes stuff known to be inferior or not relevant in current practice), then I think people who know a bit about this stuff are entitled to think something doesn't look right.

But let's say for a minute we take on good faith that the meta-analysis is as good and as comprehensive as they could have done. They've plopped out a result- 12.7 months survival and then put panco up against it and gone "brilliant, we got 16 month survival, therefore we're superior to everything". Not so fast. Because by going up against a raft of (in some case very old) studies in their analysis with known weaker/inappropriate chemo choices, isn't the push back that their better result is simply down entirely to the known better chemo they used in panco? So aren't we back to original position that the DEVICE'S efficacy remains unproven ("insufficient clinical benefit" in BSI parlance) and OSL have just shot themselves in the foot? And by the way (and thanks to my statisitican friend for pointing this out) panco's survival result actually has a pretty wide error margin- and could be as low 11 months.

And you know if this type of comparative analysis was really sufficient to get a regulatory approval, why bother any more with randomised conrtrrolled trials? Because in future any drug or device company could run as small a trial they can get away with for their product, produce a given result (a number), then concoct a meta-analysis to get a lower number result and then claim triumph because they've "beaten" everything else in that meta-analysis!