@bedger (sorry it won't let me quote your reply properly) As per...

@bedger (sorry it won't let me quote your reply properly) As per my first message, proper scientific process doesn't allow picking and choosing of sub-cohorts to make the numbers work post-hoc, that's introducing untenable bias in the experiment process and doctoring results, which means that approval will be hard to get, but you can look at these results and make some assumptions about what is going on. If you have a p value close to the p = 0.05 threshold on a post-hoc analysis it's not too compelling as evidence of effect because you might be inventing causation from correlation or witnessing something happening by chance. It could be a measurable provable effect, but not enough data to say for sure.

On the other hand, the data from the post-hoc analysis of under 65s with the addition of dexamethasone has a much lower p value and it's most obviously where effort should be targeted on a confirmatory trial. It does appear that Rem-L is working in that sub-cohort with the combined treatment.

Looking at the headline statistics, if they had have made a change to the trial inclusion criteria of patients under 65 then it is likely that the trial would have passed, but post-hoc analysis is a big no no for approval because you can usually find something compelling in the data after the trial (remembering p < 0.05% means there's a 5% chance that the result is a statistical error).

I had a few replies to my initial post dismissing my point of view, I'd encourage you to take a look at posts like this (specifically point 3 in this post):
https://hotcopper.com.au/threads/overwhelming-efficacy-the-stats-required.5579134/page-470?post_id=48366744

where I called out the risk factor of age (as did many other posters examining scientific covid papers) as a problematic issue for observing an effect in covid patients in emerging covid research. This is why the DSMB exists, to steer the trial without unblinding the results - they added an obesity criteria half way along, and they should have also added an age criteria after reading those white papers.

The good news for holders is that Novartis should have the expertise to guide Mesoblast over the line and avoid future mistakes like this, and if they design a good confirmatory trial they could probably see some nice results leading to approval in a few years time.

@Lopez - out but still following with interest.