Ann: Remestemcel-L FDA Fast Track Designation for COVID-19 ARDS, page-555

@6151 has pointed out Left-e gives a very good explanation as to the probably lowering of statistical significance of the ARDS results:

"it's possible to obtain a 43% reduction in mortality but lose statistical significance simply by lowering the baseline placebo mortality rate through improved SoC. This is outside the control of Mesoblast/rem-L. In a 300 patient trial with a placebo mortality rate of 40%, a 43% reduction in deaths in the rem-L arm gives a p-value in the .001 range. I suspect the initial assumptions were 40% mortality in the placebo group, possibly higher. However, lower the placebo mortality rate to 25% (through improved SoC) and the same 43% reduction now gives a p-value above .01. Lower the placebo death rate to 18% and the same 43% reduction with rem-L gives a p-value in the range of .05. The number of lives saved dwindles, the Yates correction begins to kick in due to low values in some cells, raising the p-value even higher, etc. In other words, as the placebo mortality rate drops, larger numbers of enrolled patients are needed to show statistical significance – even if the target 43% reduction in mortality is indeed achieved. It's why I believe they decided not to waste time continuing enrollment. Meanwhile there's a mountain of secondary end-point data percolating on 223 patients, still a sizable number. We could be in for some surprises when those data are revealed, especially the data on ventilator-free days, time in ICU, readmissions, and inflammatory markers".

I do believe there will be some very credible and informative data to be revealed at the 60 day point, but by the same logic above, one has to ask why did SI twice bang on about continued high mortality rates in webcasts. It was clearly misleading - its certainly swayed me in my continued belief in trial success despite a tardy recruitment, knowing a continued high mortality rate in the control cohort raised our chances of success. There have been so many misleading statistical representations in this pandemic that still continue (US infections per capita to date now are probably as high as 1 in 6 to date rather than 1 in 15 popular figures suggest - https://www.cidrap.umn.edu/news-perspective/2021/01/study-us-covid-cases-deaths-far-higher-reported?fbclid=IwAR0On-5ZKJrupulbcU8rHm-ANAWggnmMKI_54JkCMv9z8nJIzFr1FBIo738) I personally put faith in SI's opinion, him being on the ground so to speak (a mistake on my part). SI could not see the trial data of course, but I presumed he was surely well appraised of the situation at ground zero - clearly not.