At the interim analysis, molnupiravir reduced the risk of hospitalization or death by approximately 50%; 7.3% of patients who received molnupiravir were either hospitalized or died through Day 29 following randomization (28/385), compared with 14.1% of placebo-treated patients (53/377); p=0.0012.
That p value tells you that the sample size was sufficient. If there was a smaller difference between treatment and placebo, they may have needed a larger sample size to still show statistical significance.
Guess what? MSB trial also had impressive p values, even with small sample sizes, on patients under 65. Possible because baseline mortality was high, and difference between treated and controls was large. Problem is it wasn't in primary endpoint with 30 day mortality, but instead in secondary endpoints at 60 days and 90 days. I am not an expert on multiplicity but won't deny it could have played a role in request for another trial. Still, we're not talking about any old random secondary endpoint, but a longer term mortality endpoint. Does anyone really expect the beneficial effect to disappear when we have a 60 day mortality primary endpoint in a new trial, where we look at Dex vs Dex+Rex?
Some argue post hoc data dredging, but the under 65 analysis was prespecified and sample size was still large enough to show statistical significance. Granted, p value was p.048, as opposed to the really great p values of around 0.01 or less that we would hope for.
Analysis with dexamethasone was very impressive, with very low p values <0.01, but post-hoc analysis, nobody is denying that. Still, does this not look incredibly promising for what a new trial will show?
MSB Price at posting:
$1.63 Sentiment: Buy Disclosure: Held
(20min delay)