Hmmm. Interesting. So your assessment was NOT that Leronlimab doesn’t work, but rather that the study wasn’t powerful enough to show that it does work? In other words, if they had increased the sample size, the results would have shown statistical significance? Because as we all know by now:
- The SECONDARY end point DID show statistical significance and;
- The PRIMARY end point, though it didn’t show statistical significance, did show *clinical significance*.
So according to your calculations, a larger sample size would’ve made the Primary end point both clinically and statistically significant? That assertion raises more questions for me:
- It’s generally agreed that the reason the Primary end point didn’t reach statistical significance was not because of sample size, but because of the selected end point. At that point in the pandemic, environmental conditions had reduced the severity of the virus at the same time that medicine had developed treatment protocols with existing therapeutics that proved fairly effective against the mild-to-moderate cases that CytoDyn were studying. So in effect, CytoDyn were trying to show statistically significant effect against symptoms that were already in decline. And in addition, the end point’s measure being a subjective self-assessment by the patient also introduced variation that likely affected the final significance.
- How is it that the FDA approved, and the DSMC gave approval to continue at mid-point analysis, a trial design that never had enough patients to meet statistical significance? I’m no statistician, but it sure seems to me that that would be one of the most basic elements of a trial and one that should be gotten right at trial start. And at the very least they should be able to see a trend in the wrong direction at the mid-point analysis.
News 
Market Data 
Discover 
AI
