08-27-2020, 02:04 PM
https://www.fiercepharma.com/pharma/pfiz...ew-analyst
The Pfizer/BioNTech vaccine is halfway enrolled through its clinical trial, and at least one person's back-of-the-envelope calculation is that they could get some early data by the end of September.
It looks like they are probably going to have an iDMC take a look at an interim cut of the data (otherwise, how do you reconcile reaching full enrollment by the end of "next month" with an analysis in mid-September). If so, it would be a misnomer to say that the "first dataset should be released". What I think would happen is that a committee of experts would view the data. Let's say at that point there are 20,000 patients enrolled (10K in each arm), and in the study there have been 25 events (given that there are about 40,000 new COVID diagnoses a day, the probability that someone is diagnosed with COVID is about 1 in 8000, so right now we are probably seeing an average of 2 patients enrolled in the trial getting COVID-19 a day. So, it would be reasonable to expect about 40 cases by mid-September when the iDMC happens. If you saw 90% efficacy such that the infection rate was 0.40% (40 in 10,000) in the placebo group and 0.04% (4 in 10,000) in the treated group, the iDMC could conclude there is overwhelming evidence of efficacy and submission for an EUA would be warranted. On the other hand, if the vaccine efficacy were more like 60%, the p-value would be < 0.05, but the lower bound of the confidence interval would be about 22%, in which case the iDMC would probably say to continue the trial until it reaches its endpoint.
BC
The Pfizer/BioNTech vaccine is halfway enrolled through its clinical trial, and at least one person's back-of-the-envelope calculation is that they could get some early data by the end of September.
Quote:Mizuho Securities analyst Vamil Divan figures that vaccine timeline can only benefit Pfizer. If the company is already halfway to its goal of recruiting 30,000 patients in the mRNA vaccine trial, he figured, the company should “reach full enrollment by end of next month,” Divan wrote in a note sent to investors Thursday. That means the first dataset should be released in mid-September, potentially at an investor event the company has scheduled for that time.
It looks like they are probably going to have an iDMC take a look at an interim cut of the data (otherwise, how do you reconcile reaching full enrollment by the end of "next month" with an analysis in mid-September). If so, it would be a misnomer to say that the "first dataset should be released". What I think would happen is that a committee of experts would view the data. Let's say at that point there are 20,000 patients enrolled (10K in each arm), and in the study there have been 25 events (given that there are about 40,000 new COVID diagnoses a day, the probability that someone is diagnosed with COVID is about 1 in 8000, so right now we are probably seeing an average of 2 patients enrolled in the trial getting COVID-19 a day. So, it would be reasonable to expect about 40 cases by mid-September when the iDMC happens. If you saw 90% efficacy such that the infection rate was 0.40% (40 in 10,000) in the placebo group and 0.04% (4 in 10,000) in the treated group, the iDMC could conclude there is overwhelming evidence of efficacy and submission for an EUA would be warranted. On the other hand, if the vaccine efficacy were more like 60%, the p-value would be < 0.05, but the lower bound of the confidence interval would be about 22%, in which case the iDMC would probably say to continue the trial until it reaches its endpoint.
BC