An open letter has appeared about the recently published data from the vaccine development effort at Russia’s Gamaleya reseach center. This is of course the one that the government announced had been “approved” before even going through any Phase III trials, an even that I characterized at the time as a “ridiculous publicity stunt” (a description that I stand by). It’s worth mentioning that paper describes two n=38 studies, and it’s not even clear to me if there were more before the government announced its approval. But there may be more than just a ridiculous aspect to this.
The letter notes that the figures in the paper (illustrating antibody responses in the Phase I patients, for example) have unusual features that are difficult to explain. Specifically, there are patterns in the points for the individual patients that repeat, in almost every figure in the paper. Now, some of these are discrete variables, such as the antibody numbers that are the reciprocal of a given dilution factor, it’s still odd to see them lining up just the same way again and again (see image). And the problem shows up in the cellular data, which should be continuous variables – the same patterns seem to repeat inside the figures for different groups.
As the authors of this letter note, the actual numerical data behind all these data points was not provided in the Lancet publication, and that would be very helpful in figuring out what’s going on. But I agree with them that the patterns shown are “highly unlikely” at best. These are the sorts of repetitions that people search for in papers when they’re suspecting fraud, and that (needless to say) is not a good look for a paper of this degree of visibility and interest. I look forward to an explanation.