The first data detailing Russia’s COVID-19 vaccine—nicknamed Sputnik—was published last week (September 4) in The Lancet. Almost immediately, other scientists began to call attention to unlikely patterns in the data, asking for raw numbers to verify the study’s conclusions.
Enrico Bucci, a systems biologist and bioethicist at Temple University, published an open letter on his blog September 7 to draw The Lancet’s attention to suspected data manipulation. While he stresses that the letter is not an allegation, “the presentation of the data raises several concerns which require access to the original data to fully investigate.”
The study reports results of Russia’s Phase 1/2 clinical trial to assess the safety and effectiveness of its vaccine candidate, carried out by Moscow’s Gamaleya Institute. The tests took place...
Researchers carried out two simultaneous studies in two different hospitals, each of which involved 38 people involved in either a Phase1 or Phase 2 trial. During the Phase 1 portion, nine participants received the human adenovirus 26 (Ad26) vector and another nine received the adenovirus 5 (Ad5) vector. In Phase 2, 20 patients were given an initial dose of the Ad26 version followed by a booster of Ad5. The results show that all patients generated antibodies to the virus after 28 days, although the levels were not as high as those seen in other studies, Chemistry World reports. Side effects were said to be common, but mild.
Based on the results of this trial, Russia recently began a Phase 3 trial involving 40,000 participants. Roughly 31,000 people have already been recruited to take part since its launch on August 26, Health Minister Mikhail Murashko tells Reuters. Even without the results of this latest trial, Russia plans to ramp up production of a finished product for distribution beginning this month, a move that has alarmed scientists worldwide.
Bucci and the other cosigners of the open letter noted “strange duplicated patterns in the data, among different groups, observing different variables,” Chemistry World reports. For example, measurements of CD4 T cell counts for nine people in the study closely matched a different set of nine patients who had their CD8 T cells measured. In other instances, multiple participants had identical antibody levels.
“It’s like you enter a room with nine people and you add their ages together and find that that number is exactly the same as the combined weight of those people,” Bucci tells Chemistry World. “It is strange. But we don’t have access to the data and we can’t really assess what is going on.”
In addition, the study design was such that the statistical power to draw conclusions is limited. “They tested six different formulations, basically split among six groups of volunteers,” Bucci adds. “It is like six independent studies in a single article.”
Bucci is not the only one to stress that the study’s conclusions are tentative at best.
Luke O’Neill, an immunologist at Trinity College Dublin, tells Chemistry World that the immune response appeared weak, and that the study lacked a control group. Saad Shakir, the director of the Drug Safety Research Unit (DSRU) in the UK, spoke more plainly. “It is very preliminary. That’s the bottom line. No one has demonstrated that these [antibody] elevations will translate into protecting people”
In response to the letter’s points, Denis Logunov, a deputy director at the Gamaleya Institute, refuted the claims, stating that he provided all the raw data to the journal ahead of publication even if they were not included in the paper’s supplementary material. “The published results are authentic and accurate and were examined by five reviewers at The Lancet,” Logunov says in a statement to Reuters. “We presented specifically the data that was produced [by the trial], not the data that is supposed to please Italian experts,” he adds. Bucci and a number of other signatories are from Italy.
At least one of the paper’s reviewers has defended his analysis of the data. In an email to Reuters, Naor Bar-Zeev, an epidemiologist at John Hopkins Bloomberg School of Public Health, says that “the results are plausible, and not very different to those seen with other AdV vectored products. Bottom line, I saw no reason to doubt the legitimacy of these results over others I have read and reviewed.”
A Lancet spokesperson has stated to Reuters that they are monitoring the situation and encouraging the authors of the Russian study to respond to questions raised in the open letter.