Questions grow over AstraZeneca-Oxford coronavirus vax trials
The company has declined to address allegations that the company was cherry-picking data.
The drugmaker AstraZeneca is pushing back after experts questioned the quality of the company’s coronavirus vaccine tests.
A company spokesperson said Thursday that their vaccine clinical trials were “conducted to the highest standards.” That included oversight by a data and safety monitoring board that found “the analysis … [showed] protection from COVID-19 occurring 14 days or more after receiving two doses of the vaccine.”
The company will also submit the data to a peer-reviewed journal in the coming weeks, the spokesperson added.
Researchers have been raising concerns ever since AstraZeneca and its partner, the University of Oxford, announced Monday that their jab is approximately 70 percent effective. The news was especially welcome since this vaccine could be cheaper and easier to roll out to low and middle income countries than others, but left experts questioning the company’s findings.
The assessment issued Monday was based on data from two groups: One had nearly 9,000 people who received two full doses of the vaccine, which produced a 62 percent efficacy rate, while a smaller group of 2,741 received a half-dose followed by a full dose, which produced 90 percent efficacy.
Experts say this is a problem. For starters, the group that initially received a half-dose of the vaccine in the U.K. did so by accident, the company first told Reuters.
The issue prompted the U.S. government to investigate. Researchers like Moncef Slaoui, the outgoing head of the U.S.’s Operation Warp Speed, have wondered whether the 90 percent efficacy rate could be linked to the absence of anybody over 55 in the second trial group.
These self-reported efficacy rates could stand up to regulators’ scrutiny, but they reveal flaws in a system in which companies declare results via press release.
Paul Hunter, a medicine professor at the University of East Anglia, was among those who raised questions. He said in a statement that regulators had to review final results before anyone could be confident about the efficacy claims of the half dose regimen. He added that subgroup analyses “are always fraught with difficulties.”
The accidental difference in dosing schemes was also one of the potential issues highlighted by Sheila Bird, a researcher in biostatistics from the University of Cambridge’s Medical Research Council.
“This is a considerable complication because [the half-dose/full dose protocol] occurred in error, not [by] design,” she told POLITICO in an email Thursday.
That and other outstanding questions about the trial are what led Bird to refrain from commenting initially on Monday. In her email, she pointed to too little information on the dosing regimens; on how many people were randomized in each group; and on how this “error” affected the analysis.
“Answers await sight of written paper,” she wrote.
For its part, AstraZeneca didn’t answer questions Thursday about whether data from all participants in various clinical trials would be sent for peer review, or just the subgroups mentioned in Monday’s press release. It also declined to address allegations that the company was cherry-picking data to reach the 70 percent efficacy rate, which were detailed in a lengthy Wired article this week.
Among the issues the Wired story cited was inconsistent dosing throughout the trials. Control groups also didn’t receive the same jabs, it noted.
Those questions stem from the company’s statement that it didn’t report data from one large-scale phase 3 trial but reported on a couple of subgroups from different trials around the world. This could possibly leave out half the trial participants in its Brazil trial, noted the Wired article, written by Hilda Bastian, a health consumer advocate.
Perhaps most significant are concerns that the data could have been “cherry-picked” because AstraZeneca did not initially say which subgroups they would report for their meta-analysis of the trial and when, Bastian wrote.
The 70 percent efficacy reports are “unknowable,” she wrote. “We only have numbers on these two regimens, as opposed to everyone in the trials — and how they arrived at those percentages isn’t explained.”
Hunter is among those who sees “arguments in favor” of subgroup analyses. But he also questioned whether the AstraZeneca/Oxford trial was “adequately powered for subgroup analysis” since it doesn’t appear to be in the trial’s design plan.
These questions are soon to be in the hands of regulators, noted Bastian. But she added: “If anything short of that standard is accepted for this vaccine, it will be easy to stoke already widespread fears about corners being cut. A loss of trust would affect more than this one vaccine.”
This article is part of POLITICO’s premium policy service: Pro HealthCare. From drug pricing, EMA, vaccines, pharma and more, our specialized journalists keep you on top of the topics driving the health care policy agenda. Email for a complimentary trial.