Advice now, consent later, on COVID-19 vaccines
FDA’s vaccines advisory committee provided some advice on COVID-19 vaccines, but the real test will come when it considers an EUA
FDA’s vaccines advisory committee provided some advice on COVID-19 vaccines, but the real test will come when it considers an EUA.
FDA’s Oct. 22 vaccines advisory committee meeting was principally intended to reassure the public that career staff will make decisions about COVID-19 vaccines and will not be influenced by political preferences or timetables.
A second priority was bringing the Vaccines and Related Biological Products Advisory Committee (VRBPAC) up to speed on FDA’s thinking about criteria for authorizing or approving COVID-19 vaccines, as the committee will be called on to discuss EUA applications for individual vaccine candidates.
About six hours of the eight-hour meeting were devoted to presentations, and FDA did not ask for votes.
Although FDA sought the committee’s feedback about the agency’s guidance on emergency use authorizations (EUAs) and BLAs for COVID-19 vaccines, it is too late to make substantive changes in ongoing Phase III trials.
The most important advice the committee provided was to strongly urge vaccine sponsors to avoid unblinding Phase III trials for as long as possible, and to ask FDA to consider requiring a longer duration of safety data for products based on previously unapproved technologies.
Wary of new technologies
Several VRBPAC members said they would want to see a longer duration of safety data for a vaccine that is based on technology that has not been used in a previously approved vaccine.
David Wentworth, chief of the Virology Surveillance and Diagnosis Branch of the Influenza Division at CDC, said he found FDA’s rationale for requiring a median of two months of safety data “quite strong” in general, but he expressed reservations about mRNA vaccines and asked if FDA had considered requiring a longer follow-up period for new modalities.
Doran Fink, deputy director of FDA's Division of Vaccines and Related Products Applications, responded that FDA “did consider novelty of platform among all the variables in our considerations, but ultimately came out with our guidance” that imposes the same requirements on all vaccines, regardless of platform technology.
Michael Kurilla, director of the Division of Clinical Innovation at NIH’s National Center for Advancing Translation Sciences, and H. Cody Meissner, professor of pediatrics at Tufts University School of Medicine, also urged FDA to consider requiring a longer safety database for vaccines based on technologies like mRNA for which there is little or no previous experience.
Keep the blind as long as possible
The severity of COVID-19 has raised the question of whether it is ethical to continue placebo-controlled trials after a vaccine has received an EUA.
Panel members say the answer is yes. There was a consensus among the group that an EUA based on interim data from a Phase III trial would be acceptable only if there is a high probability that the trial will be completed.
While the committee may be willing to recommend an EUA based on the median of two months of safety data that FDA requested in its EUA guidance, it would not be comfortable doing so unless there was an assurance that several additional months of data would be collected to ensure that efficacy did not quickly wane and that no new safety issues emerged.
Because of the need for a greater level of certainty about vaccine safety and efficacy that can only be derived from a randomized, controlled trial, the committee rejected the idea that authorization of a COVID-19 vaccine based on interim data should cause a vaccine’s sponsor to unblind an ongoing trial.
However, in comments to the advisory panel, Pfizer Inc. (NYSE:PFE) stated that it believes that an EUA for its vaccine would create an ethical obligation to inform patients in its ongoing Phase III trial about whether they received the vaccine or placebo, and to offer those who received a placebo the opportunity to receive the vaccine.
If its vaccine is granted an EUA, Pfizer said it “would propose to amend our ongoing study to allow cross-over of eligible placebo subjects to the active vaccine arm if they wish to do so at any time.”
If patients made the move, it would disrupt the ability to gather long-term safety and efficacy comparisons between the vaccine and placebo arms.
An EUA for one vaccine could also create problems for trials of other vaccine candidates. Volunteers who have enrolled in a trial of a vaccine candidate that has not received an EUA are likely to want to know if they’ve received a placebo, and if so, to drop out of the trial and receive the authorized vaccine. Participants in the vaccine arm of the trial may want to know if it is safe for them to also receive the authorized vaccine.
FDA has seen this problem coming. In its guidance on EUAs for COVID-19 vaccines, FDA stated that “it does not consider availability of a COVID-19 vaccine under EUA, in and of itself, as grounds for stopping blinded follow-up in an ongoing clinical trial.”
The agency suggested that EUA requests “should include strategies that will be implemented to ensure that ongoing clinical trials of the vaccine are able to assess long-term safety and efficacy.”
FDA also envisions the need to unblind trials, stating in the guidance on BLAs for COVID-19 vaccines: “Efficacy trials should include contingency plans for continued follow up and analysis of safety and effectiveness outcomes in the event that a safe and effective vaccine becomes available (e.g., as demonstrated in a planned interim analysis or as demonstrated in another clinical trial). In that case, discussion with the agency may be necessary to address ethical arguments to break the blind and offer vaccine to placebo recipients.”
The BLA guidance proposes one way the authorization or approval of a vaccine could affect subsequent trials of COVID-19 vaccine candidates. “If the availability of a COVID-19 vaccine proven to be safe and effective precludes ethical inclusion of a placebo control group, that vaccine could serve as the control treatment in a study designed to evaluate efficacy with non-inferiority hypothesis testing.”
It is far from clear, however, that non-inferiority trials of COVID-19 vaccines would be feasible. Moreover, the difficulty of doing this could preface pressure on FDA to accept external control arms or real-world data as controls.
Demonstrating non-inferiority to an effective intervention can require very large trials or the acceptance of large confidence intervals around the results. The placebo-controlled Phase III COVID-19 trials already include at least 30,000 participants, and there will be little acceptance of uncertainty about the efficacy of a vaccine to prevent a life-threatening disease.
Speaking at the VBRPAC meeting, FDA’s Fink said that “non-inferiority trial designs require much larger sample sizes than placebo-controlled trials so feasibility will be an issue.”
Pfizer urged FDA to consider the use of real-world data to substitute for missing placebo-controlled data. It asked the agency “to be open to other scientifically and statistically sound methods to assess the long-term effectiveness and safety follow-up of recipients of our vaccine candidate (e.g., comparison of standing safety cohorts and registries).”
In its comments to FDA, the Janssen unit of Johnson & Johnson (NYSE:JNJ) expressed concerns about the effects of an EUA on its vaccine program, which is behind Pfizer’s in the clinic. Janssen said it is “evaluating a number of measures aimed at maintaining Phase 3 trial retention, should a COVID-19 vaccine become available before our trial is complete,” it stated in comments to FDA.
Janssen urged FDA and vaccine sponsors to work together to develop plans for ensuring that the unblinding does not create a de facto monopoly for the first vaccine to receive an EUA.
Janssen raised the prospect of using shared placebo data or external controls to mitigate the loss of placebo-controlled data. “Future FDA workshops on COVID-19 vaccines could include a discussion on collaboration across manufacturers on rules about unblinding ongoing trials and potentially sharing data on participants in the placebo group, and statistical approaches to analyzing results in the event that there are considerable cross-overs from the placebo arm to the treatment/vaccine arm (external controls, self-controlled designs, etc.).”
Preaching to the choir
Committee members also spent significant time discussing two topics that are unlikely to influence FDA or sponsors: the need for diversity in clinical trials and concerns that the trials are using the wrong efficacy endpoints.
When it comes to the importance of recruiting diverse clinical trial populations, VRBPAC was preaching to the choir.
FDA has emphasized the importance of enrolling Black, Hispanic and other minority groups in COVID-19 clinical trials.
Paul Offit, who serves on both the VRBPAC and a working group of NIH’s ACTIV consortium that is overseeing COVID-19 vaccine clinical trial designs, told the committee that diversity was an explicit goal and that companies have been diligent about achieving it. Offit is a professor of pediatrics at The Children’s Hospital of Philadelphia.
Vaccine companies are also on board. For example, Moderna Inc. (NASDAQ:MRNA), which delayed its trial to increase diversity, said 37% of its trial participants are from minority populations, including 10% who are African American.
The endpoints ship has sailed
Several VRBPAC members suggested that FDA’s decision to allow trials to use avoiding any symptom of COVID-19 was a mistake and instead the endpoints should be avoiding serious symptoms.
Fink argued that the symptomatic disease endpoint is sensitive. “There are multiple examples of vaccines where the data do appear to show they are most effective against more severe disease, less so against less severe disease, and even less against asymptomatic infection.”
Fink also said that it would not be feasible to require a vaccine to prevent serious disease because trials would have to be extremely large.
The acting VRBPAC chair, Arnold Monto, told the committee that “most vaccines are licensed to prevent lab-confirmed disease.” Monto, a professor of epidemiology at the University of Michigan School of Public Health, added that “things that prevent lab-confirmed infection typically prevent serious disease.”
In any case, when it comes to endpoints for COVID-19 vaccine Phase III trials, the ship has already sailed.
At the end of the meeting, Marion Gruber, director of the Office of Vaccine Research and Review at FDA’s Center for Biologics Evaluation and Research, said discussion on endpoints could not have any practical effect on ongoing trials. “We can take [the committee’s concerns] forward if we have new vaccines entering clinical studies, but it may be a little difficult for those that are already in Phase III.”