HIV and ID Observations

Figure from Edward B. Foote’s Plain Home Talk, 1896.

In case you missed it, last week the FDA granted full approval for the Novavax COVID-19 vaccine. This vaccine, which uses a more traditional protein-plus-adjuvant strategy instead of the mRNA approach of Pfizer and Moderna, is no longer in “Emergency Use Authorization (EUA)” limbo.

Here’s what that means in practical terms:

1. It shows the data the company submitted were sufficiently favorable from a safety and efficacy standpoint to merit full approval.
2. Insurance should now cover it for indicated use — something not guaranteed before.
3. The vaccine should be more broadly available. The limited availability under the EUA meant that some people who wanted an alternative to mRNA vaccines chose not to get a COVID vaccine at all.
4. The company can now legally engage in promotion (not so good) and education (good!). But it’s remarkable how few people even knew that this vaccine existed, so I’ll take this as net positive.

All good news, right? Especially since data suggest this vaccine has fewer side effects than the mRNA options. Plus, there’s a combined Novavax COVID/influenza vaccine in development that could be especially appealing to people eager for seasonal updates.

However, approval alone doesn’t tell the whole story. Much of the news coverage was on the notable limitations on its use — a strong signal we’re in a different phase of the pandemic, where regulatory decisions reflect shifting risk-benefit calculations. Specifically, the Novavax vaccine is indicated only for adults 65 and older, and for people ages 12–64 with at least one underlying condition that increases the risk of severe COVID-19.

And it isn’t just Novavax. These same limitations were soon applied to the Pfizer and Moderna mRNA vaccines as well, a coordinated move by the FDA that highlights a new, more selective era for COVID vaccine policy under new FDA leadership, and one that brings us more in line with recommendations in other countries. The rationale for these changes is summarized in a NEJM Sounding Board authored by Drs. Vinay Prasad and Marty Makary from the FDA.

This narrower indication might sound like a step backward, but it’s worth considering what it practically means: First, these changes reflect a shift in COVID clinical severity from the first 2 years of the pandemic, and not a repudiation of the vaccines’ value from the original studies. The proven efficacy of the mRNA vaccines in preventing severe disease in a population naive to SARS-CoV-2 was one of the most breathtaking scientific advances in my now lengthy career.

But now, nearly 5 years later, with essentially 100% of the population already vaccinated or previously infected or both, the virus causes less severe illness on average. As a result, the risk-benefit calculation for routine COVID boosters in younger, healthy adults has changed.

Second, it doesn’t mean younger people can’t get a COVID vaccine. The CDC list for higher risk conditions is extraordinarily inclusive: a telephone survey of nearly 450,000 randomly selected U.S. adults 18 years of age or older suggests that a high proportion would meet these criteria. Many people in the younger than 65 age group will be eligible.

And if they don’t have any of these conditions? Clinicians can still offer the vaccines off-label, much like they can for HPV (Gardasil) and herpes zoster (Shingrix), two vaccines that also launched with clearly restricted age indications placed by the FDA.

For the record, clinicians and the public alike have intuited this risk-benefit shift in COVID vaccines, and voted with their feet (and deltoid muscles). Uptake of the annual booster vaccines is low, with less than 25% of Americans receiving the shots each year. I know many otherwise quite pro-vaccine providers who recommend the vaccines to their patients with little enthusiasm. It even extends to my ID tribe and to other healthcare workers; as one colleague told me in an email as we discussed the FDA’s shift:

I stopped getting it this year because it gives me side effects that are definitely as bad as COVID itself.

Not surprisingly, the FDA’s new policy has drawn criticism from some in the medical and public health communities, including some of my colleagues. And indeed, they have some very legitimate concerns. These range from the potential for reduced access or coverage among people who still want the vaccine but now must receive it off-label, to broader worries about the message it sends regarding the importance of vaccines in general.

(Notably, the authors of the NEJM Sounding Board highlight the importance of the measles vaccine — an especially timely endorsement given the widely publicized skepticism of the head of HHS about this critically important strategy for childhood health.)

Some have also argued that decisions about vaccine policy belong with the CDC and the Advisory Committee on Immunization Practices (ACIP), not with the FDA, whose primary charge should be safety and efficacy. One vaccine expert told me this conflict reflects a broader problem with our fragmented vaccine governance — a good point. For the record, I very much continue to value ACIP’s expert input, and eagerly await it in light of this FDA decision, noting that they were already considering making similar changes.

A final point of disagreement is the importance of the observational data in informing our decisions. One of my colleagues forwarded me one such study, showing that people who got a seasonal booster were less likely to get COVID and less likely to experience serious hospitalizations. Implied in these data is that people who get the vaccine might be less likely to transmit the virus to others — no infection, no transmission. A further strength of these studies is that they reflect how vaccines perform in real-world clinical settings, not just controlled trials

The problem with observational studies like these is that people who choose to get COVID vaccines undoubtedly differ from those who do not, differing in ways that could easily influence the outcomes of interest. No amount of statistical manipulation can completely eliminate this residual confounding. And the issue of reducing virus transmission, though laudable, did not play a role in the recent approval of the RSV vaccines (which also have FDA-imposed age limits). One doesn’t administer an RSV vaccine to someone who is younger and immunocompetent because they’re spending a lot of time with older adults, or a person who is immunocompromised.

So where does that leave us? As noted by the authors of the NEJM Sounding Board, what we really need is a trial of COVID vaccines in healthy adults.

This should not be a quick study measuring post-vaccine antibody titers, but a rigorously blinded clinical trial that collects both efficacy and safety data. The results would tell us if these vaccines meaningfully reduce symptomatic disease in this lower-risk group, and whether the benefit is worth the potential side effects. Think of how much more informed our counseling of healthy adults would be! My colleague, who opted out last year and is neither over 65 or immunocompromised, would know whether getting the vaccine actually benefits her health — and also know how to counsel her patients.

(Hey, while we’re at it — what about a three-armed blinded trial comparing the available vaccines in older adults? Or in immunocompromised hosts? We can dream, can’t we?)

The FDA’s view on the need for more clinical trials, by the way, is shared by the previous director of the FDA, Dr. Robert Califf — so this isn’t just this new FDA leadership making trouble. Here’s Califf writing in JAMA earlier this year:

As the majority of US residents are declining the opportunity for COVID vaccination and many clinicians have elected not to press the issue, COVID vaccine uptake is now low enough that large RCTs [randomized clinical trials] are feasible to evaluate the efficacy and safety of new updated boosters. These trials could help to further elucidate benefits and risks in the current environment in which most people have some level of immunity due to previous infection, previous vaccination, or both.

Sounds to me like equipoise — just what we want with a clinical trial! A favorable result, with reduced symptomatic disease in the vaccine group compared to placebo, would clearly influence both guidelines and clinical practice. Conversely, if the vaccine has more adverse effects and doesn’t meaningful reduce disease, then this is also important information to know.

Back to Novavax, which fills an important niche. Not everyone wants an mRNA vaccine, whether due to side effects (raises hand!), concerns about the technology, or simple personal preference. Having an alternative that’s now fully approved — with fewer systemic side effects and a path to broader visibility — is good news for both individual choice and public health.

It’s easy to focus on what’s not included in the latest announcements — no guarantee of universal access, no new variant-specific updates, no big press conferences. But maybe the normalization of COVID vaccination policies to be more in line with international standards, including thoughtful guidance and a call for better data, is exactly what we need right now.