Opinion | Should Future COVID Boosters Include the Ancestral Strain?

The current COVID-19 mRNA vaccine boosters are bivalent, meaning they contain equal amounts of spike protein mRNA for the original Wuhan virus sequence and for the BA.4/5 Omicron variants that emerged in mid-2022. How well does this type of bivalent booster perform?

The question is relevant because FDA’s recent Vaccines and Related Biological Products Advisory Committee (VRBPAC) meeting discussed COVID-19 vaccine composition ahead of key decisions to be made later this year. For now, the FDA advisors voted to harmonize vaccine formulations; the same bivalent vaccine currently authorized will be used for both primary and booster vaccinations. But should any new vaccine continue to include the ancestral strain’s sequence? Or should it be based only on an Omicron lineage variant?

A Review of the Latest Evidence

We now have better data from a large, randomized, controlled trial conducted in the U.K. in early 2022 during the first BA.1 Omicron wave. Early in the trial, study participants were given a BA.1 monovalent booster or the original monovalent booster. But over the course of the trial, the virus continued to mutate in the real world and new Omicron subvariants emerged. It was hypothesized that a bivalent booster targeting BA.1 and the ancestral strain would elicit better cross-protection to the subvariants, and thus in the second half of the trial, study participants were given a BA.1 bivalent booster or the original monovalent booster.

Although the monovalent and bivalent COVID vaccines studied in this trial were targeted at the BA.1 Omicron variant, not the later BA.4/5 Omicron variant, the results are highly relevant. Note that although the total dose in bivalent vaccines is identical to that of the monovalent vaccines, the current formulations deliver only half a dose of vaccine against the Omicron subvariants.

The main clinical outcome, a secondary endpoint in the study, was COVID-19 as defined in the COVE trial. The BA.1 monovalent vaccine appeared to be the most effective booster of the three, followed by the BA.1 bivalent booster, and then the original monovalent booster. However, the differences were small. Relative vaccine effectiveness with the BA.1 bivalent booster was a non-significant 11.4% (95% CI -10.2 to 28.7) when compared with the original monovalent vaccine. Immunological data paralleled the clinical data. The monovalent BA.1 booster elicited marginally higher neutralizing antibody (NAb) titers against the BA.1 subvariant than did the bivalent booster, as did the BA.1 bivalent booster as compared to the original monovalent booster.

This randomized controlled trial demonstrated that the BA.1 bivalent booster is only marginally better at preventing COVID than the original monovalent booster, and that the best performer, although not by much, was the monovalent BA.1 Omicron booster. These findings are in line with what we suggested last year. Furthermore, the BA.1 bivalent booster elicited only a 1.5 times higher NAb titer than did the original monovalent booster, which is concordant with a body of literature on bivalent boosters that we previously summarized, as well as with two recent papers on the performance of the BA.4/5 bivalent booster. NAbs are the principal correlate of protection for prevention of infection. Consistent with the predictions of an influential model, slightly higher NAb titers had limited impact on the clinical endpoint.

We do not have comparable randomized trial data on the BA.4/5 bivalent boosters, so we may never know with certainty the extent to which they were an “upgrade” over the original monovalent booster. However, in recent months, federal officials and others have claimed that the bivalent boosters are “better” than the original monovalent booster. These statements are based on the results of clinical or laboratory studies performed under non-comparable conditions and not from a randomized controlled trial.

In observational studies, the clinical performance of the BA.4/5 bivalent booster has been assessed and compared with historic data on the original monovalent booster. What these studies show is the beneficial impact of the bivalent booster at reducing the incidence of a clinical endpoint. However, comparisons of how different boosters performed at different intervals since last vaccination or infection are inexact. Longer intervals between booster doses allow for greater time-dependent affinity maturation of NAbs and thus higher NAb titer boosts with an additional dose of vaccine. Such differences in intervals since last vaccination do not arise when the serum samples are collected in a randomized controlled trial, as in the phase III trial of monovalent and bivalent boosters in the U.K., or when factored into the study design. Observational studies may also be confounded by COVID-cautiousness and other behavioral differences between those who do or don’t get boosted. People who get boosted are more likely to make use of other mitigation measures like masking or testing to further reduce the risk of COVID. For example, a recent study from Israel demonstrated instantaneous protection from getting a bivalent vaccine booster over no booster, but we know it takes a few days before one could plausibly expect to see an immunological response.

Putting the Evidence Into Practice

Boosting is useful for improving protection against infection, which is particularly important for people at serious risk of severe COVID-19 due to age or underlying medical conditions. The current BA.4/5 bivalent booster performs adequately for this purpose. To the general public, it is a moot point how much “better” the bivalent booster is compared to a monovalent vaccine they can no longer obtain. However, moving forward, we need to understand how well the bivalent boosters work because the FDA will soon be discussing further changes to booster composition.

We argue, again, that there is no value to continuing with bivalent boosters — a monovalent Omicron-based vaccine is currently a better option. We also suggest that switching in a few months from targeting the BA.4/5 Omicron subvariant to a new, yet-to-emerge Omicron subvariant would yield only a small incremental improvement in either NAb titers or clinical performance. Another Omicron vaccine boost, of any composition, would act as a “dose two” for people who have received the BA.4/5 bivalent vaccine, and thereby induce broadly active, affinity-matured NAbs to Omicron-specific epitopes (the bivalent vaccine has already done this in people who were also Omicron virus-infected last year).

“Better” can mean what people want it to mean. A 6-foot 6-inch-tall basketball player may be 2 inches taller than one who is merely 6-foot 4-inches tall, but does the extra 2 inches make him a better player worth a more lucrative contract? There is more to the sport than height, and there is more to vaccine performance than a marginal — and often inappropriately exaggerated — increase in NAb titers. Vaccine composition changes are very expensive. At a time when federal resources for COVID-19 are becoming increasingly stretched, the FDA needs to give serious thought to using the available funds most wisely.

John P. Moore, PhD, is a professor of microbiology and immunology at Weill Cornell Medicine in New York City. Céline Gounder, MD, ScM, is an internist, infectious disease specialist, and epidemiologist; a senior fellow and editor-at-large for public health at the Kaiser Family Foundation and Kaiser Health News; and host of the “American Diagnosis” and “Epidemic” podcasts.