One of the topics discussed at Thursday's FDA vaccine advisory committee was the optimal boosting schedule for COVID-19 vaccines. For myriad reasons, this is not an easy task and no such schedule has yet to be made. The chief reason is because the benefits of boosting are not evenly distributed as the population is not homogenous when it comes to severe COVID-19 risk. Since the vaccines, including the bivalent version, are not very durable in preventing infection, severe disease is the principal metric we should be using to gauge vaccine effectiveness.
Given the rapid evolution of the virus and the waning of immunity against severe disease that occurs in some high-risk individuals, boosting can have major benefits in staving off hospitalizations and deaths -- the chief aim of the vaccines. However, those without risk factors for severe disease -- who have of being hospitalized even in the pre-vaccine era -- gain perhaps a few weeks to months of against mild disease before becoming susceptible to infection again. Also, because their risk of severe disease is low at baseline, these individuals will not accrue the same benefit as a higher risk individual. For this reason, I have always been of the opinion that boosting (with our current vaccine technologies) should be targeted to high-risk individuals -- as is now the case in the and , for instance -- and not universally recommended.
The federal government's recommendations for successive rounds of boosting has been confusing for even those steeped in the field. An evidence-based, predictable schedule is needed. However, that approach should not be forced into a one-size- (or even two-size)-fits-all paradigm. COVID-19 boosting schedules should be exclusively targeted to actual risk of severe disease, not convenience or simplicity of message.
Below, I will explore a few questions surrounding the "annual COVID boosters for all" debate, and discuss what alternative schedules might look like.
Annual Flu Shots vs COVID Boosters
People often ask, "Why shouldn't COVID-19 vaccination become like influenza vaccination with annual universal shots against updated strains?" This question reveals several misconceptions about influenza, the vaccine directed against it, and fundamental differences between the influenza virus and SARS-CoV-2.
Strictly speaking, I don't consider the annual influenza vaccine a booster vaccine. A booster vaccine is given when a person's immune response to the target pathogen has waned and their risk for the relevant negative outcome has risen. Each year, the influenza vaccine almost always involves a partial or complete reformulation of the prior year's shot. This is not because of waning immunity, but because the virus has evolved significantly away from the prior year's vaccine. This is akin to a new vaccine.
The annual reformulation of influenza vaccines is necessary because tracking the evolution of the virus shows that prior vaccines will not be able to prevent severe disease to the degree desired. This phenomenon is borne out by mismatched influenza seasons in which deaths and hospitalizations rise higher than in well-matched seasons.
By contrast, COVID boosters have been, until the advent of the bivalent shot, identical in composition and administered in response to waning immunity. With the notable exception of the high-risk population, they have provided little more than transient protection against infection that may last weeks (versus durability during most of an 8-month season for influenza).
Therefore, annual influenza vaccination involves the administration of a new vaccine while a booster shot is the same formulation being administered again after a period of time. While the bivalent COVID boosters did involve the addition of the mRNA of the spike protein of the BA.4/BA.5 Omicron variants, they also included the prior extinct ancestral strain's spike mRNA. According to the FDA proposal, each spring our agencies will examine circulating strains and target against those; it has yet to be determined whether future updated vaccines will continue to incorporate the ancestral strain, but the advisory committee voted unanimously to use bivalent formulations for primary vaccinations (versus using an ancestral only targeted vaccine).
With influenza, a movement toward precision medicine targets different versions of the vaccine to higher-risk populations based on risk factors for severe disease. For example, the elderly are higher-dose or adjuvanted formulations of the vaccines. These enhanced vaccines better match this group's elevated risk versus the general population.
Though taken for granted now, universal influenza vaccination has only been the norm in the U.S. . Before then, influenza vaccine -- though available to everyone -- was recommended based on a risk calculation. Indeed, that same risk-based influenza vaccination policy is in the U.K. It may be puzzling that I support universal annual influenza vaccination but not boosting of the low-risk against COVID. However, I do not believe the evidence, especially in the age of Omicron, supports the same benefits to general low-risk populations in terms of absenteeism, doctor visits, and prevention of infection for several months as does influenza vaccination of the low-risk.
The overarching point is that there are real differences in the two viruses and the two vaccines; it is not valid to gloss over the differences and artificially harmonize the vaccine schedules.
A Better COVID-19 Vaccine Schedule
As the risk for severe COVID-19 is not uniform, the benefits of current vaccines are not uniform. Some people may require zero boosters because their risk of severe disease is low; the higher risk may require one, two, or more boosters at certain intervals. However, such a schedule must be made with evidence at hand -- better evidence than guided the switch to the bivalent boosters, which are likely not more effective or cost effective than the original boosters.
Even when it comes to those at high risk for severe COVID, high-risk people are not homogenous. A person with isolated hypertension is very different than someone with multiple comorbid conditions, who is also different than someone who has had a bone marrow transplant. Each may have different vaccination needs. I could envision some being vaccinated based on antibody titers or some other biomarker, as is the case with certain other vaccines (e.g., hepatitis B).
An optimized schedule is not a fruitless task and the lackluster booster uptake -- even among the high risk -- should not dissuade the adoption of a proper schedule for boosting based on risk factors. The professional societies of physicians who care for the varied groups of the high risk -- from diabetic individuals to people living with HIV to transplant patients -- would be best suited to develop such schedules reflecting unique risk profiles amongst their patients.
Ultimately, we should seek better vaccines that are more efficacious and durable in the prevention of infection from all SARS-CoV-2 variants.
is a senior scholar at the Johns Hopkins Center for Health Security, and a practicing infectious disease, critical care, and emergency physician in Pittsburgh.
Disclosures
Adalja is a consultant, speaker, and/or advisory board member for GSK, Sanofi, Merck, and Pfizer.