A dose of Moderna’s COVID-19 vaccine is prepared for administration at Union Station in Los Angeles on Jan. 7.

Two large COVID-19 vaccine companies are applying to the Food and Drug Administration to have fourth booster shots authorized. Each has a very different approach: Moderna is applying for a shot for all adults and Pfizer wants the shot for those 65 and older. To understand which approach is correct, and who actually needs a fourth shot for COVID-19, however, we have to spend a minute understanding the immune system.

Vaccines generate neutralizing antibodies against the coronavirus in the body. These antibodies, however, can be less effective against new variants like omicron. And even if your body receives a rush of new antibodies from receiving an extra booster shot, those antibodies will wane in the following months.

That doesn’t mean continuous boosting will be necessary for all, however. The vaccines also generate something called cellular immunity, which is much longer lasting and protect many in an enduring fashion against severe disease.

Memory B cells are generated by the vaccines and have been shown to better recognize different variants of SARS-CoV-2, the virus that causes COVID-19, as they emerge. Although we do not know how long memory B cells from vaccination or infection will last in our current outbreak, survivors of the 1918 influenza pandemic were able to produce antibodies from memory B cells when their blood was exposed to the same strain nine decades later.

These vaccines also trigger the production of T cells. While B cells serve as memory banks to produce antibodies when needed, T cells amplify the body’s response to a virus and help recruit cells to attack the pathogen directly. With vaccination, we have COVID-19 specific T cells that protect us against variants, including omicron. Despite omicron’s many mutations, binding sites on the virus known as epitopes were recognized by T cells that then marked the virus for destruction.

Although we do not know how long memory T cells will last, one study that examined participants with varying degrees of initial disease severity estimated that memory T cells generated by COVID-19 infection could very well last for a lifetime. Memory T cells generated in individuals who survived a SARS infection in 2003 have been shown to be long-lasting (at least 17 years and counting) in a recent paper.

T cells also help B cells make new antibodies when needed. Therefore, whether your body needs help generating neutralizing antibodies with further vaccine doses, instead of relying on memory B cell production of antibodies, likely will be determined by your age and any comorbidities you might have.

When your antibodies wane, you can be vulnerable to reinfection — as antibodies are the first line of defense against the virus in your nose. During the omicron variant era, protection from reinfection (but not severe disease across the general population) decreased from either vaccination or previous infection, which is why booster shots were authorized in the U.S. A study from Qatar showed that protection from reinfection after natural infection declined from approximately 90% with the alpha, beta and delta variants to 56% against omicron. The protection from reinfection after two-dose vaccines declined from 89% against delta to 36% against omicron, according to a Canadian study. This is likely due to the ability of omicron to evade antibodies. In both studies, however, protection against severe disease generally endured, likely mediated by cellular immunity.

The general point of vaccination is to protect us against severe disease. Though an increase in antibodies from a vaccine boost may last only a short period (approximately four months according to one Centers for Disease Control and Prevention study), that boost will be necessary for certain groups to maintain protection against severe disease. The need for further doses of the vaccine to boost antibodies will depend on your clinical characteristics (e.g., if you’re immunocompromised or older with multiple comorbidities). For instance, in a recent CDC study, those with severe immunocompromises (patients on B cell depleting therapies or solid organ transplant recipients) and those older than 75 with four comorbidities were susceptible to severe breakthrough infections. These immunocompromised individuals have already been approved for a fourth shot and those who are older and have other medical conditions that put them at risk should obviously be next in line.

A recent study in the New England Journal of Medicine, meanwhile, showed that a fourth shot for health care workers did not improve the effectiveness of the vaccine or reduce viral loads (i.e. decrease infectiousness). That’s because health care workers are generally younger. Older patients, however, are the most likely to benefit from a booster — as requested by Pfizer.

As future variants likely arise, an inactivated whole virus vaccine with an effective adjuvant may be more protective in all individuals against any variants with multiple mutations in the spike protein. Nasal vaccines can help boost immunity in mucosal surfaces, which will help contain viral transmission. And, finally, the durable protection against severe disease from vaccines among younger, immunocompetent individuals to date gives hope for the longevity of memory cell immunity against SARS-CoV-2.

It is not sustainable to give the entire population a booster shot every four to six months, but if you’re older than 65, you should probably start receiving your fourth shot now given the ongoing viral circulation of the BA.2 subvariant. Continued scientific monitoring and rigorous data on who is most susceptible to severe breakthrough infections can tell us how to apportion boosters appropriately to keep the population safe in the future — when the virus is circulating at even high levels.

Monica Gandhi is an infectious diseases specialist and professor of medicine at UCSF.