Skip to main content

Wide World of Vaccines

April 2022

Immune Responses and Clinical Protection After Vaccine, Prior Infection or Both – Deciding on Boosters


 Christopher Harrison, MD

Column Editor: Christopher Harrison, MD | Professor of Pediatrics, UMKC School of Medicine | Clinical Professor of Pediatrics, University of Kansas School of Medicine


Data can be confusing on immune-mediated “protection” induced by SARS-CoV-2 vaccines or prior infection and recommendations on whether, when and which people need how many booster doses. Experts and nonexperts alike arrive at different conclusions about added benefit to mRNA vaccine boosters (third or fourth doses) or to vaccination before or after infection and how vaccine should be used to gain better hybrid immunity.

It would be great if we could simply measure antibodies and decide on boosters or vaccination after prior infection if the titer was too low. But no confirmed antibody threshold for protection exists, in part due to a constantly moving target of the evolving variants (looking at you, Omicron). A gold standard antibody level that predicts protection has been confounded because the antivirus effects of antibodies in plasma/serum or monoclonal antibodies are different depending on the variant. Thus, the level of protection during one wave is not reliable in the next wave. Likewise, studies use different definitions for protection (endpoints) and may measure protection at different times after the “last event” (last prior vaccine dose or infection). Establishing a protective gold standard usually requires repetitive comparisons with consistent endpoints over multiple seasons or years in similar populations with a relatively stable virus target. For SARS-CoV-2, just as we get close to a gold standard, along comes a new variant and we need to start over. 

What endpoints do the Food and Drug Administration (FDA) and its advisory board use to make decisions? They look at safety along with immunity and clinical protection from a standardized dose schedule. Note: Having some immunity, e.g., anti-SARS CoV-2 antibodies, does not necessarily equate to disease protection. Also, the endpoint might be protection against infection of any kind (even asymptomatic), only symptomatic infection, severe symptomatic infection, hospitalization or even death. Protection against any infection for more than a few months is nearly impossible for vaccines against viruses that replicate in the respiratory tract, so the FDA may require only robust and long-lasting protection against hospitalization or death.

Immune response as surrogate to protection. While protection requires both antibody and immune memory-cell responses, we usually have primarily antibody data, which can include subsets of antibodies with different roles in protection. Table 1. For respiratory pathogens, a combination of serum or plasma antibodies (IgG mostly) with mucosal antibodies (secretory IgA) should provide the best protection, but measuring secretory IgA is cumbersome and impractical on a large scale, even in research studies. Note: Vaccine does not induce secretory IgA. So, data usually relate to plasma/serum antibodies. Enzyme linked immunoassays (EIA) measure SARS-CoV-2 antibodies to whatever virus antigen/s is/are targeted in the assay. The target antigen/s can be one component, e.g., spike protein, several antigens or a mix of all SARS-CoV-2 antigens.

One problem with EIA-detected serum/plasma IgG to SARS-CoV-2 is that only some is actually protective. In fact, new data suggest that some anti-SARS-CoV-2 antibodies produced during infection can actually be pathologic, by shepherding SARS-CoV-2 virus into monocytes and triggering an explosion of inflammatory products that damage tissue, such as the lungs.1

Nevertheless, EIA-detected anti-spike IgG antibodies are important for SARS-CoV-2 immune evaluations because the spike is the main target of protective responses and is the only viral target in the vaccines currently approved for emergency use authorization (mRNA or viral vector vaccines). Prior infection also induces anti-spike antibodies, so spike antibodies also can include response to either vaccine or prior infection. Confirming prior infection requires detecting antibodies to non-spike virus parts, e.g., nucleocapsid, because current vaccines induce only anti-spike antibodies. Note: NovaVax® will be the first vaccine to use whole killed viruses and also will induce anti-nucleocapsid antibodies.

Total IgG to spike protein could eventually be a reasonable surrogate for protection against infection/disease, because the amount of total IgG (the titer) for the most part parallels anti-SARS-CoV-2 neutralizing antibody (Nab) levels, which do the heavy lifting in antibody-mediated clinical protection.2 Nabs inactivate the virus so the virus cannot infect cells or make copies of itself, thus aborting or limiting infection/disease and better correlate to protection than anti-spike IgG. Nab data are used to monitor new variants for their ability to “escape” or be less susceptible to monoclonal antibodies, or to vaccine-induced or prior infection-induced antibodies. For example, Nab data showed us that the Omicron subvariants are less effectively neutralized than Delta by several monoclonal antibodies that were recently removed from general use. Further data show that the second Omicron sub-variant BA.2 is more resistant than BA.1 to most monoclonal antibodies or antibodies from vaccines or prior infections. So, reductions in Nab activity give us better hints than anti-spike IgG that boosters might be needed.

Clinical outcomes. The FDA also considers clinical outcome data in light of Nab and anti-spike data. That is, they ask whether protection against infection/disease changes over time (at present immunity seems to wane at three to six months since the last immune triggering event). The paucity of clinical data for Omicron disease with three or four vaccine doses is the reason the FDA has yet to recommend three doses for all, or four doses even for the most vulnerable. That said, the inordinate rise in cases and persistent daily death toll motivated the FDA to issue the permissive allowance for the fourth dose in vulnerable people beyond the severely immunocompromised. For the new Omicron BA.1 and BA.2 sub-variants, clinical data are emerging from the U.S.,3 and also from Qatar and Israel where national use of boosters coupled with stringent case/disease monitoring began months ago. These data show definite benefits (cutting breakthrough cases in half for several months) but with protection beginning to wane three to four months after fourth doses.4,5

If you remain a bit uncertain about boosters, you are not alone. Some experts recommend third doses for all and fourth doses for older folks (definition of “older” ranged from >50 to >80 years old) or vulnerable people (modestly immune compromising or other underlying conditions). In contrast, others recommend third but not fourth doses even for vulnerable ages and conditions. Yet others say that boosters every six months are not needed and are unsustainable, or that we should wait for variant-specific boosters in the fall of 2022.

Those arguing against boosters seem to feel that 75% protection against Omicron BA.1- or BA.2-caused hospitalization in older and vulnerable populations is good enough. But that means older/vulnerable folks still have a 1:4 chance of hospitalization during Omicron infections – not very comforting. Add a low (<35%) protection rate against symptomatic Omicron infection six months post last event,1-3 and I can see why the FDA chose to issue the third dose recommendation and a fourth dose permissive status in the U.S.

While many studies are underway, I have surmised the following:    

  1. There is ample supply of vaccine.
  2. Boosters, whether third or fourth doses, have been very safe with profiles similar to the initial two doses. (Qatar, Israel)
  3. Omicron BA.1 and BA.2 are the most contagious variants yet and most likely to escape any existing immunity from vaccine or prior non-Omicron infection.
  4. Persons over 60 years old, or with immunocompromising/underlying conditions, are most likely to be hospitalized or die from Omicron variants, despite Omicron seemingly being less virulent than Delta.
  5. Two mRNA vaccine doses or infection with prior variants alone is not sufficiently protective against Omicron infection in almost any age, particularly at or after six months since the last vaccine dose or infection (as low as none but up to ~40% detectable protection).1-3
  6. Protection from hybrid immunity (vaccine plus infection) lasts longer than from vaccine alone or infection alone, so re-boosting those with hybrid immunity at around six months post last event is less urgent.

Taking these points into account, my sense is that the best available protection against symptomatic disease, hospitalization or death (all goals of mine) derives from hybrid immunity with at least two doses of mRNA vaccine (three or more SARS-CoV-2 antigen exposures). Next best is four doses of mRNA vaccine. After that, either three doses of mRNA vaccine, or infection plus one mRNA vaccine dose, is next. Anything less seems to not be acceptable.4

P.S. Getting booster doses reduces viral load in breakthrough infections and thereby should reduce transmission – another benefit of boosters.4 


Table 1. Antibody responses commonly used to study immunity induced by SARS-CoV-2 mRNA vaccines or by infection

Assays for antibody (Aby) to SARS-CoV-2* mRNA Vaccines** Infection Hybrid Immunity (Infection plus Vaccine) Comment
Serum/Plasma Aby X X X Easiest compartment to sample; attacks virus mostly after entry into body
IgG to spike protein X X X Spike is major target of protective immunity
IgG to nucleocapsid (NC) X X X NC not in vaccines, not known to be protective
Neutralizing Aby X*** X X*** Best serum/plasma correlate to protection
Mucosal Aby No X X Attacks virus before entry into body
Mostly secretory IgA No X X Very difficult and incompletely standardized assay

*Specificity improves over time since last event (e.g., primary vaccine series or infection), so boosting (re-exposure to antigens within vaccine or infectious virus after several-to-many months post last event) not only increases antibody titers many-fold but the antibodies are also more efficient at attaching to their targets – resulting in much higher amounts of more effective antibodies from boosting.

**mRNA vaccines induce responses only to spike protein, whereas infection induces responses to many virus components.

***Hybrid immunity and four doses of mRNA vaccine produce the highest number of cross-neutralizing antibodies to all variants.




  1. Junqueira C, Crespo Â, Ranjbar S, et al. FcγR-mediated SARS-CoV-2 infection of monocytes activates inflammation. Nature. 2022 Published online April 6, 2021. doi:10.1038/s41586-022-04702-4
  2. Khoury DS, Cromer D, Reynaldi A, et al. Neutralizing antibody levels are highly predictive of immune protection from symptomatic SARS-CoV-2 infection. Nat Med. 2021;27(7):1205-1211. doi:10.1038/s41591-021-01377-8
  3. Plumb ID, Feldstein LR, Barkley E, et al. Effectiveness of COVID-19 mRNA vaccination in preventing COVID-19–associated hospitalization among adults with previous SARS-CoV-2 infection — United States, June 2021–February 2022. MMWR Morb Mortal Wkly Rep. 2022;71(15):549-555.
  4. Abu-Raddad LJ, Chemaitelly H, Ayoub HH, et al. Effect of mRNA vaccine boosters against SARS-CoV-2 Omicron infection in Qatar. N Engl J Med. Published online March 9, 2022. doi:10.1056/NEJMoa2200797
  5. Bar-On YM, Goldberg Y, Mandel M, et al. Protection by a fourth dose of BNT162b2 against Omicron in Israel. N Engl J Med. Published online April 5, 2022. doi:10.1056/NEJMoa2201570
  6. Puhach O, Adea K, Hulo N, et al. Infectious viral load in unvaccinated and vaccinated individuals infected with ancestral, Delta or Omicron SARS-CoV-2. Nat Med. Published online April 8, 2022. doi:10.1038/s41591-022-01816-0