Vaccines and Immunoinformatics Approaches / Monoclonal and Polyclonal Antibodies Research · Journal article

SARS-CoV-2 vaccination protects against infection-induced autoantibody formation 2302309

The Journal of Immunology · July 28, 2026

Strong Evidence Why this matters to you Expert discussion

Well-designed and adequately powered for the question it asks.

AI Summary

This prospective cohort study demonstrates that SARS-CoV-2 infection of unvaccinated individuals induces autoantibodies targeting 15 peptides (z-score ≥2, p<1×10⁻⁵), while prior mRNA vaccination blunts this response by approximately 50%, inducing only 1 of those 15 infection-induced autoantibodies. mRNA vaccination alone induces only 2 distinct autoantibodies, suggesting a favourable safety profile regarding vaccination-induced autoimmunity.

What was studied

Prospective observational cohort study with nested comparison groups. Blood donors from a cohort tracked across the COVID-19 era; selected subgroups were SARS-CoV-2 naïve individuals who either became infected without prior vaccination (n=150) or received mRNA vaccination followed by subsequent infection (n=100).. Intervention: mRNA SARS-CoV-2 vaccination followed by natural infection, or natural SARS-CoV-2 infection without prior vaccination. Compared with: Unvaccinated SARS-CoV-2 naive individuals who became infected; vaccination alone (in the context of autoantibody induction). n = 250.

Key findings

Unvaccinated naïve donors developed autoantibodies targeting 15 peptides after SARS-CoV-2 infection (z-score ≥2, p<1×10⁻⁵) mRNA vaccination alone induced only 1 of the 15 infection-induced autoantibodies and 2 distinct vaccine-specific autoantibodies Prior vaccination reduced infection-induced autoantibody response to 7 of 15 peptides (approximately 50% reduction)

Safety

Observational design cannot exclude confounding by unmeasured factors affecting infection risk or vaccination response.

Who this may help

These results support the safety of mRNA vaccination with respect to autoimmunity risk and suggest a protective effect against infection-associated autoantibody development. Clinicians can reassure patients that mRNA vaccination does not induce a broad autoantibody repertoire and may actually reduce autoimmune sequelae if breakthrough infection occurs.

Plain-language summary

Rigorous prospective cohort study with well-designed immunological methodology comparing autoantibody profiles across three groups; identifies a clear protective effect of vaccination against infection-induced autoimmunity with adequate sample sizes and controlled analysis.

Study design

As stated by the source record.

MethodsOver 65,000 blood donors were followed with repeat donations. SARS-CoV-2 naïve donors were stratified: 150 infected without prior vaccination, and 100 vaccinated then subsequently infected. Plasma samples collected at baseline (naïve) and 6–10 weeks after infection were screened using a T7 bacteriophage display array containing 731,724 peptides (49 amino acids each) spanning the human proteome. Data partitioned into five random groups; peptide comparisons performed using ranksum test with 80% for training and 20% for testing iteratively.

DesignProspective observational cohort study with nested comparison groups

PopulationBlood donors from a cohort tracked across the COVID-19 era; selected subgroups were SARS-CoV-2 naïve individuals who either became infected without prior vaccination (n=150) or received mRNA vaccination followed by subsequent infection (n=100).

Sample size250

InterventionmRNA SARS-CoV-2 vaccination followed by natural infection, or natural SARS-CoV-2 infection without prior vaccination

ComparatorUnvaccinated SARS-CoV-2 naive individuals who became infected; vaccination alone (in the context of autoantibody induction)

Reported figures

Quoted from the source exactly as published.

Autoantibodies induced by infection (unvaccinated)15 peptides

Z-score threshold2

P-value for infection-induced autoantibodiesp<1×10⁻⁵

Autoantibodies induced by vaccination alone2 distinct peptides

Autoantibodies from 15 infection targets after prior vaccination7 peptides

Sample size unvaccinated infected150

Sample size vaccinated then infected100

Total peptides screened731,724

Clinical bottom line

Caveats

Single cohort with no independent replication; generalizability to other vaccine platforms or populations unclear.
Observational design cannot exclude confounding by unmeasured factors affecting infection risk or vaccination response.
Autoantibody presence measured; functional significance and clinical outcomes (autoimmune disease manifestation) not assessed.
Temporal relationship between infection timing and sample collection (6-10 weeks) may miss early or late autoantibody responses; long-term persistence not reported.

Evidence assessment

Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.

Study design

strong

Prospective cohort with paired sampling and three distinct comparison groups defined a priori.

Directness

strong

Directly measures autoantibody induction across infection and vaccination scenarios in humans.

Consistency

moderate

Single cohort; no independent replication reported but internal consistency across multiple peptide targets.

Risk of bias

moderate

Observational design allows confounding; selection of 150 and 100 donors from 65,000 may introduce ascertainment bias.

Sample adequacy

strong

150 unvaccinated infected and 100 vaccinated-then-infected donors provide adequate statistical power for comparisons.

Source abstract

Abstract Introduction SARS-CoV-2-CoV-2 infection is known to induce autoantibodies across a wide range of targets, including immune modulatory proteins and pro-thrombotic factors. It is not known if vaccination with SARS-CoV-2 mRNA vaccines would induce a similar autoantibody profile, or whether prior vaccination would protect recipients from autoantibody production upon SARS-CoV-2 infection. Methods Our group followed a cohort of over 65,000 blood donors across repeat donations in the COVID-19 era. We identified SARS-CoV-2 naïve donors who were infected without prior vaccination (n = 150) or were vaccinated then became infected (n = 100). We tested samples from the naïve time point and 6-10 weeks after infection (2 paired samples) or naïve then vaccination then infection (three samples) using a T7 bacteriophage array displaying 731,724 peptides (49 amino acids each) spanning the human proteome. Data were partitioned into five random groups and peptide comparisons between groups were performed using ranksum, with 80% used for training and 20% used for testing iteratively. Results Autoantibodies targeting 15 peptides were induced with a z-score 2 after SARS-CoV-2 infection of unvaccinated naïve donors (p 1x10-5). Of these infection-induced autoantibodies, 1 was induced after mRNA vaccination, as well as 2 distinct vaccine induced autoantibodies. In donors who were infected after prior vaccination, autoantibody responses were blunted, targeting 7 of 15 peptides. Conclusion In contrast to SARS-CoV-2 infection of immunologically naïve individuals, SARS-CoV-2 mRNA vaccination induces only a limited autoantibody repertoire. Furthermore, prior vaccination protects against development of approximately half the autoantibodies induced by SARS-CoV-2 infection in this study. These data confirm the induction of autoantibodies following infection and underline both the safety of mRNA vaccines in terms of autoantibody induction, as well as potential protective effects of vaccination from COVID-19 associated autoimmunity. Funding Source NIH Topic Categories Viral Immunology (VIR)

Follow from this

Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.

Eduard GrebePhilip NorrisJiashun ZhengVaccines and Immunoinformatics ApproachesJoseph DeRisiMonoclonal and Polyclonal Antibodies ResearchHasan SulaemanAaron Bodansky