When a family member tests positive for COVID-19, most vaccinated people assume their antibody levels will determine whether they get sick. That assumption is incomplete, and potentially consequential for how policymakers and clinicians think about vaccine performance and the immune markers used to evaluate it.
A prospective cohort study published in The Lancet Microbe, conducted through Michael Garron Hospital in Toronto and involving investigators from the University of Guelph, McMaster University, and Sinai Health’s Lunenfeld-Tanenbaum Research Institute, found that antibody levels measured shortly after household SARS-CoV-2 exposure did not predict who went on to develop infection. Instead, early differences in a specific subset of CD8+ T cells, white blood cells central to antiviral defense, were associated with whether vaccinated adults developed symptomatic COVID-19.
The AB-PROTECT study enrolled 484 adults from 339 households between October 2021 and February 2023, a period spanning the delta wave and multiple omicron sub-lineages. The study recruited asymptomatic adults who had been exposed to a household member with a confirmed SARS-CoV-2 infection, collecting blood samples as quickly as possible after exposure, typically within three days of the index case’s symptom onset. All participants had received at least two vaccine doses, with the threshold raised to three doses after omicron emerged. Of the 432 participants with sufficient follow-up data, nearly half developed SARS-CoV-2 infection within 28 days.
The researchers measured a comprehensive panel of immune markers: IgG and IgA antibodies against the spike protein and receptor-binding domain, pseudovirus and live-virus neutralization assays against multiple variants, and detailed T-cell activation and cytokine profiling. This combination of assays, evaluated in the critical window immediately following household exposure, is broader than what prior household contact studies have attempted.
Whether the researchers examined anti-spike IgG, anti-RBD IgG, IgA subtypes, pseudoneutralizing capacity, or live-virus microneutralizing titers against ancestral SARS-CoV-2 and the omicron BA.1, BA.5, and XBB.1.5 sublineages, none predicted who became infected. The predicted probability of infection changed by no more than five percentage points across the full range from the 10th to the 90th percentile of antibody levels, differences that fell well within confidence intervals spanning zero. These patterns held after stratifying by variant period and after excluding participants who developed symptoms or tested positive within 24 hours of their baseline blood draw.
The T-cell results told a different story. In a subset of 56 participants with detailed cellular immune profiling, those who went on to develop symptomatic COVID-19 had significantly lower levels of CD8+ T cells co-expressing two cytokines, granulocyte-macrophage colony-stimulating factor and tumor necrosis factor, shortly after exposure. This pattern suggests that early cytokine responses from CD8+ T cells may reflect a protective immune signature that antibody assays cannot capture.
Neutralizing antibody titers remain the primary correlate used to bridge vaccine efficacy data when head-to-head trials against newly emerged variants are not feasible. If antibody levels measured close to the time of actual household exposure do not separate those who become infected from those who do not, regulators and public health agencies may need to invest more seriously in T-cell assay standardization for use in immune bridging and correlate-of-protection frameworks.
The authors recommend that future research measure immune markers before exposure events in controlled or predictable settings, such as hospitals or congregate care facilities, where exposure intensity can be better characterized and severe outcomes monitored. That design would help determine whether the T-cell signatures identified here translate into protection against clinical severity, not only symptomatic infection.
Sources and further reading
Kandel C, Breznik JA, Taylor M, et al. Assessing SARS-CoV-2 immune markers in vaccinated asymptomatic adults after household exposure in Canada (AB-PROTECT): a prospective cohort study. The Lancet Microbe. September 15, 2026.
This article was researched and sourced by Global Biodefense editors and reported with Claude AI assistance for drafting and editing.