During the coronavirus disease-2019 (COVID-19) pandemic, severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) spike-protein targeted serology testing has played only a limited role in clinical decision-making, largely based on the advice from the United States Centers for Disease Control and Prevention (CDC) (1). CDC Guidance, updated on 16 December 2022, states, “Antibody testing is not currently recommended to assess for immunity to SARS-CoV-2 following COVID-19 vaccination or to assess the need for vaccination in an unvaccinated person” (2). Furthermore, the United States Food and Drug Administration (FDA) has not designated any Emergency Use Authorization (EUA) SARS-CoV-2 test for assessing individual immunity through antibody testing (The FDA reviews and responds to submissions from in vitro diagnostics manufacturers). Currently, the efficacy of detectable antibody levels against subsequent SARS-CoV-2 infection and adverse outcomes is incompletely understood. As a result, only a small number of studies have evaluated the risk of SARS-CoV-2 infection outcomes based on SARS-CoV-2 antibody testing, and none have evaluated this risk among subgroups of the population at the highest risk for severe adverse outcomes of SARS-CoV-2 infection, e.g., immunocompromised individuals (3–5).

The need for clinical guidelines for using SARS-CoV-2 serology testing at the individual level is most acute for immunocompromised persons and those with chronic medical conditions (6). These groups are at increased risk for serious adverse COVID-19-related outcomes, including hospitalization and death (7–10). Immunocompromised patients hospitalized with COVID-19 accounted for 12.2% of hospitalized SARS-CoV-2 infected patients but only 2.7% of the general population (11). Furthermore, immunocompromised persons are more likely to experience adverse COVID-19 outcomes, regardless of vaccination status (11). Having other chronic medical conditions and advanced age are associated with an increased vulnerability to adverse COVID-19 outcomes including age 65 years and older (6), diabetes, cardiovascular disease including hypertension, chronic kidney disease (12), and obesity (13, 14).

The objective of this study was to investigate if having detectable vs. non-detectable SARS-CoV-2 spike-protein targeted antibody levels was associated with a decreased risk of COVID-19-related adverse outcomes overall and among clinically relevant subgroups.

A total of 1,798,606 people had a SARS-CoV-2 spike-protein targeted semiquantitative serology test from 16 November 2020 to 30 December 2021 (Supplementary Figure 1). Lack of baseline insurance enrollment excluded 1,580,452 individuals, and an additional 22,708 persons were excluded for other reasons. Therefore, the study cohort included a total of 195,446 individuals: 143,091 (73.2%) individuals with detectable levels and 52,355 (26.8%) individuals with non-detectable levels of semiquantitative SARS-CoV-2 spike-protein targeted antibodies. Overall, the study cohort had a mean age of 50.6 years, was 60.2% female, and was most commonly from the Northeastern (39.0%) and Southern (36.4%) regions of the United States. Only 24.9% of individuals had a record of receiving at least one dose of a COVID-19 vaccine. Of study individuals with any documented vaccination prior to exposure (SARS-CoV-2 spike-protein antibody testing), 95.5% (46,373 of 48,711) had a detectable antibody. In contrast, only 65.9% (96,718 of 146,735) of study individuals without documented vaccination prior to exposure had detectable antibody levels. Having an immunocompromising medical condition or another vulnerable medical condition associated with a heightened risk of severe COVID-19 was present in 7.0 and 45.9% of the cohort, respectively.

Table 1 shows the baseline characteristics of the study cohort stratified by the exposure groups. Before propensity score matching, baseline covariates were generally well-balanced between exposure groups (ASD ≤0.10), with some exceptions (e.g., age, region of the United States, year/season of cohort entry, and COVID-19 vaccination status). After propensity score matching, all baseline covariates were well-balanced between exposure groups, indicating adequate control of measured confounders.

Healthcare providers seek guidance on how to evaluate an individual's SARS-CoV-2 risk, especially for those with high-risk conditions, i.e., immunocompromised and vulnerable patients. Kaufman et al. have postulated that SARS-CoV-2 spike-protein targeted serology test results may be clinically useful, notably among these high-risk individuals concerning the subsequent risk of adverse outcomes (22). This position was based on a literature review that revealed (1) individuals at increased risk for severe outcomes following SARS-CoV-2 infection were less likely to develop a robust antibody response following infection and vaccination (23) and (2) studies showing that people with non-detectable or low levels of SARS-CoV-2 spike-protein targeted antibodies were more likely to have subsequent adverse events, i.e., hospitalization with COVID-19 and death, than those with higher levels (8, 24). Furthermore, SARS-CoV-2 spike-targeted antibody titer levels correlate with protection against subsequent SARS-CoV-2 infection or reinfection (25). This study confirms the two prior key findings by demonstrating that people with detectable SARS-CoV-2 spike-targeted antibody levels, as well as those with higher vs. lower detectable antibody levels, have a lower risk of COVID-19 serious outcomes in overall and adds novel findings among subgroups of patients at increased risk of SARS-CoV-2 infection (7–10).

These observations are consistent with other studies that have demonstrated similar associations between specific medical conditions and SARS-CoV-2 spike-protein targeted serology results and between serology results and subsequent outcomes (3–5). A recent study of cancer vs. non-cancer (control) patients in the United Kingdom showed that detectable levels of SARS-CoV-2 spike-protein targeted antibodies were associated with protection against subsequent SARS-CoV-2 infection and serious outcomes (26). Among patients with cancer, a non-detectable vaccine antibody response was associated with more than three times the risk of subsequent SARS-CoV-2 infection and more than six times the risk of a COVID-19-associated hospitalization. The authors suggest that patients with non-detectable antibody levels may benefit from additional vaccine doses, prophylactics, and early treatment.

In the United States, the three COVID-19 vaccines available during the study period were manufactured by BioNtech/Pfizer, Janssen/Johnson & Johnson, and Moderna. SARS-CoV-2 vaccination effectively reduced serious COVID-19 outcome events even though the benefit of vaccination is generally less effective among immunocompromised persons as compared to immunocompetent individuals (27). In this study, 24.9% (48,711 of 195,446) of individuals received at least one vaccine dose. In the matched cohort, the HR for subsequent infection among those with detectable vs. non-detectable SARS-CoV-2 spike-protein targeted antibody levels was 0.56 (95% CI 0.53–0.58) and 0.57 (95% CI 0.44–0.72) for the vaccinated and unvaccinated groups, respectively. In the matched cohort of this study, the HR for serious composite outcomes of hospitalization with COVID-19 or all-cause mortality among those with detectable vs. non-detectable SARS-CoV-2 spike-protein targeted antibody levels was 0.07 (95% CI 0.02–0.28) and 0.20 (95% CI 0.16–0.27) for the vaccinated and unvaccinated groups, respectively. Although the CI are overlapping, this observation supports the clinical value of having detectable antibodies while still recognizing that other immune mechanisms of protection are likely involved in the protection against SARS-CoV-2 infection and COVID-19 outcomes. Furthermore, the direction of the HR supports the concept that vaccination was likely useful in reducing serious outcomes.

Future studies may be useful in delineating the interplay of humoral and cellular immune system components in protection against SARS-CoV-2 infection and its consequences. Tixagevimab/cilgavimab was the only combination therapeutic authorized to date by both the FDA [effective 8 December 2021, revoked 26 January 2023 (28)] and the European Medicine Agency (EMA) (effective 25 March 2022) for pre-exposure prophylaxis of COVID-19. This was especially relevant in the population with immunocompromising conditions who fail to mount a detectable antibody response after multiple vaccine doses. The authors of the primary tixagevimab/cilgavimab study note, “The limitations of our trial include the low number of events in smaller but important subgroups, including immunocompromised persons, so that efficacy in these groups could not be estimated” (29). The FDA subsequently recommended high dosing of tixagevimab/cilgavimab after a significant number of patients in the immunosuppressed group were found to have breakthrough infections (30). Young-Xu et al. at Veteran Affairs Healthcare Systems, found that compared to 251,756 propensity-matched immunocompromised or at-risk historical controls, 1,848 tixagevimab/cilgavimab-treated patients had a lower incidence of SARS-CoV-2 infection, COVID-19 hospitalization, and all-cause mortality (31). Until newer effective prophylactic drugs are approved, respiratory tract masking may be especially valuable within the high-risk population, e.g., with immunocompromising conditions, when visiting healthcare facilities where there are potentially SARS-CoV-2-infected patients (32). The CDC suggests that masks can provide an extra level of protection against SARS-CoV-2 infection and its resulting severe events (6). Early antiviral treatments may be beneficial as well, especially in the immunocompromised population (33). Again, the effectiveness of these multiple infection mitigation measures with current and future SARS-CoV-2 variants is worthy of investigation. Given our findings among immunocompromised persons, the study's findings support the application of these suggested COVID-19 mitigation measures in high-risk populations, particularly those who are SARS-CoV-2 seronegative. Additional studies may indicate the specific potential benefit of additional vaccine dosing and other mitigation efforts to reduce the risk of adverse outcomes in individuals with non-detectable SARS-CoV-2 antibody levels.

The strengths of this study include the use of a large-scale real-world database with information aggregated from diverse sources, inclusive of multiple laboratory antibody test methods, analysis of subsequent SARS-CoV-2 infections, COVID-19 hospitalizations and all-cause mortality, and multivariate modeling with confounding control. Tracking changes in semiquantitative SARS-CoV-2 spike-protein targeted antibody levels over time within an at-risk individual may provide insights into the durability of the antibody response and assist in determining the subsequent risk of infection (5). Alternative assays that measure antibody neutralization of novel spike protein(s) or cellular-based adaptive immunity assays are being studied for their associated clinical utility but they are not yet widely commercially available (34). Of note, this study included SARS-CoV-2 spike-protein targeted antibody tests and did not include rapid antigen tests or nucleocapsid antibody tests. Differences in antibody generation post-infection have been observed with SARS-CoV-2 nucleocapsid antibody tests. SARS-CoV-2 Omicron variants impacted the performance of some NAAT assay methods. The FDA updates the few assay methods that are adversely affected by viral mutations (35). None of those SARS-CoV-2 NAAT assay methods were used in this study. Therefore, SARS-CoV-2 variants had no reported impact on the performance of the SARS-CoV-2 spike protein targeted antibody or NAAT testing in this study.

The study had some limitations. First, the evaluation time period was early in the pandemic (November 2020 to December 2021) and included a portion of the time when COVID-19 vaccines were not yet widely adopted. Similarly, home laboratory testing that identifies SARS-CoV-2 infections was not captured though also not yet common during the study period. Second, information on the medical reason for the requested SARS-CoV-2 serology testing was unavailable. Third, the COVID-19 infection outcome was largely driven by infections identified in the health insurance claims data. SARS-CoV-2 PCR/NAAT tests performed by LabCorp and Quest Diagnostics, although substantial in aggregate number, represent less than 20% of all total SARS-CoV-2 PCR/NAAT conducted in the United States during the study period (36–38). Finally, deaths were infrequent, precluding studying associations between SARS-CoV-2 antibody levels and mortality alone.

In summary, this large United States-based real-world evidence-based study utilized linked medical claims and clinical laboratory data to examine associations between SARS-CoV-2 spike-protein antibody levels and clinical outcomes. The study demonstrated that people with detectable levels of SARS-CoV-2 spike-protein targeted antibodies had a lower risk of subsequent SARS-CoV-2 infections and serious composite outcomes (hospitalization with an associated SARS-CoV-2 infection or all-cause mortality). This observed effect was seen in the overall population and also within clinically relevant subgroups, including the immunocompromised population. Analyses of individuals with detectable antibodies >250 vs. < 250 BAU/ml generated directionally consistent results, albeit with a less potent magnitude of effect. Thus, federal policymakers and clinicians may find SARS-CoV-2 spike-protein targeted serology testing to be a useful adjunct in counseling immunocompromised persons and other higher at-risk individuals about adverse outcomes and apply appropriate actions to mitigate such risks.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

All authors were involved in study design, analysis, interpretation of data, writing of the manuscript, and the decision to submit the manuscript for publication. All authors contributed to the article and approved the submitted version.