TL;DR
This study investigated how antibody levels after receiving the Novavax NVX-CoV2373 COVID-19 vaccine correlate with the risk of developing severe COVID-19. The key finding is that higher antibody levels, particularly neutralizing antibodies against the Delta variant, were strong predictors of reduced risk for severe disease, suggesting that low antibody responses indicate vulnerability.
What Problem Does This Paper Address?
While the NVX-CoV2373 vaccine demonstrated high efficacy against symptomatic COVID-19 in initial trials, understanding why it protects and identifying individuals who might remain at higher risk is crucial. Specifically, the authors aimed to determine if antibody levels measured after vaccination could serve as "immune correlates of risk" (CoRs) for severe COVID-19, especially during the emergence of new variants like Delta. Previous analyses in the PREVENT-19 trial had established CoRs for any severity of COVID-19 but lacked sufficient data for severe outcomes. This paper addresses this gap by extending the follow-up period and analyzing a larger number of severe cases to establish these critical correlates.
Methods
The researchers conducted a "stage 2" analysis using data from the PREVENT-19 trial, focusing on participants who received the NVX-CoV2373 vaccine and were initially SARS-CoV-2 negative. They employed a case-cohort sampling design to efficiently analyze antibody responses. Key methods included:
- Antibody Assays: Measuring both binding antibody (IgG) concentrations and neutralizing antibody (nAb) titers against both ancestral and Delta variants of the SARS-CoV-2 Spike protein. These were measured at approximately 2 weeks after the second vaccine dose (D35), representing peak antibody levels.
- Immune Correlates of Risk (CoR) Analysis: Using statistical models (inverse probability sampling weighted Cox regression) to assess how D35 antibody levels predicted the risk of severe COVID-19 and Delta COVID-19 over an extended follow-up period (up to 10 months post-D35).
- Exposure-Proximal Correlates Analysis: A novel approach to account for antibody waning over time. This method models the instantaneous risk of infection based on a hypothetical daily measurement of antibody levels, providing a more dynamic understanding of protection.
Datasets
The study utilized data from the PREVENT-19 trial, a randomized, placebo-controlled, phase 3 efficacy trial of the NVX-CoV2373 recombinant protein vaccine. The analysis specifically focused on:
- Per-protocol baseline SARS-CoV-2-negative original vaccine arm participants: This cohort was selected to ensure that the observed outcomes were due to vaccine efficacy and not pre-existing infection or placebo effects.
- Breakthrough cases: Participants who developed COVID-19 despite vaccination, including a subset with severe outcomes.
- Noncases: Participants who remained infection-free throughout the follow-up period.
The study specifically analyzed data from 287 breakthrough Delta cases (including 8 severe cases) and 446 noncases for the D35 antibody marker analysis.
Key Insights and Ideas
- Antibodies as Strong Predictors of Severe Disease: The most striking insight is the clear inverse correlation between antibody levels and the risk of severe COVID-19. Higher antibody titers at peak response (D35) significantly reduced the likelihood of developing severe illness.
- Delta Variant Antibodies are Key: Antibodies specifically targeting the Delta variant showed a particularly strong association with protection against severe Delta COVID-19, highlighting the importance of variant-specific immune responses.
- Distinct Antibody Profiles in Severe Cases: A notable finding was the clear separation in antibody levels between severe cases and the general vaccinated population. Severe cases consistently had much lower antibody levels, often falling into the lowest tertiles of antibody concentrations.
- Vulnerability Indicated by Low Response: The study strongly suggests that individuals with low antibody responses post-vaccination are more vulnerable to severe COVID-19.
- Time-Varying Correlates: The exposure-proximal analysis demonstrated that even as antibody levels naturally decline over time, the current level of antibodies still inversely correlates with the risk of Delta COVID-19, indicating a dynamic protective effect.
Key Results
- Peak Antibody Levels and Severe COVID-19: For Delta-specific neutralizing antibody titers (nAb-ID50), the geometric mean was significantly lower in severe cases (9.6 AU/mL) compared to noncases (209.5 AU/mL). The frequency of detectable nAb-ID50 was also much lower in severe cases (62.5%) versus noncases (98.3%).
- Hazard Ratios for Severe COVID-19: Each 10-fold increase in D35 Delta nAb-ID50 was associated with a 0.13-fold decrease in the hazard of severe COVID-19 (95% CI: 0.03, 0.57). Similarly, for D35 Delta Spike IgG, the hazard ratio was 0.11 (95% CI: 0.05, 0.26).
- Cumulative Incidence: The estimated cumulative incidence of severe COVID-19 through 10 months was substantially lower at higher antibody percentiles. For example, at the 95th percentile of Delta nAb-ID50 (2522 AU/mL), the risk was 0.0002, compared to 0.0338 at the 2.5th percentile (8.4 AU/mL).
- Delta COVID-19 Risk: Antibody levels also inversely correlated with the risk of any-severity Delta COVID-19, though the effect was less pronounced than for severe disease.
Limitations and Future Work
- Small Number of Severe Cases: The authors acknowledge that the analysis of severe COVID-19 endpoints was limited by a small number (eight) of cases with available D35 antibody data, which impacts statistical precision.
- Lack of Randomized Comparison Arm for Correlates: Correlates of protection analyses could not be performed due to the absence of a placebo arm in the extended follow-up.
- Exclusion of Cellular Responses: The analysis focused solely on antibody responses, excluding cellular immunity (T cell responses), which may also play a role in preventing severe disease.
- Future Research: The findings suggest that future vaccination strategies, booster recommendations, and pre-exposure prophylaxis could potentially be informed by antibody levels, particularly for high-risk populations. Further research is needed to confirm these findings after booster vaccinations and against emerging variants like Omicron.
Summary updated