Abstract
Though eleven novel COVID-19 vaccines have demonstrated efficacy, additional affordable, scalable, and deliverable vaccines are needed to meet unprecedented global demand. With placebo-controlled efficacy trials becoming infeasible due to the roll out of licensed vaccines, a correlate of protection is urgently needed to provide a path for regulatory approval of novel vaccines. To assess whether antibody titers may reasonably predict efficacy, we evaluated the relationship between efficacy and in vitro neutralizing and binding antibodies of 7 vaccines for which sufficient data have been generated. Once calibrated to titers of human convalescent sera reported in each study, a robust correlation was seen between neutralizing titer and efficacy (ρ= 0.79) and binding antibody titer and efficacy (ρ = 0.93), despite geographically diverse study populations subject to different forces of infection and circulating variants, and use of different endpoints, assays, convalescent sera panels and manufacturing platforms. This correlation is strengthened by substituting post-hoc analyses for efficacy against the ancestral strain (D614G), where available. Together with an accumulating body of evidence from natural history studies and animal models, these results support the use of post-immunization antibody titers as the basis for establishing a correlate of protection for COVID-19 vaccines.
Introduction
The global response to the COVID-19 pandemic has seen the rapid development and deployment of several highly effective vaccines. However, the need for a diverse portfolio of vaccines—including those that are more affordable, easier to manufacture, and stable at higher temperatures—remains critical to achieving global health equity.
As the first generation of vaccines achieved regulatory approval and widespread distribution, the traditional pathway for testing new candidates—large-scale, placebo-controlled Phase 3 trials—became ethically problematic and practically difficult. In regions where effective vaccines are available, it is no longer justifiable to withhold them from trial participants. To overcome this hurdle, researchers must identify a "Correlate of Protection" (CoP): a measurable biomarker that reliably indicates whether a person is protected. Establishing such a correlate would allow for "immunobridging" trials, where new vaccines could be authorized based on their ability to elicit a specific immune response rather than waiting for thousands of natural infections to occur.
Methods: The Challenge of Calibration
A major obstacle in identifying a CoP for COVID-19 has been the lack of standardization across different clinical trials. Each vaccine manufacturer used different laboratory assays, units of measurement, and reference panels to quantify immune responses, making direct comparisons impossible.
To address this, the authors conducted a meta-analysis of seven major vaccines: Pfizer/BioNTech, Moderna, Gamaleya (Sputnik V), AstraZeneca/Oxford, Sinovac, Novavax, and Janssen (Johnson & Johnson). They employed a calibration technique to normalize data across studies. Instead of comparing raw antibody counts, they expressed vaccine-induced antibody levels as a ratio relative to the levels found in "human convalescent sera" (HCS)—samples from individuals who had naturally recovered from COVID-19—measured within the same study using the same assay. This standardized "fold-increase" or "ratio" allowed for a direct comparison of the relative potency of each vaccine's immune response regardless of the specific lab test used.
Results: Neutralizing and Binding Antibodies
The study evaluated two primary types of antibody responses: virus-neutralizing antibodies (VNA), which prevent the virus from entering cells, and binding antibodies (measured via Spike IgG ELISA), which attach to the virus's spike protein.
Neutralizing Antibodies
When calibrated against convalescent sera, neutralizing antibody titers showed a strong correlation with vaccine efficacy (ρ = 0.79). This relationship suggests that the neutralizing capacity of the blood is a reliable indicator of how well a vaccine will prevent symptomatic disease.
Binding Antibodies
Surprisingly, binding antibodies (Spike IgG) demonstrated an even stronger correlation with efficacy (ρ = 0.93). The linear model based on binding antibodies explained approximately 94.2% of the variance in efficacy across the different vaccine platforms. This finding is significant because ELISA-based binding assays are generally easier to standardize and perform at scale than functional neutralization assays.
Discussion and Implications
The high degree of correlation observed across diverse vaccine platforms (mRNA, viral vector, protein subunit, and inactivated virus) suggests a common mechanism of protection. These results provide a scientific foundation for regulatory agencies to consider antibody titers as a primary endpoint for the approval of second-generation vaccines.
Limitations and Future Work
While these findings are robust, several caveats remain. First, while antibodies are excellent predictors, other components of the immune system, such as T-cells, likely contribute to long-term protection and prevention of severe disease. Second, the data primarily reflect responses to the ancestral SARS-CoV-2 strain. As new variants emerge, the specific thresholds for protection may shift, requiring updated calibration. Finally, the authors emphasize the urgent need for laboratories to adopt the WHO International Standard for antibody units (IU/ml or BAU/ml) to make future data even more comparable across global studies.
Summary updated