The immune system doesn't respond the same way in everyone. Vaccines train it, but some people's bodies mount a much stronger antibody response than others. Researchers have long suspected this variation matters, but it's been hard to pin down whether it predicts something clinically serious: not just any COVID-19 infection, but severe COVID-19, the kind that lands someone in the hospital, on a ventilator, or worse.
The problem has always been numbers. Severe COVID-19 is, thankfully, rare among vaccinated people, and rare events are statistically difficult to study. Researchers need a meaningful number of severe cases before they can draw reliable conclusions about what predicts them, and most vaccine trials simply never accumulate enough.
A new analysis, led by Dr. Youyi Fong and colleagues at Fred Hutchinson Cancer Center and published in Clinical Infectious Diseases, got a rare opportunity to answer this question directly. Follow-up in the PREVENT-19 trial happened to stretch across the emergence of the Delta variant, and Delta turned out to cause enough breakthrough infections, including 11 severe cases, to finally allow a real look at whether antibody levels predict severe disease, not just infection in general.
The results were striking. Among participants who went on to develop severe COVID-19, average neutralizing antibody levels measured two weeks after vaccination were more than 20-fold lower than in people who never got infected at all. 98% of people who stayed healthy had detectable neutralizing antibodies against Delta; among those who developed severe disease, only about 63% did.
Among participants in this trial, nearly all severe COVID-19 cases occurred among those with the lowest antibody levels. None occurred among people with the highest antibody levels, though the number of severe cases was small. In statistical terms, every 10-fold increase in antibody titer was associated with an 85–90% reduction in the risk of severe disease. That's a far stronger relationship than researchers had previously seen between antibody levels and infection of any severity.
Antibody levels were much less predictive of whether someone developed a breakthrough infection than whether that infection became severe. Even participants with high antibody levels sometimes caught Delta. Severe disease, though, was concentrated almost entirely among people with the lowest post-vaccination antibody levels.
“Prior to this study, we have always thought that immune responses would be more predictive of risk of severe COVID than COVID of any severity,” said Dr. Fong, “But it is difficult to study severe COVID because the frequency is much lower and the sample size is an issue. In this study, we were fortunate to have over 10 severe cases, partly because the circulating strain at the time was Delta, which allowed us to quantify the association between immune responses and risk of severe COVID.” Dr. Peter Gilbert, the study's senior author, added: “The results supported the hypothesis with flying colors, indicating a remarkably strong association of neutralizing antibody titers with severe COVID, and suggesting that low titers are an indicator of vulnerability of serious disease.”
Timing made this possible, though it's not just about case counts. Other major vaccine efficacy trials, including the Johnson & Johnson and AstraZeneca trials, did look for the same signal in severe COVID-19, but found a much weaker separation between severe cases and everyone else. The Novavax vaccine produced unusually strong antibody responses in most people, making the contrast with the small number of severe cases especially clear.
The findings suggest that low post-vaccination antibody levels could serve as a warning sign, pointing to people who face higher risk of severe outcomes and could benefit from booster doses or additional protective measures. That's potentially useful for guiding vaccination strategy in higher-risk populations.
The trial's timeline also raises a question beyond the paper's core findings, as Dr. Fong points out: “One interesting aspect of the study that is not directly related to the main conclusions is that in this study period, the Delta wave gave way to the Omicron wave (end of 2021, beginning of 2022). It is not totally clear what could be done, but it is such a unique dataset.”
The same trial that captured the Delta signal also captured the moment the virus began shifting toward Omicron, a natural experiment that hasn't yet been fully explored. What's already clear is more immediately useful. Strong antibody responses after vaccination tracked closely with protection from severe illness, and low levels after vaccination could help identify who stands to benefit most from a booster.
This research was supported by the National Institute of Allergy and Infectious Diseases (NIAID), the Coronavirus Prevention Network (CoVPN), the Biomedical Advanced Research and Development Authority (BARDA), and Novavax.
Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium Members Dr. Peter Gilbert contributed to this research.
Fong, Y., Huang, Y., Huang, Y., Woo, W., McGarry, A., Áñez, G., Dunkle, L. M., Cho, I., Houchens, C. R., Martins, K., Jayashankar, L., Castellino, F., Petropoulos, C. J., Leith, A., Haugaard, D., Webb, W., Lu, Y., Yu, C., Carpp, L. N., Randhawa, A. K., … (2026). Analysis of Antibody Markers as Immune Correlates of Risk of Severe COVID-19 in the PREVENT-19 Efficacy Trial of the NVX-CoV2373 Recombinant Protein Vaccine. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America, 82(4), e863–e874.
Science Spotlight writer Darya Moosavi is a postdoctoral research fellow within Johanna Lampe's research group at Fred Hutch. Darya studies the nuanced connections between diet, gut epithelium, and gut microbiome in relation to colorectal cancer using high-dimensional approaches.