Modeling and predicting individual variation in COVID-19 vaccine-elicited antibody response in the general population

• DOI: 10.1371/journal.pdig.0000497
• Publication Date: 2024-05-03T17:20:31Z
• Authors: Naotoshi Nakamura; Yurie Kobashi; Kwang Su Kim; Hyeongki Park; Yuta Tani; Yuzo Shimazu; Tianchen Zhao; Yoshitaka Nishikawa; Fumiya Omata; Moe Kawashima; Makoto Yoshida; Toshiki Abe; Yoshika Saito; Yuki Senoo; Saori Nonaka; Morihito Takita; Chika Yamamoto; Takeshi Kawamura; Akira Sugiyama; Aya Nakayama; Yudai Kaneko; Yong Dam Jeong; Daiki Tatematsu; Marwa Akao; Yoshitaka Sato; Shoya Iwanami; Yasuhisa Fujita; Masatoshi Wakui; Kazuyuki Aihara; Tatsuhiko Kodama; Kenji Shibuya; Shingo Iwami; Masaharu Tsubokura
• Keywords: Pandemic; Preparedness; Antibody response; Antibody titer

As we learned during the COVID-19 pandemic, vaccines are one of most important tools in infectious disease control. To date, an unprecedentedly large volume high-quality data on vaccinations have been accumulated. For preparedness future pandemics beyond COVID-19, these valuable datasets should be analyzed to best shape effective vaccination strategy. We collecting longitudinal from a community-based cohort Fukushima, Japan, that consists 2,407 individuals who underwent serum sampling two or three times after two-dose with either BNT162b2 mRNA-1273. Using individually reconstructed time courses vaccine-elicited antibody response based mathematical modeling, first identified basic demographic and health information contributed main features dynamics, i.e., peak, duration, area under curve. showed dynamics were partially explained by underlying medical conditions, adverse reactions vaccinations, medications, consistent findings previous studies. then applied factors recently proposed computational method optimally fit “antibody score”, which resulted integer-based score can used as basis for identifying higher lower titers information. The easily calculated themselves practitioners. Although sensitivity this is currently not very high, future, more become available, it has potential identify vulnerable populations encourage them get booster vaccinations. Our model extended any kind therefore form policy decisions regarding distribution strengthen immunity pandemics.