Exposure route, not dose, is the primary driver of infection patterns for a respiratory virus.

Infectious disease severity, shedding, and within-host kinetics can vary substantially with exposure dose, inoculation route, and host factors. However, the relative contributions of these variables to infection heterogeneity remain poorly quantified because small sample sizes in controlled in vivo experiments limit their statistical power and scientific scope. Here, by compiling and jointly analyzing the largest published database of nonhuman primate challenge experiments (107 studies; 721 animals), we show that exposure route drives SARS-CoV-2 infection kinetics more strongly than dose, age, sex, or species. Aerosol inoculation yields kinetics distinct from all other respiratory exposure routes. Route and dose jointly determine which tissues become infected, and 50% infectious doses vary greatly depending on the route of exposure and on the tissue sampled. Our findings underscore the central role of exposure route in pathogenesis and transmission, and highlight the untapped potential for quantitative, cross-study syntheses to extract additional insights from costly animal experiments.