Abstract Numerous somatic mutations have been implicated in cancer. However, but the evolutionary dynamics that shape tumor initiation and progression remain incompletely understood. In particular, systematic quantification of how mutation rates and selective pressures change across tumor stages has not previously been achieved. Unlike mutation prevalence, measures of selection intensity directly estimate the survival and proliferative advantage conferred by each variant, providing a means to distinguish drivers of initiation from those of progression. We analyzed thousands of tumor genomes, exomes, and targeted gene sequencing datasets across multiple cancer types to evaluate how mutation rate and selection intensity change from organogenesis to primary tumor development and metastatic spread. We calculated cancer effect sizes—population-genetic scaled selection coefficients—for driver mutations and genes. Mutation spectra were broadly conserved across stages within cancer types, but underlying mutation rates generally escalated with progression. In contrast, selection intensity shifted in gene- and stage-specific patterns. For example, BRAF V600E exhibited orders-of-magnitude stronger selection in metastasis compared to primary melanoma, SPOP mutations played an early initiating role in prostate cancer by enhancing selection on cooperating genes, and RET alterations were strongly selected during metastatic progression in thyroid cancer. These findings reveal that tumor evolution is shaped not only by mutation rate but also by dynamic, stage-specific selective pressures. Systematic quantification of selection intensity provides insight into the gene mutations that are most critical at each stage of cancer development, informing precision strategies for therapeutic targeting matched to disease stage and progression. Citation Format: Moein Rajaei, Jeffrey P. Townsend. Variation in selection intensity and mutation rates during tumor evolution across cancer types abstract. In: Proceedings of the AACR Special Conference in Cancer Research: Cancer Evolution: The Dynamics of Progression and Persistence; 2025 Dec 4-6; Albuquerque, NM. Philadelphia (PA): AACR; Cancer Res 2025;85 (23Suppl): Abstract nr B039.
Jeffrey P. Townsend (Thu,) studied this question.