步骤特异性选择的定量分析揭示结肠癌中驱动突变的有序获得

Colorectal cancer is a leading cause of cancer mortality worldwide. Colon adenocarcinoma develops through sequential evolutionary steps originating in adenomatous polyps. Many driver mutations of colon cancer and their relations have been hypothesized. However, quantitative timing and strength of selection across specific stages of progression remains unknown. Somatic single nucleotide variants from colon adenomas and adenocarcinomas were analyzed using a step-specific likelihood model that treats adenoma formation and malignant progression as consecutive steps within a shared evolutionary trajectory. Analyses were stratified by POLE / POLD1 mutation and DNA mismatch repair (MMR) status to evaluate background-specific effects. We identified mutations under positive selection during initial formation of adenomas and later in malignant progression to adenocarcinomas, and also quantified how each successive driver changes selection on the next. Many trends of step-specific selection emerged that were consistent across POLE / POLD1 -mutant and MMR-deficient tumors, such as high early selection on APC and strong late selection on CTNNB1 and TP53 . However, step-specific selection varied across the steps of progression in mutants of other genes, including KRAS , BRAF , and FBXW7 . Inferred evolutionary trajectories point to early disruption of WNT and TGF-beta signaling, followed by later acquisition of TP53, MAPK, and PI3K pathway mutations. These findings provide a data-driven, high-resolution temporal sequence of the selective pressures shaping tumorigenesis. They identify potential biomarkers for early detection, and reveal how polymerase and mismatch-repair backgrounds alter evolutionary paths by purely altering gene- and site-specific mutation rates. Consistency with experimentally identified early drivers ( APC , KRAS ) and late drivers ( TP53 ) validates the approach and demonstrates its suitability for extending step-specific analyses to many additional cancer types. Ultimately, such high-resolution evolutionary trajectories enable precision-medicine targeted drug treatments to be exquisitely tailored to individual patients and facilitate treatments that not only target extant variants but also anticipate additional therapies that could be useful to forestall or prevent the next driver mutation and consequent burgeoning tumor burden.