Jones GD, Brandt WS, Shen R, et al. - Among patients with fully resected stages I to III lung adenocarcinoma (LUAD), researchers undertook this prospective cohort analysis to determine the tumor genomic factors independently related to recurrence, as well as to construct a computational machine-learning prediction model (PRecur) to ascertain if the combination of genomic and clinicopathologic characteristics could predict risk of recurrence better than the TNM system. They included 426 patients treated from January 1, 2008, to December 31, 2017, at a single large cancer center and selected in consecutive samples. Changes in SMARCA4e and TP53 and fraction of genome altered were shown to be independently related to relapse-free survival (RFS). The performance of the PRecur prediction model for prediction of RFS was better than that of the TNM-based model. Overall, findings indicate that an enhanced risk stratification as well as prediction of recurrence following surgical resection of early-stage LUAD can be achieved by combining tumor genomics and clinicopathologic characteristics. Enhanced detection of cases at risk for recurrence may improve accrual to adjuvant therapy clinical trials.