Genomic instability, postoperative recurrence and therapeutic vulnerabilities in resectable non‑small cell lung cancer (Review)

Resectable non‑small cell lung cancer (NSCLC) is managed largely according to anatomical stage, pathological risk and actionable driver alterations, yet these factors do not fully explain postoperative recurrence. Genomic instability may contribute to recurrence by promoting clonal diversification, intratumoral heterogeneity, occult dissemination, persistence of residual tumor cells, and immune escape. In the present review, chromosomal instability (CIN), copy‑number complexity, whole‑genome doubling, DNA repair defects, replication stress, and extrachromosomal DNA (ecDNA) were critically evaluated using a three‑axis translational framework encompassing biological consequences, potential clinical roles, and strength of evidence. Current evidence suggests that clonal diversity and copy‑number complexity have the clearest near‑term prognostic rationale. By contrast, CIN and whole‑genome doubling are supported more strongly by evolutionary and mechanistic rather than prospective clinical evidence. Defects in DNA repair, replication stress, and ecDNA represent potential therapeutic vulnerabilities, but their clinical relevance remains to be established. To date, no treatment‑predictive biomarkers based on genomic instability have been identified for resectable NSCLC. Direct clinical evidence linking any specific genomic instability feature to the presence or longitudinal dynamics of postoperative molecular residual disease (MRD) remains limited. Postoperative circulating tumor DNA‑defined MRD provides prognostic information more directly related to residual disease but remains assay‑dependent and should not be considered a genomic‑instability phenotype. Therefore, features of genomic instability should remain investigational and should not replace established clinical, pathological, or molecular decision‑making. Their near‑term value lies in refining biological risk models and generating testable hypotheses for biomarker‑defined perioperative trials.