Abstract 4015: Whole-genome duplication predicts metastasis risk and timing across human cancers

Abstract Introduction: Whole-genome duplication (WGD) is a pervasive event in cancer evolution, conferring tolerance to large-scale genomic alterations and reshaping selective pressures. While WGD is linked to aneuploidy and poor prognosis, its impact on metastatic risk across cancer types remains unclear. We evaluated whether WGD detected in the primary tumor predicts subsequent metastasis risk and timing in a pan-cancer cohort. Methods: We analyzed 33,349 tumors across 29 cancer types profiled by MSK-IMPACT (2014-2024). Metastatic site and timing were derived from NLP-based annotations integrating pathology and radiology reports. Analyses were restricted to stage I-III primary tumors without prior metastasis. Landmark analyses at 1 and 3 years compared distinct metastatic sites among patients with sufficient follow-up, excluding cancer types with fewer than 50 eligible patients at the 3-year landmark. Within each cancer type, nonparametric cumulative incidence functions were estimated with death as a competing event. WGD was inferred from FACETS allele-specific copy numbers, defining WGD-positive as >50% of autosomes with major copy number ≥2. Results: In total, 6,434 primary tumors met inclusion criteria, comprising 5,170 WGD- and 1,264 WGD+ cases. Across cancer types, WGD+ tumors showed a significantly higher likelihood of metastasis within three years compared with WGD- tumors (60.4% vs. 42.0%, p<0.001), a pattern consistent within individual cancer types. WGD+ tumors also exhibited a greater metastatic burden, with a higher proportion of patients developing metastases to ≥3 sites (20.4% vs. 11.3%, p<0.001). WGD+ tumors were more likely to develop brain metastases overall (4.7% vs. 2.6%, p=0.003). One-third (32.2%) of WGD+ tumors lacked TP53 alterations, indicating genome doubling can arise independently of canonical TP53 loss. WGD significantly predicted metastatic progression in both TP53-altered (HR=1.10, p=0.043) and TP53-wildtype tumors (HR=1.38, p<0.001), with a larger effect size observed in the TP53-wildtype subset. Conclusions: In this pan-cancer, primary-only analysis, WGD emerged as an early and consistent predictor of metastatic progression across tumor types, with effects persisting in TP53-wildtype disease. These findings position WGD as a unifying hallmark of cancer progression and highlight how large-scale genomic-clinical integration can reveal key determinants of tumor evolution. Citation Format: Jierui Xu, Henry Walch, Maria Perry, Christopher Fong, Justin Jee, Karl Pichotta, Adam Price, Allison L. Richards, Chaitanya Bandlamudi, Mark Donoghue, Nikolaus Schultz, Walid K. Chatila. Whole-genome duplication predicts metastasis risk and timing across human cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 4015.