Circulating tumor DNA predicts venous thromboembolism in patients with cancers

Background: Despite rapid advances in liquid biopsy for circulating tumor DNA (ctDNA), its prognostic value for venous thromboembolism (VTE) in patients with cancer is underexplored, particularly in underserved and minoritized populations. Objectives: To evaluate the role of ctDNA in risk stratification for cancer-associated VTE. Methods: We analyzed data from 1038 cancer patients who underwent ctDNA measurement for oncologic care at a large safety-net hospital system in the United States. We investigated the association between ctDNA and VTE after adjusting for cancer type, stage, treatment, and time from initial diagnosis using Fine-Gray models. We further assessed the discrimination of the genetic, clinical-only, and combined models using the area under the time-dependent receiver operating characteristic curve (AUC). Results: The presence of pathogenic ctDNA was independently associated with VTE after adjusting for clinical variables. Independent of tumor type, the number of pathogenic ctDNA mutations was predictive of future VTE risk (adjusted subdistribution hazard ratios of 2.75, 1.94, and 1.38 for >= 3, 2, and 1 pathogenic mutation, respectively, compared with none; P < .0001). The association was primarily driven by mutations in KRAS, PTEN, CDKN2A, NF1, and EGFR genes. Compared with the clinical-only model (AUC, 0.71; 95% CI, 0.64-0.76), the combined clinical and ctDNA model had a numerically higher time-dependent AUC (AUC, 0.74; 95% CI, 0.67-0.80). Conclusion: ctDNA testing may serve as an adjunctive tool to clinical risk assessment models in cancer patients to improve personalized VTE risk assessment and management.