Abstract
Cancer patients are at increased risk of venous thromboembolism (VTE), representing one of the main causes of death. In this paper, we used the ONCOTHROMB12-01 study cohort for the analysis of 18-month overall survival (OS) and cancer-associated venous thrombosis (CAT) cumulative incidence. The OS is described for patients with and without CAT across age, sex, cancer type, stage and presence of recurrent VTE using the Kaplan-Meier method and log-rank test. Moreover, a Cox regression model with stepwise feature selection is trained to identify predictors of prognosis. Results show an 18-month CAT cumulative incidence of 16.5 % (CI 13.1-21.0 %) and a median OS for patients with CAT of 9.1 months (CI 5.9-12.5). Furthermore, CAT is associated with shorter survival in lung, colorectal and pancreatic cancer in advanced stages, regardless of age and sex; VTE recurrence also reduces survival significantly. The model found ECOG performance status of 2, cancer type, metastatic stage, VTE, mucinous histology and complete tumor resection as meaningful predictors; the latter being the only protective factor. In conclusion, the diagnosis of CAT has a profound impact on OS in the ONCOTHROMB12-01 study regardless of the clinical characteristics of patients.