Risk of cancer after incident venous thromboembolism or cerebral venous thrombosis: Evidence from an Italian administrative healthcare database cohort (2002-2023).

BACKGROUND: Unprovoked venous thromboembolism (VTE) is a marker of occult malignancy. While this association is well established for deep vein thrombosis (DVT) and pulmonary embolism (PE), evidence for cerebral venous thrombosis (CVT) remains limited. We evaluated the incidence and risk of newly diagnosed cancer following DVT/PE and CVT. METHODS: Using administrative healthcare databases from the Lombardy region (2002-2023), we included subjects aged ≥30 years hospitalised for incident DVT/PE or CVT and age- and sex-matched controls. Cancer incidence was assessed by cumulative incidence functions (CIFs) and standardized incidence ratios (SIRs), with subdistribution hazard ratios (sHRs). RESULTS: 109,371 DVT/PE (mean age 73 years; 57% female) and 1929 CVT patients (mean age 53 years; 68% female) were included. During follow-up, cancer was diagnosed in 10% and 4% of DVT/PE and CVT patients, respectively, corresponding to a higher cancer risk in DVT/PE (sHR 1.73 95% CI, 1.68-1.79) and CVT patients (sHR 1.42 95% CI 1.08-1.89) in comparison with matched controls. Incidence differences narrowed over time; SIRs of 3.85 [95% CI 2.11-7.26] and 4.61 [95% CI 4.37-4.87] at 1 year, and 1.47 [95% CI 1.10-1.95] and 2.12 [95% CI 2.06-2.19] at 5 years, in CVT and DVT/PE, respectively. In comparison with control group, the incidence of cancer was always higher in the DVT/PE group while for CVT patients for hematologic and central nervous system (CNS) cancers. CONCLUSIONS: Similar to unprovoked VTE, CVT should be considered a marker of occult cancer, particularly hematologic and CNS malignancies, with the highest vigilance warranted during the first year.