Complications in patients with cancer-associated thrombosis on tinzaparin by cancer site and frailty: a pooled analysis.

Cancer-associated thrombosis (CAT) is a frequent and severe complication of malignancy. Although patients undergo anticoagulation, risks of recurrent venous thromboembolism (VTE) and major bleeding (MB) remain substantial, but data are limited according to specific tumor sites and in patients with frailty. Our objective was to evaluate 6-month incidences of recurrent VTE, MB, and death in patients with CAT treated with tinzaparin, and to assess the influence of tumor site and frailty on these risks. Individual patient-level data from 1413 patients were pooled from 3 prospective cohort studies and the tinzaparin arm of a randomized controlled trial. Frailty was defined as age of ≥75 years, creatinine clearance of <50 mL/min, body weight of ≤50 kg, or Eastern Cooperative Oncology Group performance status of ≥2. Cumulative incidences were estimated using Kalbfleisch and Prentice method and plotted to illustrate the benefit-risk balance by tumor site and frailty. At 6 months, cumulative incidences of recurrent VTE and MB were 6.2% and 3.4%, respectively. Recurrent VTE incidence surpassed MB incidence in gastrointestinal, lung, genitourinary, and gynecological cancers, whereas MB risk exceeded VTE risk in breast cancer. Among patients with frailty criteria, recurrent VTE risk increased with the number of frailty factors, whereas MB remained relatively stable. In patients with CAT receiving tinzaparin, VTE recurrence, unlike bleeding risk, varies according to tumor site and patient frailty. In frail patients with CAT treated with standard full-dose tinzaparin, these findings do not support routine dose modification.