Mutational status and thromboembolic risk in lung, gastrointestinal and breast cancer patients.

Abstract

e23147 Background: Cancer-Associated Thrombosis (CAT) is one of the most threatening complications of cancer. Recent evidences suggested a link between the molecular profile of solid tumors and the incidence of CAT. The aim of this study was to explore the relationship between the mutational status of breast, lung and gastrointestinal cancer patients and the risk of CAT. Methods: We retrospectively evaluated the molecular profile, analysed as per clinical practice, of all consecutive patients hospitalized at the National Cancer Institute’s Department in Milan from October 2017 to November 2018. Patients with previous thromboembolic events and patients under anticoagulant therapy at cancer diagnosis were excluded. Due to death as competing risk, the Fine and Gray proportional regression model was used to detect statistical association and estimate relative risk. Results: The resulting cohort consisted of 484 patients, of whom 47% had gastrointestinal cancers, 18% had lung cancer and 15% had breast cancer. Molecular investigations were available for 375 (77%) patients; in particular, a 50-gene Next Generation Sequencing (NGS) panel was performed on 148 (31%) patients. After a median follow up of 17 months, 118 patients (24%) exhibited clinical manifestations of thrombosis (i.e. deep vein thrombosis, pulmonary thromboembolism, splanchnic thrombosis, disseminated intravascular coagulation, arterial thrombosis) and 117 (24%) patients deceased without thrombotic events. A statistically significant association was observed between incidence of CAT and presence of TP53 (HR 0.50, p = 0.04), c-KIT (HR 4.30, p = 0.041), and SMAD4 (HR 3.19, p = 0.029) mutations. No significant association was detected for KRAS and MET gene mutations, even if HRs were >2. Conclusions: In this study, the mutational status of TP53, SMAD4 and c-KIT genes was statistically associated to the risk of thrombosis. Due to methodological limits and low prevalence of mutations, large prospective studies are warranted, with the aim of better defining the role of oncogenes in CAT risk.