1823P Characterization of thrombosis risk in patients with cancer: Preliminary results

Venous thromboembolism (VTE) is a frequent complication of cancer. Nowadays, cancer patients are increasingly treated in the ambulatory care setting, thus most of the VTE events occur outside the hospital. The incidence of VTE has previously been reported to be highest in the first 12 months after cancer diagnosis. However, there are few prospective studies evaluating the risk of thrombosis in the Iberian population with cancer. Patients were included between February 2018 and December 2019 in a prospective observational study. Patients with breast in adjuvant, metastatic breast in treatment only with hormone therapy, non-metastatic and metastatic hormone-sensitive prostate, leukemia, cancer of the head and neck, larynx, melanoma and thyroid or receiving pharmacological VTE prophylaxis were excluded. In this analysis we report data of VTE and bleeding at 3 months. The Khorana risk model was also explored in our population. Among 1,336 cancer patients, 86 (6.4%) were diagnosed with VTE at 3 months. Sites of solid tumors with the highest rates of VTE included biliary (13.5%), pancreas (9.6%), NSCLC (7.8%), urothelial (7.5%), stomach (6.0%) and colorectal (5.8%). VTE tended to be more frequent in metastatic or locally advanced solid cancers (7.7% vs 3.6%; p=0.07). In 284 patients with central venous catheter, 8 (2.8%) had upper limbs VTE. Out of 86 VTE patients, 9 (10.5%) had venous compression by mass (p=0.04) and 12 (14%) reduced mobilization (p<0.001). 613 (45.9%) and 205 (15.3%) had a Khorana score ≥2 or ≥3 respectively. Khorana score showed low capacity to distinguish between patients who experienced/did not experience a VTE event at three months (AUC 0.53). Clinically relevant bleeding occurred in 30 (2.2%) patients. The most hemorrhagic tumors were stomach (5.7%), pancreas (4.9%) or gynecological (4.4%). The cancers who were particularly prone to VTE in our sample were pancreas, biliary or lung. Although our patients did not use pharmacological thromboprophylaxis, the hemorrhagic risk was relevant in upper digestive tumors. Khorana score is not a predictive tool for thrombosis in the Iberian population with cancer.