A VALIDATION, WITH NEW CLINICAL APPLICABILITY, OF A CLINICAL-GENETIC RISK MODEL THAT PREDICTS THROMBOSIS WITH HIGH SENSITIVITY IN PATIENTS WITH LYMPHOMA

The incidence of venous thromboembolic events (VTE) in patients with lymphoma is estimated around 10%, increasing the morbidity and mortality for this group of patients. Khorana scale has been developed to assess VTE risk in cancer patient but it is not useful for lymphoma patients. Recently, a new risk score has been developed to predict VTE in this specific population (Throly). Likewise, the incorporation of genetic variables into a VTE risk model improved VTE risk assessment in solid tumors (TiC-ONCO). In this context, the aim of our study was to evaluate if the same approach of TiC-ONCO is applicable to patients with lymphoma. Between 2014 and 2018, 254 lymphomas were diagnosed in our center, 33 of them presented VTE. The F5 rs6025, F5 rs4524, F13 rs5985, SERPINA10 rs2232698, in addition to clinical variables (immobilization, tumor type, Ann Arbor score, mediastinum extension, and personal thrombotic history) were significant associated to VTE risk and used to build the TiC-ONCO-associated Lymphoma score (TiC-Lympho). The Khorana, ThroLy and TiC-Lympho scores were compared through predictive value and area under the curve (AUC). We observed a cumulative incidence of VTE of 9% (Fig.1). The results of sensitivity, specificity, predictive value and AUC of the Khorana, Throly, and TiC-Lympho are shown in Table 1 and Table 2. Keywords: immunochemotherapy; non-Hodgkin lymphoma (NHL); prognostic indices.