Prediction of venous thromboembolism in newly diagnosed patients treated for lymphoid malignancies: validation of the Khorana Risk Score

The utility of the venous thromboembolism (VTE) risk assessment model known as the Khorana Risk Score (KRS) in patients with lymphoid malignancies receiving outpatient chemotherapy is not defined. We evaluated the association of the KRS with VTE in patients treated for diffuse large B cell lymphoma (DLBCL) or Hodgkin lymphoma (HL). Retrospective analyses were performed in 428 patients, 241 of whom were newly diagnosed with DLBCL and 187 of whom had HL. During the initial therapy, 64 (15%) patients developed VTE and 56 died during follow-up. More VTE events occurred in patients with DLBCL than in patients with HL. According to the KRS, 364 (85%) and 64 (15%) patients were considered to be at intermediate risk and high risk of VTE development, respectively. The high-risk KRS patients were more often diagnosed with HL than DLBCL (19 vs. 10%, P = 0.0143). The KRS did not discriminate between high-and intermediate-risk patients with respect to VTE occurrence (17 vs. 15%, P = 0.5868). In our patients, the KRS did not adequately predict VTE (positive predictive value 15%, negative predictive value 82% and C statistic 0.51). In the multivariate analysis, bulky disease (OR 2.34; 95% CI 1.62-3.36, P < 0.0001), poor prognostic disease (OR 1.32; 95% CI 1.01-1.74, P = 0.049) and DLBCL histological subtype (OR 1.61; 95% CI 1.17-2.19, P = 0.003) were all significantly associated with the VTE development. In this cohort of patients with lymphoid malignancies, the KRS did not adequately stratify or predict VTE events in patients at a higher risk of VTE. This finding suggests the need for the development of a disease-specific VTE assessment model.