Development of PREVEMM: A Venous Thromboembolism Risk Score for Newly Diagnosed Multiple Myeloma Patients

BACKGROUND: Venous thromboembolism (VTE) is an important complication in newly diagnosed multiple myeloma (NDMM), particularly during the first year after diagnosis. Existing risk assessment models show limited predictive accuracy. This study developed a novel risk scoring system using complementary approaches to improve VTE risk prediction in NDMM. METHODS: We retrospectively analyzed 225 NDMM patients from 4 hospitals (2014-2019) to develop a prediction score for first-year VTE. The model was built using Least Absolute Shrinkage and Selection Operator (LASSO) regularization with variables showing P < .25 in univariate analysis. Performance was assessed using area under the curve (AUC) and Hosmer-Lemeshow test. IMPEDE-VTE and SAVED scores were calculated for comparison. RESULTS: Among 225 patients (median age 68 years), 32 (14%) developed VTE, predominantly pulmonary embolism (71%) within the first year after diagnosis. The LASSO model had a c-statistic of 0.86 and identified ECOG ≥ 1, high R-ISS stage, high-dose dexamethasone (> 160 mg per cycle), Lactate dehydrogenase (LDH) > 370 IU/L, immunoparesis, spinal cord compression, and paralysis as independent predictors (AUC: 0.86). LASSO-derived PREVEMM score significantly outperformed existing tools (AUC: 0.88 vs 0.61 for both IMPEDE-VTE and SAVED). CONCLUSION: PREVEMM score demonstrates superior VTE risk stratification capabilities, supporting its potential clinical utility for individualized thromboprophylaxis in NDMM.