Background: Guidelines support pharmacologic thromboprophylaxis in MM patients identified as “high-risk” for VTE. Tools for VTE risk assessment in MM are contradictory and have not been validated. Such risk stratification would allow for use of thromboprophylaxis in MM patients at high-risk of VTE while avoiding anticoagulant exposure in patients at low risk. We aimed to develop and validate a risk prediction model for VTE in MM. Methods: We identified patients starting chemotherapy for MM within the Veterans Administration between 9/1999 and 12/2013. Using a split-sample method, we randomly created derivation (2/3) and validation cohorts (1/3). Variables associated with an increased risk of VTE within 6-months of chemotherapy start (univariate p≤ 0.05) and those with a p≤ 0.10 and an effect consistent with prior literature were offered into a backward stepwise model. Variables were removed until remaining variables predicted VTE (p< 0.05 OR p≤ 0.30 plus consistency with prior literature). A Myeloma Clot Score (MCS) was developed based on the parameter estimates. Results: The derivation cohort included 3036 patients of who 371 developed VTE within 6-months of starting chemotherapy. The MCS (Table 1) had a c-statistic of 0.68 in the derivation cohort and similar in the validation cohort. Incidence of VTE over the 6-month study period by score is in Table 1. In the validation cohort, the Hosmer-Lemeshow test was nonsignificant, showing adequate calibration. Conclusions: We developed and validated a risk model for predicting VTE in patients with MM starting chemotherapy. This MCS could be used to select patients who are likely to benefit from thromboprophylaxis.

*Patients diagnosed 1999-2007