D-Dimer Improves Risk Prediction of Venous Thromboembolism in Patients with Multiple Myeloma

Introduction: Venous thromboembolism (VTE) is a common cause of morbidity and mortality among patients with multiple myeloma (MM). Thromboprophylaxis is a safe and effective way to decrease VTE in other high-risk populations. Recently, a clinical prediction model was developed to identify patients with newly diagnosed MM starting chemotherapy at highest risk of VTE. The model, IMPEDE VTE, found the following clinical risk factors for VTE: immunomodulatory drugs, body mass index, recent pathologic fracture of the femur or hip, erythropoietin stimulating drugs, dexamethasone, doxorubicin, Asian ethnicity/race, history of VTE, tunneled line or central venous catheter, and existing use of aspirin or anticoagulation. External validation of the model has yielded a c-statistic for VTE risk prediction of 0.64 to 0.65. Laboratory parameters can predict VTE in some patients with cancer. Accordingly so, the addition of laboratory parameters to IMPEDE VTE has the ability to improve model performance. Thus, we sought to determine the association between soluble P-selectin and D-dimer with the development of VTE in patients with newly diagnosed MM starting chemotherapy.

Methods: We identified 545 patients from the Washington University in St. Louis MM banking protocol, with available plasma from time of diagnosis (2007-2019). Thirty-eight cases of VTE were identified within 6 months following treatment initiation. An additional 137 patients were randomly selected as controls. D-dimer and soluble P-selectin ELISA assays were performed on the banked plasma by Eve Technologies, who was blinded to case vs. control status. Both assays were performed in duplicate and results averaged. All additional variables were collected through manual chart abstraction. IMPEDE VTE scores were calculated as we previously described (Sanfilippo et al.). The association of D-dimer and soluble P-selectin with VTE risk was assessed using Cox regression, adjusting for IMPEDE VTE score.

Results: The median age of all 545 patients was 65 (range 32-79), 54% were male, and 85% were white. All patients received novel chemotherapy agents for first-line MM therapy and 66% underwent autologous stem cell transplant. Of the 38 cases with VTE, 20 patients had deep vein thrombosis, 17 had pulmonary emboli, and 1 patient had concurrent events. The median time from chemotherapy initiation to VTE was 51 days (range 4-193). The median IMPEDE VTE score was 5 (range -1 to 12). Each unit increase in IMPEDE VTE score was associated with a 21% increase in risk for VTE (HR 1.21; 95% CI 1.04-1.40; p = 0.01).

Median D-dimer was 11,795 ng/mL (range 280-144,832). Each 1000 ng/mL unit increase in D-dimer was associated with a 2% increase risk for VTE after controlling for IMPEDE VTE score (aHR 1.02; 95% CI 1.01-1.04; p < 0.001). Patients in the highest quartile of D-dimer levels, above the 75th percentile, had a 2-fold increase in risk of VTE after adjusting for IMPEDE VTE score (aHR 2.04; 95% CI 1.03-4.02; p = 0.04).

Median soluble P-selectin was 189 ng/ml (range 23-638). There was no association between soluble P-selectin level and VTE risk in patients with MM.

Conclusions and Relevance: D-dimer is predictive of VTE in patients with MM starting chemotherapy. The combination of D-dimer and the IMPEDE VTE score can improve identification of patients at high risk of VTE and therefore allow for selection of primary thromboprophylaxis among patients with MM.