Venous Thromboembolism Risk in Patients with Newly Diagnosed Multiple Myeloma Treated with Carfilzomib or Bortezomib in Combination with Lenalidomide and Dexamethasone

Background: Multiple myeloma (MM) significantly increases the risk of venous thromboembolism (VTE). Additionally, treatments for MM, including immunomodulators and steroids, further increase the risk. Prior literature indicates that the regimen carfilzomib, lenalidomide, and dexamethasone (KRd) may have a higher incidence of thromboembolic events compared with bortezomib, lenalidomide, and dexamethasone (VRd). The ENDURANCE trial which randomized newly diagnosed patients to KRd or VRd suggested a modest increase in VTE in KRd from 2% to 5%. However, a subsequent retrospective study in a real-world population in the United Kingdom (Piedra, et al, Br J Haematol, 2021) reported an increase from 5% to 16%. The baseline VTE incidence from these two prior reports is substantially different and the relative impact of clinical trial eligibility and geographic factors is uncertain. Consequently, we aimed to evaluate VTE risk in KRd compared with VRd at a large academic medical center in the United States. Methods: A retrospective review of patients with newly diagnosed MM presenting for referrals at a single institution between January 2019 and December 2020 was conducted. Patients were followed for VTE events for 6 months following treatment initiation. Those who changed induction therapy were censored at that time. The primary endpoint was comparison of VTE rate between the two cohorts. Secondary analyses included Cox Proportional Hazards modeling of VTE risk. IMPEDE VTE risk score (Sanfilippo, et al, Am J Hematol, 2019) was calculated and controlled for in multivariate analysis. Results: There were a total of 209 patients included in the analysis. The median age was 64 (range 27-85), 63% were male, and 82% were White. 69 patients (33%) received KRd induction and 140 (67%) received VRd. The median number of cycles of each administered was 4. 84% of the KRd cohort underwent autologous stem cell transplantation (ASCT) during first remission compared to 81% of the VRd cohort (p = 0.703). 90% were commenced on thromboprophylaxis, 79% with aspirin and 11% with an alternative such as warfarin or direct oral anticoagulants. Baseline VTE risk was similar between the two cohorts, the median IMPEDE VTE score was 4 for each (p = 0.668). There was a trend for greater thromboprophylaxis in the KRd cohort, with 96% receiving aspirin or alternative compared to 87% of those in the VRd cohort (p = 0.138). Overall, 18 patients (9%) had a VTE event during the follow-up period, including 5 (7%) in the KRd cohort and 13 (9%) in the VRd cohort (p = 0.795). The median time to VTE was 35 days (range 2-153) in the KRd cohort and 67 days (range 21-173) in the VRd cohort. There was no statistically significant difference in VTE risk between treatment with KRd compared to VRd (HR 0.74; 95% CI 0.26-2.08; p = 0.568). This did not change after adjusting for thromboprophylaxis (aHR 0.93; 95% CI 0.32-2.68; p = 0.885) or IMPEDE VTE risk score (aHR 0.79; 95% CI 0.28-2.22; p = 0.649). Overall, thromboprophylaxis was associated with a 76% reduction in VTE risk (aHR 0.24; 95% CI 0.08-0.69; p = 0.008); this was similar among patients on KRd (HR 0.11) and on VRd (0.28). Conclusions: In this cohort of 209 patients, KRd was not associated with an increase in VTE compared to VRd, contrary to prior reports. The overall incidence of VTE in our cohort was higher than the ENDURANCE trial. Differences between our study and Piedra et al's may be explained by the duration of therapy, duration of follow up, and use of ASCT. Alternatively, this may represent different distributions of thrombotic risk factors in the underlying population or more selective use of KRd based on perceived risk. Further research exploring the relationship between exposure to carfilzomib and thrombosis is needed.