Abstract
Background:
Multiple Myeloma (MM) is associated with increased risk for venous thromboembolism (VTE). Treatment, such as dexamethasone, immunomodulatory drugs (IMID), alkylating agents, and doxorubicin, alter hemostatic pathways and thus promote thrombogenesis 1. MM patients with VTE have a 3-fold increase in mortality compared to those without VTE, so identifying those at risk and aiming to prevent VTE events is important 2. Several clinical VTE risk prediction scores have been developed, including the SAVED score, IMPEDE VTE score, and more recently the PRISM score 2,4,5. The National Comprehensive Cancer Network suggests that patient with MM on IMID therapy should be on aspirin, or therapeutic anticoagulation for those at "high risk"3. However, it remains unclear which risk model, if any, should be used.Our objective was to validate the three published risk assessment tools in a community setting and assess the predictive ability of each.
Methods:
We conducted a retrospective chart review of all patients with newly diagnosed multiple myeloma who started chemotherapy at Gundersen Health System (La Crosse, WI) between 2010 and 2020 who had at least 6 months of follow up documented. Patients with prior indication for ongoing therapeutic anticoagulation or a diagnosis of VTE within 6 months prior to starting therapy were excluded. Total scores for IMPEDE VTE, SAVED and PRISM scores were calculated from the chemotherapy start date. Statistical analysis included Chi-square, Fisher's exact and Wilcoxon rank sum tests, and Kaplan Meier survival analysis. A p-value ≤ 0.05 was considered significant and all analysis was completed in SAS version 9.4.
Results:
Our cohort contained 123 patients diagnosed with MM. Average age was 68 years (SD 12.1, range 37-92). Our study included 68 (55%) males and 55 (45%) females with 121 (98%) being White/Caucasian. The mean BMI of patients was 29.4 kg/m2 (SD 7.0, range 18.6-54.4). Kaplan Meier survival analysis showed a 5-year survival rate of 53.1% (95% CI [42.7%, 63.4%]). In the entire cohort, 10 (8.1%) patients were diagnosed with VTE (as compared to 5.8% in IMPEDE study, 8.7% in SAVED study and 8.2% in PRISM study) with 80.0% occurring within 6 months of treatment start date. Aspirin was the most frequently used agent for thromboprophylaxis with 88 (86.3%) patients receiving either 81, 162, or 325 mg of aspirin. IMID therapy was given to 76 (61.8%) patients, 114 (92.7%) received dexamethasone and 114 (92.7%) received proteasome inhibitors. Amongst those on IMIDs, 72 (94.7%) patients received prophylaxis, most commonly aspirin. Abnormal metaphase cytogenetics were noted in 104 (85.4%) patients.
Neither the IMPEDE VTE (p=0.6), SAVED (p=0.9) nor PRISM risk scores (p=0.3) were able to statistically predict VTE outcome in our patient population. Using the IMPEDE score, 7 patients in the intermediate risk group and 3 patients in the low-risk group had a VTE. In the SAVED model, 5 patients in the low-risk group and 5 patients in the high-risk group had a VTE. Using the PRISM risk score, all 10 of the patients with VTE were in the intermediate risk group. Most patients who were on IMID therapy fell into the intermediate risk group on the IMPEDE VTE and PRISM scoring systems, and the SAVED score had an approximately equal patient distribution between the high risk and low risk group.
Conclusions:
Our patients with multiple myeloma had similar rates of VTE as compared to the published models, with the majority occurring in the first 6 months of chemotherapy. In total, our patients on IMID therapy received appropriate prophylaxis with aspirin. Overall, 8.1% of our patients had a VTE event. However, none of the three risk models were able to predict the development of VTE. In fact, many of the VTE events occurred in patients who were felt to be low or intermediate risk. While the sample size is small and from a single health system, we had excellent follow up and ability to closely examine each chart for treatment and outcomes. Further efforts should focus on collaboration across institutions to increase the sample size, to validate and compare existing models. The majority of myeloma treatment occurs in the community; thus, it is important to ensure the findings are reproducible in that patient population.
Disclosures
No relevant conflicts of interest to declare.
Circulating Plasma Cells as a Biomarker to Predict Newly Diagnosed Multiple Myeloma Prognosis: Developing Nomogram Prognostic Models