Validation and Exploratory Refinement of the HFA-ICOS Score for Cardiovascular Risk in Proteasome Inhibitor-Treated Multiple Myeloma: Single-Center Retrospective Study

Background: Proteasome inhibitors (PIs) are integral in multiple myeloma (MM) treatment but carry a substantial risk of cardiovascular adverse events (CVAEs). The Heart Failure Association–International Cardio-Oncology Society (HFA-ICOS) risk score was designed to identify patients at risk of cardiotoxicity, but its performance in MM remains uncertain. Methods: We retrospectively evaluated 98 patients with MM or primary amyloidosis treated with PIs between 2019 and 2024 to externally validate and refine this score. Results: CVAEs occurred in 22 patients (22.5%), predominantly heart failure. The original HFA-ICOS model demonstrated modest predictive accuracy (AUC 0.66) and sensitivity (50%), with frequent risk overclassification. Carfilzomib exposure (HR 4.68, 95% CI 1.47–14.90; p = 0.009) and elevated NT-proBNP before cycle 2 (>300 pg/mL; HR 3.13, 95% CI 1.10–8.93; p = 0.033) independently predicted CVAEs. A dynamic model incorporating these parameters and adjusting the age threshold to ≥65 years was associated with improved discrimination (AUC 0.72, p = 0.032), model fit (ΔAIC = −4), and CVAE-free survival stratification (p = 0.026), achieving 90.9% sensitivity. Conclusions: These findings indicate that the original HFA-ICOS score has limited prognostic value in PI-treated MM patients. Incorporating early NT-proBNP monitoring, carfilzomib exposure, and refined age categorization may improve risk prediction and support more personalized cardiovascular surveillance strategies in cardio-oncology. However, this refined dynamic model should be regarded as exploratory and requires validation in larger independent cohorts before it can be considered for clinical application.