BLOOD PRESSURE VARIABILITY AS PREDICTOR OF CANCER THERAPY-RELATED CARDIOVASCULAR TOXICITY IN PATIENTS WITH MULTIPLE MYELOMA

Abstract

Objective: Blood pressure variability (BPV) is an independent predictor of cardiovascular (CV) events and overall mortality. The potential role of BPV in defining CV risk of patients undergoing anticancer therapy with a potential cancer therapy-related cardiovascular toxicity (CTR-CVT) is currently unknown. The aims of this study were: (i) to assess the predictive value of BPV indexes for CTR-CVT development and defining the corresponding cut-off values; (ii) to identify subgroups of subjects based on BPV, through clustering analysis. Design and method: Patients with Multiple Myeloma, who were eligible to anticancer therapy with Carfilzomib, were consecutively enrolled and underwent a baseline evaluation, including office blood pressure (BP) measurement, 24h-Ambulatory Blood pressure Monitoring (ABPM), PWV, and Echocardiography. Short-term BPV was assessed through standard deviation (SD), weighted standard deviation (wSD), coefficient of variation (CoV), average real variability (ARV), and variability independent of the mean (VIM), evaluated on ABPM measurements. Results: The study population included 124 patients (50% female). During follow-up, CTR-CVT occurred in 70 subjects (56.5% of the total population), including both BP and CV events. Night-time BPV was associated to CTR-CVT, independently by BP, age, smoking, and comorbidities as diabetes and renal dysfunction (night-time systolic CoV: adjusted OR 1.11 [1.01-1.22]; night-time systolic VIM adjusted OR 1.17 [1.03-1.35]; night-time diastolic VIM adjusted OR 1.18 [1.01-1.40]). Cut-off for these BPV parameters were identified as predictor of CTR-CVT occurrence: 10.5 for night-time systolic CoV (sensibility 73%, specificity 47%); 7.8 and 6.4 for systolic and diastolic night-time VIM (sensibility 85% for both; specificity 35% and 39%, respectively). Clustering analysis identified a subgroup of subjects characterized by the highest BPV, who had a greater prevalence of events during follow-up, but no differences in other conventional CV risk factors, as BP, dyslipidaemia, diabetes, renal function, and smoking habit, as well as left ventricular mass and global longitudinal strain. Conclusions: Short-term BPV is an independent predictor of CTR-CVT in patients with Multiple Myeloma undergoing therapy with Carfilzomib. The inclusion of these indexes improves CV risk prediction in these subjects and higher BPV is a distinctive feature predicting adverse events, on top of conventional CV risk factors.