Beyond the left ventricle: CMR assessment of right ventricular dysfunction during cancer therapy

Abstract

Cancer therapy can cause left and right ventricular (RV) cancer therapy-related cardiac dysfunction (CTRCD). How patients transition from having normal myocardial function to isolated left ventricular (LV), RV, or biventricular (BV) CTRCD, and the associations of these phenotypes with post-therapy prognostic measures remain unknown. We aimed to define patterns of ventricular dysfunction during anthracycline and trastuzumab therapy. This ancillary study of the EMBRACE-MRI study included women with early-stage HER2-positive breast cancer treated with sequential anthracycline and trastuzumab, prospectively enrolled between 2014 and 2021 at our Hospitals. LV and RV function were assessed by CMR pre-anthracyclines, post-anthracyclines, at 3 and 6 months during trastuzumab therapy, and early post-completion of trastuzumab. All CMR analyses were performed using CVI42. Cardiopulmonary exercise testing was performed early post-completion of trastuzumab. LV CTRCD was defined based on the CREC criteria while RV CTRCD was defined as a >10% reduction in RVEF from baseline to <49%. Patients were classified as having isolated LV CTRCD, isolated RV CTRCD, or biventricular (BV) CTRCD based on the first CTRCD event. Descriptive statistics were used to summarize baseline clinical characteristics, the timing of CTRCD, and CMR data. Between-group comparisons were evaluated using Wilcoxon rank-sum tests for continuous variables. Among the 153 patients (median age 51.4, IQR: [43.5-57.5] years), the first CTRCD event included LV, RV, or BV CTRCD in 36 (24%), 4 (3%) and 6 (4%) patients respectively (Figure 1). The first instance of RV, LV, and BV CTRCD was 3 months, 3 to 6 months, and 12 months after trastuzumab initiation respectively. Patients also transitioned between phenotypes, with one patient with initial RV CTRCD developing BV CTRCD. Baseline demographics characteristics were similar between groups with no difference in cumulative anthracycline dose. However, patients with RV/BV CTRCD had a lower baseline RVEF (58 [55-59] % vs 61 [58-64]%, p=0.02) and more frequently received right-sided radiation (62% vs 35%) (Figure 2). Post-completion of trastuzumab, LV CTRCD patients had lower VO2 max (17.1 vs 19.2 mL/kg/min, p=0.059) and significantly reduced predicted VO2 max (73% vs 89%, p=0.014) compared to RV/BV CTRCD. In this cohort of HER2-positive breast cancer patients treated with anthracycline and trastuzumab LV CTRCD was the commonest type of CTRCD, however RV/BV CTRCD occurred in 7% of patients with distinct temporal patterns. LV CTRCD was associated with lower post-treatment functional capacity and RV CTRCD may be more common in those receiving right-sided radiation. These findings underscore the importance of systematic BV monitoring throughout cancer therapy. The long-term prognostic implications of LV vs RV CTRCD is a focus on ongoing investigation.Flow chart, transition between states  Baseline characteristics