Looking beyond LVEF: cardiac and skeletal muscle function in long-term breast cancer survivors with normal versus reduced peak oxygen uptake

Background Peak oxygen uptake (VO2peak) is a marker of cardiovascular disease (CVD) and heart failure (HF) risk that is gaining interest in breast cancer survivors (BCS) treated with cardiotoxic therapy. However, the factors contributing to reduced VO2peak in this population remain unclear. Purpose To compare exercise magnetic resonance imaging (MRI) derived measures of cardiac and skeletal muscle function in BCS with normal (≥85% predicted) or reduced (<85% predicted) VO2peak. Methods 31 older (≥60 yrs) BCS previously treated (≥1 year) with cardiotoxic therapy (anthracycline chemotherapy and/or trastuzumab) without established CVD were recruited. VO2peak was measured from a maximal cardiopulmonary exercise test (cycle ergometer), and used to classify BCS into normal (≥85% predicted) or reduced VO2peak (<85% predicted) based on age, sex and height predicted values. Ventricular volumes, hemodynamics, and arterio-venous oxygen difference (a-vO2diff) was assessed with MRI at rest and during sub-maximal stepping exercise at ~60% of VO2peak. MRI-derived measures of peak calf muscle VO2peak (mVO2peak), blood flow and a-vO2diff (popliteal vein) were also assessed during an incremental to maximal plantar flexion exercise test. Groups were compared using analysis of covariance with adjustment for age. Results BCS with reduced VO2peak (n=18; 71±8% predicted) vs normal VO2peak (n=13; 99±13% predicted) were older (72±5 vs 67±3 yrs, P=0.004), tended to be leaner (BMI: 25.6±4.1 vs 28.3±5.6 kg/m2, P=0.14) and were evaluated a similar time post-treatment (15±6 vs 13±5 yrs, P=0.23). The reduced VO2peak group had markedly lower VO2peak (22.2±4.5 vs 17.3±3.6 mL/kg/min, P=0.002), despite similar peak respiratory exchange ratio (1.32±0.13 vs 1.30±0.13, P=0.56) and maximal heart rate (HR; 97±13 vs 101±10% predicted, P=0.40). At rest, both groups had comparable left-ventricular ejection fraction (LVEF; 59±5 vs 62±8%, P=0.30), but BCS with reduced VO2peak had smaller resting end-diastolic volume index (68±8 vs 77±15 mL/m2, P=0.033), and as a consequence, stroke volume index (SVi; 40±4 vs 47±6 mL/m2, P<0.001). During stepping exercise (Fig 1), BCS with reduced VO2peak had similar a-vO2diff (+0.3 mL/dL, P=0.75) but significantly decreased cardiac index (CI; -0.6 L/min/m2, P=0.031) secondary to decreased SVi (-7.3 mL/m2, P=0.035), as HR (-2 b/min, P=0.70) and LVEF (-1%, P=0.61) were comparable. In contrast, during small muscle mass exercise (Fig 2) there were no differences in calf mVO2peak (+2 mL/min, P=0.71), muscle blood flow (+5 mL/min, P=0.94) or a-vO2diff (+0.5 mL/dL, P=0.32). Conclusions Reduced VO2peak in long-term BCS is secondary to smaller ventricular size and blunted exercise cardiac augmentation, rather than impairments in skeletal muscle (at the small muscle level). This highlights the importance of further studies to evaluate the prognostic significance of reduced VO2peak for future CVD and HF events in BCS.