Cardioprotection against cardiac dysfunction from breast cancer chemotherapy: a bayesian and frequentist network meta-analysis of randomised controlled trials

Background Anthracyclines (ANT) are the cornerstone of multiple chemotherapy regimens but at the risk of potential cardiotoxicity. Previous RCTs have tested the prophylactic effects of multiple cardioprotective agents to prevent ANT-related cardiotoxicity. Unfortunately, attempts to combine RCT findings in previous meta-analyses have been heterogeneous, creating further uncertainty. There remains an unmet need to determine the role of cardio-protective agents in breast cancer. Purpose To assess the comparative efficacy of cardioprotective drugs in patients with breast cancer using both Bayesian and frequentist analyses of randomised controlled trials Methods We performed a systematic review using four databases (CENTRAL, Cochrane Reviews, MEDLINE, SCOPUS), to find RCTs evaluating cardio protective drugs in breast cancer patients without prior ANT exposure. The population included was anthracycline naïve, and trials were excluded if cardio-protective agents were commenced post anthracycline treatment. The primary outcome was a mean change in LVEF pre and post ANT dosing. Results were pooled with both Bayesian and frequentist approaches using random effects models in R statistical software. Results We identified 12 RCTs from 2807 search results (n=1126, Age 51 years, ANT dose 412m/m2, baseline LVEF 62.6%) with comparisons including beta-blockers (BB) (n=9), Angiotensin Converting Enzyme inhibitors (ACEi)/Angiotensin Receptor Blockers (ARB) (n=3), combination BB + AA (n=2), spironolactone (n=1) and statins (n=1). All included trials had either intermediate or high risk of bias, with marked heterogeneity in ANT dosing and LVEF monitoring. Overall, our Bayesian network meta-analysis showed no statistically significant difference in mean LVEF preservation between AA (1.3%, 95% credible interval [−0.20, 2.9]), BB (0.77, [−0.21, 1.8]), AABB (0.84 [−1.1, 2.8]), SPR (0.72, [−2.3, 3.7]) or statin (0.60, [−2.4, 3.6]) when compared against placebo. After ranking for efficacy, ACEi/ARBs achieved the most protection against LVEF decline of 1.3% [95% CI: −0.2, 2.9] although still not significant. Conversely, frequentist analysis showed benefit in using AA (Standardised Mean Difference (SMD) 1.32% [0.32, 2.33]) and BB (SMD 0.76% [0.12, 1.4]). Conclusion Bayesian analysis demonstrated no difference in LVEF with cardio-protective agents. In contrast, frequentist analysis showed that AA and BB may provide significant cardio-protection. The quality of RCT data to date is limited by a high risk of bias and significant heterogeneity between RCA reporting. Larger trials with clear population definition are required to determine whether any drug class provides benefit in this setting.