Sex-Related Differences in Therapeutic Response to Mineralocorticoid Receptor Antagonists in Heart Failure: Summarizing Trial Evidence

This article refers to ‘Sex differences in mineralocorticoid receptor antagonist trials: a pooled analysis of three large clinical trials’ by X. Rossello et al., published in this issue on pages 834–844. Sex-related differences in cardiovascular diseases have been overlooked for long time, but it is now becoming more and more evident that improving their understanding is key to precision medicine's success. Heart failure (HF) represents the paradigm of a medical condition with profound differences between sexes covering the whole syndrome, from aetiology to treatment. The overall lifetime risk of HF appears comparable in men vs. women but HF prevalence seems to be higher in women.1 Phenotypic presentation typically differs between the two sexes, with higher prevalence of HF with preserved ejection fraction (HFpEF) in women and of HF with reduced ejection fraction (HFrEF) in men. Also, the comorbidity burden shows sex-related disparities, with women more likely to be older, with higher New York Heart Association (NYHA) class and natriuretic peptide levels, affected by hypertension and chronic kidney disease, whereas males more likely to suffer from diabetes and ischaemic heart disease.2, 3 Despite the lower quality of life and the more severe symptoms, which have been repeatedly reported in women vs. men with HF, women appear to ‘benefit’ from a prognostic advantage which is consistent across the ejection fraction spectrum.3 Sex-related disparities in HF treatment need also to be acknowledged. With few exceptions, the recruitment of females in the largest HF randomized controlled trials (RCT) has been limited, raising concerns about generalizability of RCT evidence to the overall HF population. Several real-world data analyses have shown lower use of HF guideline-recommended therapies in women vs. men, with women more likely to receive diuretics.3-6 Sex-related differences in pharmacokinetics and pharmacodynamics might influence the net benefit of HF medications and the occurrence of side effects in women vs. men, explaining, at least partially, the under-treatment often observed in women.2 Based on these considerations, the need for solid data supporting an appropriate use of HF evidence-based therapies in women has become a burning issue. A previous meta-analysis of RCTs showed angiotensin-converting enzyme inhibitors (ACEi) and beta-blockers being effective regardless of sex.7 Sex-related differences in therapeutic response have been reported in a post-hoc analyses of the DIG trial, with digoxin being associated with higher risk of any death in women.8 In the HFrEF trial PARADIGM-HF the benefit of sacubitril/valsartan in terms of mortality/morbidity improvement was consistent regardless of sex, whereas in the more ‘nebulous’ world of HFpEF, a pre-specified subgroup analysis of the PARAGON-HF trial suggested sacubitril/valsartan to reduce cardiovascular mortality/total HF hospitalizations in women but not men.9, 10 Mineralocorticoid receptor antagonists (MRAs) have IA recommendation for the treatment of HFrEF patients remaining symptomatic despite treatment with an ACEi and a beta-blocker and, among the currently used HF medications, are those which have also shown potential efficacy in HFpEF.11 Although with the several limitations linked with the use of subgroup analyses (e.g. low statistical power), no sex-related differences in outcomes have been reported for MRAs in HFrEF.12, 13 A previous post-hoc analysis of TOPCAT-Americas in HFpEF/HF with mid-range ejection fraction suggested spironolactone to reduce all-cause mortality in females but not in males.14 In this issue of the journal, Rossello et al.15 pooled data from the three largest RCTs on MRAs in HF in order to assess potential sex-related differences in treatment response by an adequately powered analysis. The first important finding was that differences in most of the important patient characteristics according to sex were consistent with those reported in large registries, i.e. women were older, had higher NYHA class, were more likely suffering from hypertension and impaired renal function but less likely to report ischaemic heart disease, were less likely to receive renin–angiotensin system inhibitors but more likely to receive diuretics.3 Similarly, the outcome analysis was also consistent with results from registries, i.e. better survival in women vs. males across the ejection fraction spectrum.3 Since transferring evidence from RCTs to the real world may be limited by the differences between trial populations and patients encountered in routine practice, the study by Rossello et al. offers the opportunity to confirm previous evidence but in a different setting. In a trial setting, the Authors could convincingly demonstrate similar treatment effect for MRA across sexes, i.e. 31% relative risk reduction of cardiovascular death/HF hospitalization in women vs. 29% in males. This analysis had the great strength to be based on individual patient-level data, which allowed also to adjust for potential confounders and perform subgroup analysis. As expected, the risk reduction associated with MRA use was greater in HFrEF compared to HFpEF. However, the ≈20% relative risk reduction shown in both men and women with HFpEF (i.e. TOPCAT-Americas population) was not statistically significant, which is probably explained by the smaller sample size of the HFpEF vs. HFrEF population considered in this study. Notably, in HFpEF there was no statistically significant interaction between sex and treatment, leading to conclude that the benefit associated with MRA use in the overall TOPCAT-Americas analysis has to be interpreted as consistent in both men and women.11 This is a key result since the proportion of women is even higher in HFpEF vs. HFrEF. Findings from Rossello et al. partially contradict a recent post-hoc analysis from the TOPCAT-Americas, which reported spironolactone to reduce all-cause mortality in females but not in males.14 The Authors attributed these discrepancies to chance-finding reflecting the small number of events in TOPCAT-Americas rather than a true influence of sex. However, different underlying mechanisms are involved in the pathogenesis of HFpEF vs. HFrEF and influence of sex-related factors might exert a stronger effect in HFpEF vs. HFrEF pathophysiology,2 with obvious consequences on treatment response. In the PARAGON-HF trial, for instance, a significant reduction in the occurrence of the primary outcome was observed in women but not in men, whereas no significant interaction between sacubitril/valsartan and sex in terms of outcomes emerged in the PARADIGM-HF trial.9, 10 Notably, the meta-analysis by Rossello et al. reassures on the lack of sex-related differences in the occurrence of spironolactone-associated side effects. Perceived/actual risk of hyperkalaemia is one of the major explanations for MRA underuse in clinical practice, especially in women who are more likely to suffer from chronic kidney disease. Therefore, these are important data that support an appropriate use of MRA in women as well as in men in daily clinical practice. Finally, the lack of sex-related differences in treatment response was also observed in important subgroups, such as older patients and those with impaired renal function, who are more likely to be under-treated in real-world care. The lack of patients with ejection fraction of 36–44%, the overall limited number of women vs. men, and of HFpEF vs. HFrEF patients, with HFpEF patients entirely enrolled in only one trial, are important limitations that need to be acknowledged for the correct interpretation of this meta-analysis. Use of HF treatments, in particular in the RALES trial, was different compared to contemporary care (i.e. high use of digitalis and diuretics, low use of beta-blockers, no use of HF devices), which is also an important point to be considered. In conclusion, the Authors approached one of the major issues in the contemporary management of HF by using an individual patient-level meta-analysis which is in the top of the “level of evidence” pyramid in evidence-based healthcare. This is an important study which, however, still suffers from the known limitations affecting RCTs, i.e. the large under-representation of women. MRA therapy represent an undisputed cornerstone for the treatment of HFrEF but its use in daily clinical practice is still limited.3 This study supports MRA use regardless of sex and reassures about the lack of sex-related differences in the occurrence of important side effects. By highlighting the importance of MRAs in both men and women, this analysis carries the potential of promoting the correct prescribing of this life-saving treatment for HFrEF. Ongoing RCTs on HFpEF such as the Spironolactone Initiation Registry Randomized Interventional Trial in Heart Failure with Preserved Ejection Fraction (SPIRRIT, NCT02901184) and the SPIRonolactone In the Treatment for Heart Failure (SPIRIT-HF, EudraCT 2017–000697-11) will provide more data on sex-related differences in treatment response in HFpEF. Conflict of interest: none declared.