Two-Year Outcomes of the MITRA-FR Trial: Towards an Integrated Approach in the Evaluation of Patients with Secondary Mitral Regurgitation

This article refers to ‘Percutaneous repair or medical treatment for secondary mitral regurgitation: outcomes at 2 years’ by B. Iung et al., published in this issue on pages xxx. Transcatheter mitral valve repair (TMVR) with the MitraClip (MC) system (Abbott Vascular, Santa Clara, CA, USA) was introduced in 2003 as an alternative to open-heart surgery for patients with severe symptomatic mitral regurgitation (MR).1, 2 Based on observational studies and the randomized EVEREST II trial comparing TMVR with MC to surgery,3 it received CE mark in 2008 followed by Food and Drug Administration approval for the treatment of primary MR in 2013. EVEREST II enrolled predominantly patients with primary MR, and long-term follow-up data throughout 5 years confirmed similar survival in both groups. However, the composite endpoint of MR > 2+ and need for reintervention (mitral valve surgery or reoperation) occurred more frequently among patients who underwent TMVR. Notwithstanding, when the same endpoint was analysed according to the underlying MR aetiology, patients with secondary MR had comparable outcomes, irrespective of their treatment group. Secondary MR is a complex disease entity and the result of several intertwined pathophysiological processes including left ventricular, atrial and annular dilatation, dyssynchronous contraction as well as papillary muscle dysfunction. In addition, the severity of secondary MR is variable and highly dependent on pre- and afterload conditions, which in turn affects reproducible echocardiographic grading. Several observational studies and one randomized trial failed to demonstrate a benefit of treating secondary MR at the time of surgical revascularization in patients with ischaemic cardiomyopathy.4-6 Therefore, secondary MR has been regarded for a long time as a marker of impaired prognosis7 rather than a specific target for interventions. Against this background, the therapeutic potential of TMVR in the specific setting of secondary MR required investigation in dedicated studies. Recently, a large-scale registry including 616 patients with secondary MR showed sustainable MR and symptom improvement, as well as a significant decrease of left ventricular volumes throughout 1 year.8 Acute changes of the mitral annulus geometry after clipping, in particular a decrease of the antero-posterior dimension, have been described as a potential trigger of positive left ventricular remodelling.9, 10 Several randomized clinical trials have been designed with this objective in mind, and MITRA-FR and COAPT are the first to address this gap in evidence. The at first glance diverging primary endpoint results of MITRA-FR11 and COAPT12 reported in 2018 were perhaps the most striking expression of the complexity of this disease so far. While both studies compared guideline-directed medical therapy (GDMT) alone to TMVR with MC combined with GDMT, the rate of death or unplanned hospitalization did not differ in MITRA-FR at 1 year, whereas COAPT reported a relative reduction of 47% for heart failure (HF) hospitalizations and 38% for mortality in favour of TMVR with MC during a 2-year follow-up period. Among differentiating features put forward to rationalize the dichotomy in outcomes, the two-fold longer follow-up period, as well as the larger sample size of the COAPT trial have been repeatedly voiced in numerous editorials and viewpoints.13-17 In this context, the 2-year results of the MITRA-FR trial published in this issue of the Journal were highly anticipated, and the authors ought to be applauded for the effort to extend data collection in the framework of an investigator-initiated trial with limited resources.18 At 2 years, follow-up data regarding the primary endpoint (all-cause mortality and unplanned rehospitalization for HF) were available in 95% of the 307 randomized patients. The primary endpoint occurred in 63.8% of patients who underwent TMVR with MC combined with GDMT and 67.1% in the GDMT alone group [hazard ratio (HR) 1.01; 95% confidence interval (CI) 0.77–1.34]. There was no significant difference when both components of the composite endpoint were analysed separately; the rate of mortality was 34.9% vs. 34.2% (HR 1.02; 95% CI 0.70–1.50), while the rate of unplanned hospitalization for HF was 55.9% vs. 61.8% (HR 0.97; 95% CI 0.72–1.30) in patients who underwent TMVR combined with GDMT compared to the GDMT alone group. Results were consistent for all pre-specified subgroups except for a larger effect of TMVR in patients with a baseline creatinine <1.5 mg/dL. Interestingly, a post-hoc analysis of the cumulative incidence of unplanned hospitalizations for HF – the primary endpoint of the COAPT trial – found a non-significant trend in favour of the TMVR group (HR 0.87; 95% CI 0.56–1.35) that was mainly driven by a reduced incidence of unplanned hospitalizations during the second year after treatment. Notwithstanding, MITRA-FR failed to show benefit of TMVR with MC compared to GDMT alone throughout 2 years, maintaining its original conclusion that TMVR does not unfold a therapeutic benefit in all secondary MR patients. In others words: patient selection must aim to identify individuals who benefit most from TMVR and define favourable clinical scenarios. Several similarities between studies are noteworthy including patient age (70 ± 10 years in MITRA-FR vs. 72 ± 12 years in COAPT), left ventricular ejection fraction (33 ± 7% vs. 31 ± 10%) and proportion of patients with ischaemic cardiomyopathy (59% vs. 61%). Moreover, only small differences were reported concerning the type and composition of GDMT at the time of inclusion, while the use of cardiac resynchronization therapy was slightly higher in COAPT (36% vs. 27% in MITRA-FR). Other variables including peri-procedural adverse events or procedural success may not be reliably compared owing to discrepant definitions and the use of diverging MR severity grading schemes in the two trials. As displayed in Figure 1, patients assigned to GDMT had very similar prognosis irrespective of the trial (in both cases the composite of death or unplanned hospitalization occurred in about 67%). Thus, the diverging study outcomes may be exclusively explained by a fundamentally different response to TMVR. In this regard, MITRA-FR certainly provides useful guidance to refine future selection criteria for TMVR. Patients with lower MR grade and severe left ventricular dilatation may not benefit from the intervention, as well as those in whom failure of maximally tolerated GDMT (± cardiac resynchronization therapy) to reduce symptoms or secondary MR severity has not been clearly demonstrated. Thus, early evaluation of patients with secondary MR within the Heart Team in close collaboration with imaging specialists to ensure proper longitudinal assessment as well as with HF specialists to establish and adjust GDMT is of paramount importance prior to considerations for TMVR. Of note, beside prognostic outcomes, symptomatic improvement certainly constitutes the most important goal for patients evaluated for TMVR, particularly when considering the dismal prognosis of HF associated with secondary MR.7 Although data from the MITRA-FR trial do not allow definitive conclusions in this regard, TMVR may still be offered as symptomatic therapy in selected patients who do not fulfil the COAPT inclusion criteria and are not candidates for circulatory support or heart transplantation. Of note, in COAPT, improvement in quality of life occurred uniformly irrespective of MR severity according to effective regurgitant orifice area (cut-off for stratification: 40 mm2) or left ventricular end-diastolic volume index (cut-off for stratification: 94 mL/m2).21 Although the planned 5-year follow-up of MITRA-FR as well as a meta-analysis of individual patient data of both trials may provide valuable insights, future research will also need to focus on the identification of the appropriate timing for intervention. Promising parameters include better characterization of the relationship between secondary MR severity and ventricular dimensions, the detection of myocardial fibrosis potentially precluding reverse remodelling after treatment, as well as the use of strain echocardiography as a better indicator of left ventricular function in patients with MR. Conflict of interest: S.W. reports research and educational grants to the institution from Amgen, Abbott, Bayer, BMS, Boston Scientific, Biotronik, CSL Behring, Edwards Lifesciences, Medtronic, Polares, and Sinomed. F.P. has no conflict of interest.