In MITRA-FR, ‘Disproportionate’ MR Theory Doesn’t Pan Out

In the latest effort to find some common ground between the neutral MITRA-FR trial and the positive COAPT study, investigators for the former have reanalyzed patients according to baseline severity of mitral regurgitation (MR) and/or degree of left ventricular (LV) dysfunction. To no avail.

“In the MITRA-FR trial, we could not identify a subset of patients defined based on the degree of the regurgitation, LV remodeling, or on their combination, including those deemed as having disproportionate MR, that might have benefited from transcatheter correction using the MitraClip system,” write investigators led by David Messika-Zeitoun, MD, PhD (University of Ottawa Heart Institute, Canada).

“Our understanding of the two trials is that COAPT is showing that the clip works and MITRA-FR is telling us yes, but not in everybody,” Messika-Zeitoun elaborated to TCTMD. Given recent theories relating to the degree of MR and LV dysfunction that might best respond to transcatheter mitral valve replacement (TMVR) they decided to do this post hoc analysis in the MITRA-FR cohort. “We would have been happy to see a subset have benefit from the clip,” he said. “We didn't.”

As previously reported by TCTMD, the very different outcomes seen in the MITRA-FR and COAPT trials have prompted 2 years of speculation. A dominant theory to emerge has been that the very stringent COAPT enrollment criteria led to the inclusion of patients most likely to benefit—in particular, patients with symptomatic, severe MR and LV dysfunction but with left ventricles not so “blown” that improving MR would not be sufficient to show a benefit. In a post hoc analysis, COAPT patients with the lowest effective regurgitant orifice (≤ 30 mm2 ) and the largest LV size (LV end-diastolic volume index ≥ 96 mL/m2) appeared to get the most benefit from clip therapy, supporting Grayson and Packer’s proposed “proportionate and disproportionate” MR, in which MitraClip therapy is more likely to prove successful in patients whose MR severity is disproportionate to the degree of left ventricular enlargement and remodeling.

As the investigators note in this paper, published in the JACC: Cardiovascular Imaging, the popularity of Grayson and Packer’s theory has led to calls for the MITRA-FR group to do the current analysis.

Neutral Results

To this end, MITRA-FR investigators looked at the trial’s primary endpoint of all-cause death/unplanned HF hospitalizations at 24 months among patients with different degrees of MR severity, regurgitant volume and regurgitant fraction, LV remodeling (end-diastolic and end-systolic diameters and volumes), as well as these parameters in combination, according to treatment group.

They found that the impact was “neutral” in subsets with the most severe as well as the mildest MR, as well as for subsets with more or less LV remodeling as defined according to diastolic/systolic diameters or volumes. Even in patients with a ratio of MR severity/LV end-systolic volume consistent with the notion of disproportionate MR, no benefit to the MitraClip was seen.

“There were no significant interactions between trial group and any of the subsets defined based on the degree of the regurgitation, the degree of LV remodeling/dysfunction or the combination of these parameters with respect to the composite of all-cause death or unplanned hospitalization for heart failure at 24 months in both intention-to-treat and per-protocol analyses,” Messika-Zeitoun et al conclude.

The authors take pains to detail the strengths and limitations of their analysis, noting that measurements were done by a central core lab, using prospectively collected, adjudicated data from randomized patients. On the other hand, the analysis was not prespecified and the numbers of patients in the analyzed subgroups were small. “However,” they write, “our results were consistent in all subsets while in contrast, in COAPT, the positive effect of TMVR was consistent in most echocardiographic subsets. Nevertheless, we cannot exclude that the MITRA-FR population might have presented with uniformly advanced disease, precluding identification of a subset that might have benefited from TMVR.”

But at a minimum, they say, the MITRA-FR analysis calls into question the use of the proportionate/disproportionate “framework” for understanding who might benefit from the MitraClip.

“I think this tells us that it’s probably more complex than just this ratio and my understanding now is that patient selection is not only large MR and small LV, but maybe we are not taking into account the right ventricle and the [tricuspid regurgitation]: other parameters that are bad prognostic markers for the clip,” said Messika-Zeitoun. “I think it is more complex than just finding a ratio and sending the patient for a clip.”

COAPT co-PI Gregg Stone, MD (Icahn School of Medicine at Mount Sinai, New York, NY), commenting on the analysis for TCTMD, countered that the MITRA-FR trial “was not large enough to try to tease out the modest number of ‘COAPT-like’ patients that night have benefitted—the confidence intervals are just too wide in these small subgroups for meaningful conclusions. In addition, there are potential differences in the long-term use of medical therapies between the two arms in MITRA-FR, and differences in MitraClip technique, and long-term effectiveness compared with COAPT that might have influenced the discordant results.”

Addressing the medical-therapy point with TCTMD, Messika-Zeitoun acknowledged that a shortcoming of MITRA-FR is that investigators did not collect information on drug dose. “But if you look just at the class of medications, patients were very well-treated and to be honest, it's real life. . . . In real life you do what you think is the best and after that you do the intervention. So we are more in keeping with what is real practice than was COAPT,” he said.

We need to continue our work to identify the patient who would benefit most from the clip. David Messika-Zeitoun

In the paper, the authors also make the point that the thresholds used in their analysis do not necessarily represent disproportionate MR. “Disproportionate MR might be defined by higher degree of MR/LV size ratio than those observed in the MITRA-FR trial, and we could not test the 0.20 threshold due to limited sample size. However, to perform meaningful statistical analysis with reasonable sample size in all subsets we had to use the median values observed in our population.”

Stone, likewise, said via email: “Unfortunately, there is not yet an accepted method to identify disproportionate versus proportionate versus nonsevere MR in the individual patient, as opposed to a population. MITRA-FR uniformly enrolled a series of patients with less severe MR and more advanced LV dysfunction (greater LV dilatation) than COAPT.”

From Stone’s perspective, “the major lessons haven’t changed: MitraClip treatment of a COAPT-eligible population who remain symptomatic despite maximally-tolerated guideline directed medical therapy by operators achieving a high success rate, low complication rate, and durable efficacy will improve prognosis in a large proportion of patients with heart failure and severe MR. Conversely, no benefits should be expected in patients with lesser degrees of MR (< 3+ by the US ASE criteria) or with more severe LV dysfunction (eg LV ESD > 7 cm).”

Messika-Zeitoun, on the other hand, believes more still needs to be learned about patient selection. Investigators for both trials are hoping some answers may come from a meta-analysis combining their data, a collaboration under active discussion between the groups. A third trial, said Messika-Zeitoun, also would be helpful, although he agreed that MitraClip’s manufacturer, Abbott, is unlikely to fund such a study, having achieved such positive results with COAPT, its pivotal trial. MITRA-FR, by contrast, was financed by the French Ministry of Health and Research National Program, with Abbott Vascular providing devices as well as support for investigators’ meetings.

“I don't want to be too negative,” Messika-Zeitoun insisted. “We were not able to identify a subset that would benefit from the clip based on MITRA-FR, and for me the message is that we need to work harder to identify the patient who might benefit from the clip and not treat everybody. It's unfortunate that this analysis didn't [identify this group], but it's still an important result.”