Lessons From MAVERICK-HCM

Hypertrophic cardiomyopathy (HCM) is a primary myocardial disease associated with heart failure, atrial fibrillation, activity intolerance, and fatigue (1). MYH7 and MYBPC3 encode proteins that enable cardiac contractility, and mutations in these genes are found in approximately one-half of HCM patients (2). HCM is often considered synonymous with obstructive hypertrophic cardiomyopathy (HOCM), in which left ventricular outflow tract (LVOT) obstruction is a primary disease driver. Defined as resting or provoked LVOT gradient of at least 30 mm Hg, it is estimated that at least one-half of HCM is HOCM. For those with HOCM, septal reduction through surgical myectomy, often accompanied by mitral valve repair, can alleviate symptoms of activity intolerance and improve quality of life. Percutaneous alcohol ablation is also a viable strategy to relieve LVOT gradient. However, not all HCM is obstructive in nature. Broader clinical and genetic screening identified many HCM patients without LVOT obstruction. Clinically, nonobstructive HCM is seen as concentric hypertrophy with impaired filling in the absence of a prominent gradient at rest or with exertion. There are few pharmacological treatments for HOCM and even fewer for nonobstructive HCM, and mainstay drug therapy aims to reduce heart rate, often with beta-adrenergic and/or nondihydropyridine calcium-channel blockade. Often, the only option for intractable heart failure is transplant. In this issue of the Journal, a new report by Ho et al. (3) described the results of a dose-finding study with mavacamten, a small molecule allosteric inhibitor of myosin. MAVERICK-HCM (Mavacamten in Adults With Symptomatic Non-Obstructive Hypertrophic Cardiomyopathy) identified a reduction in 2 key biomarkers, N-terminal pro–B-type natriuretic peptide (NT-proBNP) and cardiac troponin I, over the 16 weeks of dosing.

Both HOCM and nonobstructive HCM are characterized as having a hypercontractile left ventricular myocardium. The molecular underpinnings of hypercontractility relate to an energetically inefficient myosin. Active myosin is a composite of 2 myosin heads with intertwined tails. During the myosin force production cycle, there is an auto-inhibited state, sometimes referred to as a super-relaxed state. With certain myosin mutations, the HCM sarcomere spends less time in this inhibited state, leading to hyperactivation and excess ATP utilization (4,5). MYH7 mutations linked to HCM are enriched in the converter domain of b-myosin heavy chain, where they directly impair the super-relaxed state (6,7). Specific MYBPC3 mutations similarly reduce the time myosin is inactivated, promoting hyperactivity (8). The small molecule mavacamten stabilizes this inhibited state, effectively lengthening the time that myosin is inactive (9). Mavacamten shifts the kinetics of the actin-activated phosphate release step of myosin activation, thus decreasing ATPase activity and essentially slowing myosin’s interaction with actin. Although there are multiple myosins, mavacamten is specific for β-myosin heavy chain, which is expressed in cardiac and slow skeletal muscle. Physiologically, cardiac function is reduced with mavacamten.

Mavacamten was previously evaluated in HOCM in the open-label PIONEER-HCM (Pilot Study Evaluating MYK-461 in Subjects With Symptomatic Hypertrophic Cardiomyopathy and Left Ventricular Outflow Tract Obstruction; NCT02842242) (10). PIONEER-HCM was a prospective, phase 2 study to evaluate pharmacokinetics and pharmacodynamics. Participants (n = 21) had resting LVOT gradients of ≥30 or ≥50 mm Hg of provoked gradient. A reduction in gradient was observed, and 8 participants had a reduction of LVOT to <30 mm Hg. A lowering of serum NT-proBNP was seen. Notably, LVEF declined −6% to −15%, and this is consistent with the expected mode of a myosin inactivator. These findings were sufficient to prompt a larger phase 3 trial, which is underway in the EXPLORER-HCM (Clinical Study to Evaluate Mavacamten [MYK-461] in Adults With Symptomatic Obstructive Hypertrophic Cardiomyopathy; NCT03470545) study.

The new study, MAVERICK-HCM, was designed to evaluate dosing of mavacamten in HCM without obstructive physiology, since the drug has unusual extended pharmacokinetic properties. A total of 59 patients were enrolled and received mavacamten over 16 weeks (n = 19 allocated to serum drug concentration of ~200 ng/ml, n = 21 allocated to ~500 ng/ml, and n = 19 to placebo), followed by an 8-week washout period. Importantly, 58% of patients enrolled were women. The MAVERICK-HCM study met its primary objective to assess the safety and tolerability of mavacamten in symptomatic non-obstructive HCM patients. Indeed, the drug was well-tolerated, with no differences in reported serious adverse events (SAE) between treatment groups (10% on mavacamten and 21% on placebo). A common SAE was atrial fibrillation (5% in both groups), and most other side effects (>70%) were mild in nature.

A key secondary outcome was reduction in bio-markers associated with increased wall stress and myocardial injury. Typical for a phase 2 study testing pharmacokinetic properties, the concentration data distribution is plotted, and the magnitude of SD can often exceed the size of the mean. Therefore, the most appropriate summary descriptor is the geometric mean as opposed to the arithmetic mean, which was used in this clinical trial. In the pooled-mavacamten group, the NT-proBNP geometric mean decreased by 53% versus 1% in the placebo group (−435 pg/ml vs. −6 pg/ml; p = 0.0005). Similarly, the geometric mean for cardiac troponin I also decreased by 34% in the pooled-mavacamten group compared to a 4% increase in the placebo group, which was statistically significant as well (p = 0.009).

Although the study was not designed or powered to evaluate clinical benefit, the investigators used this opportunity to include exploratory functional endpoints, including peak VO₂ or New York Heart Association (NYHA) functional class. Among a high-risk subgroup with elevated myocardial injury bio-markers (cardiac troponin I >99th percentile) or elevated diastolic filling pressures (average E/e′ >14 on echocardiogram), one-third of mavacamtentreated patients met the composite functional endpoint—defined as achieving: 1) an improvement of at least 1.5 ml/kg/min in peak VO₂ with a reduction of ≥NYHA functional class; or 2) an improvement of 3.0 ml/kg/min or more in peak VO₂ with no worsening in NYHA functional class—compared with none in the placebo group (p = 0.03).

As an exploratory study, the authors were cautious not to overinterpret the data. Not unlike Russian nesting dolls, where a series of smaller and smaller dolls are hidden inside a single doll, what may seem quite simple is actually very complex. A 1-sized approach does not fit all, and high-risk subgroups of an already highly selected population (59 nonobstructive-HCM patients enrolled after screening over 150 patients) may not be reflective of physiology across the entire spectrum of nonobstructive disease. However, the biomarker data are striking and suggest a physiological benefit. We anxiously await the results from the phase 3 follow-on study of mavacamten in nonobstructive HCM.

This study promotes the idea of precision medicine or titration of drug to effect, and this may represent the forefront of care for the patient with HCM. In this study, 12.5% of patients (n = 5) had a drop in their left ventricular ejection fraction (LVEF) to ≤45%, which led to stopping the drug per pre-specified protocol. All 5 patients had recovery of LVEF and experienced no long-term negative effects. Going forward, investigators and clinicians may look to clinical markers such as LVEF or more subtle markers of left ventricular systolic dysfunction such as global longitudinal strain, when determining dosing strategies, similar to cardio-oncology protocols.

Of the 21 patients in PIONEER-HCM, 10 carried genetic variants with 5 known pathogenic variants in the genes most commonly implicated in HCM. In MAVERICK-HCM, genetic results were similar, with 55% (n = 22 of 40) of patients having a pathogenic variant in common HCM genes and 18% (n = 7 of 40) having a variant of uncertain significance. In the future, we may see targeted pharmacological approaches in genetically at-risk family members of all forms of HCM patients in addition to gene-therapy treatments. These findings represent a rigorous, well-designed phase 2 study that has offered a glimmer of hope in a high-risk nonobstructive HCM population with limited available therapies.