Mavacamten in Symptomatic Nonobstructive Hypertrophic Cardiomyopathy (ODYSSEY-HCM)

Authors, Journal, Affiliations, Type, DOI

• Authors: Milind Y. Desai, Anjali T. Owens, Theodore Abraham, Iacopo Olivotto, Pablo Garcia-Pavia, Renato D. Lopes, Perry Elliott, et al., for the ODYSSEY-HCM Investigators
• Journal: New England Journal of Medicine
• Affiliations: Cleveland Clinic (lead); co-investigators from University of Pennsylvania, UCSF, Florence Italy, Madrid, Duke, University College London, Graz, Heidelberg, Japan, Portugal, Brazil; sponsored by Bristol Myers Squibb
• Type: Phase 3, international, double-blind, placebo-controlled randomized clinical trial
• DOI: 10.1056/NEJMoa2505927

Overview

ODYSSEY-HCM was a phase 3 RCT (N=580; 201 centers; 22 countries) evaluating mavacamten in adults with symptomatic nonobstructive HCM (LVOTO <30 mmHg rest, <50 mmHg provocation). Both co-primary endpoints were missed: peak VO2 improvement was 0.47 ml/kg/min greater with mavacamten (P=0.07) and KCCQ-CSS improvement was 2.7 points greater (P=0.06). NT-proBNP was substantially reduced by mavacamten (geometric mean ratio 0.41 vs placebo), demonstrating a dissociation between biomarker benefit and clinical efficacy. Safety was a concern: LVEF <50% occurred in 21.5% of mavacamten patients vs 1.7% placebo, and serious CHF events in 6.6% vs 1.7%. The trial establishes that mavacamten is ineffective as currently approved in nonobstructive HCM.

Keywords

Hypertrophic cardiomyopathy, nonobstructive HCM, mavacamten, cardiac myosin inhibitor, peak oxygen uptake, KCCQ, LVEF, NT-proBNP, exercise capacity, randomized controlled trial

Key Takeaways

Background and Rationale

• Mavacamten is approved for symptomatic obstructive HCM based on EXPLORER-HCM and VALOR-HCM trials showing LVOTO reduction, symptom improvement, and reduced eligibility for septal reduction therapy
• Nonobstructive HCM affects 30–50% of all HCM patients; it has no approved pharmacotherapy; beta-blockers, verapamil/diltiazem, and diuretics are used off-label with limited efficacy
• A prior phase 2 trial (MAVERICK-HCM) showed significant NT-proBNP and troponin I reductions with mavacamten in nonobstructive HCM; a post hoc analysis identified a subgroup with improved VO2/KCCQ — this provided the rationale for ODYSSEY-HCM

Trial Design

• Phase 3, double-blind, parallel-group, placebo-controlled, 48-week RCT at 201 HCM referral centers in 22 countries (December 2022 – March 2024)
• N=580: 289 mavacamten, 291 placebo (1:1 randomization)
• Key eligibility: NYHA II–III, LVOTO <30 mmHg rest and <50 mmHg on provocation, LVEF ≥60%, KCCQ-CSS ≤85, NT-proBNP above upper limit of normal, respiratory exchange ratio ≥1.0 on CPET
• Mavacamten started at 5 mg/day; dose adjusted (1, 2.5, 5, 10, or 15 mg) based on LVEF by core laboratory echocardiography
• Regimen interruption required if LVEF <50%; permanent discontinuation if LVEF ≤30%
• Stratified by NYHA class, exercise test modality, and beta-blocker use

Patient Characteristics

• Mean age 56±15 years; 46% women; BMI 28.2±5.0
• Mean HCM duration 10.3 years; 43.3% family history of HCM; 5.9% prior septal reduction therapy
• 78.1% on beta-blockers; mean KCCQ-CSS 56.9 (moderately symptomatic)
• Mean LVEF 65.8%; mean LV mass index 122.3 kg/m²; mean E/e' ratio 13.3; mean LAVi 43.5 ml/m² (advanced diastolic dysfunction)
• Median NT-proBNP 917.5 ng/L; median hs-troponin T 18.4 ng/L (elevated biomarkers confirming disease severity)
• Mean peak VO2 18.2 ml/kg/min (68% predicted); 30.2% NYHA III
• Final dose distribution at week 48 (mavacamten): 1 mg 3.1%, 2.5 mg 6.2%, 5 mg 30.8%, 10 mg 27.7%, 15 mg 21.5% — 49.2% on higher doses (10–15 mg)

Primary Efficacy Results — Both Endpoints Missed

• Peak VO2: +0.52 ml/kg/min with mavacamten vs +0.05 ml/kg/min with placebo; between-group difference +0.47 ml/kg/min (95% CI −0.03 to 0.98; P=0.07; threshold for significance P<0.01)
• KCCQ-CSS: +13.1 points with mavacamten vs +10.4 with placebo; between-group difference +2.7 points (95% CI −0.1 to 5.6; P=0.06; threshold for significance P<0.04)
• Neither primary endpoint met its pre-specified significance threshold; graphical testing plan was stopped and no secondary endpoint P-values are reported

Secondary Endpoints (No P-values; Descriptive Only)

• NT-proBNP: Geometric mean ratio mavacamten/placebo = 0.41 (95% CI 0.36–0.47) — 59% reduction; large and clinically meaningful biomarker benefit
• HCMSQ shortness-of-breath score: Between-group difference −0.7 (95% CI −1.2 to −0.2) — modest symptom benefit favoring mavacamten
• VE/VCO₂ slope: Between-group difference −0.81 (95% CI −2.21 to 0.60) — non-significant trend favoring mavacamten
• NYHA class improvement ≥1: 60.1% mavacamten vs 57.8% placebo (small risk difference)

Safety — Notable Concerns

• LVEF <50%: 21.5% mavacamten vs 1.7% placebo — more than 10-fold higher rate; LVEF returned to ≥50% in all but 3 patients after regimen interruption
• LVEF ≤30%: 2.4% mavacamten vs 0% placebo
• Regimen interruption: 25.7% mavacamten vs 7.6% placebo; permanent discontinuation 10.7% vs 5.8%
• Serious CHF events: 6.6% mavacamten vs 1.7% placebo — 4× higher CHF hospitalisation rate
• Atrial tachyarrhythmias (serious events): 4.2% mavacamten vs 3.4% placebo (similar)
• 5 of 19 (26%) patients who developed CHF had concurrent LVEF <50%
• Higher dose use in this trial (49% on 10–15 mg) compared with obstructive HCM trials may have contributed to increased LVEF reduction

Discussion and Mechanistic Interpretation

• The different effect of mavacamten in obstructive vs nonobstructive HCM suggests that LVOTO relief is the primary mechanism of clinical benefit in obstructive HCM; without obstruction, this mechanism is absent
• In nonobstructive HCM, dominant symptom drivers are: diastolic dysfunction, LV noncompliance, subendocardial ischemia, microvascular disease, chronotropic incompetence — none of which are corrected by cardiac myosin inhibition
• Inhibiting an overactive sarcomere without the "compensatory" benefit of LVOTO relief may worsen net cardiac function — explaining both the CHF signal and the absence of clinical benefit
• The large NT-proBNP reduction (consistent with reduced wall stress) did not translate to clinical benefit, confirming that biomarker improvement does not reliably predict clinical efficacy in nonobstructive HCM
• Both groups had marked placebo effect (KCCQ-CSS +10.4 points), potentially reflecting intensified care and arrhythmia management during trial enrollment
• Nonobstructive HCM patients in this trial had more advanced disease than obstructive HCM trials (longer disease duration, more AF, worse KCCQ)

Limitations of the Document

• Predominantly White patient population — results may not generalize to other racial/ethnic groups
• Some patients with HFpEF may have been misclassified as nonobstructive HCM despite stringent criteria
• 8.3% permanently discontinued assigned regimen — may have attenuated or inflated differences
• 48-week duration precludes long-term efficacy/safety assessment
• Trial used treatment-policy estimand (all available data regardless of discontinuation), which may underestimate protocol-adherent effect
• Large placebo response observed in KCCQ-CSS (10.4 points) — may be due to intensified care and monitoring during the trial

Key Concepts Mentioned

• concepts/LVOTO — primary mechanism of benefit in obstructive HCM; absent in nonobstructive HCM
• concepts/Septal-Reduction-Therapy — context: mavacamten avoids SRT in obstructive HCM; no SRT role in nonobstructive HCM
• concepts/Sarcomere-Biology — mavacamten mechanism as cardiac myosin inhibitor

Key Entities Mentioned

• entities/Mavacamten — drug under study; established in obstructive HCM; now shown ineffective in nonobstructive HCM
• entities/HCM — nonobstructive subtype, 30–50% of all HCM patients

Wiki Pages Updated

• wiki/sources/mavacamten-odysseyhcm-nejm-2025.md — created
• wiki/entities/Mavacamten.md — ODYSSEY-HCM results added; new contradiction added
• wiki/entities/HCM.md — Nonobstructive HCM treatment section updated
• wiki/sourceindex.md — entry added
• wiki/wikiindex.md — Mavacamten entity description updated