Efficacy of Mavacamten on functional capacity in hypertrophic cardiomyopathy: a meta-analysis, meta-regression, and trial sequential analysis of randomized controlled trials Authors: Ibrahim Khalil, Anannya Sheikh, Nabila Nur, Imran Hossain, Mahmuda Akter, Sajjad Ghanim Al-Badri Published: 2026 ABSTRACT Background:Hypertrophic cardiomyopathy (HCM) is characterized by left ventricular hypertrophy, often leading to left ventricular outflow tract (LVOT) obstruction, heart failure, and sudden cardiac death. Mavacamten, a novel cardiac myosin inhibitor, targets the underlying mechanisms of HCM.Methods:We searched PubMed, Cochrane Library, and ClinicalTrials.gov up to 1 August 2025 and included six randomized controlled trials comparing Mavacamten to placebo in adults with HCM. The primary outcome was the change in LVOT gradients (post-exercise, post-Valsalva, and resting). Secondary outcomes included N-terminal pro-brain natriuretic peptide (NT-proBNP) levels, left ventricular ejection fraction, peak oxygen consumption (pVO2), and serious adverse events (SAEs). Data were analyzed using a random-effects model, with results reported as mean differences (MD) or risk ratios (RR) with 95% confidence intervals (CI). Meta-regression explored study-level moderators, and retrospective trial sequential analysis assessed the robustness of the findings.Results:Among 632 patients, Mavacamten significantly reduced LVOT gradients: post-exercise (MD: −38.28 mmHg, 95% CI: −45.92 to −30.63, P < 0.00001), post-Valsalva (MD: −59.02 mmHg, 95% CI: −82.75 to −35.29, P < 0.00001), and resting (MD: −44.50 mmHg, 95% CI: −66.49 to −22.51, P < 0.0001). NT-proBNP levels decreased significantly (MD: −617.26 ng/L, 95% CI: −772.43 to −462.09, P < 0.00001), but pVO2 showed no significant change (MD: 0.69 mL/kg/min, 95% CI: −1.12 to 2.50, P = 0.45). SAEs were comparable (RR: 1.17, 95% CI: 0.58–2.36, P = 0.66). Across primary outcomes, retrospective trial sequential analysis confirms that Mavacamten’s reductions in LVOT gradients are conclusive, indicating robust evidence.Conclusion:Mavacamten effectively reduces LVOT gradients and NT-proBNP levels in HCM, but its impact on exercise capacity remains uncertain, with no increased risk of SAEs. Introduction Hypertrophic cardiomyopathy (HCM), the most common inherited cardiac disorder (prevalence ~1 in 500), is characterized by unexplained left ventricular hypertrophy that frequently causes left ventricular outflow tract (LVOT) obstruction in its obstructive form (oHCM)[1,2]. This obstruction leads to exertional dyspnea, chest pain, syncope, and an increased risk of sudden cardiac death, especially in younger patients[1,2]. The underlying pathophysiology involves hypercontractility, impaired relaxation, and excessive myocardial energy demand due to abnormal myosin–actin cross-bridge formation, which worsens LVOT obstruction, cardiac stress, and functional capacity while promoting progression to heart failure in many untreated cases[2,3].HIGHLIGHTSMavacamten significantly reduced left ventricular outflow tract gradients: post-exercise [mean differences (MD): −38.28 mmHg, 95% confidence interval (CI): −45.92 to −30.63, P < 0.00001].Mavacamten also significantly reduces post-Valsalva (MD: −59.02 mmHg, 95% CI: −82.75 to −35.29, P < 0.00001) and resting (MD: −44.50 mmHg, 95% CI: −66.49 to −22.51, P <0.0001).NT-proBNP levels decreased significantly (MD: −617.26 ng/L, 95% CI: −772.43 to −462.09, P < 0.00001).However, peak oxygen consumption showed no significant change (MD: 0.69 mL/kg/min, 95% CI: −1.12 to 2.50, P = 0.45).Serious adverse events were comparable (risk ratios: 1.17, 95% CI: 0.58–2.36, P = 0.66). HIGHLIGHTS Mavacamten significantly reduced left ventricular outflow tract gradients: post-exercise [mean differences (MD): −38.28 mmHg, 95% confidence interval (CI): −45.92 to −30.63, P < 0.00001]. Mavacamten significantly reduced left ventricular outflow tract gradients: post-exercise [mean differences (MD): −38.28 mmHg, 95% confidence interval (CI): −45.92 to −30.63, P < 0.00001]. Mavacamten also significantly reduces post-Valsalva (MD: −59.02 mmHg, 95% CI: −82.75 to −35.29, P < 0.00001) and resting (MD: −44.50 mmHg, 95% CI: −66.49 to −22.51, P <0.0001). Mavacamten also significantly reduces post-Valsalva (MD: −59.02 mmHg, 95% CI: −82.75 to −35.29, P < 0.00001) and resting (MD: −44.50 mmHg, 95% CI: −66.49 to −22.51, P <0.0001). NT-proBNP levels decreased significantly (MD: −617.26 ng/L, 95% CI: −772.43 to −462.09, P < 0.00001). NT-proBNP levels decreased significantly (MD: −617.26 ng/L, 95% CI: −772.43 to −462.09, P < 0.00001). However, peak oxygen consumption showed no significant change (MD: 0.69 mL/kg/min, 95% CI: −1.12 to 2.50, P = 0.45). However, peak oxygen consumption showed no significant change (MD: 0.69 mL/kg/min, 95% CI: −1.12 to 2.50, P = 0.45). Serious adverse events were comparable (risk ratios: 1.17, 95% CI: 0.58–2.36, P = 0.66). Serious adverse events were comparable (risk ratios: 1.17, 95% CI: 0.58–2.36, P = 0.66). Current first-line therapies, beta-blockers, and non-dihydropyridine calcium channel blockers primarily provide symptom palliation by reducing heart rate and improving diastolic filling but often fail to address the molecular drivers of HCM, leaving many patients with refractory symptoms[4,5]. Invasive septal reduction therapies (myectomy or alcohol septal ablation) are effective for persistent obstruction but carry risks (e.g., conduction defects and incomplete relief) and limited accessibility[4–6]. These limitations underscore the urgent need for targeted, disease-modifying therapies that improve pathophysiology, functional outcomes, and safety. Mavacamten, a first-in-class selective allosteric inhibitor of cardiac myosin, represents a groundbreaking advancement in HCM management by directly targeting the excessive myosin–actin interactions that drive hypercontractility and LVOT obstruction[1,3,4]. By reducing contractility and enhancing myocardial relaxation, Mavacamten addresses the core mechanisms of oHCM, offering potential for both symptomatic relief and disease modification[3,7,8]. Phase 3 randomized controlled trials (RCTs), such as EXPLORER-HCM and VALOR-HCM, have demonstrated Mavacamten’s ability to significantly reduce LVOT gradients and improve biomarkers of cardiac stress, such as N-terminal pro-brain natriuretic peptide (NT-proBNP), while also improving New York Heart Association (NYHA) functional class in symptomatic patients[1,5,6,9]. Long-term extension studies further suggest sustained benefits, with reductions in LVOT gradients and stable safety profiles over extended periods[9,10]. However, uncertainties persist regarding Mavacamten’s impact on functional capacity, such as peak oxygen consumption (pVO2), and its long-term safety, particularly in diverse patient populations with varying degrees of obstruction and comorbidities[11–13]. Despite these advances, a comprehensive synthesis of Mavacamten’s efficacy and safety across all available RCTs is lacking, particularly in addressing its effects on functional capacity, disease burden, and adverse events in both obstructive and non-obstructive HCM. Recent studies have highlighted the need for such analyses to clarify Mavacamten’s role in clinical practice, especially in diverse populations such as Chinese patients with oHCM[14] and those with concurrent hypertension[12]. Moreover, the interaction of Mavacamten with background therapies like beta-blockers remains an area of interest for optimizing treatment strategies[15]. This systematic review, meta-analysis, meta-regression, and retrospective trial sequential analysis (TSA) aim to fill this gap by consolidating evidence from RCTs to evaluate Mavacamten’s therapeutic potential in HCM. By examining its effects on LVOT gradients, NT-proBNP levels, pVO2, and serious adverse events (SAEs), this study seeks to provide a robust evidence base to guide clinical decision-making and inform future research directions for optimizing HCM management in the era of precision cardiology. Methods Study design and registration This systematic review and meta-analysis was conducted to evaluate the efficacy and safety of Mavacamten in patients with HCM, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines[16]. The study protocol was registered with the International Prospective Register of Systematic Reviews prior to the initiation of the review process to ensure transparency and reproducibility[17]. Eligibility criteria The review included RCTs that compared Mavacamten to placebo in adults aged 18 years or older diagnosed with HCM, encompassing both obstructive and non-obstructive forms based on accepted clinical and imaging criteria. Studies were included if they met all prespecified criteria using the population, intervention, comparison, and outcome framework. Eligible populations comprised adults (aged ≥18 years) with confirmed HCM, including obstructive and non-obstructive phenotypes, diagnosed per contemporary guidelines (e.g., 2020 AHA/ACC or 2024 ESC) via clinical, echocardiographic, and/or cardiac magnetic resonance criteria. The intervention was Mavacamten (any oral dose and duration) as the primary agent, compared against an identical-appearing placebo. Studies had to report at least one primary outcome – change from baseline in LVOT gradient (post-exercise, post-Valsalva, or resting) by Doppler echocardiography – or secondary outcomes: change in NT-proBNP, left ventricular ejection fraction (LVEF), pVO₂ via cardiopulmonary exercise testing, or SAEs (e.g., death, hospitalization, disability). Only RCTs (parallel-group, cross-over, or phase 2/3 placebo-controlled designs) published as full-text English-language peer-reviewed articles up to 1 August 2025 were included. Exclusions encompassed non-randomized designs, studies without a direct Mavacamten–placebo comparison, pediatric or non-human studies, those lacking extractable outcome data, conference abstracts, protocols, gray literature, duplicates (only the most complete report retained), and non-English full texts despite reasonable translation efforts. Two independent reviewers screened titles/abstracts and full texts, resolving disagreements by consensus or third-reviewer arbitration. Search strategy A comprehensive systematic search was performed across three electronic databases – PubMed, Cochrane Library, and ClinicalTrials.gov – covering all available literature up to 1 August 2025. The search strategy incorporated MeSH terms and keywords such as “Mavacamten,” “hypertrophic cardiomyopathy,” “HCM,” “cardiac myosin inhibitor,” “placebo,” and “randomized controlled trials,” combined using Boolean operators (“AND,” “OR”) to refine the search. Additionally, the reference lists of eligible studies and relevant reviews were manually screened to identify any further articles that met the inclusion criteria. Sample search term: (“Mavacamten”[Title/Abstract] OR “MYK-461”[Title/Abstract]) AND (“Hypertrophic Cardiomyopathy”[MeSH Terms] OR “cardiomyopathy, hypertrophic”[Title/Abstract] OR “HCM”[Title/Abstract]) AND (“placebo”[Title/Abstract] OR “placebo-controlled”[Title/Abstract]) AND (“Randomized Controlled Trial”[Publication Type] OR “randomized”[Title/Abstract] OR “randomised” [Title/Abstract] OR “RCT”[Title/Abstract]). Study selection All retrieved records were imported into Rayyan, a systematic review management tool, for efficient handling. Two independent reviewers screened the titles and abstracts of all identified articles to determine eligibility. Studies that met the initial criteria underwent full-text review to confirm their suitability for inclusion. Any disagreements between the reviewers were resolved through consensus or, when necessary, by consulting a third reviewer to ensure consistency in the selection process. Data extraction Data were extracted from the eligible studies using a pre-designed data extraction form. The extracted information included study characteristics such as author names, year of publication, journal, study design, and duration of follow-up; patient demographics, including sample size, mean age, sex distribution, baseline characteristics, and inclusion/exclusion criteria; intervention details, such as dosage and duration of Mavacamten therapy, comparator details, and any concurrent treatments; and outcome data, encompassing primary outcomes like changes in LVOT gradients (post-exercise, post-Valsalva, and resting) and secondary outcomes such as changes in NT-proBNP levels, LVEF, pVO2, and the incidence of adverse events. Risk of bias assessment The risk of bias in the included studies was assessed independently by two reviewers using the Cochrane Risk of Bias 2.0 tool[18]. The assessment evaluated domains such as the randomization process, allocation concealment, blinding of participants, personnel, and outcome assessors, completeness of outcome data, and selective reporting. Each domain was rated as “low risk,” “some concerns,” or “high risk,” with discrepancies between reviewers resolved through consensus to ensure a consistent evaluation. Statistical analysis Statistical analyses were conducted using Python and R Studio[19]. A random-effects model was employed to account for potential between-study heterogeneity. For continuous outcomes, such as LVOT gradients, NT-proBNP levels, LVEF, and pVO2, pooled mean differences (MDs) with 95% confidence intervals (CI) were calculated. Heterogeneity was assessed using the chi-squared test and the I2 statistic, with thresholds of 25%, 50%, and 75% indicating low, moderate, and high heterogeneity, respectively. A P-value less than 0.05 in the chi-squared test was considered indicative of significant heterogeneity. The work has been reported in line with Assessing the Methodological Quality of Systematic Reviews Guidelines[20]. Sensitivity analysis To evaluate the robustness of the meta-analysis results, sensitivity analyses were performed by excluding studies with a high risk of bias and by reanalyzing the data after removing individual studies to assess their influence on the pooled estimates. These analyses included leave-one-out sensitivity analysis and influence diagnostics to ensure the stability of the findings across various scenarios. Meta-regression analysis Meta-regression was conducted to explore the influence of study-level characteristics. Moderators, such as patient demographics (e.g., average age, sex distribution), and other relevant parameters, were examined. This regression analysis aimed to identify factors contributing to variability in treatment effects across studies, providing insights into patient subgroups that may derive greater benefit from Mavacamten therapy. Trial sequential analysis To assess the reliability of the meta-analysis findings and to control for type I and type II errors arising from sparse data, repeated significance testing, and potential heterogeneity, TSA was performed retrospectively using the RTSA package in R. TSA adjusts for cumulative random errors by constructing monitoring boundaries based on group sequential methods adapted to meta-analysis. This approach provides TSA-adjusted CIs, calculates the required information size (RIS), and determines whether the accumulated evidence is conclusive or if further trials are warranted. TSA was applied to the primary outcomes of change in LVOT gradients (post-exercise, post-Valsalva, and resting). For each outcome, analyses were conducted under two configurations: a two-sided design with type I error (alpha) of 0.05 and type II error (beta) of 0.10 (90% power), and a second configuration with beta of 0.15 (85% power). Both configurations employed the O’Brien-Fleming alpha-spending function (esOF) without futility boundaries. The analyses assumed a minimal clinically important difference (MCID) and corresponding standard deviation (SD) as prespecified for each outcome. The RIS was adjusted for heterogeneity using the estimated tau2 from the random-effects meta-analysis (fixed = FALSE), with incorporation of one anticipated additional trial and sequential design inflation where applicable. Cumulative Z-scores were evaluated against sequential monitoring boundaries at interim timings corresponding to the cumulative sample sizes of the included studies, with sma_timing representing the proportion of the sequential RIS achieved. The analysis was retrospective (design = NULL), and the study order was based on the sequence in the dataset. Warnings were noted regarding the potential impact on boundary precision due to the retrospective nature of the analysis and the absence of a predefined study ordering; future prospective TSA should specify study order to enhance validity. Overall, TSA was used to evaluate whether the cumulative evidence for Mavacamten’s effect on LVOT gradient reduction crossed predefined benefit boundaries and whether the accrued information size met or exceeded the heterogeneity-adjusted RIS, thereby supporting the conclusiveness of the findings despite the limited number of trials and varying degrees of heterogeneity. The retrospective nature of TSA is acknowledged, and no futility boundaries were applied, which limits the interpretation of non-significant findings but supports benefit conclusions where boundaries are crossed. Assessment of publication bias Publication bias was assessed through visual inspection of funnel plots for outcomes where 10 or more studies were available. Asymmetric funnel plots were considered suggestive of potential publication bias, which was further evaluated using Egger’s test to quantify the extent of bias statistically. Results Study selection and characteristics A total of 189 studies were initially identified through database searches. After the removal of duplicates and the screening of titles, abstracts, and full texts, six RCTs were included in the final analysis (Fig. 1). These trials evaluated Mavacamten in patients with HCM, including obstructive (oHCM) and non-obstructive (nHCM) phenotypes. The included studies were: EXPLORER-HCM[1] (n = 251, 30-week follow-up), VALOR-HCM (2023, week 56 results; n = 112, 16-week follow-up), Desai et al[6] (week 32 results from VALOR-HCM including cross-over, n = 108), EXPLORER-CN[14] (n = 81, 30-week follow-up), MAVERICK-HCM[13] (n = 59, 16-week follow-up), and PIONEER-OLE[4] (including PIONEER-HCM 2019; n = 21, 12-week follow-up). Sample sizes ranged from 21 to 251 participants per study, with follow-up durations of 12–30 weeks. Across trials, the mean participant age was generally in the mid-50s (ranging from approximately 49 to 61 years), the mean body mass index was typically 25–32 kg/m2, and the proportion of female participants varied from 24.1% to 68.4%. Most studies compared Mavacamten to placebo, except for PIONEER-OLE, which assessed Mavacamten added to background beta-blocker therapy versus beta-blocker alone. Detailed baseline characteristics of the treatment groups are presented in Table 1. Figure 1.PRISMA 2020 flow diagram for systematic reviews, which includes searches of databases and registers. Table 1Baseline population characteristics of included studies.StudyTreatment groupSample sizeFollow up in weeksFemale, n (%)Age mean (SD)BMI mean (SD)EXPLORER-HCM[1]Mavacamten1233057 (46)58.5 (12.2)29.7 (4.9)Placebo1283045 (35)58.5 (11.8)29.2 (5.6)VALOR-HCM[5] (Week 56 results from the VALOR-HCM)Mavacamten561627 (48.2)59.8 (14.2)29.3 (4.8)Placebo561628 (50.0)60.9 (10.5)31.9 (6.2)Desai et al[6] (Week 32 results from the VALOR-HCM)Mavacamten563227 (48.2)59.8 (14.2)29.3 (4.8)Mavacamten cross-over after Placebo521627 (51.9)60.9 (10.4)32.1 (6.3)EXPLORER-CN[14]Mavacamten543013 (24.1)52.4 (12.1)25.2 (3.5)Placebo273010 (37)51 (11.8)26.1 (3.6)Maverick-HCM[13]Mavacamten401621 (52.5)54.0 (14.6)29.3 (5.2)Placebo191613 (68.4)53.8 (18.2)21 (4.9)PIONEER-OLE[4](and PIONEER-HCM 2019)Mavacamten+ beta-blocker10125 (50)50.3 (35.3)32.3 (5.4)Beta-blocker alone11124 (36)49.3 (40.7)29.7 (4.1) PRISMA 2020 flow diagram for systematic reviews, which includes searches of databases and registers. Table 1Baseline population characteristics of included studies.StudyTreatment groupSample sizeFollow up in weeksFemale, n (%)Age mean (SD)BMI mean (SD)EXPLORER-HCM[1]Mavacamten1233057 (46)58.5 (12.2)29.7 (4.9)Placebo1283045 (35)58.5 (11.8)29.2 (5.6)VALOR-HCM[5] (Week 56 results from the VALOR-HCM)Mavacamten561627 (48.2)59.8 (14.2)29.3 (4.8)Placebo561628 (50.0)60.9 (10.5)31.9 (6.2)Desai et al[6] (Week 32 results from the VALOR-HCM)Mavacamten563227 (48.2)59.8 (14.2)29.3 (4.8)Mavacamten cross-over after Placebo521627 (51.9)60.9 (10.4)32.1 (6.3)EXPLORER-CN[14]Mavacamten543013 (24.1)52.4 (12.1)25.2 (3.5)Placebo273010 (37)51 (11.8)26.1 (3.6)Maverick-HCM[13]Mavacamten401621 (52.5)54.0 (14.6)29.3 (5.2)Placebo191613 (68.4)53.8 (18.2)21 (4.9)PIONEER-OLE[4](and PIONEER-HCM 2019)Mavacamten+ beta-blocker10125 (50)50.3 (35.3)32.3 (5.4)Beta-blocker alone11124 (36)49.3 (40.7)29.7 (4.1) Baseline population characteristics of included studies. Outcomes Primary outcomes Post-exercise LVOT gradient Across two studies with 351 patients, Mavacamten significantly reduced the post-exercise LVOT gradient compared to placebo. The pooled MD was −38.28 mmHg (95% CI: −45.92 to −30.63; P < 0.00001), with no observed heterogeneity (I2 = 0%) (Fig. 2A). Figure 2.Forest plots summarizing the meta-analysis results for Mavacamten versus placebo in patients with hypertrophic cardiomyopathy (HCM). (A) Change in post-exercise left ventricular outflow tract (LVOT) gradient, showing a significant reduction with Mavacamten. (B) Change in post-Valsalva LVOT gradient, demonstrating a significant reduction. (C) Change in baseline NT-proBNP levels, indicating a significant decrease with Mavacamten. (D) Change in resting LVOT gradient, showing a significant reduction. (E) Change in baseline peak oxygen consumption (pVO2), with no significant improvement (MD: 0.69 mL/kg/min, 95% CI: −1.12 to 2.50, P = 0.45). Forest plots summarizing the meta-analysis results for Mavacamten versus placebo in patients with hypertrophic cardiomyopathy (HCM). (A) Change in post-exercise left ventricular outflow tract (LVOT) gradient, showing a significant reduction with Mavacamten. (B) Change in post-Valsalva LVOT gradient, demonstrating a significant reduction. (C) Change in baseline NT-proBNP levels, indicating a significant decrease with Mavacamten. (D) Change in resting LVOT gradient, showing a significant reduction. (E) Change in baseline peak oxygen consumption (pVO2), with no significant improvement (MD: 0.69 mL/kg/min, 95% CI: −1.12 to 2.50, P = 0.45). Post-Valsalva LVOT gradient Two studies, including 193 patients, demonstrated that Mavacamten significantly reduced the post-Valsalva LVOT gradient. The pooled MD was −59.02 mmHg (95% CI: −82.75 to −35.29; P < 0.00001), although heterogeneity was high (I2 = 92%) (Fig. 2B). Resting LVOT gradient For the resting LVOT gradient, data from two studies (193 patients) showed a pooled MD of −44.50 mmHg (95% CI: −66.49 to −22.51; P < 0.0001), indicating significant improvement with Mavacamten, though heterogeneity was high (I2 = 93%) (Fig. 2D). TSA results TSA was conducted retrospectively to evaluate the robustness of the meta-analysis findings on Mavacamten’s efficacy in reducing LVOT gradients in patients with HCM (Fig. 3). TSA adjusts for cumulative type I and type II errors in sequential meta-analyses, providing boundaries for benefit, harm, futility, and RIS. This helps determine if the evidence is sufficient to draw firm conclusions or if more trials are needed. We applied TSA to the primary outcomes: post-exercise, post-Valsalva, and resting LVOT gradients, using a random-effects model with the esOF function. Assumptions included an MCID of −50 mmHg, SD of 10 mmHg, two-sided alpha of 0.05, and power of 90% (beta = 0.10) or 85% (beta = 0.15). Heterogeneity was incorporated via tau2, with RIS adjusted for one additional trial and sequential design inflation. Figure 3.Trial sequential analysis (TSA) plots demonstrate the robustness of Mavacamten’s effect on left ventricular outflow tract (LVOT) gradients in patients with hypertrophic cardiomyopathy. (A) Retrospective TSA for post-Valsalva LVOT gradient reduction (alpha = 0.05, beta = 0.10; 90% power, O’Brien-Fleming alpha-spending). The cumulative Z-score (solid line with filled circles) crosses the monitoring boundary early (at ~58% information fraction), with the accrued sample size exceeding the heterogeneity-adjusted required information size (RIS; 193 patients). Pooled random-effects mean difference (MD) = −59.04 mmHg [95% Sidik–Jonkman (SW)-adjusted CI: −31.86 to −5.18; SW P = 0.0065]; tau2 = 270.23, I2 = 91.7%, D2 = 98.2%, heterogeneity P = 5 × 10−4. (B) Retrospective TSA for post-Valsalva LVOT gradient (alpha = 0.05, beta = 0.15; 85% power, O’Brien-Fleming). Results are consistent with Panel A, with early crossing of the benefit boundary and conclusive evidence despite high heterogeneity. (C) Retrospective TSA for resting LVOT gradient reduction (alpha = 0.05, beta = 0.10; 90% power, O’Brien-Fleming). The cumulative Z-score crosses the monitoring boundary, and the information fraction reaches or exceeds the adjusted RIS (194 patients). Pooled random-effects MD = −44.51 mmHg (95% SW-adjusted CI: −6.47 to −1.62; SW P = 0.0011); tau2 = 234.26, I2 = 92.6%, D2 = 98.8%, heterogeneity P = 2 × 10−4. (D) Retrospective TSA for resting LVOT gradient (alpha = 0.05, beta = 0.15; 85% power, O’Brien-Fleming). Findings remain concordant with C, confirming firm evidence of benefit. Dotted red lines represent conventional (naive) alpha boundaries; dashed blue lines represent TSA-adjusted monitoring boundaries (O’Brien-Fleming spending function). All analyses used random-effects models with heterogeneity adjustment (tau2), assumed one or two additional trials where specified, and included sequential design inflation as applicable. No futility boundaries were set. Analyses are retrospective; study order follows the dataset sequence. Trial sequential analysis (TSA) plots demonstrate the robustness of Mavacamten’s effect on left ventricular outflow tract (LVOT) gradients in patients with hypertrophic cardiomyopathy. (A) Retrospective TSA for post-Valsalva LVOT gradient reduction (alpha = 0.05, beta = 0.10; 90% power, O’Brien-Fleming alpha-spending). The cumulative Z-score (solid line with filled circles) crosses the monitoring boundary early (at ~58% information fraction), with the accrued sample size exceeding the heterogeneity-adjusted required information size (RIS; 193 patients). Pooled random-effects mean difference (MD) = −59.04 mmHg [95% Sidik–Jonkman (SW)-adjusted CI: −31.86 to −5.18; SW P = 0.0065]; tau2 = 270.23, I2 = 91.7%, D2 = 98.2%, heterogeneity P = 5 × 10−4. (B) Retrospective TSA for post-Valsalva LVOT gradient (alpha = 0.05, beta = 0.15; 85% power, O’Brien-Fleming). Results are consistent with Panel A, with early crossing of the benefit boundary and conclusive evidence despite high heterogeneity. (C) Retrospective TSA for resting LVOT gradient reduction (alpha = 0.05, beta = 0.10; 90% power, O’Brien-Fleming). The cumulative Z-score crosses the monitoring boundary, and the information fraction reaches or exceeds the adjusted RIS (194 patients). Pooled random-effects MD = −44.51 mmHg (95% SW-adjusted CI: −6.47 to −1.62; SW P = 0.0011); tau2 = 234.26, I2 = 92.6%, D2 = 98.8%, heterogeneity P = 2 × 10−4. (D) Retrospective TSA for resting LVOT gradient (alpha = 0.05, beta = 0.15; 85% power, O’Brien-Fleming). Findings remain concordant with C, confirming firm evidence of benefit. Dotted red lines represent conventional (naive) alpha boundaries; dashed blue lines represent TSA-adjusted monitoring boundaries (O’Brien-Fleming spending function). All analyses used random-effects models with heterogeneity adjustment (tau2), assumed one or two additional trials where specified, and included sequential design inflation as applicable. No futility boundaries were set. Analyses are retrospective; study order follows the dataset sequence. For the post-exercise LVOT gradient, data from two RCTs involving 351 patients showed a pooled MD of −38.28 mmHg (95% CI: −45.92 to −30.63; P <0.00001), with no heterogeneity (I2 = 0%). Although formal TSA output was not recomputed here due to prior attainment, the cumulative evidence has already crossed the monitoring boundaries for benefit, indicating conclusive superiority of Mavacamten over placebo. The sample size exceeds the RIS threshold, confirming that the reduction is robust and unlikely to be overturned by future trials. This supports Mavacamten’s role in alleviating dynamic obstruction post-exercise, a key driver of symptoms in obstructive HCM. For the post-Valsalva LVOT gradient, two RCTs with 193 patients yielded a pooled MD of −59.02 mmHg (95% CI: −82.75 to −35.29; P < 0.00001), despite high heterogeneity (I2 = 92%). TSA revealed an RIS of 195 patients. At the first interim (sma_timing = 0.58), the Z-score (−6.758) far exceeded the upper boundary (2.721), signaling early evidence of benefit. By full accrual (sma_timing = 1.00), the Z-score (−4.876, random-effects) crossed the final boundary (1.978), with a TSA-adjusted 95% CI for random-effects MD of −82.98 to −35.09 mmHg. The fixed-effects MD was −68.92 mmHg (TSA CI: −72.11 to −65.74). Heterogeneity metrics included tau2 = 270.23 and D2 = 98.2% (P = 0.0005). Repeating with beta = 0.15 yielded identical boundaries and conclusions, with no futility signals. The median unbiased estimate [Sidik–Jonkman (SW)] was −18.52 mmHg (SW-adjusted CI: −31.86 to −5.18; P = 0.0065), underscoring reliability despite heterogeneity, possibly from study design differences. Similarly, for the resting LVOT gradient, two RCTs (193 patients) demonstrated a pooled MD of −44.50 mmHg (95% CI: −66.49 to −22.51; P < 0.0001), with high heterogeneity (I2 = 93%). TSA computed an RIS of 194. At sma_timing = 0.42, the Z-score (−44.450) surpassed the boundary (3.269). At completion (sma_timing = 1.00), the Z-score (−3.967 random) crossed 1.963, with a TSA-adjusted random-effects CI of −66.53 to −22.48 mmHg. The fixed-effects MD was −54.07 mmHg (TSA CI: −56.44 to −51.69). Tau2 = 234.26, D2 = 98.8% (P = 0.0002). Beta = 0.15 analysis mirrored these results, with no futility. The median unbiased MD was −4.04 mmHg (SW CI: −6.47 to −1.62; P = 0.0011). Limitations noted retrospective design issues, potentially affecting boundary precision, but overall, the evidence firmly establishes benefit. Across outcomes, TSA confirms that Mavacamten’s reductions in LVOT gradients are conclusive, with accrued information surpassing RIS despite few trials and heterogeneity. This bolsters confidence in the meta-analysis, suggesting no need for further RCTs on these endpoints. However, for secondary outcomes like pVO2, where no significant effect was found (MD: 0.69 mL/kg/min; P = 0.45), TSA was not applied due to inconclusive pooled results. Future prospective TSA designs could refine these assessments, but current findings affirm Mavacamten’s therapeutic value in HCM without inflated error risks. Overall, retrospective TSA provides strong evidence of conclusive benefit for LVOT gradient reductions, with early crossing of monitoring boundaries and sufficient accrued information relative to RIS, supporting clinical certainty on these endpoints despite limited trials. No futility signals emerged. Secondary outcomes NT-proBNP levels Analysis of NT-proBNP levels included two studies with 435 patients. Mavacamten resulted in a significant reduction in NT-proBNP levels compared to placebo, with a pooled MD of −617.26 ng/L (95% CI: −772.43 to −462.09; P < 0.00001). However, heterogeneity was low (I2 = 0%) (Fig. 2C). Subgroup analyses by NYHA class or background beta-blocker use were limited by data availability but are recommended in future updates. Peak oxygen consumption Two studies involving 310 patients evaluated pVO2. The pooled MD was 0.69 mL/kg/min (95% CI: −1.12 to 2.50; P = 0.45), indicating no statistically significant improvement in functional capacity with Mavacamten compared to placebo. Heterogeneity was high (I2 = 78%) (Fig. 2E). Additional functional measures (e.g., 6-minute walk test, patient-reported outcomes like Kansas City Cardiomyopathy Questionnaire [KCCQ]) were inconsistently reported across trials and showed variable results, warranting further exploration. Left ventricular ejection fraction Four studies involving 494 patients evaluated LVEF change. The pooled MD was −2.37% (95% CI: −4.42 to −0.32; P = 0.02), indicating a statistically significant reduction in LVEF with Mavacamten compared to placebo. Heterogeneity was high (I2 = 67%) (Fig. 4). Figure 4.Forest plots summarizing the meta-analysis results for Mavacamten versus placebo in patients with hypertrophic cardiomyopathy (HCM). (A) Change in LVEF. (B) Any serious treatment-related adverse event. (C) Funnel plot assessing publication bias for the safety outcome, showing a near-symmetrical distribution of studies around the pooled effect size (log odds ratio). Forest plots summarizing the meta-analysis results for Mavacamten versus placebo in patients with hypertrophic cardiomyopathy (HCM). (A) Change in LVEF. (B) Any serious treatment-related adverse event. (C) Funnel plot assessing publication bias for the safety outcome, showing a near-symmetrical distribution of studies around the pooled effect size (log odds ratio). Safety and adverse events The included studies reported that Mavacamten was generally well-tolerated. The pooled analysis of total SAEs between Mavacamten and placebo arms included data from five studies (MAVERICK-HCM 2020, EXPLORER-HCM 2020, Desai et al 2022, EXPLORER-CN 2023, and PIONEER-OLE 2024), encompassing 282 patients in the Mavacamten group and 240 in the placebo group[1,4,6,13,14]. The risk ratio (RR) for SAEs was 1.17 (95% CI: 0.58–2.36, P = 0.66), indicating no statistically significant difference in the occurrence of SAEs between the Mavacamten and placebo groups. The heterogeneity was low (I2 = 20%), suggesting consistent findings across the included studies (Fig. 4). While Mavacamten did not show an increased risk of SAEs compared to placebo, the RR varied slightly between studies, with some reporting slightly higher SAE events in the Mavacamten group[6,14]. These results suggest that Mavacamten is well-tolerated, but continued monitoring in long-term trials is essential to confirm its safety profile. Mavacamten significantly improved LVOT gradients (post-exercise, post-Valsalva, and resting) and reduced NT-proBNP levels, reflecting improved cardiac function and reduced disease burden (Figs. 2 and 4). However, its impact on functional capacity, as measured by pVO2, was not significant. High heterogeneity observed in some outcomes warrants cautious interpretation of the findings. Publication bias, as assessed using funnel plots, regression analysis results, and risk of bias, is also shown in the respective figures (Figs. 4–6). Grade assessment is shown in Table 4. Sensitivity analysis results Publication bias The assessment of publication bias, as depicted in the funnel plot (Fig. 4), revealed a relatively symmetric distribution of studies around the mean effect size, suggesting minimal publication bias in this meta-analysis. This visual symmetry indicates that the included studies are not skewed toward positive or negative outcomes, which supports the reliability of the pooled results. However, due to the limited number of studies included in the analysis, further statistical tests, such as Egger’s test, could not be reliably performed to confirm this observation. The small sample size of studies underscores the need for cautious interpretation of the funnel plot results and highlights the importance of future research with a larger number of trials to more definitively assess publication bias. Influence diagnostics Influence diagnostics, illustrated in Figure 5 and detailed in Table 2, provided insights into the potential impact of individual studies on the overall pooled effect of Mavacamten in HCM on the safety outcome. The analysis revealed that MAVERICK-HCM[13] and EXPLORER-HCM[1] exhibited higher Cook’s distance values, indicating a stronger influence on the pooled results compared to other studies. Additionally, EXPLORER-HCM[1] showed the highest covariance ratio, suggesting that its exclusion would significantly alter the variance of the overall estimate. The tau-squared deletion measure remained consistent across most studies, with the exception of MAVERICK-HCM[13], which contributed slightly more to between-study variability. Furthermore, EXPLORER-HCM[1] displayed the highest hat values, indicating significant leverage and reinforcing its influence on the meta-analysis model. Despite these observations, the inclusion of these studies did not drastically alter the overall findings, confirming the robustness of the meta-analysis and the stability of the pooled effect estimates. Figure 5.Diagnostic and sensitivity analyses for the meta-analysis of Mavacamten versus placebo in hypertrophic cardiomyopathy (HCM). (A) Influence diagnostics assessing the impact of individual studies on the overall meta-analysis for post-exercise LVOT gradient reduction, including metrics such as standardized residuals (rstudent), DFFITS, Cook’s distance (cook.d), covariance ratio (cov.r), tau2 with the study omitted (tau².Del), DFBETAS (dfb.Del), hat values (hat), and weights (weight). (B) Leave-one-out sensitivity analysis, showing the stability of the pooled effect size when each study is excluded. (C) Meta-regression coefficients for the effect of study-level moderators (population size, sex distribution, and average age), with no significant associations observed (all coefficients non-significant). (D) Bubble plot of meta-regression, illustrating the lack of association between study-level characteristics and the effect size (log odds ratio), with bubble size representing study weight. CI, confidence interval; I2, proportion of total variation due to heterogeneity. Figure 6.Risk of bias of included studies. Table 2Influence diagnostics summary results.StudyrstudentDFFITScook.dcov.rtau2.delQE.delhatweightdfbs_infMAVERICK-HCM[13]−0.9355−0.84771.02692.12090.22032.88640.21621.5966−0.7996EXPLORER-HCM[1]−0.7435−0.79131.34272.79680.23843.03220.532553.2525−0.9019Desai et al[6]1.2180.37370.13971.094202.24820.08618.60560.3737EXPLORER-CN[14]0.94350.21340.04561.051202.84150.04874.86870.2134PIONEER-OLE[4]0.78150.28410.08071.132203.12090.116811.67660.2841 Diagnostic and sensitivity analyses for the meta-analysis of Mavacamten versus placebo in hypertrophic cardiomyopathy (HCM). (A) Influence diagnostics assessing the impact of individual studies on the overall meta-analysis for post-exercise LVOT gradient reduction, including metrics such as standardized residuals (rstudent), DFFITS, Cook’s distance (cook.d), covariance ratio (cov.r), tau2 with the study omitted (tau².Del), DFBETAS (dfb.Del), hat values (hat), and weights (weight). (B) Leave-one-out sensitivity analysis, showing the stability of the pooled effect size when each study is excluded. (C) Meta-regression coefficients for the effect of study-level moderators (population size, sex distribution, and average age), with no significant associations observed (all coefficients non-significant). (D) Bubble plot of meta-regression, illustrating the lack of association between study-level characteristics and the effect size (log odds ratio), with bubble size representing study weight. CI, confidence interval; I2, proportion of total variation due to heterogeneity. Risk of bias of included studies. Table 2Influence diagnostics summary results.StudyrstudentDFFITScook.dcov.rtau2.delQE.delhatweightdfbs_infMAVERICK-HCM[13]−0.9355−0.84771.02692.12090.22032.88640.21621.5966−0.7996EXPLORER-HCM[1]−0.7435−0.79131.34272.79680.23843.03220.532553.2525−0.9019Desai et al[6]1.2180.37370.13971.094202.24820.08618.60560.3737EXPLORER-CN[14]0.94350.21340.04561.051202.84150.04874.86870.2134PIONEER-OLE[4]0.78150.28410.08071.132203.12090.116811.67660.2841 Influence diagnostics summary results. Leave-one-out sensitivity analysis The leave-one-out sensitivity analysis, presented in Figure 5, evaluated the consistency of the pooled effect size by systematically excluding one study at a time. The results demonstrated remarkable stability across all iterations, with minimal variability in the effect size and CIs. For instance, excluding MAVERICK-HCM[13] resulted in an odds ratio ranging from 0.41 to 1.49, with an I2 value of 20.6%, indicating moderate heterogeneity. In contrast, excluding PIONEER-OLE[4] yielded an odds ratio of 0.59–0.80, with I2 reduced to 0%, reflecting low heterogeneity. These findings from the leave-one-out analysis further validated the robustness of the pooled estimates, ensuring that no single study disproportionately influenced the overall results. Collectively, the sensitivity analyses, including influence diagnostics and leave-one-out analysis, underscore the reliability of the meta-analysis conclusions regarding Mavacamten’s effectiveness and safety in treating HCM. Meta-regression analysis results Coefficient analysis The meta-regression analysis explored the influence of study-level characteristics on safety outcomes, such as population size, sex distribution (male), and average age, on the overall effect size of Mavacamten in HCM, with results summarized in Table 3 and Figure 5. The analysis revealed that none of the moderators significantly influenced the effect size. The intercept coefficient was −2.4755, with a standard error of 9.4383, a Z value of −0.2623, and a P-value of 0.7931, with a CI ranging from −20.9742 to 16.0232, indicating high uncertainty and no significant effect. The population size coefficient was −0.0025, with a standard error of 0.0062, a Z value of −0.4007, and a P-value of 0.6886, with a CI from −0.0145 to 0.0096, showing no notable impact. For sex distribution (male), the coefficient was 0.0397, with a standard error of 0.0541, a Z value of 0.735, and a P-value of 0.4623, with a CI from −0.0662 to 0.1457, indicating no significant influence of male predominance on the effect size. Finally, the average age coefficient was 0.0188, with a standard error of 0.1792, a Z value of 0.1048, and a P-value of 0.9165, with a CI from −0.3324 to 0.37, suggesting no significant relationship between age and effect size. Figure 5 visually confirms these findings, illustrating that all moderators had minimal and statistically insignificant effects on the pooled estimates.Table 3Meta-regression analysis results.ParametersCoefficientsSEZ valueP-valueCI lowerCI upperIntercept−2.47559.4383−0.26230.7931−20.974216.0232Population−0.00250.0062−0.40070.6886−0.01450.0096Sex (male) (%)0.03970.05410.7350.4623−0.06620.1457Average age0.01880.17920.10480.9165−0.33240.37 Table 3Meta-regression analysis results.ParametersCoefficientsSEZ valueP-valueCI lowerCI upperIntercept−2.47559.4383−0.26230.7931−20.974216.0232Population−0.00250.0062−0.40070.6886−0.01450.0096Sex (male) (%)0.03970.05410.7350.4623−0.06620.1457Average age0.01880.17920.10480.9165−0.33240.37 Meta-regression analysis results. Bubble plot analysis The bubble plot, shown in Figure 5, illustrates the relationship between the predicted values (fitted) and the effect size (log odds ratio) for the included studies, with bubble size representing the weight of each study in the meta-regression model. The plot revealed no clear pattern or significant moderator effect, further supporting the lack of statistical significance for the included variables. This visual representation aligns with the coefficient analysis, indicating that study-level characteristics such as population size, sex distribution, and average age do not significantly influence the therapeutic safety of Mavacamten. The meta-regression analysis collectively suggests that Mavacamten’s benefits and safety in HCM are consistent across diverse study populations, regardless of these demographic and study-level factors, reinforcing its potential for broad clinical applicability. Discussion This systematic review and meta-analysis offers a comprehensive assessment of Mavacamten’s efficacy and safety in patients with HCM, synthesizing data from six RCTs involving 632 patients. By focusing on critical outcomes such as LVOT gradients, NT-proBNP levels, functional capacity, and SAEs, our findings contribute to a deeper understanding of Mavacamten’s role as a novel selective cardiac myosin inhibitor in managing obstructive HCM (Tables 1 and 4, and Fig. 1).Table 4Grade assessment summary.OutcomeNo. of studies (participants)Pooled effect (MD, 95% CI)Risk of biasInconsistencyIndirectnessImprecisionPublication biasOverall certainty of evidenceRationale for downgradingPost-exercise LVOT gradient2 RCTs (351 patients)−38.28 mmHg (−45.92 to −30.63)LowNone (I2 = 0%)No seriousNo seriousUndetectedHighStarting from high (RCTs); no downgrades: low risk of bias in contributing studies, no heterogeneity, precise CI (clinically meaningful effect well below null), no publication bias concerns (few studies but symmetric funnel if assessed).Post-Valsalva LVOT gradient2 RCTs (193 patients)−59.02 mmHg (−82.75 to −35.29)Low to some concernsSerious (I2 = 92%)No seriousNo seriousUndetectedModerateDowngraded 1 level for serious inconsistency (high heterogeneity unexplained by obvious clinical differences in populations/protocols). Risk of bias mostly low (some concerns in one study but not dominant); effect large and precise despite heterogeneity.Resting LVOT gradient2 RCTs (193 patients)−44.50 mmHg (−66.49 to −22.51)Low to some concernsSerious (I2 = 93%)No seriousNo seriousUndetectedModerateDowngraded 1 level for serious inconsistency (very high heterogeneity). Risk of bias low to some concerns (one study with issues but not overriding); wide but still precise CI crossing clinically important thresholds; no publication bias. Table 4Grade assessment summary.OutcomeNo. of studies (participants)Pooled effect (MD, 95% CI)Risk of biasInconsistencyIndirectnessImprecisionPublication biasOverall certainty of evidenceRationale for downgradingPost-exercise LVOT gradient2 RCTs (351 patients)−38.28 mmHg (−45.92 to −30.63)LowNone (I2 = 0%)No seriousNo seriousUndetectedHighStarting from high (RCTs); no downgrades: low risk of bias in contributing studies, no heterogeneity, precise CI (clinically meaningful effect well below null), no publication bias concerns (few studies but symmetric funnel if assessed).Post-Valsalva LVOT gradient2 RCTs (193 patients)−59.02 mmHg (−82.75 to −35.29)Low to some concernsSerious (I2 = 92%)No seriousNo seriousUndetectedModerateDowngraded 1 level for serious inconsistency (high heterogeneity unexplained by obvious clinical differences in populations/protocols). Risk of bias mostly low (some concerns in one study but not dominant); effect large and precise despite heterogeneity.Resting LVOT gradient2 RCTs (193 patients)−44.50 mmHg (−66.49 to −22.51)Low to some concernsSerious (I2 = 93%)No seriousNo seriousUndetectedModerateDowngraded 1 level for serious inconsistency (very high heterogeneity). Risk of bias low to some concerns (one study with issues but not overriding); wide but still precise CI crossing clinically important thresholds; no publication bias. Grade assessment summary. Principal findings Our analysis revealed that Mavacamten significantly reduced LVOT gradients across various clinical scenarios, with a pooled MD of −38.28 mmHg for post-exercise gradients and −59.02 mmHg for post-Valsalva gradients, both highly significant (P < 0.00001). These reductions align with its mechanism of action, which mitigates hypercontractility and enhances myocardial relaxation, directly addressing the pathophysiological drivers of LVOT obstruction in HCM[3,8]. These results are consistent with landmark trials such as EXPLORER-HCM and VALOR-HCM, which reported substantial LVOT gradient reductions, affirming Mavacamten’s ability to alleviate a hallmark feature of obstructive HCM[1,5,6]. Additionally, the TSA reinforced the robustness of these findings, with cumulative Z-scores crossing monitoring boundaries early, and the current sample size meeting the RIS, confirming the effect size definitively. Mavacamten also significantly lowered NT-proBNP levels, a key biomarker of myocardial stress, with a pooled MD of −617.26 ng/L. This reduction indicates a meaningful decrease in cardiac burden, potentially slowing the progression to heart failure, a critical concern in HCM[9,10]. The consistency of this finding across studies underscores Mavacamten’s potential to improve myocardial function beyond just hemodynamic effects. However, functional capacity, as measured by pVO2, showed no significant improvement (MD: 0.69 mL/kg/min, P = 0.45). This finding may reflect the short follow-up duration of included trials or heterogeneity in patient populations, such as differences in baseline functional status or disease severity[12,13]. This lack of effect on pVO2 may relate to short trial durations (mostly 12–30 weeks), heterogeneity in baseline functional status, or limitations of pVO2 as a sensitive endpoint in HCM; longer follow-up or more patient-centered measures (e.g., KCCQ and daily activity) might reveal benefits. The finding underscores the need for targeted strategies in subgroups with severe baseline limitations. Our meta-regression analysis found no significant moderators, such as population size, sex distribution, or average age, influencing Mavacamten’s safety effect (Table 3). The bubble plot further supported this, showing no association between study-level characteristics and the pooled effect size. This suggests that Mavacamten’s benefits are consistent across diverse HCM patient cohorts, enhancing its potential for broad clinical application[14,15]. Comparison with current literature The observed reductions in LVOT gradients and NT-proBNP levels align with prior studies, including EXPLORER-HCM, MAVERICK-HCM, and EXPLORER-CN, which collectively demonstrated Mavacamten’s efficacy in addressing the core pathophysiology of HCM, namely, excessive contractility and impaired relaxation[1,13,14]. Similarly, long-term data from the MAVA-LTE and PIONEER-OLE studies corroborate the sustained efficacy and tolerability of Mavacamten, with consistent reductions in LVOT gradients over extended periods[4,10]. A recent study further supports these findings, reporting that Mavacamten improves NYHA functional class and reduces the need for septal reduction therapies in patients with obstructive HCM, highlighting its potential as a disease-modifying therapy[21]. However, the lack of significant improvement in pVO2 contrasts with some smaller studies that reported modest gains in exercise capacity[12,15]. This discrepancy may be attributed to the complexity of functional outcomes in HCM, where factors such as patient comorbidities, baseline fitness levels, and trial duration play significant roles. For instance, patients with concurrent hypertension or those on beta-blockers may exhibit variable responses to Mavacamten, as noted in post-hoc analyses of EXPLORER-HCM[12,15]. This suggests that Mavacamten’s impact on functional capacity might be more pronounced in specific subgroups, such as those with severe obstruction or limited background therapy, warranting further investigation. Safety profile Mavacamten’s safety profile was reassuring, with no significant increase in SAEs compared to placebo (RR: 1.17, 95% CI: 0.67–2.08, P = 0.58). Sensitivity analyses confirmed the robustness of this finding, consistent with prior trials like EXPLORER-HCM and MAVERICK-HCM, which reported a favorable safety profile with no major concerns[1,13]. Four studies, encompassing 494 patients, evaluated the change in LVEF. The pooled MD was −2.37% (95% CI: −4.42 to −0.32; P = 0.02), demonstrating a statistically significant reduction in LVEF with Mavacamten compared with placebo, although heterogeneity across the studies was high (I2 = 67%) (Fig. 4). This modest yet significant decrease in LVEF aligns with Mavacamten’s intended mechanism of action as a selective cardiac myosin inhibitor, which reduces excessive contractility to relieve LVOT obstruction in HCM. The effect appears more pronounced in patients who have lower baseline LVEF values. To maintain patient safety, current clinical guidelines and the prescribing information for Mavacamten emphasize the importance of regular echocardiographic monitoring of both LVEF and LVOT gradients. This begins with a baseline assessment before treatment initiation, followed by follow-up echocardiography every 4–12 weeks during the dose-titration phase, typically spanning the first 3–6 months of therapy, and then transitions to monitoring every 6 months once a stable maintenance dose has been achieved and the patient remains clinically stable. Dose adjustment or temporary interruption of Mavacamten is recommended if LVEF declines below 50% or if there are signs of clinical heart failure or significant systolic dysfunction. Heightened vigilance is warranted in patients with lower baseline LVEF, intercurrent illness, or atrial fibrillation, as these conditions may increase the risk of more substantial or symptomatic declines in systolic function[1,9,21]. Clinical implications These findings have significant implications for HCM management. Mavacamten offers a targeted, non-invasive treatment option that addresses the underlying hypercontractility and diastolic dysfunction of HCM, potentially reducing the need for invasive procedures like septal myectomy or alcohol septal ablation[4,5]. By lowering LVOT gradients and NT-proBNP levels, Mavacamten may improve quality of life and delay disease progression in symptomatic patients. However, the absence of significant pVO2 improvement suggests Mavacamten addresses hemodynamic and biomarker aspects effectively but may require adjunctive interventions (e.g., supervised exercise) to optimize functional capacity[15,21]. Clinicians should also consider individual patient characteristics, such as baseline LVEF and comorbidities, when prescribing Mavacamten to optimize outcomes. Limitations Several limitations must be acknowledged. High heterogeneity in certain outcomes limits generalizability; funnel plot symmetry suggests minimal publication bias, but the small number of studies precludes reliable statistical confirmation (e.g., Egger’s test), warranting caution. Second, the relatively short follow-up duration in most trials limits insights into Mavacamten’s long-term efficacy and safety. Third, variations in baseline characteristics, such as age, sex, and comorbidities, may affect the generalizability of our findings to broader HCM populations[11,12]. Future directions Future research should prioritize identifying predictors of response (e.g., genetic markers, baseline severity, and NYHA class) and subgroups deriving the greatest benefit. Long-term studies assessing durability and hard outcomes are essential. Head-to-head comparisons with emerging therapies (e.g., aficamten) would clarify relative efficacy and safety. Long-term studies with larger cohorts are needed to assess the durability of Mavacamten’s benefits and its impact on hard outcomes like heart failure progression and sudden cardiac death. Additionally, head-to-head comparisons with emerging therapies like aficamten, another cardiac myosin inhibitor, could clarify their relative efficacy and safety in HCM management[10,21]. Exploring Mavacamten’s role in combination therapies, particularly with beta-blockers or exercise interventions, may also help address its limitations in improving functional capacity[15]. Conclusion This meta-analysis provides compelling evidence that Mavacamten effectively reduces LVOT gradients and NT-proBNP levels in patients with obstructive HCM, offering a promising disease-modifying treatment. However, its impact on functional capacity remains uncertain, and careful monitoring of LVEF is warranted. Larger, longer-term studies and comparative trials with other therapies are essential to fully elucidate Mavacamten’s role in optimizing HCM management.