Lester, Steven J.

Aims:Increased left ventricular (LV) wall thickness is frequently encountered in transthoracic echocardiography (TTE). While accurate and early diagnosis is clinically important, given the differences in available therapeutic options and prognosis, an extensive workup is often required to establish the diagnosis. We propose the first echo-based, automated deep learning model with a fusion architecture to facilitate the evaluation and diagnosis of increased left ventricular (LV) wall thickness. Methods and Results: Patients with an established diagnosis of increased LV wall thickness (hypertrophic cardiomyopathy (HCM), cardiac amyloidosis (CA), and hypertensive heart disease (HTN)/others) between 1/2015 and 11/2019 at Mayo Clinic Arizona were identified. The cohort was divided into 80%/10%/10% for training, validation, and testing sets, respectively. Six baseline TTE views were used to optimize a pre-trained InceptionResnetV2 model. Each model output was used to train a meta-learner under a fusion architecture. Model performance was assessed by multiclass area under the receiver operating characteristic curve (AUROC). A total of 586 patients were used for the final analysis (194 HCM, 201 CA, and 191 HTN/others). The mean age was 55.0 years, and 57.8% were male. Among the individual view-dependent models, the apical 4-chamber model had the best performance (AUROC: HCM: 0.94, CA: 0.73, and HTN/other: 0.87). The final fusion model outperformed all the view-dependent models (AUROC: HCM: 0.93, CA: 0.90, and HTN/other: 0.92). Conclusion: The echo-based InceptionResnetV2 fusion model can accurately classify the main etiologies of increased LV wall thickness and can facilitate the process of diagnosis and workup.

Aims Increased LV wall thickness is frequently encountered in transthoracic echocardiography (TTE). While accurate and early diagnosis is clinically important, given the differences in available therapeutic options and prognosis, an extensive workup is often required to establish the diagnosis. We propose the first echo-based, automated deep learning model with a fusion architecture to facilitate the evaluation and diagnosis of increased left ventricular (LV) wall thickness. Methods and Results Patients with an established diagnosis of increased LV wall thickness (hypertrophic cardiomyopathy (HCM), cardiac amyloidosis (CA), and hypertensive heart disease (HTN)/others) between 1/2015 to 11/2019 at Mayo Clinic Arizona were identified. The cohort was divided into 80%/10%/10% for training, validation, and testing sets, respectively. Six baseline TTE views were used to optimize a pre-trained InceptionResnetV2 model. Each model output was used to train a meta-learner under a fusion architecture. Model performance was assessed by multiclass area under the receiver operating characteristic curve (AUROC). A total of 586 patients were used for the final analysis (194 HCM, 201 CA, and 191 HTN/others). The mean age was 55.0 years, and 57.8% were male. Among the individual view-dependent models, the apical 4 chamber model had the best performance (AUROC: HCM: 0.94, CA: 0.73, and HTN/other: 0.87). The final fusion model outperformed all the view-dependent models (AUROC: CA: 0.90, HCM: 0.93, and HTN/other: 0.92). Conclusion The echo-based InceptionResnetV2 fusion model can accurately classify the main etiologies of increased LV wall thickness and can facilitate the process of diagnosis and workup.

Introduction: Multiple supervised machine learning (ML) models have been used to assess the outcomes in patients who underwent transcatheter aortic valve replacement (TAVR) procedure, however, supervised ML requires significant expert time for annotation and labeling. Unsupervised ML can provide insight into the dataset without this process. Hypothesis: We hypothesized that unsupervised k-means ML can identify distinct phenogroups in TAVR patients that can predict prognosis without a priori knowledge. Methods: A retrospective study of patients who underwent TAVR (Jan 2012-Dec 2017) at Mayo Clinic sites was performed. Baseline demographics, EKG, cardiac computed tomography, and echocardiographic data were abstracted. Variables with ≥ 20% missing data were excluded. Prior to the k-means clustering algorithm, principal component analysis was used for dimensionality reduction. Input variables were analyzed according to k-means determined clusters. Kaplan-Meier survival analysis was used to assess the survival of each cluster. Results: A total of 1071 consecutive patients were included in the final analysis. A total of 141 unique variables were used. The mean age was 80.7±8.6 years, 58.2% were male. Patients were divided into 2 distinct phenogroups. In the two phenogroups, cluster 0 (n=639) has significantly better survival than cluster 1 (n= 432) (log-rank p=0.0002) (Figure 1). Cluster 1 has features of low flow-low gradient (LFLG) hemodynamic profile, including lower systolic blood pressure, mean aortic valve gradient, peak aortic valve velocity, and worse LV ejection fraction. Conclusions: Unsupervised ML can identify distinct phenotype features with prognostic significance in TAVR patients. Our results suggest that patients with LFLG hemodynamic features are intrinsically different from other TAVR patients and require more attention to avoid adverse outcomes.

Introduction: Unsupervised machine learning (ML) can provide insight into a dataset without significant expert time for annotation and labeling, but was never used to assess the heterogeneous MitraClip population. Hypothesis: We hypothesized that unsupervised k-means ML could identify prognostically-distinct phenogroups in MitraClip patients without a priori knowledge of the dataset. Methods: Patients who underwent MitraClip (June 2014-September 2020) at Mayo Clinic sites were identified from the Mayo institutional NCDR database for baseline and follow-up data. Variables with ≥ 50% missing data were excluded. A k-means algorithm was used for clustering analysis. Input variables were grouped according to 4 distinct k-means determined clusters. Kaplan-Meier survival analysis was used to assess the survival of each cluster. Results: A total of 389 consecutive patients were included in the final analysis, and 95 unique variables were used. The mean age was 80.3±8.7 years; 256 (65.8%) were male. The mean STS MV Replace score was 9.1±5.7. Fifty-five (14.5%) patients died during the mean follow duration (185 days). Kaplan-Meier analysis showed significant survival differences among clusters (Figure 1). The cluster (cluster 2, n=120) with best survival performance (all-cause mortality 3.3%) had features of non-calcified mitral leaflets, nonischemic cardiomyopathy, and less than moderate tricuspid regurgitation. Cluster 3 (n=66) had the worst survival performance (all-cause mortality 27.3%) and opposite features than cluster 2. Conclusions: Unsupervised ML can identify distinct phenotype clusters with prognostic significance in MitraClip patients. Patients with non-calcified mitral leaflets, nonischemic cardiomyopathy, and less than moderate tricuspid regurgitation are intrinsically different from other MitraClip patients and should be considered the potential optimal candidates.

Introduction: Obstructive hypertrophic cardiomyopathy (oHCM) consists of a constellation of LV hypercontractility, LV outflow tract obstruction (LVOTO), and impaired relaxation, often resulting in dyspnea and reduced exercise capacity. In healthy dogs, the novel myosin modulator mavacamten (MAVA), decreased LV hypercontractility and improved compliance. In the PIONEER-HCM clinical study, improvements in post-exercise LVOT gradient, exercise capacity, and symptoms were reported. To examine whether MAVA could also limit residual myosin-actin cross-bridges during diastole, and improve LV compliance, its effects on diastolic indices were evaluated. Methods: In PIONEER-HCM, an open-label, multi-site, prospective study, oHCM subjects received MAVA for 12 weeks at either 10-20 mg/d (n = 11, Cohort A; 10 completed) or at 2-5 mg/d (n = 10, Cohort B). Echo-Doppler indices of LV filling and relaxation were compared at baseline to week 12. The p-values are from Wilcoxon signed rank tests, evaluating the distribution of within-subject 12-wk changes from baseline around a null of zero. Results: MAVA reduced hypercontractility in a dose-dependent fashion, with relief of LVOTO in all subjects achieving plasma MAVA concentration >350 ng/ml (data previously reported). There was also an increase in mitral annular velocity during early diastole (e’ lat ). There was a concomitant reduction in E/e’ lat and increase in LV end-diastolic volume (LVEDV) consistent with improved LV compliance (Table). MAVA improved dyspnea score with a trend towards reduction in NT-proBNP. Conclusions: Administration of MAVA is associated with improvement in measures of myocardial relaxation (e’ lat ) and compliance in parallel with reduction in LVOT gradient. These findings are consistent with preclinical findings reported in healthy dogs and support additional evaluation of MAVA in patients with obstructive or non-obstructive HCM.

Introduction: Hypertrophic cardiomyopathy (HCM) carries substantial lifelong morbidity. Mutations in sarcomeric proteins that lead to HCM cause excess contractility, poor relaxation, and inefficient energy utilization. Although obstructive physiology is a prominent feature of disease, approximately 30% of HCM patients experience symptoms in the context of non-obstructive HCM (nHCM). Management options are extremely limited with little other than cardiac transplantation to offer severely affected patients. Mavacamten (MAVA), a first-in-class small molecule, is a selective modulator of cardiac myosin that inhibits the cardiac myosin ATPase to reversibly attenuate cardiac contractility. In addition to its effects on contractility, MAVA has been shown to improve LV compliance in preclinical and clinical studies. The Phase 2 MAVERICK-HCM study is designed to evaluate safety and tolerability, and to provide pilot data on the effects of MAVA on biomarkers, cardiac function, and exercise capacity in nHCM. Methods: This double-blind study will enroll 60 individuals with nHCM (Left ventricular outflow tract gradient <30 mmHg; resting or provoked), NYHA Class II or III, and LVEF ≥55%. Subjects will be randomized 1:1:1 to one of two target plasma drug concentrations (Group 1: ~200 ng/mL and Group 2: ~500 ng/mL) or placebo for 16 weeks. The starting dose of MAVA is 5 mg, with one-step dose titration at Week 6 based on plasma drug concentration. Predefined criteria, including LVEF, will guide study drug discontinuation if indicated. Cardiopulmonary exercise testing will be performed at baseline and Week 16 to assess the impact on exercise capacity. The primary endpoint is the frequency and severity of treatment-emergent adverse events. The study design also includes multiple secondary endpoints including echocardiographic measures of diastolic function, NT-proBNP levels, patient reported outcomes, and physical activity by wearable accelerometer. An independent data monitoring committee will oversee safety. Conclusions: MAVERICK-HCM is testing a novel therapy to improve symptom burden in nHCM; a patient cohort currently lacking effective treatment. By targeting a core molecular mechanism of disease, MAVA has the potential to alter natural history.

Introduction: Obstructive hypertrophic cardiomyopathy (oHCM) is characterized by dynamic LV outflow tract (LVOT) obstruction resulting from ventricular hypertrophy and mitral valve systolic anterior motion (SAM), and by elevated left ventricular (LV) filling pressures, all of which result in a significant morbidity despite medical therapy. Mavacamten (Mava) is a novel inhibitor of cardiac myosin that significantly reduces LVOT gradients in oHCM, noted initially in the phase 2 PIONEER and more recently in the phase 3 EXPLORER-HCM studies. Additional echocardiographic analyses were performed to investigate the effect of Mava on focused measures of cardiac structure and function in oHCM. Methods: EXPLORER-HCM (NCT03470545) was a multicenter, double-blind, placebo-controlled, phase 3 study that randomized symptomatic oHCM patients (LV ejection fraction ≥55% and resting and/or provoked LVOT gradient ≥50 mmHg) 1:1 to Mava vs placebo for 30 weeks. Echocardiograms were performed serially every 2-4 weeks over 30 weeks. Results: In total, 251 patients with oHCM were enrolled and 244 patients completed the study (mean age 58.5; 40.6% female). Compared to placebo, 30-week treatment with Mava led to significant reductions in LA volume index (-7.5 ml/m 2 , p<0.0001), lateral E/e’ (-3.8, p<0.0001), septal E/e’ (-3.4, p<0.0001), and LV mass index (-15.5 g/m 2 , p<0.0001) ( Table ). Significantly more Mava-treated patients achieved resolution of mitral valve SAM compared to placebo (80.9% vs 34.0% of patients; difference of 46.8%; p<0.0001) and mitral regurgitation (MR) (9.0% vs 0.0% of patients; difference of 9.0%; p=0.0006). Conclusions: Treatment with Mava for 30 weeks led to improvement in LV hypertrophy and markers of left-sided filling pressures (LA volume and E/e’). Furthermore, Mava treatment was associated with reduction of SAM and MR. These findings suggest that Mava has beneficial effects on several adverse pathophysiologic processes that are hallmarks of oHCM.

Background: Obstructive hypertrophic cardiomyopathy (oHCM) is characterized by unexplained left ventricular (LV) hypertrophy associated with dynamic LV outflow tract obstruction. Current medical therapies are nonspecific and have limited efficacy in relieving symptoms. Mavacamten is a first-in-class targeted inhibitor of cardiac myosin, which has been shown to reduce LV outflow tract obstruction, improve exercise capacity, and relieve symptoms of oHCM in the PIONEER-HCM phase 2 study. Methods: EXPLORER-HCM is a multicenter, phase 3, randomized, double-blind, placebo-controlled trial to investigate the efficacy and safety of mavacamten in treating symptomatic oHCM. Eligible adults with oHCM and New York Heart Association Functional Class II or III are randomized 1:1 to receive once-daily, oral mavacamten, or matching placebo for 30 weeks. The primary composite functional end point is clinical response at week 30 compared to baseline defined as either (1) an increase in peak oxygen consumption ≥1.5 mL/kg/min and reduction of at least one New York Heart Association class; or (2) an improvement of ≥3.0 mL/kg/min in peak oxygen consumption with no worsening of New York Heart Association class. Secondary end points include change in postexercise LV outflow tract gradient, New York Heart Association class, peak oxygen consumption, and patient-reported outcomes assessed by the Kansas City Cardiomyopathy Questionnaire and a novel HCM-specific instrument. Exploratory end points aim to characterize the effect of mavacamten on multiple aspects of oHCM pathophysiology. Conclusions: EXPLORER-HCM is a phase 3 trial in oHCM testing a first-in-class, targeted strategy of myosin inhibition to improve symptom burden and exercise capacity through reducing LV outflow tract obstruction. Results of this trial will provide evidence to support the first disease-specific treatment for HCM. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03470545.

Abstract Background In a phase 2 PIONEER-HCM study, pts with symptomatic, obstructive hypertrophic cardiomyopathy (oHCM) showed improvement in left ventricular outflow tract (LVOT) obstruction, exercise capacity, and symptoms after 12 wk of treatment with the novel myosin modulator, mavacamten (Mava). Purpose To examine the long-term safety and effectiveness of Mava in PIONEER-OLE study Methods PIONEER-OLE (NCT03496168) is an ongoing 2-y multicenter study for adults with symptomatic oHCM who completed PIONEER-HCM (NCT02842242). The starting dose of Mava is 5 mg/d with titration at wk 6 to an individualized therapeutic dose (5, 10, or 15 mg). Evaluations are at wk 4, 6, 8, 12 and every 12 wk thereafter to monitor LV ejection fraction (LVEF), LVOT gradient, New York Heart Association (NYHA) class, NT-proBNP, drug concentration, and safety. Results 13 pts (mean age, 57.8 y; 9 male; 12 on beta-blockers) were enrolled. Mean baseline LVOT obstruction and LVEF, and wk 12 changes from baseline, were similar to those in PIONEER-HCM (Table). Mava significantly reduced resting and provoked LVOT gradients and NT-proBNP at wk 12 and 24 compared with baseline (P<0.004). Of 10 pts who reached wk 24, 8 reported improvement in NYHA class (1 improved Class III to II; 7 improved Class II to I), and 2 pts remained Class II. Mava has been well tolerated up to 40 wk; 31 adverse events (AEs; 22 mild, 5 moderate) were reported in 8 pts; 1 pt had 3 severe and 1 serious AE (cholangiocarcinoma); all AEs were unrelated to study drug. Results from PIONEER-OLE Parameter PIONEER-HCMa PIONEER-OLE Baseline Wk 12 Baselineb Wk 12 Change at Wk 12 Wilcoxon Signed Wk 24 Change at Wk 24 Wilcoxon Signed Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Rank Mean ± SD Mean ± SD Rank (n=13) (n=13) (n=13)c (n=12)c (n=12)c P value (n=10)c (n=10)c P value LVOT Rest gradient, mmHg 69.7±53.9 27.8±31.3 67.3±42.8 12.0±5.4 −57.9±43.2 0.0005 10.5±4.8 −66.6±42.4 0.0020 LVOT Valsalva gradient, mmHg 93.7±55.6 36.8±37.5 89.9±30.7 (n=12) 23.6±20.0 −66.4±35.3 (n=11) 0.0020 21.1±11.5 −67.3±33.5 (n=9) 0.0039 LVEF, % 73.0±5.6 64.6±10.5 72.0±4.9 67.6±7.2 −4.4±5.5 0.0269 68.2±6.5 −3.2±3.3 0.0195 NT-proBNP, pg/mL 1601.3±2782 (n=12) 684±980 1836±2886 181±211 −1759±2789 0.0005 170±225 −2128±3104 0.0039 Data extraction date January 24, 2019. aCombined results shown for pts from PIONEER-HCM originally in cohort A (n=5) and cohort B (n=8). bBaseline in PIONEER-OLE occurred 6–18 months after completion of PIONEER-HCM. cNumber of pts with data available for analysis, unless otherwise specified. Conclusion Despite management with current therapies, pts enrolled in PIONEER-OLE with similar levels of obstruction and hypercontractility as in PIONEER-HCM. In this longest observation period, Mava significantly reduced obstruction (LVOT gradient) in pts with oHCM beyond standard HCM therapy, while maintaining normal LVEF and improving symptoms. Acknowledgement/Funding MyoKardia