Characterization and Application of Novel Exercise Recovery Patterns That Reflect Cardiac Performance: A Substudy of the SEQUOIA-HCM Trial

Abstract

Background: Post-exercise oxygen uptake recovery (VO₂Rec) is slow in advanced heart failure. We sought to establish easily derived VO₂Rec measures and evaluate their cardiospecificity and prognostic relevance in patients with dyspnea on exertion. We further sought to determine VO₂Rec modifiability proportional to changes in cardiac function with disease-specific treatment of obstructive hypertrophic cardiomyopathy.

Methods: VO₂Rec patterns were evaluated in relation to cardiac performance and the primary outcome of heart failure hospitalization or death in a referral cohort with dyspnea on exertion undergoing cardiopulmonary exercise testing with hemodynamic monitoring (MGH-ExS [Massachusetts General Hospital Exercise Study]). We then investigated longitudinal measures of VO₂Rec in the pivotal phase 3 randomized controlled trial SEQUOIA-HCM (Safety, Efficacy, and Quantitative Understanding of Obstruction Impact of Aficamten in Hypertrophic Cardiomyopathy) of aficamten versus placebo for 24 weeks in participants with symptomatic obstructive hypertrophic cardiomyopathy. For both cohorts, VO₂Rec was uniformly measured as time for VO₂ to decline by >0%, 12.5% (VO₂T_12.5%), 25%, and 50% of peak VO₂.

Results: Among 814 MGH-ExS patients (58±16 years of age, 58% women), those with a longer VO₂T_12.5% (≥35 versus <35 seconds) demonstrated elevated exercise pulmonary capillary wedge pressure to cardiac output slope (P<0.0001) with no difference in peripheral oxygen extraction (P=0.11). For each 15-second increase in VO₂T_12.5%, the hazard ratio for heart failure hospitalization and all-cause death was 1.54 (95% CI, 1.35-1.76; P<0.001). In SEQUOIA-HCM participants with cardiopulmonary exercise testing at baseline and week 24 (n=263, 59.1±2.9 years of age, 41% women), baseline VO₂T_12.5% was 45±20 seconds and improved 8 seconds (95% CI, -12 to -5 seconds; P<0.001) with aficamten treatment compared with placebo at 24 weeks. Participants treated with aficamten versus placebo were more likely to improve VO₂T_12.5% by ≥15 seconds (odds ratio [OR], 3.7 [95% CI, 1.9-6.9]; number needed to treat=4.8). Shortening of VO₂T_12.5% correlated with reduced NT-proBNP (N-terminal pro-B-type natriuretic peptide), high-sensitivity cardiac troponin I, and left ventricular outflow tract gradient (all P<0.005).

Conclusions: This study established VO₂T_12.5% as a new measure that reflects cardiac performance during exercise and predicted heart failure event-free survival. Furthermore, VO₂T_12.5% improved proportional to improvements in left ventricular outflow tract gradient and cardiac biomarkers in response to aficamten treatment, a cardiospecific therapy for obstructive hypertrophic cardiomyopathy. The simplicity and physiological relevance of VO₂T_12.5% support its regular inclusion in cardiopulmonary exercise testing protocols evaluating cardiac function during exercise.

Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT05186818.

Keywords: aficamten; cardiac myosin; cardiomyopathy, hypertrophic; exercise test; heart failure, diastolic; post-exercise recovery.

Figure 1.

Establishing reliable measurements of VO₂ recovery. Oxygen uptake (VO₂) recovery of one individual in SEQUOIA-HCM (Safety, Efficacy, and Quantitative Understanding of Obstruction Impact of Aficamten in Hypertrophic Cardiomyopathy) at baseline and after 24 weeks of aficamten as a representative example of measuring VO₂ recovery, defined by varying percentages of peak VO₂ (VO₂T_0%, VO₂T_12.5%, VO₂T_25%, and VO₂T_50%). VO₂T indicates time to oxygen uptake.

Figure 2.

VO₂T_12.5% vs other noninvasive cardiopulmonary exercise testing measures in relation to the diagnosis of heart failure with preserved ejection fracture. A, Participants with a time to recover 12.5% of peak VO2 (VO₂T_12.5%) ≥35 seconds demonstrated a higher pulmonary capillary wedge pressure to cardiac output slope than VO₂T_12.5% <35 seconds, with the group mean exceeding the diagnostic thresholds for heart failure of 2.0 mm Hg/L per minute. B, However, there was no difference in percent predicted peak peripheral extraction (peak peripheral extraction corrected for hemoglobin concentration) between groups. C, Peak oxygen uptake, oxygen uptake/work slope, and minute ventilation/CO₂ uptake nadir identified participants with heart failure with preserved ejection fracture and an area under the concentration-time curve (SE) of 0.60 (0.02), 0.65 (0.02), and 0.62 (0.03), respectively. VO₂T_12.5% identified heart failure with preserved ejection fracture with an area under the concentration-time curve of 0.72 (0.02). D, Participant demographics, including age, sex, and body mass index, and other noninvasive cardiopulmonary exercise testing measures explained 23% of the variance in pulmonary capillary wedge pressure to cardiac output slope. Addition of VO₂T_12.5% further improves the hemodynamic models to explain 27% of the total variance. *P<0.05 vs both 0 to 15 seconds and 15 to 25 seconds, †P<0.05 vs 25 to 35 seconds, ‡P<0.05 vs 45 to 200 seconds (A). *P<0.05 vs VO₂T_12.5% in DeLong test (C). *P<0.001 vs the previous model in the likelihood ratio test (D). Error bars represent SD of the mean (A and B). AUC indicates area under the curve; Ca-vO₂, arteriovenous oxygen content difference indicative of peripheral oxygen extraction; CPET, cardiopulmonary exercise testing; HFpEF, heart failure with preserved ejection fracture; PCWP/CO, pulmonary capillary wedge pressure to cardiac output slope; VO₂, oxygen uptake; VO2T12.5%, time to recover 12.5% of peak VO2; and VE, minute ventilation.

Figure 3.

Two-minute recovery of VO₂ components in a subset of the referral cohort. A, Patients with prolonged time to recovery 12.5% of peak VO2 had lower cardiac output recovery 2 minutes after exercise (41 vs 13% recovered) B, There was no difference in the rate of peripheral oxygen extraction recovery (100 vs 97%) between groups. C and D, Schematic of relative changes in oxygen uptake recovery component variables during the post-exercise recovery period derived from 2-minute post-exercise recovery data. Data plotted are median with interquartile range. Ca-vO₂ indicates arteriovenous oxygen content difference; CO, cardiac output; HFpEF-HDI, heart failure preserved ejection fraction; HR, heart rate; oHCM, obstructive hypertrophic cardiomyopathy; SV, stroke volume; VO₂Rec, oxygen uptake recovery; VO₂T_12.5%, time to recover 12.5% of peak VO2; and VO₂, oxygen uptake.

Figure 4.

Application of time to oxygen uptake recovery of 12.5% to identify individuals at high-risk of heart failure hospitalization and all-cause death. A, Time to recover 12.5% of peak VO2 (VO2T12.5%) with a cut point of 35 seconds identified 230 patients at elevated risk of heart failure hospitalization and all-cause death. B, Stratification of the Massachusetts General Hospital cohort by percent predicted peak VO₂ and VO2T12.5% identified a cohort of 150 patients with impaired peak oxygen uptake and prolonged VO2T12.5% that had significantly elevated risk of adverse outcomes. The 3 other groups demonstrated a similar risk of heart failure hospitalization and all-cause death. C, VO2T12.5% further stratified patients with NT-proBNP (N-terminal pro-B-type natriuretic peptide) ≤125 pg/mL. The light-weighted lines represent the 95% CI. HF indicates heart failure; MGH, Massachusetts General Hospital; VO₂T_12.5%, time to recover 12.5% of peak VO2; and VO₂, oxygen uptake.

Figure 5.

Application of VO2T12.5% to measure cardiac impairment in SEQUOIA-HCM. A, Responder analyses of clinically meaningful changes in time to recover 12.5% of peak VO2 (VO₂T_12.5%) with aficamten vs placebo at 24 weeks. Data are represented as a proportion of participants experiencing minimal change (<5 seconds), moderate (≥5 seconds), or large (≥15 seconds) deteriorations or improvements in VO₂T_12.5% per treatment group. The number needed to treat for any improvement was 4.6, and the number needed to treat for large improvement was 4.8. A decrease in VO₂T_12.5% from baseline to week 24 corresponded to (B) a decrease in NT-proBNP (N-terminal pro-B-type natriuretic peptide), (C) hs-cTnI (high-sensitivity cardiac troponin I), and (D) left ventricular outflow tract gradient at rest and with (E) Valsalva in participants with obstructive hypertrophic cardiomyopathy. LVOT-G, left ventricular outflow tract gradient; NNT number needed to treat; SEQUOIA-HCM, Safety, Efficacy, and Quantitative Understanding of Obstruction Impact of Aficamten in Hypertrophic Cardiomyopathy; VO₂, oxygen uptake; VO₂Rec, oxygen uptake recovery.