Source: European Medicines Agency (EU) Revision Year: 2023 Publisher: Bristol-Myers Squibb Pharma EEIG, Plaza 254, Blanchardstown Corporate Park 2, Dublin 15, D15 T867, Ireland
Pharmacotherapeutic group: Cardiac therapy, Other cardiac preparations
ATC code: C01EB24
Mavacamten is a selective, allosteric, and reversible cardiac myosin inhibitor. Mavacamten modulates the number of myosin heads that can enter power-generating states, thus reducing (or in HCM normalizing) the probability of force-producing systolic and residual diastolic cross-bridge formation. Mavacamten also shifts the overall myosin population towards an energy-sparing, but recruitable, super-relaxed state. Excess cross-bridge formation and dysregulation of the super-relaxed state of myosin are mechanistic hallmarks of HCM, which can result in hyper-contractility, impaired relaxation, excess energy consumption, and myocardial wall stress. In HCM patients, cardiac myosin inhibition with mavacamten normalises contractility, reduces dynamic LVOT obstruction, and improves cardiac filling pressures.
In the EXPLORER-HCM study, mean (SD) resting LVEF was 74% (6) at baseline in both treatment arms, reductions in mean absolute change from baseline in LVEF was -4% (95% CI: -5.3, -2.5) in the mavacamten arm and 0% (95% CI: -1.2, 1.0) in the placebo arm over the 30-week treatment period. At Week 38, following an 8-week interruption of mavacamten, mean LVEF was similar to baseline for both treatment arms.
In the EXPLORER-HCM study, patients achieved reductions in mean resting and provoked (Valsalva) LVOT gradient by week 4 which were sustained throughout the 30-week study duration. At week 30, the mean change from baseline in resting and Valsalva LVOT gradients were -39 (95% CI: -44.0, -33.2) mmHg and -49 (95% CI: -55.4, -43.0) mmHg, respectively, for mavacamten arm and -6 (95% CI: -10.5, -0.5) mmHg and -12 (95% CI: -17.6, -6.6) mmHg, respectively, for the placebo arm. At week 38, following 8 weeks of mavacamten washout, mean LVEF and LVOT gradients were similar to baseline for both treatment arms.
In HCM, the QT interval may be intrinsically prolonged due to the underlying disease, in association with ventricular pacing, or in association with medicinal products with potential for QT prolongation commonly used in the HCM population. An exposure-response analysis across all clinical studies in HCM patients has shown a concentration-dependent shortening of the QTcF interval with mavacamten. The mean placebo corrected change from baseline in oHCM patients was -8.7 ms (upper and lower limit of the 90% CI -6.7 ms and -10.8 ms, respectively) at the median steady-state Cmax of 452 ng/mL. Patients with longer baseline QTcF intervals tended to display the greatest shortening.
Consistent with nonclinical findings in normal hearts, in one clinical study in healthy subjects sustained exposure to mavacamten at supratherapeutic levels leading to marked depression of systolic function was associated with QTc prolongation (<20 ms). No acute QTc changes have been observed at comparable (or higher) exposures after single doses. The findings in healthy hearts are attributed to an adaptive response to the cardiac mechanical/functional changes (marked mechanical LV depression) occurring in response to myosin inhibition in hearts with normal physiology and LV contractility.
The efficacy of mavacamten was evaluated in a double-blind, randomised, placebo-controlled, parallel-arm, multicentre, international, Phase 3 study enrolling 251 adult patients with NYHA class II and III oHCM, LVEF ≥55%, and LVOT peak gradient ≥50 mmHg at rest or with provocation at time of oHCM diagnosis and Valsalva LVOT gradient ≥30 mmHg at screening. The majority of patients received background HCM treatment for a total of 96% in mavacamten arm (beta blockers 76%, calcium channel blockers 20%) and of 87% in the placebo arm (beta blockers 74%, calcium channel blockers 13%).
Patients were randomised in a 1:1 ratio to receive either a starting dose of 5 mg of mavacamten (123 patients) or matching placebo (128 patients) once daily for 30 weeks. The dose was periodically adjusted to optimise patients' response (decrease in LVOT gradient with Valsalva manoeuvre), maintain LVEF ≥50%, and was also guided by plasma concentrations of mavacamten. Within the dose range of 2.5 mg to 15 mg, a total of 60 patients received 5 mg and 40 patients received 10 mg. During the study, 3 of 7 patients on mavacamten had LVEF <50% prior to the week 30 visit and temporarily interrupted their dose; 2 patients resumed treatment at the same dose and 1 patient had the dose reduced from 10 mg to 5 mg.
Treatment assignment was stratified by baseline NYHA class (II or III), current treatment with beta blockers (yes or no), and type of ergometer (treadmill or exercise bicycle) used for assessment of peak oxygen consumption (pVO2). Patients on background dual treatment with beta blocker and calcium channel blocker treatment or disopyramide or ranolazine were excluded. Patients with known infiltrative or storage disorder causing cardiac hypertrophy that mimicked oHCM, such as Fabry disease, amyloidosis, or Noonan syndrome with LV hypertrophy, were also excluded.
The baseline demographic and disease characteristics were balanced between mavacamten and placebo. The mean age was 59 years, 54% (mavacamten) vs 65% (placebo) were male, mean body mass index (BMI) was 30 kg/m², mean heart rate 63 bpm, mean blood pressure 128/76 mmHg, and 90% were Caucasian. At baseline, approximately 73% of randomised subjects were NYHA class II and 27% were NYHA class III. The mean LVEF was 74%, and the mean Valsalva LVOT was 73 mmHg. 8% had prior septal reduction therapy, 75% were on beta -blockers, 17% were on calcium channel blockers, 14% had history of atrial fibrillation, and 23% with implantable cardioverter defibrillator (23%). In EXPLORER-HCM there were 85 patients aged 65 years or older, 45 patients were dosed with mavacamten.
The primary outcome measure included a change at week 30 in exercise capacity measured by pVO2 and symptoms measured by NYHA functional classification, defined as an improvement of pVO2 by ≥1.5 mL/kg/min and an improvement in NYHA class by at least 1 OR an improvement of pVO2 by ≥3.0 mL/kg/min and no worsening in NYHA class.
A greater proportion of patients treated with mavacamten met the primary and secondary endpoints at week 30 compared to placebo (see table 4).
Table 4. Analysis of the primary composite and secondary endpoints from EXPLORER-HCM study:
| Mavacamten N=123 | Placebo N=128 | |
|---|---|---|
| Patients achieving primary endpoint at week 30, n (%) | 45 (37%) | 22 (17%) |
| Treatment difference (95% CI) | 19.4 (8.67, 30.13) | |
| p-value | 0.0005 | |
| Change from baseline post-exercise LVOT peak gradient at week 30, mmHg | N=123 | N=128 |
| Mean (SD) | -47 (40) | -10 (30) |
| Treatment difference* (95% CI) | -35 (-43, -28) | |
| p-value | <0.0001 | |
| Change from baseline to week 30 in pVO2, mL/kg/min | N=123 | N=128 |
| Mean (SD) | 1.4 (3) | -0.05 (3) |
| Treatment difference* (95% CI) | 1.4 (0.6, 2) | |
| p-value | <0.0006 | |
| Patients with improvement of NYHA class ≥ 1 at week 30 | N=123 | N=128 |
| N, (%) | 80 (65%) | 40 (31%) |
| Treatment difference (95% CI) | 34 (22, 45) | |
| p-value | <0.0001 | |
| Change from baseline to week 30 in KCCQ-23 CSS† | N=92 | N=88 |
| Mean (SD) | 14 (14) | 4 (14) |
| Treatment difference* (95% CI) | 9 (5, 13) | |
| p-value | <0.0001 | |
| Baseline | N=99 | N=97 |
| Mean (SD) | 71 (16) | 71 (19) |
| Change from baseline to week 30 in HCMSQ SoB domain score‡ | N=85 | N=86 |
| Mean (SD) | -2.8 (2.7) | -0.9 (2.4) |
| Treatment difference* (95% CI) | -1.8 (-2.4, -1.2) | |
| p-value | <0.0001 | |
| Baseline | N=108 | N=109 |
| Mean (SD) | 4.9 (2.5) | 4.5 (3.2) |
* Least-squares mean difference
† KCCQ-23 CSS = Kansas City Cardiomyopathy Questionnaire-23 Clinical Summary Score. The KCCQ-23 CSS is derived from the Total Symptoms Score (TSS) and the Physical Limitations (PL) score of the KCCQ-23. The CSS ranges from 0 to 100, with higher scores representing better health status. A significant treatment effect on the KCCQ-23 CSS favouring mavacamten was first observed at week 6 and remained consistent through week 30.
‡ HCMSQ SoB = Hypertrophic Cardiomyopathy Symptom Questionnaire Shortness of Breath. The HCMSQ SoB domain score measures frequency and severity of shortness of breath. The HCMSQ SoB domain score ranges from 0 to 18, with lower scores representing less shortness of breath. A significant treatment effect on the HCMSQ SoB favouring mavacamten was first observed at week 4 and remained consistent through week 30.
A range of demographic characteristics, baseline disease characteristics, and baseline concomitant medicinal products were examined for their influence on outcomes. Results of the primary analysis consistently favoured mavacamten across all subgroups analysed.
The efficacy of mavacamten was evaluated in a Phase 3, double-blind, randomised, 16-week placebo-controlled trial in 112 patients with symptomatic oHCM who were septal reduction therapy (SRT) eligible. Patients with severely symptomatic drug-refractory oHCM, and NYHA class III/IV or class II with exertional syncope or near syncope were included in the study. Patients were required to have LVOT peak gradient ≥50 mmHg at rest or with provocation, and LVEF ≥60%. Patients must have been referred or under active consideration within the past 12 months for SRT and had been actively considering scheduling the procedure.
Patients were randomised 1:1 to receive treatment with mavacamten or placebo once daily. The dose was periodically adjusted within the dose range of 2.5 mg to 15 mg to optimise patient’s response.
The baseline demographic and disease characteristics were balanced between mavacamten and placebo. The mean age was 60.3 years, 51% were male, mean BMI was 31 kg/m², mean heart rate 64 bpm, mean blood pressure 131/74 mmHg, and 89% were Caucasian. At baseline, approximately 7% of randomised subjects were NYHA class II and 92% were NYHA class III. 46% were on beta-blockers monotherapy, 15% were on calcium channel blockers monotherapy, 33% were on a mixed combination of beta -blockers, calcium channel blockers, and 20% were on disopyramide alone or in combination with other treatment. In VALOR-HCM there were 45 patients aged 65 years or older, 24 patients were dosed with mavacamten.
Mavacamten was shown to be superior to placebo in meeting the primary composite endpoint at week 16 (see table 5). The primary endpoint was a composite of
The treatment effects of mavacamten on LVOT obstruction, functional capacity, health status, and cardiac biomarkers were assessed by change from baseline through week 16 in post-exercise LVOT gradient, proportion of patients with improvement in NYHA class, KCCQ-23 CSS, NT-proBNP, and cardiac troponin I. In the VALOR-HCM study, hierarchical testing of secondary efficacy endpoints showed significant improvement in the mavacamten group compared to the placebo group (table 5).
Table 5. Analysis of the primary composite and secondary endpoints from VALOR-HCM study:
| Mavacamten N=56 | Placebo N=56 | |
|---|---|---|
| Patients achieving primary composite endpoint at week 16, n (%) | 10 (17.9) | 43 (76.8) |
| reatment difference (95% CI) | 58.9 (44.0, 73.9) | |
| p-value | <0.0001 | |
| Patient decision to proceed with SRT | 2 (3.6) | 2 (3.6) |
| SRT-eligible based on guideline criteria | 8 (14.3) | 39 (69.6) |
| SRT status not evaluable (imputed as meeting primary endpoint) | 0 (0.0) | 2 (3.6) |
| Change from baseline post-Exercise LVOT peak gradient at week 16, (mmHg) | N=55 | N=53 |
| Mean (SD) | -39.1 (36.5) | -1.8 (28.8) |
| Treatment difference* (95% CI) | -37.2 (-48.1, -26.2) | |
| p-value | <0.0001 | |
| Patients with improvement of NYHA class ≥ 1 at week 16 | N=55 | N=53 |
| N, (%) | 35 (62.5%) | 12 (21.4%) |
| Treatment difference (95% CI) | 41.1 (24.5%, 57.7%) | |
| p-value | <0.0001 | |
| Change from baseline to week 16 in KCCQ-23 CSS† | N=55 | N=53 |
| Mean (SD) | 10.4 (16.1) | 1.8 (12.0) |
| Treatment difference* (95% CI) | 9.5 (4.9, 14.0) | |
| p-value | <0.0001 | |
| Baseline | N=56 | N=56 |
| mean (SD) | 69.5 (16.3) | 65.6 (19.9) |
| Change from baseline to week 16 in NT-proBNP | N=55 | N=53 |
| ng/L geometric mean ratio | 0.35 | 1.13 |
| Geometric mean ratio mavacamten/placebo (95% CI) | 0.33 (0.27, 0.42) | |
| p-value | <0.0001 | |
| Change from baseline to week 16 in Cardiac Troponin I | N=55 | N=53 |
| ng/L geometric mean ratio | 0.50 | 1.03 |
| Geometric mean ratio mavacamten/placebo (95% CI) | 0.53 (0.41, 0.70) | |
| p-value | <0.0001 | |
* Least-squares mean difference.
† KCCQ-23 CSS=Kansas City Cardiomyopathy Questionnaire-23 Clinical Summary Score. The KCCQ-23 CSS is derived from the Total Symptoms Score (TSS) and the Physical Limitations (PL) score of the KCCQ-23. The CSS ranges from 0 to 100, with higher scores representing better health status.
In the VALOR-HCM study, secondary endpoint of NT-proBNP, at week 16 (see table 5) showed a sustained reduction from baseline after mavacamten treatment compared to placebo that was similar to that seen in EXPLORER-HCM at week 30. Exploratory analysis of left ventricular mass index (LVMI) and left atrial volume index (LAVI) showed reductions in the mavacamten treated patients compared to placebo in EXPLORER-HCM and VALOR-HCM.
The European Medicines Agency has deferred the obligation to submit the results of studies with CAMZYOS in one or more subsets of the paediatric population in treatment of HCM (see section 4.2 for information on paediatric use).
Mavacamten is readily absorbed with a median tmax of 1 hour (range: 0.5 to 3 hours) after oral administration with an estimated oral bioavailability of approximately 85% within the clinical dose range. The increase in mavacamten exposure is generally dose proportional after once daily doses of mavacamten (2 mg to 48 mg).
After a single dose of 15 mg mavacamten, Cmax and AUCinf are 47% and 241% higher, respectively, in CYP2C19 poor metabolisers compared to normal metabolisers. Mean half-life is prolonged in CYP2C19 poor metabolisers compared to normal metabolisers (23 days versus 6 to 9 days, respectively).
Inter-subject PK variability is moderate, with a coefficient of variation for exposure of approximately 30-50% for Cmax and AUC.
A high fat, high calorie meal delayed absorption resulting in a median tmax of 4 hours (range: 0.5 to 8 hours) in the fed state compared to 1 h in the fasted state. Administration with meals resulted in a 12% decrease in AUC0-inf, however this decrease is not considered clinically significant. Mavacamten may be administered with or without meals.
As mavacamten is titrated based on clinical response (see section 4.2), simulated steady state exposures are summarized using individualised dosage by phenotype (table 6).
Table 6. Simulated average steady state concentration by dose and CYP2C19 phenotype in patients titrated to effect based on Valsalva LVOT and LVEF:
| Dose | Median concentration (ng/ml) | ||||
|---|---|---|---|---|---|
| Poor metabolisers | Intermediate metabolisers | Normal metabolisers | Rapid metabolisers | Ultra-rapid metabolisers | |
| 2.5 mg | 451.9 | 274.0 | 204.9 | 211.3 | 188.3 |
| 5 mg | 664.9 | 397.8 | 295.4 | 311.5 | 300.5 |
Plasma protein binding of mavacamten is 97-98% in clinical studies. The blood-to-plasma concentration ratio is 0.79. The apparent volume of distribution (Vd/F) ranged from 114 L to 206 L. Specific studies to assess distribution of mavacamten have not been conducted in humans, however data are consistent with a high volume of distribution.
Based upon 10 male subjects dosed for up to 28 days, the amount of mavacamten distributed to the semen was considered to be low.
Mavacamten is extensively metabolised, primarily through CYP2C19 (74%), CYP3A4 (18%), and CYP2C9 (7.6%) based on in-vitro reaction phenotyping. Metabolism is expected to be driven through all three pathways, and primarily through CYP2C19 in CYP2C19 intermediate, normal, rapid and ultra-rapid metabolisers. Three metabolites have been detected in human plasma. The exposure of the most abundant metabolite MYK-1078 in human plasma was less than 4% of the exposure of mavacamten, and the other two metabolites had exposures less than 3% of the exposure of mavacamten indicating these would have minimal to no impact on the overall activity of mavacamten. In CYP2C19 poor metabolisers mavacamten is metabolised primarily by CYP3A4. No data are available on the metabolite profile in CYP2C19 poor metabolisers.
Based on pre-clinical data, for a dose up to 5 mg in CYP2C19 poor metabolisers and for a dose up to 15 mg in CYP2C19 intermediate to ultra-rapid metabolisers, mavacamten is not an inhibitor of CYP 1A2, 2B6, 2C8, 2D6, 2C9, 2C19, or 3A4 at clinically relevant concentrations.
In vitro data indicate that mavacamten is not an inhibitor of major efflux transporters (P-gp, BCRP, BSEP, MATE1, or MATE2-K) or major uptake transporters (organic anion transporting polypeptides [OATPs], organic cation transporters [OCTs], or organic anion transporters [OATs]) at therapeutic concentrations for a dose up to 5 mg in CYP2C19 poor metabolisers and for a dose up to 15 mg in CYP2C19 intermediate to ultra-rapid metabolisers.
Mavacamten is cleared from plasma primarily by metabolism through cytochrome P450 enzymes. Terminal half-life is 6 to 9 days in CYP2C19 normal metabolisers and 23 days for CYP2C19 poor metabolisers. Half-life is estimated to be, 6 days for CYP2C19 ultra-rapid metabolisers, 8 days for CYP2C19 rapid metabolisers, and 10 days for CYP2C19 intermediate metabolisers.
Drug accumulation occurs with an accumulation ratio about 2-fold for Cmax and about 7-fold for AUC in CYP2C19 normal metabolisers. The accumulation depends on the metabolism status for CYP2C19 with the largest accumulation observed in CYP2C19 poor metabolisers. At steady-state, the peak-to-trough plasma concentration ratio with once daily dosing is approximately 1.5.
Following a single 25 mg dose of 14C labelled mavacamten in CYP2C19 normal metabolisers, 7% and 85% of the total radioactivity was recovered in the faeces and urine of CYP2C19 normal metabolisers, respectively. Unchanged active substance accounted for approximately 1% and 3% of the administered dose in the faeces and urine, respectively.
Polymorphic CYP2C19 is the main enzyme involved in the metabolism of mavacamten. An individual carrying two normal function alleles is a CYP2C19 normal metaboliser (e.g., *1/*1). An individual carrying two non-functional alleles is a CYP2C19 poor metaboliser (e.g., *2/*2, *2/*3, *3/*3). The incidence of CYP2C19 poor metaboliser phenotype ranges from approximately 2% in Caucasian to 18% in Asian populations.
Exposure to mavacamten increased approximately dose proportionally between 2 mg and 48 mg and is expected to result in dose proportional exposure increase across the therapeutic range of 2.5 mg to 5 mg in CYP2C19 poor metabolisers and 2.5 mg to 15 mg in CYP2C19 intermediate to ultra-rapid metabolisers.
No clinically significant differences in the PK of mavacamten were observed using population PK modelling based on age, sex, race or ethnicity.
A single dose PK study was conducted in patients with mild (Child-Pugh class A) or moderate (Child-Pugh class B) hepatic impairment, as well as a control group with normal hepatic function. Mavacamten exposures (AUC) increased 3.2-fold and 1.8-fold in patients with mild and moderate impairment, respectively, compared to patients with normal hepatic function. There was no effect of hepatic function on Cmax, consistent with no change in the rate of absorption and/or volume of distribution. The amount of mavacamten excreted in urine in all 3 studied groups was 3%. A dedicated PK study has not been conducted in patients with severe (Child-Pugh class C) hepatic impairment.
Approximately 3% of a mavacamten dose is excreted in the urine as parent substance. A population PK analysis, which comprised eGFR down to 29.5 mL/min/1.73m², demonstrated no correlation between renal function and exposure. A dedicated PK study has not been conducted in patients with severe renal impairment (eGFR <30 mL/min/1.73m²).
Non-clinical data reveal no special hazard for humans based on conventional studies of safety pharmacology, repeated dose toxicity, genotoxicity and carcinogenic potential. Toxicology findings were generally related to adverse reductions on cardiac function consistent with exaggerated primary pharmacology in healthy animals. These effects occurred at clinically relevant exposures.
In reproductive toxicity studies, there was no evidence of effects of mavacamten on mating and fertility in male or female rats or in the viability and fertility of offspring of dams at any dose tested. However, plasma exposures (AUC) of mavacamten at the highest doses tested were less than in humans at the maximum recommended human dose (MRHD).
Mavacamten adversely affected embryo-foetal development in rats and rabbits. When mavacamten was administered orally to pregnant rats during the period of organogenesis, decreased mean foetal body weight, increases in post implantation loss, and foetal malformations (visceral and skeletal) were observed at clinically relevant exposures. Visceral malformations involved heart malformation in foetuses, including one total situs inversus, while skeletal malformations were manifested mostly as increased incidences of fused sternebrae.
When mavacamten was administered orally to pregnant rabbits during the period of organogenesis, visceral and skeletal malformations were noted, consisting of malformations of the great vessels (dilatation of pulmonary trunk and/or aortic arch), cleft palate and higher incidences of fused sternebrae. Maternal plasma exposure levels (AUC) at the no effect dose level for embryo-foetal development in both species were less than those in humans at the MRHD.
In a pre- and post-natal development study, administration of mavacamten to pregnant rats from gestation day 6 to lactation/post-partum day 20 did not result in adverse effects in the dams or offspring exposed daily from before birth (in utero) through lactation. The maternal exposure was less than the MRHD. No information is available on the excretion of mavacamten in animal milk.
© All content on this website, including data entry, data processing, decision support tools, "RxReasoner" logo and graphics, is the intellectual property of RxReasoner and is protected by copyright laws. Unauthorized reproduction or distribution of any part of this content without explicit written permission from RxReasoner is strictly prohibited. Any third-party content used on this site is acknowledged and utilized under fair use principles.
