| Characteristic | N = 351 |
|---|---|
| Age | 61 (11) |
| Sex | |
| female | 12 (34%) |
| male | 23 (66%) |
| Ethnicity | |
| Asian or Asian British | 8 (23%) |
| Black, Black British, Caribbean or African | 1 (2.9%) |
| Mixed or multiple ethnic groups | 1 (2.9%) |
| White | 22 (63%) |
| Any other ethnic group | 3 (8.6%) |
| Hypertension | 10 (29%) |
| Diabetes Type 2 | 3 (8.6%) |
| Dyslipidaemia | 3 (8.6%) |
| Smoking | 0 (0%) |
| Atrial Fibrillation | 6 (17%) |
| Stroke | 1 (2.9%) |
| Family History of SCD | 3 (9.1%) |
| Family History of Cardiomyopathy | 0 (0%) |
| Gene Positive | 11 (31%) |
| Gene Classification | |
| non pathogenic | 0 (0%) |
| VUS | 4 (36%) |
| pathogenic | 5 (45%) |
| likely pathogenic | 2 (18%) |
| likely non-pathogenic | 0 (0%) |
| not_applicable | 0 (0%) |
| missing | 0 (0%) |
| Variant | |
| 24 (69%) | |
| CSPR3 | 1 (2.9%) |
| MYBPC3 | 5 (14%) |
| MYH7 | 4 (11%) |
| TNNI3 | 1 (2.9%) |
| NSVT | 7 (22%) |
| CYP450 Metaboliser | |
| poor | 1 (2.9%) |
| intermediate | 8 (23%) |
| normal | 13 (37%) |
| rapid | 11 (31%) |
| ultrarapid | 2 (5.7%) |
| Bisoprolol | 22 (63%) |
| Verapamil | 12 (35%) |
| CIED | |
| no | 27 (77%) |
| PPM | 1 (2.9%) |
| ICD_transvenous | 4 (11%) |
| ICD_subcut | 2 (5.7%) |
| CRTP | 0 (0%) |
| CRTD | 0 (0%) |
| ILR | 1 (2.9%) |
| ICD indication | |
| prim_prevention | 6 (100%) |
| sec_prevention | 0 (0%) |
| Haemoglobin | 141 (17) |
| Haematocrit | 0.42 (0.05) |
| White Cell Count | 7.70 (2.37) |
| creat | 86 (19) |
| eGFR_epi_2021 | 82 (16) |
| Urea | 6.07 (1.67) |
| NT-proBNP | 1,277 (1,397) |
| hs Troponin T | 15 (8) |
| LVEDV Indexed (trabecular) | 76 (14) |
| LVEF (trabecular) | 78 (7) |
| LV Mass Indexed (trabecular) | 112 (37) |
| rvedv_i | 82 (13) |
| rvef | 62 (11) |
| mapse_avg | 13.18 (2.90) |
| LA Area Indexed | 17.8 (4.7) |
| RA Area Indexed | 11.6 (3.4) |
| Global Native T1 | 1,028 (42) |
| Global T2 | 49 (2) |
| Max LVOT gradient (mmHg) | 102 (38) |
| 1 Mean (SD); n (%) | |
Reverse Remodelling in Hypertrophic Cardiomyopathy with Mavacamten
Kostas Moschonas
Contents
- Background
- LV mass regression
- Topology of regional regression in wall thickness
- Tissue composition: ECV, cellular and matrix volumes
- Drivers of regional regression in wall thickness
Mavacamten MoA
\(\uparrow\) probability of \(SRX\) state
\(\downarrow\) rate of \(P_i\) release \(\rightarrow\) \(\downarrow\) ATPase activity
Nat Rev Cardiol 2022; 19
Clinical Effect
In obstructive HCM
- Improves symptoms and increases \(pVO_2\)
- \(\downarrow\) NT-proBNP, LV mass, LAVi
In non-obstructive HCM
- Does not improve symptoms or \(pVO_2\)
- \(\downarrow\) NT-proBNP, LV mass, LAVi
Lancet 2020; 396
NEJM 2025; 393
HCM is a regional disease
Study Design
Adults with HCM and:
- NYHA II or III
- \(Max\,LVOT\,gradient\geq 50 mmHg\), at rest or provocation
- \(LVEF \geq 55\%\)
Mavacamten started at \(2.5\,mg\)
Up-titrated according to LVOT gradient and symptoms.
Clinical follow-up monthly for first 3-4 months.
CMRs at baseline, 2 and 6 months.
Flow diagram
Baseline characteristics
Overall effect
NYHA Class
LVOT gradient
| Baseline | 2 Months | 6 Months | |
|---|---|---|---|
| LVOT grad (mmHg) | 102 | 69 | 50 |
| % change | 32% | 51% | |
| p<0.001 | p<0.001 |
NT-proBNP
| Baseline | 2 Months | 6 Months | |
|---|---|---|---|
| NT-proBNP (pg/mL) | 758 | 344 | 180 |
| % change | 55% | 76% | |
| p<0.001 | p<0.001 |
Mavacamten dose
| Dose at 6 months (mg) | N (%) |
|---|---|
| 2.5 | 10 (31%) |
| 5 | 16 (50%) |
| 10 | 6 (19%) |
LV Mass
LV Mass Regression
| Baseline | 2 Months | 6 Months | |
|---|---|---|---|
| LVMi (g/m2) | 113 | 103 | 96 |
| % change | -8.8% | -15% | |
| p<0.001 | p<0.001 |
Heterogeneous response
Response not fully captured by maximal wall thickness
| Baseline | 2 Months | 6 Months | |
|---|---|---|---|
| MWT (mm) | 19 | 18 | 17.1 |
| p<0.001 | p<0.001 | ||
| LVEDVi (mL/m2) | 76 | 76 | 75.1 |
Topology of hypertrophy regression
Baseline regional hypertrophy
Regional regression
Regional regression at 2 months
| Segment | Baseline_Z | Post_Z | Delta_Z | CI | P_adj | Sig | N_Hyper_Base | N_Norm | Pct_Norm |
|---|---|---|---|---|---|---|---|---|---|
| Basal Ant | 3.60 | 3.08 | -0.52 | (-0.95, -0.10) | 0.001 | *** | 21 | 7 | 33% |
| Basal AntSep | 4.27 | 3.61 | -0.66 | (-1.02, -0.31) | 0.000 | *** | 22 | 6 | 27% |
| Basal InfSep | 4.47 | 3.87 | -0.60 | (-0.98, -0.22) | 0.000 | *** | 22 | 3 | 14% |
| Basal Inf | 5.39 | 4.46 | -0.93 | (-1.51, -0.35) | 0.000 | *** | 26 | 7 | 27% |
| Basal InfLat | 2.16 | 1.07 | -1.09 | (-1.80, -0.39) | 0.001 | ** | 9 | 6 | 67% |
| Basal AntLat | 3.34 | 2.68 | -0.66 | (-1.17, -0.16) | 0.000 | *** | 17 | 7 | 41% |
| Mid Ant | 4.36 | 3.79 | -0.57 | (-1.16, 0.02) | 0.000 | *** | 20 | 4 | 20% |
| Mid AntSep | 4.08 | 3.43 | -0.65 | (-1.11, -0.19) | 0.000 | *** | 18 | 5 | 28% |
| Mid InfSep | 4.87 | 4.23 | -0.64 | (-1.12, -0.16) | 0.000 | *** | 22 | 3 | 14% |
| Mid Inf | 5.31 | 4.38 | -0.93 | (-1.51, -0.35) | 0.000 | *** | 26 | 6 | 23% |
| Mid InfLat | 3.23 | 1.87 | -1.36 | (-1.99, -0.72) | 0.000 | *** | 14 | 7 | 50% |
| Mid AntLat | 3.98 | 3.20 | -0.78 | (-1.37, -0.20) | 0.000 | *** | 18 | 4 | 22% |
| Apical Ant | 3.49 | 3.19 | -0.30 | (-0.84, 0.24) | 0.001 | *** | 16 | 3 | 19% |
| Apical Sep | 3.14 | 2.51 | -0.63 | (-1.29, 0.03) | 0.003 | ** | 13 | 3 | 23% |
| Apical Inf | 3.37 | 2.48 | -0.89 | (-1.60, -0.18) | 0.000 | *** | 15 | 3 | 20% |
| Apical Lat | 3.02 | 2.77 | -0.25 | (-0.79, 0.29) | 0.001 | *** | 13 | 2 | 15% |
Regional regression at 6 months
| Segment | Baseline_Z | Post_Z | Delta_Z | CI | P_adj | Sig | N_Hyper_Base | N_Norm | Pct_Norm |
|---|---|---|---|---|---|---|---|---|---|
| Basal Ant | 3.60 | 2.83 | -0.77 | (-1.21, -0.33) | 0.001 | *** | 20 | 7 | 35% |
| Basal AntSep | 4.27 | 3.23 | -1.04 | (-1.42, -0.67) | 0.000 | *** | 21 | 6 | 29% |
| Basal InfSep | 4.47 | 3.31 | -1.16 | (-1.56, -0.76) | 0.000 | *** | 20 | 6 | 30% |
| Basal Inf | 5.39 | 3.61 | -1.78 | (-2.39, -1.18) | 0.000 | *** | 22 | 7 | 32% |
| Basal InfLat | 2.16 | 0.96 | -1.20 | (-1.94, -0.46) | 0.001 | ** | 9 | 6 | 67% |
| Basal AntLat | 3.34 | 2.34 | -1.00 | (-1.53, -0.48) | 0.000 | *** | 17 | 8 | 47% |
| Mid Ant | 4.36 | 3.10 | -1.26 | (-1.88, -0.64) | 0.000 | *** | 19 | 6 | 32% |
| Mid AntSep | 4.08 | 2.91 | -1.18 | (-1.65, -0.70) | 0.000 | *** | 17 | 4 | 24% |
| Mid InfSep | 4.87 | 3.55 | -1.33 | (-1.83, -0.82) | 0.000 | *** | 20 | 5 | 25% |
| Mid Inf | 5.31 | 3.61 | -1.71 | (-2.31, -1.10) | 0.000 | *** | 23 | 8 | 35% |
| Mid InfLat | 3.23 | 1.67 | -1.56 | (-2.23, -0.90) | 0.000 | *** | 14 | 8 | 57% |
| Mid AntLat | 3.98 | 2.67 | -1.32 | (-1.93, -0.70) | 0.000 | *** | 16 | 5 | 31% |
| Apical Ant | 3.49 | 2.52 | -0.97 | (-1.55, -0.40) | 0.001 | *** | 13 | 4 | 31% |
| Apical Sep | 3.14 | 2.13 | -1.01 | (-1.70, -0.31) | 0.003 | ** | 11 | 5 | 45% |
| Apical Inf | 3.37 | 2.02 | -1.35 | (-2.10, -0.60) | 0.000 | *** | 12 | 6 | 50% |
| Apical Lat | 3.02 | 2.05 | -0.97 | (-1.54, -0.40) | 0.001 | *** | 11 | 5 | 45% |
Do non-hypertrophied segments regress?
Defined as baseline z-score < 3
Compared to hypertrophied segments
Defined as baseline z-score >3
Tissue Composition
Conventional tissue characterisation measures
| Baseline | 6 Months | |
|---|---|---|
| T1 (ms) | 1028 | 1021 (NS) |
| T2 (ms) | 49 | 49.3 (NS) |
| LGE (g) | 12.8 | 10.7 (NS) |
| LGE (% of LVM) | 11.6 | 11.4 (NS) |
Regional baseline LGE
ECV fraction
| Baseline | 6 Months | |
|---|---|---|
| ECV (%) | 28.8 | 28.6 (NS) |
No difference across segments
Cellular & Matrix Volume
\[ \begin{align} \text{Cellular Vol.} &= \frac{\text{LVM}_i}{1.055} \times (1 - \text{ECV}) \\[1em] \text{Matrix Vol.} &= \frac{\text{LVM}_i}{1.055} \times \text{ECV} \end{align} \]
Regression of Cellular & Matrix Volume
| Baseline | 6 Months | p-value | |
|---|---|---|---|
| Cellular volume | 30.9 | 26.7 | <0.001 |
| Matrix volume | 12.6 | 11 | <0.001 |
What drives hypertrophy regression?
Do segments with a lot of LGE regress?
Here defined as > 50% enhanced area
| record_id | segment_name | lge_pea | wt_mm_baseline | wt_mm_cmr2 | delta_z |
|---|---|---|---|---|---|
| HH02 | Mid Anterior | 50.4 | 16.8 | 14.0 | -2.7 |
| HH03 | Basal Anteroseptum | 50.1 | 18.0 | 18.2 | 0.1 |
| MHCM06 | Basal Inferior | 62.8 | 19.8 | 16.8 | -2.6 |
| MHCM06 | Basal Anterior | 57.1 | 14.9 | 14.6 | -0.2 |
| MHCM14 | Mid Anteroseptum | 59.7 | 29.5 | 21.2 | -5.2 |
| MHCM14 | Mid Inferoseptum | 52.7 | 30.1 | 20.7 | -6.2 |
| MHCM23 | Mid Inferolateral | 64.2 | 20.1 | 15.1 | -5.5 |
| MHCM23 | Mid Anterior | 57.6 | 22.0 | 20.1 | -1.7 |
| MHCM23 | Mid Anterolateral | 50.2 | 20.1 | 18.5 | -1.7 |
Mean delta z-score: -2.85
95% CI -5.01 to -0.7
P = 0.0138
What drives reverse remodelling?
Linear Mixed Effects Model:
Random effects: intercept and slope for baseline z-score.
Conclusion
Thank you
Supplemental
Regional regression of Cell Volume
| Segment | Baseline | CMR2 | Delta | Lower | Upper | P_Val | Sig |
|---|---|---|---|---|---|---|---|
| 1 | 30.64979 | 26.10763 | -4.542154 | -6.553534 | -2.530775 | 1.269333e-04 | *** |
| 2 | 30.38893 | 25.82103 | -4.567905 | -6.486260 | -2.649550 | 7.022208e-05 | *** |
| 3 | 31.16680 | 26.71115 | -4.455651 | -6.405322 | -2.505979 | 1.120028e-04 | *** |
| 4 | 30.61707 | 26.80265 | -3.814417 | -5.917493 | -1.711341 | 1.136745e-03 | * |
| 5 | 31.12135 | 26.97990 | -4.141457 | -6.329761 | -1.953153 | 7.720340e-04 | *** |
| 6 | 31.72916 | 27.04550 | -4.683663 | -6.728937 | -2.638389 | 1.093029e-04 | *** |
| 7 | 30.83176 | 25.93268 | -4.899080 | -7.038289 | -2.759872 | 1.222145e-04 | *** |
| 8 | 30.67890 | 26.08010 | -4.598801 | -6.724719 | -2.472883 | 2.237167e-04 | *** |
| 9 | 30.85193 | 26.11540 | -4.736530 | -6.718763 | -2.754297 | 7.810059e-05 | *** |
| 10 | 31.22928 | 26.58548 | -4.643799 | -6.773320 | -2.514279 | 2.097054e-04 | *** |
| 11 | 31.65179 | 26.75740 | -4.894381 | -6.920580 | -2.868183 | 7.065887e-05 | *** |
| 12 | 31.54261 | 26.31388 | -5.228734 | -7.199154 | -3.258313 | 2.322199e-05 | *** |
| 13 | 31.11877 | 26.03088 | -5.087894 | -7.046260 | -3.129528 | 2.449826e-05 | *** |
| 14 | 30.93117 | 26.60223 | -4.328942 | -6.042774 | -2.615111 | 3.536829e-05 | *** |
| 15 | 31.24273 | 26.29616 | -4.946578 | -6.963845 | -2.929310 | 4.976545e-05 | *** |
| 16 | 30.96388 | 26.34040 | -4.623482 | -6.522039 | -2.724925 | 5.412365e-05 | *** |
| Contrast | Est. | SE | P-val |
|---|---|---|---|
| Septum - Inferior | −0.308 | 0.180 | 0.318 |
| Septum - Anterior | 0.193 | 0.180 | 0.706 |
| Septum - Lateral | −0.447 | 0.156 | 0.022 |
| Inferior - Anterior | 0.501 | 0.201 | 0.063 |
| Inferior - Lateral | −0.139 | 0.180 | 0.866 |
| Anterior - Lateral | −0.640 | 0.180 | 0.002 |
Regional regression of Matrix Volume
| Segment | Baseline | CMR2 | Delta | Lower | Upper | P_Val | Sig |
|---|---|---|---|---|---|---|---|
| 1 | 13.14903 | 11.46544 | -1.683589 | -2.484649 | -0.88252891 | 2.805517e-04 | *** |
| 2 | 13.40988 | 11.75837 | -1.651512 | -2.565334 | -0.73769096 | 1.183321e-03 | * |
| 3 | 12.63202 | 10.86025 | -1.771763 | -2.541411 | -1.00211365 | 1.051335e-04 | *** |
| 4 | 13.18175 | 10.75813 | -2.423614 | -3.132993 | -1.71423460 | 6.192978e-07 | *** |
| 5 | 12.67746 | 10.59032 | -2.087136 | -2.928832 | -1.24544048 | 4.335353e-05 | *** |
| 6 | 12.06966 | 10.53041 | -1.539249 | -2.228000 | -0.85049811 | 1.460265e-04 | *** |
| 7 | 12.94800 | 11.35170 | -1.596299 | -2.417062 | -0.77553632 | 6.705049e-04 | *** |
| 8 | 13.11106 | 11.17724 | -1.933823 | -2.725103 | -1.14254332 | 6.527289e-05 | *** |
| 9 | 12.93566 | 11.16203 | -1.773631 | -2.668355 | -0.87890659 | 5.554829e-04 | *** |
| 10 | 12.54999 | 10.70350 | -1.846485 | -2.521882 | -1.17108761 | 1.779285e-05 | *** |
| 11 | 12.13676 | 10.53019 | -1.606571 | -2.463538 | -0.74960431 | 9.114019e-04 | *** |
| 12 | 12.24630 | 10.96921 | -1.277089 | -2.237164 | -0.31701328 | 1.183959e-02 | * |
| 13 | 12.68004 | 11.53972 | -1.140319 | -2.213922 | -0.06671516 | 3.845356e-02 | * |
| 14 | 12.86765 | 10.98789 | -1.879757 | -2.977944 | -0.78156933 | 1.867079e-03 | * |
| 15 | 12.55608 | 11.28200 | -1.274079 | -2.529050 | -0.01910786 | 4.690668e-02 | * |
| 16 | 12.83493 | 11.24273 | -1.592199 | -2.811444 | -0.37295472 | 1.296667e-02 | * |
| Contrast | Est. | SE | P-val |
|---|---|---|---|
| Septum - Inferior | 0.308 | 0.180 | 0.318 |
| Septum - Anterior | −0.193 | 0.180 | 0.706 |
| Septum - Lateral | 0.447 | 0.156 | 0.022 |
| Inferior - Anterior | −0.501 | 0.201 | 0.063 |
| Inferior - Lateral | 0.139 | 0.180 | 0.866 |
| Anterior - Lateral | 0.640 | 0.180 | 0.002 |
Mass & LVOT at 6 months
Relationship with NT-proBNP
Part 2
Does ECV influence hypertrophy regression?
Independent of reduction in LVOT gradient
NT-proBNP (linear axis)
Why this is not regression to the mean
1/ Random error scales with size
Segments with absolute WT in mm would have regressed more.
2/ ANCOVA
If a septal segment and an inferior segment started at the exact baseline Z-score, would they end up equal?
No, the septum would end-up significantly thicker:
Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: cmr2 ~ baseline + region + (1 | record_id)
Data: df_ancova
REML criterion at convergence: 470.7
Scaled residuals:
Min 1Q Median 3Q Max
-2.43308 -0.56016 -0.04212 0.54723 2.30726
Random effects:
Groups Name Variance Std.Dev.
record_id (Intercept) 1.3011 1.1407
Residual 0.4381 0.6619
Number of obs: 186, groups: record_id, 31
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.26483 0.31443 66.54916 -0.842 0.40265
baseline 0.71283 0.04344 131.86913 16.411 < 2e-16 ***
regionSeptum 0.31733 0.10817 163.73348 2.934 0.00383 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) baseln
baseline -0.710
regionSeptm -0.426 0.3073/ Direct comparison
Compared 2 competing models to predict % change in absolute wall thickness in mm: baseline mm vs baseline z-score.
[1] "AIC for MM Model: 3662.4"[1] "AIC for Z Model: 3586.6"4/ Magnitude of the effect
RTM typically accounts for small corrections, e.g. test-retest variability of 5-10%
We observe very large reductions in z-scores, 2 to 3.
3.2/ Individual model output
Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: pct_change ~ wt_mm_baseline + (1 | record_id)
Data: df_currency
REML criterion at convergence: 3654.4
Scaled residuals:
Min 1Q Median 3Q Max
-3.7134 -0.5232 -0.0301 0.5649 3.2861
Random effects:
Groups Name Variance Std.Dev.
record_id (Intercept) 70.33 8.386
Residual 83.66 9.147
Number of obs: 492, groups: record_id, 31
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -1.8316 2.2838 110.7480 -0.802 0.424
wt_mm_baseline -0.6140 0.1208 489.6540 -5.083 5.3e-07 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr)
wt_mm_basln -0.730Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: pct_change ~ z_score_baseline + (1 | record_id)
Data: df_currency
REML criterion at convergence: 3578.6
Scaled residuals:
Min 1Q Median 3Q Max
-3.4322 -0.5653 -0.0154 0.5687 3.3123
Random effects:
Groups Name Variance Std.Dev.
record_id (Intercept) 76.61 8.753
Residual 70.83 8.416
Number of obs: 492, groups: record_id, 31
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -1.3887 1.8255 43.7965 -0.761 0.451
z_score_baseline -2.3229 0.2206 450.0816 -10.528 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr)
z_scor_bsln -0.464