Long-Term Effect of Tafamidis on Clinical Parameters and Prognostic Predictors in Patients With Transthyretin Amyloid Cardiomyopathy

Abstract

Background: Accurate prediction of prognosis in transthyretin amyloid cardiomyopathy (ATTR-CM) is crucial for optimal treatment selection, including tafamidis, the only approved therapy for ATTR-CM. Although tafamidis has been proven to improve prognosis, the long-term serial changes in comprehensive parameters related to ATTR-CM, including cardiac biomarkers and imaging parameters, under tafamidis remain unknown.

Methods and Results: In this study, we used Cox regression analysis on data from 258 consecutive patients diagnosed with ATTR-CM at Kumamoto University to determine prognostic factors. During clinical follow-up, the serial changes in parameters were compared between tafamidis-treated and tafamidis-naïve patients. An elevated high-sensitivity cardiac troponin T (hs-cTnT) level at baseline was identified as a stronger independent predictor of all-cause death compared with left ventricular ejection fraction (LVEF) and extracellular volume. During follow-up (median: 24.4 months), estimated glomerular filtration rate and LVEF declined significantly with time in both cohorts. Notably, serum hs-cTnT and B-type natriuretic peptide levels were significantly elevated in the tafamidis-naïve cohort compared to baseline, but this increase was prevented by tafamidis treatment.

Conclusions: Of the ATTR-CM-related parameters investigated, an increased hs-cTnT level at baseline was a promising determinant of poor prognosis. Long-term tafamidis treatment prevented a deterioration in cardiac biomarkers, and the measurement of these markers may enable appropriate monitoring of disease progression.

Transthyretin amyloid cardiomyopathy (ATTR-CM) is a progressive and fatal infiltrative cardiomyopathy caused by transthyretin (TTR) amyloid deposition into the myocardium, leading to refractory heart failure, conduction disturbance, and arrhythmia. The number of patients being diagnosed with ATTR-CM is increasing as a result of increased awareness of the condition among physicians following on from the emergence of disease-modifying therapy¹ and establishment of non-invasive diagnostic criteria.^2,3

Currently, tafamidis, a kinetic stabilizer of TTR tetramers, is the only approved disease-modifying agent for ATTR-CM; its effectiveness was established in the Transthyretin Amyloidosis Cardiomyopathy Clinical Trial (ATTR-ACT)¹ and validated in recent studies.^4–6 Tafamidis is very expensive, and its long-term continuation (>18 months) is essential to see any effect on survival outcomes;¹ thus, accurate prediction of long-term prognosis and the subsequent selection of appropriate candidates for tafamidis treatment at the time of diagnosis are imperative. However, there remains a paucity of evidence comparing the utility of prognostic determinants among the recently established ATTR-CM-related parameters, including cardiac imaging,^7,8 echocardiography,^9,10 functional capacity,^7,11,12 and cardiac biomarkers,^13,14 in a large series of patients with ATTR-CM. Further, although the ATTR-ACT and other studies demonstrated a favorable effect of tafamidis on parameters primarily up to 30 months after tafamidis initiation,^15–18 the effect of tafamidis on parameters, including cardiac biomarkers and conventional and speckle tracking echocardiographic data, over a longer time frame is not fully known.

Given the nature of ATTR-CM, disease progression is generally observed, especially in untreated patients. However, the optimal marker to monitor disease progression remains unclear. In 2021, an expert consensus document from the European Society of Cardiology (ESC) was published that suggested the importance of using a multiparametric approach to monitor ATTR-CM disease progression according to the 3 domains, namely: clinical and functional; laboratory; and imaging and electrocardiographic (ECG).¹⁹ The consensus document proposed thresholds for laboratory, echocardiographic, and functional capacity data to define disease progression.¹⁹ However, that consensus document was based on experience, and relevant associations between disease progression, as assessed by the multiparametric approach, and prognosis have not yet been validated using real-world data.

Thus, the aim of the present study aimed was to explore potential predictors of survival among a comprehensive list of ATTR-CM-related parameters in a large cohort of patients with ATTR-CM. In addition, the study evaluated serial changes in parameters after long-term tafamidis use and the prognostic impact of the deterioration of multiple parameters in tafamidis-treated ATTR-CM patients.

Methods

Study Population

Figure 1 shows a flowchart for this retrospective single-center observational study. A total of 328 consecutive patients diagnosed with ATTR-CM at Kumamoto University Hospital between November 2013 and August 2023 were enrolled. Of these, 71 patients were excluded because they met the following exclusion criteria: (1) participated in the ATTR-ACT trial and were assigned to the tafamidis 20 mg group (n=2); (2) participated in a clinical trial of novel ATTR-CM treatment other than tafamidis (n=22); (3) discontinued tafamidis after initiation (n=8); (4) underwent aortic valve replacement after ATTR-CM diagnosis (n=10); and (5) were followed up for <6 months (n=28). Patients who had been administered tafamidis were assigned to the “tafamidis cohort,” and those who had not been administered tafamidis were assigned to the “tafamidis-naïve cohort.”

Figure 1.

Flowchart for this study. ATTR-ACT, Transthyretin Amyloidosis Cardiomyopathy Clinical Trial; ATTR-CM, transthyretin amyloid cardiomyopathy; AVR, aortic valve replacement; BNP, B-type natriuretic peptide; ECG, electrocardiogram; HF, heart failure; hs-cTnT, high-sensitivity cardiac troponin T; NYHA, New York Heart Association; TAVI, transcatheter aortic valve implantation; TTE, transthoracic echocardiography.

In Analysis 1, the remaining 258 patients with ATTR-CM were analyzed (tafamidis cohort, n=151; tafamidis-naïve cohort, n=107). In the entire tafamidis cohort, TTR amyloid deposition was pathologically confirmed, and the absence of TTR gene mutations was confirmed by genetic testing. Patients in the tafamidis-naïve cohort were diagnosed with ATTR-CM based on either non-invasive diagnostic criteria or invasive (biopsy) criteria as per the current Japanese guideline.³ Fifty-five (51%) patients in the tafamidis-naïve cohort underwent genetic testing, which revealed the absence of TTR gene mutation.

Analysis 2 was performed on the tafamidis cohort (n=151) after excluding 46 patients due to missing data at the 1-year follow-up (n=24, including 7 patients who died within 1 year of starting tafamidis), incomplete assessment at the 1-year follow-up visit (n=7), or missing follow-up data after the 1-year follow-up visit (n=15). The association between prognosis and disease progression (see below) was evaluated in the remaining 105 tafamidis-treated patients.

The study protocol conforms to the principles and ethical guidelines outlined in the Declaration of Helsinki and its amendments. The locally appointed Human Ethics Committee of Kumamoto University approved this study’s protocol (Approval no. 1590). The need for informed consent from patients was waived due to the low-risk nature of this retrospective study and the inability to directly obtain consent from all patients. We promoted this study protocol at Kumamoto University Hospital and on our website (http://www.kuadai-junnai.com) and provided patients the opportunity to withdraw from the study.

Clinical Follow-up Protocol

Baseline clinical demographics, past medical history, New York Heart Association (NYHA) class, laboratory and cardiac imaging data, conventional and speckle tracking echocardiography data, right heart catheterization data, and medication use for heart failure were obtained at the time of tafamidis initiation in the tafamidis cohort and at the time of diagnosis in the tafamidis-naïve cohort.

In Analysis 1, follow-up started on the day of tafamidis initiation (tafamidis cohort) or on the day of diagnosis with ATTR-CM (tafamidis-naïve cohort). In 5 patients, tafamidis therapy was not initiated at the time of diagnosis, but was eventually administered. These 5 patients were allocated to the tafamidis-naïve cohort, with the date of tafamidis initiation regarded as the latest follow-up date.

In Analysis 2, clinical follow-up was initiated after the 1-year follow-up visit. The primary outcome was the incidence of all-cause death, and the secondary outcome was the incidence of cardiovascular death and heart failure hospitalization during the follow-up period. These data were collected from medical records and telephone interviews with the patients or family members. Heart failure hospitalization was defined as an unplanned hospitalization due to heart failure, for which heart failure treatment was required.

Definition of Disease Progression

In Analysis 2, as the recent expert consensus has proposed,¹⁹ disease progression at 1 year after tafamidis initiation was assessed based on the following 3 separate domains: clinical and functional; biomarker; and imaging and ECG. Deterioration in the clinical and functional domain was determined based on the occurrence of heart failure-related hospitalization during the first year after tafamidis administration and/or a deterioration in NYHA functional class. Worsening of the biomarker domain was defined as a ≥30% relative increase in either high-sensitivity cardiac troponin T (hs-cTnT) or B-type natriuretic peptide (BNP) concentrations from baseline and/or an advance in the biomarker staging that was previously reported by us using hs-cTnT, BNP, and estimated glomerular filtration rate (eGFR)¹⁴ due to the absence of N-terminal pro BNP (NT-proBNP) measurement, because BNP is more commonly measured in Japan. Deterioration in the imaging and ECG domain was defined as a new-onset conduction disturbance on ECG, an advance in the diastolic dysfunction grade,²⁰ and/or the following changes in echocardiographic parameters: an increase in interventricular septal thickness in diastole (IVSd) ≥2 mm, an increase in left ventricular global longitudinal strain (LV-GLS) ≥1%, a decrease in left ventricular ejection fraction (LVEF) ≥5%, and/or a decrease in stroke volume ≥5 mL. A new-onset conduction disturbance was defined as a new-onset complete right or left bundle branch block or intraventricular conduction disturbance with a QRS duration >120 ms. Consequently, we calculated the disease progression score by adding 1 point for each domain criterion fulfilled (total score 0–3 points).

Data Correction Methods

Serum hs-cTnT concentrations were measured using the Elecsys 2010 Troponin T HS kit (Roche Diagnostics, Indianapolis, IN, USA). Plasma BNP concentrations were measured using the MI02 Shionogi BNP (Abbott Japan, Matsudo, Japan). Conventional echocardiography was performed using an Epiq 7G (Philips, Bothell, WA, USA). Left ventricular chamber size, wall thickness, left atrial volume, the early transmitral velocity to mitral annular early diastolic velocity ratio (E/e′), and LVEF, calculated by the modified Simpson’s method, were measured. Two-dimensional strain analysis based on speckle tracking echocardiography was performed by an operator who was blinded to the clinical data using a post-processing software program (2D Strain Analysis; TOMTEC Imaging Systems, Unterschleissheim, Germany) as per the previously reported protocols.^10,21,22

^{99 m}Tc-labelled pyrophosphate scintigraphy was performed using a GE Discovery 670 dual-head single-photon emission computed tomography camera with low-energy, high-resolution collimators (GE Healthcare, Waukesha, WI, USA). Three hours after radiotracer administration, the cardiac uptake of ^{99 m}Tc-labelled pyrophosphate was evaluated by using both planar and single-photon emission computed tomography images. The heart-to-contralateral ratio was quantitatively calculated by the total counts in a region of interest over the heart to background counts in a similarly sized region of interest over the contralateral chest. Cardiac magnetic resonance (CMR) imaging data were corrected using a 3.0-T magnetic resonance imaging scanner (Ingenia CX, R5.4; Philips Healthcare, Best, Netherlands). T₁ mapping in a mid-ventricular short-axis section of the left ventricle was performed using the modified Look-Locker inversion recovery sequence to analyze the native T₁ value (normal value in our institution, 1,200–1,260 ms). Extracellular volume (ECV) was calculated using ECV maps produced from analyzing pre- and post-contrast images with a post-processing workstation (Ziostation2; Ziosoft, Tokyo, Japan), as described previously.²³

Right heart catheterization was performed using a 6-Fr balloon-tipped Swan-Ganz catheter (Edwards Lifesciences, Irvine, CA, USA) in a clinically stable condition at the time of ATTR-CM diagnosis. Under fluoroscopy after calibration with the zero-level set at the mid-thoracic line, pulmonary capillary wedge pressure (PCWP) and pulmonary arterial pressure were recorded. A blood sample was collected from the pulmonary artery to evaluate mixed venous oxygen saturation. The cardiac index was measured using thermodilution and the Fick principle.

Statistical Analyses

Normally distributed continuous variables are presented as the mean ± SD and were compared using Students t-test. Non-normally distributed continuous variables are presented as the median with interquartile range (IQR) and were compared using the Wilcoxon signed-rank test. Categorical data are presented as frequencies and percentages and were compared using a Chi-squared test. The significance of differences in multiple repeated clinical values between baseline and after diagnosis in the tafamidis-naïve cohort or tafamidis administration in the tafamidis cohort were analyzed by one-way repeated measures analysis of variance if the variable was normally distributed and by the Friedman test if the variable was not normally distributed. To compare changes in hs-cTnT concentrations over time between groups, we used a linear mixed-effects model. Time was treated as a categorical variable and included as a fixed effect, along with treatment and their interaction. A random intercept was included for each patient to account for within-patient correlations. Least squares means were estimated for each time point within each treatment group to facilitate comparison. Multiple comparisons were adjusted using the Dunnett method, with baseline as the reference.

Kaplan-Meier survival analysis was used to compare the occurrence of all-cause death during the follow-up period between subgroups. To identify the determinants of all-cause and cardiovascular deaths, Cox regression analysis was conducted to compute hazard ratios (HRs) and 95% confidence intervals (95% CIs). In multivariable analysis, we selected independent covariates from the univariable analysis. To examine potential differential prognostic impacts of hs-cTnT concentrations across subgroups, we calculated the P value for interaction by incorporating interaction terms between hs-cTnT concentrations and subgroup variables into the proportional hazards model. Receiver operating characteristic (ROC) curve analysis was performed to assess the ability of the parameters to predict the appearance of all-cause death and identify the optimal cut-off value for the prediction of all-cause death. To investigate the association between hs-cTnT or BNP concentrations and mortality, we used a Cox proportional hazards model with restricted cubic splines, incorporating 4 knots to allow for the modeling of potential non-linear relationships. In selecting the number of knots for the cubic spline curve, we used the Akaike information criterion to determine the optimal number of knots. In addition, the reference point for the HR was set at each median value of the hs-cTnT and BNP concentration.

Statistical analyses were performed using JMP version 17 (SAS Institute, Cary, NC, USA), SPSS version 26 (IBM Corp., Armonk, NY, USA), and R version 4.0.5 (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was set at a two-tailed P value of <0.05.

Results

Baseline Characteristics and Serial Changes in Parameters After Tafamidis Treatment

Baseline characteristics of the tafamidis cohort, the tafamidis-naïve cohort, and the entire cohort are presented in Table 1. The entire cohort was predominantly male (84%), and the mean age was 78.2±6.0 years. Approximately one-third of patients had a history of prior hospitalization due to heart failure before ATTR-CM diagnosis. Diuretics were prescribed in 72% of the entire cohort. Almost half the tafamidis-naïve cohort was non-invasively diagnosed with ATTR-CM. Compared with the tafamidis-naïve cohort, the tafamidis cohort was significantly younger, had lower frequencies of NYHA Class III and receiving β-blockers, lower levels of BNP, left atrial peak longitudinal strain (LS), and right atrial peak LS, and significantly higher values of the heart-to-contralateral ratio, ECV, and right ventricular global LS. To confirm the generalizability of the cohort, the baseline characteristics between the studied cohort (n=258) and excluded cohort (n=36; excluded due to discontinuation of tafamidis and follow-up period ≤6 months) were compared, revealing no significant difference between the 2 groups (Supplementary Table 1).

Table 1.

Baseline Demographics of Participants

Variables	n, data available	Entire cohort (n=258)	Tafamidis cohort (n=151)	Tafamidis-naїve cohort (n=107)	P value
Age (years)	258	78.2±6.0	76.2±5.4	81.0±5.8	<0.01
Male sex	258	217 (84)	130 (86)	87 (81)	0.30
BMI (kg/m2)	258	22.9±3.1	23.2±3.2	22.5±3.0	0.09
Definite diagnosis	258	203 (79)	151 (100)	52 (49)	<0.01
NYHA Class ≥III	258	102 (40)	47 (31)	55 (51)	<0.01
Prior HF hospitalization	258	93 (36)	48 (32)	45 (42)	0.09
Hypertension	258	145 (56)	78 (52)	67 (63)	0.08
Diabetes	258	59 (23)	36 (24)	23 (21)	0.66
Atrial fibrillation	258	144 (56)	86 (57)	58 (54)	0.66
hs-cTnT (ng/mL)	256	0.054 [0.036–0.083]	0.054 [0.037–0.080]	0.056 [0.034–0.096]	0.68
BNP (pg/mL)	257	225 [114–369]	191 [106–282]	285 [129–452]	<0.01
eGFR (mL/min/1.73 m2)	258	49.8±15.4	50.6±14.0	48.6±17.3	0.30
IVSd (mm)	258	15.3±2.5	15.7±2.4	14.8±2.5	<0.01
LVDd (mm)	258	41.2±6.1	40.6±5.3	42.0±7.0	0.08
LVEF (%)	258	50.8±10.3	50.4±10.2	51.4±10.6	0.46
LAVI (mL/m2)	125	54.5±16.2	54.2±14.7	56.1±19.3	0.55
E/e′	257	17.6±6.4	17.3±5.5	18.0±7.4	0.39
LV-GLS (%)	199	−8.9±2.8	−8.6±2.6	−9.4±2.9	0.06
LA peak LS (%)	62	9.7±5.3	8.9±5.4	13.1±3.0	0.01
RV-GLS (%)	61	−12.4±4.9	−11.6±4.6	−16.0±4.4	<0.01
RA peak LS (%)	61	13.3±8.2	11.7±8.0	20.4±5.2	<0.01
H/CL ratio	216	1.82±0.30	1.89±0.33	1.73±0.23	<0.01
Native T1 (ms)	125	1,429±78	1,425±60	1,439±116	0.40
ECV (%)	109	55.2±14.2	57.0±13.9	49.7±13.7	0.02
PCWP mean (mmHg)	137	16.3±7.0	16.2±7.2	16.8±6.4	0.70
PAP mean (mmHg)	137	24.0±7.9	23.8±7.9	24.8±7.8	0.54
Cardiac index (mL/min/m2)
Thermodilution method	134	2.13±0.45	2.13±0.45	2.16±0.44	0.73
Fick principle	114	2.14±0.54	2.15±0.56	2.09±0.45	0.65
SvO2 (%)	109	66.7±6.4	67.2±6.3	65.1±6.5	0.17
ACEi/ARB	258	105 (41)	62 (41)	43 (40)	0.89
ARNI	258	18 (7)	12 (8)	6 (6)	0.46
β-blocker	258	71 (28)	33 (22)	38 (36)	0.02
Diuretics	258	187 (72)	110 (73)	77 (72)	0.88
Dose of loop diureticsA	258	26.9±16.0	24.5±13.5	30.4±18.5	0.01
MRA	258	104 (40)	67 (44)	37 (35)	0.11
SGLT2 inhibitor	258	36 (14)	30 (20)	6 (6)	<0.01

Unless indicated otherwise, data are given as the mean ± SD, median [interquartile range], or n (%). ^ALoop diuretic dose equivalent to that of furosemide. ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor-neprilysin inhibitor; BMI, body mass index; BNP, B-type natriuretic peptide; ECV, extracellular volume; E/e′, ratio between E-wave velocity and the early diastolic velocity of the septum at the level of the mitral annulus; eGFR, estimated glomerular filtration rate; HF, heart failure; H/CL ratio, heart-to-contralateral ratio; hs-cTnT, high-sensitivity cardiac troponin T; IVSd, interventricular septal thickness in diastole; LA peak LS, left atrial peak longitudinal strain; LAVI, left atrial volume index; LVDd, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; LV-GLS, left ventricular global longitudinal strain; MRA, mineralocorticoid receptor antagonist; NYHA, New York Heart Association; PAP, pulmonary arterial pressure; PCWP, pulmonary capillary wedge pressure; RA peak LS, right atrial peak longitudinal strain; RV-GLS, right ventricular global longitudinal strain; SGLT2, sodium-glucose cotransporter 2; SvO₂, mixed venous oxygen saturation.

Serial changes in parameters after diagnosis in the tafamidis-naïve cohort or after tafamidis treatment in the tafamidis cohort are summarized in Table 2. eGFR and LVEF levels declined significantly with time in both cohorts. In contrast, no significant changes were observed in IVSd, E/e′, or speckle tracking echocardiographic data other than right atrial peak LS in the tafamidis-naïve cohort. Serum hs-cTnT and BNP concentrations were significantly elevated in the tafamidis-naïve cohort. Tafamidis treatment prevented the increase in both BNP and hs-cTnT concentrations. Notably, hs-cTnT concentrations were significantly decreased at 1 year after tafamidis initiation, and this reduction was sustained up to 3 years after tafamidis administration (Figure 2).

Table 2.

Serial Changes in Parameters From Baseline to After Diagnosis or Tafamidis Administration

Tafamidis cohort	n, data available	Baseline	After tafamidis administration			P value
			1 year	2 years	3 years
hs-cTnT (ng/mL)	51	0.047 [0.033–0.076]	0.041 [0.032–0.064]	0.035 [0.030–0.067]	0.038 [0.028–0.068]	<0.01
BNP (pg/mL)	57	208 [109–277]	171 [112–271]	181 [122–284]	188 [96–297]	0.33
eGFR (mL/min/1.73 m2)	56	54.1±13.8	52.2±14.6	49.8±12.8	47.3±13.0	<0.01
IVSd (mm)	57	16.1±2.2	15.8±2.0	16.1±2.2	15.8±2.3	0.30
LVDd (mm)	57	41.1±4.5	41.7±4.4	41.9±5.2	42.1±4.5	0.08
LVEF (%)	57	50.6±10.5	50.8±10.3	48.4±10.0	47.7±10.0	<0.01
LAD (mm)	57	40.7±6.1	41.9±6.4	41.7±6.1	42.7±5.6	0.01
E/e′	57	17.7±6.0	16.9±5.8	16.8±5.9	16.9±6.1	0.55
LV-GLS (%)	49	−8.4±2.8	−8.5±2.9	−8.0±2.7	−8.0±2.8	0.21
LA peak LS	43	8.8±5.3	8.0±3.8	7.5±4.2	6.7±4.3	0.06
RV-GLS (%)	42	−11.4±4.4	−11.5±4.1	−11.2±4.2	−10.4±4.9	0.35
RA peak LS	42	11.7±8.3	11.2±7.6	11.0±8.1	9.3±8.1	0.27
Tafamidis-naїve cohort	n, data available	Baseline	After diagnosis			P value
			1 year	2 years	3 years
hs-cTnT (ng/mL)	9	0.025 [0.021–0.036]	0.025 [0.017–0.044]	0.030 [0.020–0.055]	0.038 [0.024–0.059]	<0.01
BNP (pg/mL)	8	190 [121–308]	164 [88–311]	253 [78–485]	435 [96–564]	<0.01
eGFR (mL/min/1.73 m2)	13	50.0±11.0	48.9±13.4	47.8±14.1	43.2±14.6	<0.01
IVSd (mm)	12	13.8±2.2	13.9±2.3	13.9±2.4	14.7±2.1	0.12
LVDd (mm)	12	43.0±8.6	44.5±6.7	42.9±6.2	41.6±7.1	0.04
LVEF (%)	12	54.5±11.9	56.5±10.4	55.9±10.5	52.1±7.4	0.08
LAD (mm)	12	36.2±5.6	38.3±3.9	38.8±4.6	39.8±5.2	0.12
E/e′	12	15.8±5.2	15.3±5.4	15.5±7.6	15.0±4.2	0.93
LV-GLS (%)	12	−12.1±3.5	−12.4±2.2	−11.2±2.3	−10.6±2.6	0.16
LA peak LS	12	13.1±3.0	14.9±4.8	14.2±4.9	13.3±5.4	0.54
RV-GLS (%)	11	−16.0±4.4	−16.8±3.6	−14.8±4.0	−14.1±3.8	0.22
RA peak LS	11	20.4±5.2	23.2±6.2	19.9±4.7	16.5±5.6	0.02

Unless indicated otherwise, data are given as the mean ± SD or median [interquartile range]. LAD, left atrium diameter. Other abbreviations as in Table 1.

Figure 2.

Least-squares mean differences from baseline to each annual follow-up year of high-sensitivity cardiac troponin T (hs-cTnT) concentrations in the tafamidis and tafamidis-naïve cohorts. Circles show the least-squares mean; error bars indicate 95% confidence intervals. Multiple comparisons of the significance of differences in changes in hs-cTnT concentrations were performed using the Dunnett test.

Factors Predictive of Mortality in Patients With ATTR-CM

In Analysis 1, over a median follow-up period of 24.4 months (IQR 13.3–38.3 months), 81 (31%) patients died from any cause and 65 (25%) patients died due to cardiovascular-related causes. Eighty-four (33%) patients experienced hospitalization for heart failure. The incidence rate of heart failure hospitalization was 16.33 per 100 person-years. Univariable Cox regression analysis revealed that higher age and hs-cTnT, BNP, LV-GLS, mean PCWP, mean pulmonary arterial pressure levels, NYHA Class III, lower eGFR and LVEF levels, and a lack of tafamidis treatment were significantly associated with an increased risk of all-cause death (Table 3). Multivariable analysis adjusting the effect of independent covariates in the univariable analysis identified hs-cTnT at baseline (HR 1.14 [95% CI 1.02–1.28, P=0.02] per 0.01-ng/mL increase) and tafamidis treatment (HR 0.16; 95% CI 0.04–0.61; P<0.01) as independent determinants of all-cause mortality. Similar analysis regarding the predictors of all-cause death in the tafamidis cohort alone also revealed that increased hs-cTnT at baseline predicted poor prognosis, even after adjusting for ECV and PCWP (Supplementary Table 2). In contrast, multivariable Cox regression analysis revealed that the factors significantly predictive of cardiovascular-related deaths were higher age (HR 1.31 [95% CI 1.02–1.95; P=0.03] per 1-year increase) and ECV (HR 1.16 [95% CI 1.03–1.32; P=0.02] per 1% increase; Supplementary Table 3).

Table 3.

Cox Regression Analysis for All-Cause Mortality of Patients With Transthyretin Amyloid Cardiomyopathy

Variables	Univariable			Multivariable
	HR	95% CI	P value	HR	95% CI	P value
Age (per 1-year increase)	1.12	1.08–1.18	<0.01	1.08	0.98–1.19	0.13
Male sex	0.86	0.47–1.55	0.61
NYHA Class ≥III	3.36	2.11–5.36	<0.01	1.01	0.37–2.75	0.99
Atrial fibrillation	1.46	0.93–2.30	0.10
hs-cTnT (per-0.01 ng/mL increase)	1.13	1.10–1.17	<0.01	1.14	1.02–1.28	0.02
BNP (per 10-pg/mL increase)	1.01	1.01–1.02	<0.01	1.01	0.99–1.02	0.08
eGFR (per 1-mL/min/1.73 m2 increase)	0.95	0.94–0.97	<0.01	1.03	0.98–1.07	0.24
IVSd (per 1-mm increase)	1.02	0.93–1.11	0.72
LVDd (per 1-mm increase)	1.00	0.97–1.04	0.91
LVEF (per 1% increase)	0.98	0.96–0.99	0.02	0.96	0.91–1.01	0.12
LV-GLS (per 1% increase)	1.10	1.02–1.20	0.02	0.99	0.82–1.21	0.98
LA peak LS (per 1% increase)	1.07	0.90–1.31	0.45
RV-GLS (per 1% increase)	0.93	0.78–1.12	0.41
RA peak LS (per 1% increase)	1.07	0.98–1.18	0.15
Native T1 (per 100-ms increase)	1.22	0.78–1.81	0.36
ECV (per 1% increase)	1.02	0.99–1.06	0.13
PCWP mean (per 1-mmHg increase)	1.07	1.02–1.12	<0.01	0.94	0.79–1.13	0.52
PAP mean (per 1-mmHg increase)	1.08	1.03–1.12	<0.01	1.07	0.92–1.24	0.39
Cardiac index (per 1-L/min/m2 increase)
Thermodilution method	0.85	0.37–1.88	0.70
Fick principle	1.02	0.47–2.07	0.96
SvO2 (per 1% increase)	0.96	0.90–1.03	0.23
Tafamidis treatment	0.24	0.15–0.40	<0.01	0.16	0.04–0.61	<0.01

CI, confidence interval; HR, hazard ratio. Other abbreviations as in Table 1.

Prognostic Impact of hs-cTnT Levels in ATTR-CM

ROC curve analysis assessing the ability of LVEF, ECV, and hs-cTnT to predict all-cause death revealed that the hs-cTnT level was a superior predictor (area under the curve [AUC] 0.79; 95% CI 0.68–0.89; P<0.01) to ECV (AUC 0.58; 95% CI 0.44–0.72; P=0.29) and LVEF (AUC 0.61; 95% CI 0.49–0.74; P=0.07; Figure 3A). Furthermore, ROC curve analysis revealed that the optimal cut-off hs-cTnT value to predict all-cause mortality was 0.061 ng/mL (sensitivity 73%; specificity 74%). After dividing the entire cohort into subgroups according to an hs-cTnT cut-off value of 0.061 ng/mL, Kaplan-Meier curves demonstrated that the subgroup with hs-cTnT levels ≥0.061 ng/mL had a significantly higher probability of all-cause death than the group with hs-cTnT <0.061 ng/mL (log-rank P<0.01; Figure 3B). Restricted cubic spline curves showed a linear association between hs-cTnT (Figure 3C) and BNP (Figure 3D) concentrations and the occurrence of all-cause death. Subgroup analysis revealed the consistent utility of hs-cTnT concentrations for predicting poor prognosis across subgroups according to age, sex, NYHA class, the presence of prior heart failure hospitalization, tafamidis treatment, eGFR, LVEF, native T₁, and ECV (Figure 4). The prognostic ability of hs-cTnT concentrations was particularly pronounced in the subgroups with female sex, NYHA Class I/II, eGFR above the median, ECV below the median, and tafamidis treatment.

Figure 3.

Predictors of all-cause death in the entire transthyretin amyloid cardiomyopathy (ATTR-CM) cohort. (A) Receiver operating characteristic (ROC) curve analysis assessing the ability of left ventricular ejection fraction (LVEF), extracellular volume (ECV), and high-sensitivity cardiac troponin T (hs-cTnT) to predict all-cause death (n=109). AUC, area under the curve; CI, confidence interval. (B) Kaplan-Meier curves showing the all-cause death-free survival rate in patients with hs-cTnT concentrations ≥0.061 and <0.061 ng/mL. (C,D) Estimated cubic spline curves of the association between baseline hs-cTnT (C) and B-type natriuretic peptide (BNP; D) concentrations and the risk of all-cause death. Blue lines indicate the hazard ratio, and the gray shaded areas indicate 95% CIs.

Figure 4.

Subgroup analysis of the prognostic impact of high-sensitivity cardiac troponin T (hs-cTnT) concentrations for all-cause death. Interaction terms between hs-cTnT concentrations and subgroup variables were included in the proportional hazards model to calculate the P value for interaction, assessing potential differential effects across subgroups. CI, confidence interval; ECV, extracellular volume; eGFR, estimated glomerular filtration rate; HFH, heart failure hospitalization; HR, hazard ratio; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association.

Association Between Multiparameter Deterioration and Prognosis in Tafamidis-Treated Patients

In Analysis 2, disease progression 1 year after tafamidis initiation was assessed by evaluating clinical and functional, biomarker, and imaging and ECG parameters (Figure 5A); the frequency of deterioration in each domain was 20%, 45%, and 67%, respectively. The disease progression score was calculated by adding 1 point as each domain criterion was met (total score 0–3 points), and patients in the tafamidis-treated group were divided into subgroups according to the calculated score: 19%, 40%, 35%, and 6% of patients had a disease progression score of 0, 1, 2, and 3 points, respectively. Importantly, Kaplan-Meier curve analysis demonstrated that the patients had an incremental cumulative risk of all-cause death as the disease progression score increased (log-rank P<0.01; Figure 5B). Moreover, patients meeting multiple domain criteria (i.e., disease progression score 2–3 points) had a significantly higher probability of all-cause death than those with a disease progression score of 0–1 (log-rank P<0.01; Figure 5C). The findings of sensitivity analysis using a threshold defining the worsening of BNP as a ≥40% relative increase in the BNP concentration from baseline, as proposed by the joint scientific statement on natriuretic peptides in heart failure,²⁴ were consistent with those using a threshold of a ≥30% relative increase in the BNP concentration.

Figure 5.

Multiparameter assessment in tafamidis-treated patients with transthyretin amyloid cardiomyopathy (ATTR-CM). (A) Multiparametric criteria for determining the disease progression score at the 1-year follow-up in tafamidis-treated patients with ATTR-CM, and the frequency of deterioration in each domain and disease progression score. BNP, B-type natriuretic peptide; ECG, electrocardiogram; HF, heart failure; hs-cTnT, high-sensitivity cardiac troponin T; IVSd, interventricular septal thickness in diastole; LVEF, left ventricular ejection fraction; LV-GLS, left ventricular global longitudinal strain; NYHA, New York Heart Association; SV, stroke volume. (B) Kaplan-Meier curve analysis showing differences in all-cause death-free survival according to disease progression score. (C) Kaplan-Meier curve analysis showing differences in all-cause death-free survival rates according to disease progression scores of 0–1 or 2–3.

Discussion

This study demonstrated that the hs-cTnT concentration is a promising determinant predictive of mortality that performs better than other parameters related to ATTR-CM. Tafamidis treatment showed consistent effectiveness in improving prognosis even after adjusting for baseline ATTR-CM-related parameters; it persistently lowered the hs-cTnT concentration during treatment. Moreover, deterioration in multiple parameters, as proposed by the ESC expert consensus,¹⁹ was observed in 41% of tafamidis-treated patients despite tafamidis treatment for 1 year, and was significantly associated with poor prognosis.

Accurate prediction of long-term prognosis in ATTR-CM is crucial for appropriate selection of disease-modifying therapy; however, to the best of our knowledge, there is a paucity of evidence that comprehensively compares the prognostic utility of ATTR-CM-related parameters, including those recently established (i.e., ECV, native T₁, and speckle tracking echocardiographic and hemodynamic parameters), in a large series of patients with ATTR-CM. In this study, we found that the baseline hs-cTnT concentration was a promising predictor of mortality compared with other parameters reflective of ATTR-CM. The hs-cTnT concentration is a highly specific and sensitive marker of myocardial damage, and increases in hs-cTnT are representative of ATTR-CM and strongly associated with poor prognosis in patients with ATTR-CM.^4,14,25,26 As we reported previously, hs-cTnT is significantly positively correlated with the amount of amyloid deposition in the myocardium.²⁷ Furthermore, a marked decrease in hs-cTnT concentrations with tafamidis treatment was observed in previous studies,^4,28 as well as in the present tafamidis cohort, but not in our tafamidis-naïve cohort. An unfavorable change in hs-cTnT despite tafamidis treatment had prognostic significance,²⁸ suggesting that hs-cTnT may be a widely applicable and reliable marker for predicting prognosis, regardless of tafamidis administration. Although the underlying mechanism by which tafamidis decreases hs-cTnT concentrations remains unclear, it is speculated that tafamidis suppresses further accumulation of TTR amyloid fibrils in the myocardium,²⁹ resulting in attenuation of ongoing myocardial damage caused by amyloid oligomers. Further, the present study revealed that elevated hs-cTnT concentrations resulted in a higher risk of all-cause death, whereas the ECV value on CMR was an independent predictor of cardiovascular death, but not all-cause death. The former can be influenced by various systemic conditions other than cardiac injury,³⁰ and the latter may more precisely and specifically reflect the amyloid-derived tissue damage in the myocardium.³¹

Although the appropriate identification of disease progression in each individual, regardless of disease stage, is of great importance, a protocol to monitor disease progression has not been well established given the difficulty in arriving at a universal definition of disease progression. This is possibly attributed to the heterogeneity of disease severity at diagnosis, comorbidities, concomitant treatment in each individual, and methods used to assess ATTR-CM-related parameters at different institutions. The recent ESC consensus document¹⁹ proposed a novel definition to monitor disease progression with a multifactorial method and provided clear cut-off values; however, the approach and cut-off values for each parameter are yet to be validated using real-world data. Recent studies identified deterioration in NT-proBNP as a useful marker to stratifying disease progression in ATTR-CM,^32,33 whereas the clinical significance of the deterioration of ATTR-CM-related parameters other than biomarkers remains uncertain. Real-world data in the present study validated the utility of a multiparameter assessment for disease progression proposed by the ESC expert consensus document, and simultaneously highlighted the importance of serial monitoring multiple factors, not just biomarkers, during follow-up in ATTR-CM.

Cardiac imaging represented by CMR and bone scintigraphy is necessary to ideally evaluate the properties of myocardial tissue, but it is difficult to perform these procedures longitudinally multiple times due to the high cost and requirement for multidisciplinary instruments and expertise. Notably, the 3-domain approach we used to assess disease progression, involving functional capacity and cardiac biomarkers, echocardiography, and ECG, can be easily and repeatedly performed without high cost; hence, this follow-up method could be widely acceptable for long-term assessment of ATTR-CM regardless of the size of the hospital. Several novel anti-amyloid compounds are in advanced stages of development and are expected to become available in the near future. The findings of the present study regarding disease progression would be consistent with those for anti-amyloid agents other than tafamidis. After the approval of novel anti-amyloid agents, appropriate identification and stratification of disease progression would enable the identification of patients requiring alternative agents with different mechanisms or the consideration of administering combination therapy.

This study has some limitations. First, this was a single-center retrospective study. Second, baseline and follow-up quantitative data on exercise capacity and quality of life, including the Kansas City Cardiomyopathy Questionnaire and 6-minute walking distance, were not assessed. Third, in Analysis 2, only tafamidis-treated patients with assessment at the 1-year follow-up were analyzed and patients who died within 1 year after starting tafamidis treatment were excluded, which could have resulted in selection bias.

In conclusion, of the ATTR-CM-related parameters, including those recently established, increased hs-cTnT at baseline was a promising determinant of poor prognosis. Long-term tafamidis treatment attenuated the increase in cardiac biomarkers. Deterioration in multiple parameters, as proposed by a recent expert consensus, despite tafamidis treatment for 1 year was significantly associated with worse prognosis, which may enable the appropriate monitoring of ATTR-CM disease progression and optimization of disease-modifying therapies.

Acknowledgments

We would like to thank Editage for editing and reviewing this manuscript for English language.

Sources of Funding

This study was supported by a Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research (Grant no. 23K19595 to N.K.; Grant no. 22K08134 to Y.I.).

Disclosures

Y.I. has received remuneration for lectures from Pfizer Japan Inc., Daiichi Sankyo Company Limited, and Mitsubishi Tanabe Pharma Corporation. S.T. and K.T. have received remuneration for lectures from Pfizer Japan Inc. M.U. has received remuneration for lectures and research funding from Pfizer Japan Inc. and Alnylam. E.Y. and K.T. are members of Circulation Journal’s Editorial Team. The remaining authors declare they have no conflicts of interest.

IRB Information

The study protocol was approved by the Human Ethics Committee of Kumamoto University (Approval no: 1590).

Data Availability

The deidentified participant data will not be shared.

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-24-0733

References

• 1.   Maurer MS, Schwartz JH, Gundapaneni B, Elliott PM, Merlini G, Waddington-Cruz M, et al. Tafamidis treatment for patients with transthyretin amyloid cardiomyopathy. N Engl J Med 2018; 379: 1007–1016, doi:10.1056/NEJMoa1805689.
• 2.   Garcia-Pavia P, Rapezzi C, Adler Y, Arad M, Basso C, Brucato A, et al. Diagnosis and treatment of cardiac amyloidosis: A position statement of the ESC Working Group on Myocardial and Pericardial Diseases. Eur Heart J 2021; 42: 1554–1568, doi:10.1093/eurheartj/ehab072.
• 3.   Kitaoka H, Izumi C, Izumiya Y, Inomata T, Ueda M, Kubo T, et al. JCS 2020 guideline on diagnosis and treatment of cardiac amyloidosis. Circ J 2020; 84: 1610–1671, doi:10.1253/circj.CJ-20-0110.
• 4.   Takashio S, Morioka M, Ishii M, Morikawa K, Hirakawa K, Hanatani S, et al. Clinical characteristics, outcome, and therapeutic effect of tafamidis in wild-type transthyretin amyloid cardiomyopathy. ESC Heart Fail 2023; 10: 2319–2329, doi:10.1002/ehf2.14380.
• 5.   Ghoneem A, Bhatti AW, Khadke S, Mitchell J, Liu J, Zhang K, et al. Real-world efficacy of tafamidis in patients with transthyretin amyloidosis and heart failure. Curr Probl Cardiol 2023; 48: 101667, doi:10.1016/j.cpcardiol.2023.101667.
• 6.   Wang J, Chen H, Tang Z, Zhang J, Xu Y, Wan K, et al. Tafamidis treatment in patients with transthyretin amyloid cardiomyopathy: A systematic review and meta-analysis. EClinicalMedicine 2023; 63: 102172, doi:10.1016/j.eclinm.2023.102172.
• 7.   Lane T, Fontana M, Martinez-Naharro A, Quarta CC, Whelan CJ, Petrie A, et al. Natural history, quality of life, and outcome in cardiac transthyretin amyloidosis. Circulation 2019; 140: 16–26, doi:10.1161/CIRCULATIONAHA.118.038169.
• 8.   Martinez-Naharro A, Kotecha T, Norrington K, Boldrini M, Rezk T, Quarta C, et al. Native T₁ and extracellular volume in transthyretin amyloidosis. JACC Cardiovasc Imaging 2019; 12: 810–819, doi:10.1016/j.jcmg.2018.02.006.
• 9.   Chacko L, Martone R, Bandera F, Lane T, Martinez-Naharro A, Boldrini M, et al. Echocardiographic phenotype and prognosis in transthyretin cardiac amyloidosis. Eur Heart J 2020; 41: 1439–1447, doi:10.1093/eurheartj/ehz905.
• 10.   Usuku H, Takashio S, Yamamoto E, Yamada T, Egashira K, Morioka M, et al. Prognostic value of right ventricular global longitudinal strain in transthyretin amyloid cardiomyopathy. J Cardiol 2022; 80: 56–63, doi:10.1016/j.jjcc.2022.02.010.
• 11.   Cheng RK, Levy WC, Vasbinder A, Teruya S, De Los Santos J, Leedy D, et al. Diuretic dose and NYHA functional class are independent predictors of mortality in patients with transthyretin cardiac amyloidosis. JACC CardioOncol 2020; 2: 414–424, doi:10.1016/j.jaccao.2020.06.007.
• 12.   Hashimoto T, Ikuta K, Yamamoto S, Yoshitake T, Suenaga T, Nakashima S, et al. Right ventricular to pulmonary artery uncoupling is associated with impaired exercise capacity in patients with transthyretin cardiac amyloidosis. Circ J 2025; 89: 31–40, doi:10.1253/circj.CJ-24-0402.
• 13.   Gillmore JD, Damy T, Fontana M, Hutchinson M, Lachmann HJ, Martinez-Naharro A, et al. A new staging system for cardiac transthyretin amyloidosis. Eur Heart J 2018; 39: 2799–2806, doi:10.1093/eurheartj/ehx589.
• 14.   Nakashima N, Takashio S, Morioka M, Nishi M, Yamada T, Hirakawa K, et al. A simple staging system using biomarkers for wild-type transthyretin amyloid cardiomyopathy in Japan. ESC Heart Fail 2022; 9: 1731–1739, doi:10.1002/ehf2.13847.
• 15.   Hanna M, Fine NM, Gundapaneni B, Sultan MB, Witteles RM. Improvements in efficacy measures with tafamidis in the Tafamidis in Transthyretin Cardiomyopathy clinical trial. JACC Adv 2022; 1: 100148, doi:10.1016/j.jacadv.2022.100148.
• 16.   Shah SJ, Fine N, Garcia-Pavia P, Klein AL, Fernandes F, Weissman NJ, et al. Effect of tafamidis on cardiac function in patients with transthyretin amyloid cardiomyopathy: A post hoc analysis of the ATTR-ACT randomized clinical trial. JAMA Cardiol 2024; 9: 25–34, doi:10.1001/jamacardio.2023.4147.
• 17.   Ochi Y, Kubo T, Baba Y, Sugiura K, Miyagawa K, Noguchi T, et al. Early experience of tafamidis treatment in Japanese patients with wild-type transthyretin cardiac amyloidosis from the Kochi Amyloidosis cohort. Circ J 2022; 86: 1121–1128, doi:10.1253/circj.CJ-21-0965.
• 18.   Ichikawa Y, Oota E, Odajima S, Kintsu M, Todo S, Takeuchi K, et al. Impact of tafamidis on echocardiographic cardiac function of patients with transthyretin cardiac amyloidosis. Circ J 2023; 87: 508–516, doi:10.1253/circj.CJ-22-0683.
• 19.   Garcia-Pavia P, Bengel F, Brito D, Damy T, Duca F, Dorbala S, et al. Expert consensus on the monitoring of transthyretin amyloid cardiomyopathy. Eur J Heart Fail 2021; 23: 895–905, doi:10.1002/ejhf.2198.
• 20.   Nagueh SF, Smiseth OA, Appleton CP, Byrd BF 3rd, Dokainish H, Edvardsen T, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography: An update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2016; 29: 277–314, doi:10.1016/j.echo.2016.01.011.
• 21.   Oike F, Usuku H, Yamamoto E, Yamada T, Egashira K, Morioka M, et al. Prognostic value of left atrial strain in patients with wild-type transthyretin amyloid cardiomyopathy. ESC Heart Fail 2021; 8: 5316–5326, doi:10.1002/ehf2.13621.
• 22.   Usuku H, Yamamoto E, Sueta D, Shinriki R, Oike F, Tabata N, et al. A new staging system using right atrial strain in patients with immunoglobulin light-chain cardiac amyloidosis. ESC Heart Fail 2024; 11: 1612–1624, doi:10.1002/ehf2.14710.
• 23.   Engblom H, Kanski M, Kopic S, Nordlund D, Xanthis CG, Jablonowski R, et al. Importance of standardizing timing of hematocrit measurement when using cardiovascular magnetic resonance to calculate myocardial extracellular volume (ECV) based on pre- and post-contrast T₁ mapping. J Cardiovasc Magn Reson 2018; 20: 46, doi:10.1186/s12968-018-0464-9.
• 24.   Tsutsui H, Albert NM, Coats AJS, Anker SD, Bayes-Genis A, Butler J, et al. Natriuretic peptides: Role in the diagnosis and management of heart failure: A scientific statement from the Heart Failure Association of the European Society of Cardiology, Heart Failure Society of America and Japanese Heart Failure Society. J Card Fail 2023; 29: 787–804, doi:10.1016/j.cardfail.2023.02.009.
• 25.   Grogan M, Scott CG, Kyle RA, Zeldenrust SR, Gertz MA, Lin G, et al. Natural history of wild-type transthyretin cardiac amyloidosis and risk stratification using a novel staging system. J Am Coll Cardiol 2016; 68: 1014–1020, doi:10.1016/j.jacc.2016.06.033.
• 26.   Arima N, Ochi Y, Kubo T, Murakami Y, Nishino K, Yamamoto H, et al. Prospective multicenter screening with high-sensitivity cardiac troponin T for wild-type transthyretin cardiac amyloidosis in outpatient and community-based settings. Circ J 2025; 89: 24–30, doi:10.1253/circj.CJ-24-0479.
• 27.   Morioka M, Takashio S, Nakashima N, Nishi M, Fujiyama A, Hirakawa K, et al. Correlation between cardiac images, biomarkers, and amyloid load in wild-type transthyretin amyloid cardiomyopathy. J Am Heart Assoc 2022; 11: e024717, doi:10.1161/JAHA.121.024717.
• 28.   Kuyama N, Takashio S, Oguni T, Yamamoto M, Hirakawa K, Ishii M, et al. Cardiac biomarker change at 1 year after tafamidis treatment and clinical outcomes in patients with transthyretin amyloid cardiomyopathy. J Am Heart Assoc 2024 13: e034518, doi:10.1161/JAHA.124.034518.
• 29.   Yu AL, Chen YC, Tsai CH, Chao CC, Su MY, Juang JJ, et al. Tafamidis treatment decreases ^{99 m}Tc-pyrophosphate uptake in patients with hereditary Ala97Ser transthyretin amyloid cardiomyopathy. JACC Cardiovasc Imaging 2023; 16: 866–867, doi:10.1016/j.jcmg.2022.12.016.
• 30.   De Michieli L, Cipriani A, Iliceto S, Dispenzieri A, Jaffe AS. Cardiac troponin in patients with light chain and transthyretin cardiac amyloidosis: JACC: CardioOncology state-of-the-art review. JACC CardioOncol 2024; 6: 1–15, doi:10.1016/j.jaccao.2023.12.006.
• 31.   Sado DM, Flett AS, Banypersad SM, White SK, Maestrini V, Quarta G, et al. Cardiovascular magnetic resonance measurement of myocardial extracellular volume in health and disease. Heart 2012; 98: 1436–1441, doi:10.1136/heartjnl-2012-302346.
• 32.   Law S, Petrie A, Chacko L, Cohen OC, Ravichandran S, Gilbertson JA, et al. Change in N-terminal pro-B-type natriuretic peptide at 1 year predicts mortality in wild-type transthyretin amyloid cardiomyopathy. Heart 2022; 108: 474–478, doi:10.1136/heartjnl-2021-319063.
• 33.   Ioannou A, Cappelli F, Emdin M, Nitsche C, Longhi S, Masri A, et al. Stratifying disease progression in patients with cardiac ATTR amyloidosis. J Am Coll Cardiol 2024; 83: 1276–1291, doi:10.1016/j.jacc.2023.12.036.