TTR Stabilizers are Associated with Improved Survival in Patients with Transthyretin Cardiac Amyloidosis

Abstract

Background

Transthyretin cardiac amyloidosis (TTR-CA) is caused by dissociation of transthyretin (TTR) into monomers, which misassemble into amyloid fibrils. TTR stabilizers act at the dimer-dimer interface to prevent dissociation. We investigated differences in survival among patients with TTR-CA on stabilizer medications compared to those not on stabilizers.

Methods

A retrospective study of patients with TTR-CA presenting to a single center was conducted. Baseline characteristics were compared between those treated with stabilizers and those not treated with stabilizers. Cox proportional hazards modeling assessed for univariate predictors of the composite outcome of death or orthotopic heart transplant (OHT). Multivariable Cox proportional hazards assessed whether stabilizer treatment was independently associated with improved death or OHT after controlling for significant univariate predictors.

Results

120 patients (mean age 75±8, 88% male) were included: 29 patients who received stabilizers and 91 patients who did not. Stabilizer use was associated with a lower risk of the combined endpoint of death or OHT (HR 0.32, 95% CI 0.18-0.58, p<0.0001). Subjects treated with stabilizers were more likely to be of White race (93% vs. 55%, p<0.001), classified as NYHA Class I-II (79% vs. 38%, p=0.002), less likely to have a mutation (10% vs. 36%, p=0.010), have lower troponin I (median 0.06 vs. 0.12 ng/mL, P=0.002), and higher LVEF (49% vs. 40%, p=0.011), suggesting earlier stage of disease. In Multivariable Cox analysis, the association between stabilizer and death or OHT persisted when adjusted for all non-collinear univariate predictors with p<0.05 (HR 0.37, 95% CI 0.19-0.75, p=0.003).

Conclusion

TTR stabilizers are associated with decreased death and OHT in TTR-CA. These results need to be confirmed by ongoing randomized clinical trials.

Transthyretin cardiac amyloidosis (TTR-CA) is a significant cause of progressive heart failure with preserved ejection fraction (HFpEF)¹. Transthyretin (TTR) is a 127- amino acid protein produced by the liver that circulates as a homotetramer. However, destabilization of the TTR protein from either inherited mutations (ATTRm) or the aging process in wild-type disease (ATTRwt) promotes its dissociation into monomers, which misassemble into amyloid fibrils and deposit in tissues and organs. In the heart, deposition of amyloid fibrils in the extracellular space leads to diastolic dysfunction, progressing to heart failure with a restrictive physiology and eventual death^{1, 2}.

More than 120 mutations in the TTR gene affect persons of all ages with ATTRm. The most common allele in the United States, caused by a substitution of valine to isoleucine at position 122 (Val122Ile), is found in 3.4% of black persons^2–5. ATTRwt predominantly affects older adult men⁶ and has been detected at autopsy in 32% of patients with HFpEF greater than 75 years of age without antemortem suspicion of amyloid deposition^{7, 8} and in 13.3% of hospitalized patients with HFpEF and left ventricular wall thickness of 12mm or greater¹.

Previously, treatment options for TTR-CA were limited to supportive care with a minority receiving heart transplant⁹, an option that often is not feasible in elderly patients with advanced disease. Stabilization of the TTR homotetramer has become a target for novel therapies to prevent progression of TTR amyloidosis^{2, 10, 11}. The approach was first considered after the discovery that a benign polymorphism in the TTR protein, Thr119Met, effectively stabilized the protein in the context of a trans-destabilizing mutation¹². Multiple TTR kinetic stabilizer structures have been studied. Diflunisal, a non-steroidal anti-inflammatory drug (NSAID), and tafamidis, an analog of diflunisal without NSAID properties, are the only two small molecule stabilizers of the TTR tetramer that have been shown to date to be safe in animal studies and human clinical trials^13–24.

Long-term effects of stabilizers in TTR-CA have yet to be demonstrated in large cohorts. In heart failure patients, where chronic NSAID use may lead to worsening renal dysfunction and precipitate heart failure, diflunisal, at low doses, has been shown to be safe in both ATTRwt and ATTRm patients over the short to intermediate term, and potentially slow disease progression^{18, 25}. Phase II trials of tafamidis demonstrate a high degree of stabilization of the TTR tetramer and the absence of clinically significant changes in most clinical, biochemical, electrocardiographic and echocardiographic parameters, consistent with the potential for tafamidis to slow cardiac disease progression^26–28. A phase III double blind, placebo controlled, randomized trial evaluating the efficacy, safety and tolerability of tafamidis is anticipated to report results in 2018²⁹. We conducted a retrospective analysis of patients with TTR-CA seen at the Center for Advanced Cardiac Care (CACC) at Columbia University Medical Center in order to evaluate for differences in survival in patients treated with stabilizer medications compared to those not on stabilizers.

Methods

The data, analytic methods, and study materials will not be made available to other researchers for purposes of reproducing the results or replicating the procedure. The study was approved the Columbia University Medical Center Institutional Review Board.

Study population and study design

We conducted a retrospective analysis of 160 patients with TTR-CA presenting to CACC between 2001 and 2016 (Figure 1). Cardiac amyloidosis was defined by: (1) endomyocardial biopsy-proven congophilic deposits, (2) histological documentation of congophilic deposition in at least 1 non-cardiac organ and echocardiographic evidence of infiltrative cardiomyopathy without another cause of left ventricular hypertrophy, and/or (3) positive Tc⁹⁹ pyrophosphate (PYP) scan with grade 2 or 3 myocardial uptake or a heart to contralateral ratio of >1.5³⁰. Evidence for TTR as the precursor protein causing amyloid was confirmed by histochemical analysis or mass spectroscopy. Both ATTRwt and ATTRm patients were included. TTR-CA patients were subdivided into two groups: those treated with stabilizers (tafamidis or diflunisal) and those not treated with stabilizers.

Figure 1.

Study Population and Breakdown by Treatment Strategy

Among the 160 patients, 40 patients were excluded from further analysis for the following reasons: if they received a medication other than tafamidis or diflunisal for treatment of TTR-CA (n=10; Revusiran (n=9), Doxycycline (n=1)), or were part of a randomized blinded clinical study of either tafamidis (n=29)²⁹ or Ionis TTR-Rx (n=1). In the remaining 120 patients, 13 were on diflunisal and 16 were on tafamidis (including 9 in the open label tafamidis trial, NCT00935012)²⁶. Baseline characteristics were assessed and included cardiac biomarkers (troponin I and brain natriuretic peptide [BNP]), estimated glomerular filtration rate (eGFR), modified BMI (mBMI; serum albumin multiplied by BMI), and functional status by New York Heart Association (NYHA) Class. Echocardiography was used to assess the left ventricular structure and function as described by the AHA guidelines³¹. Electrocardiography at the time of TTR-CA diagnosis analyzed for presence of arrhythmias, conduction abnormalities, low QRS voltage, and pseudoinfarct (anterior or lateral) pattern. Right heart catheterization (RHC) data from time of diagnosis was obtained. Patients were followed for a combined primary outcome of all-cause mortality or orthotopic heart transplant (OHT) from time of first diagnosis until March 30^th, 2017, ascertained from the electronic medical record, contacting primary physicians, and by cross-referencing the social security death index. We include OHT as a “death equivalent” because patients who underwent OHT would have otherwise died without transplant.

Statistical Analysis

Analyses were performed using Statistical Analysis Software version 9.4 (SAS Institute, Inc). Continuous variables were presented as mean±SD or median (IQR) if highly skewed and categorical variables were summarized as counts (percentages). The Χ² test or the Fisher exact test (when cell counts were small) were used to compare categorical variables, and the Wilcoxon rank sum test was used to compare continuous variables, between patients who were treated and patients who were not treated with stabilizers. Univariate Cox proportional hazards models were created to identify factors associated with death or OHT. Troponin and BNP were log transformed to adjust for skewness. Multivariable Cox proportional hazards modeling was subsequently performed to determine whether stabilizer treatment was independently associated with reduced death or OHT after controlling for all non-collinear univariate predictors of death or OHT with p<0.05: race, presence of mutation, NYHA class, Log₁₀ BNP, and LVEF by 1% increase. Kaplan-Meier survival analysis using log-rank test was performed to compare patients treated with a stabilizer to those not treated with a stabilizer for the composite end point of death or OHT. Supplemental analyses for the outcome of death alone were similarly conducted. Finally, Kaplan-Meier survival analysis using log-rank test was performed to compare outcomes between patients treated with tafamidis and diflunisal (Supplemental Figure 3).

Results

Baseline characteristics (Table 1)

Table 1.

Demographic and Clinical Variables Among Patients with TTR-CA comparing “On Stabilizer” vs. “Not on Stabilizer”

Demographic	Overall (N=120)	Not on Stabilizer (N=91)	On Stabilizer (N=29)	P-value
Age, yrs	75 ± 8	75 ± 8	75 ± 9	0.704
Male sex, N (%)	105 (87.5%)	77 (84.6%)	28 (96.6%)	0.114
Race, N (%)
White	77 (64.2%)	50 (55.0%)	27 (93.0%)
Black	35 (29.2%)	34 (37.3%)	1 (3.5%)	<0.001
Other	8 (6.6%)	7 (7.7%)	1 (3.5%)
NYHA class, N (%)
I	6 (5.0%)	4 (4.4%)	2 (6.9%)
II	53 (44.2%)	32 (35.1%)	21 (72.4%)	0.002
III	54 (45.0%)	48 (52.8%)	6 (20.7%)
IV	7 (5.8%)	7 (7.7%)	0 (0%)
Era diagnosed
Early (2001-2010)	40 (33%)	24 (26%)	16 (55%)
Middle (2010-2014)	40 (33%)	36 (40%)	4 (14%)	<0.001
Late (2014-2016)	40 (33%)	31 (34%)	9 (31%)
ATTR mutation, N (%)	36 (30.0%)	33 (36.3%)	3 (10.3%)	0.010
SBP, mmHg	113 ± 15	113 ± 16	115 ± 15	0.317
DBP, mmHg	68 ± 10	68 ± 9	70 ± 11	0.306
Biochemistry
Troponin I, ng/mL (median, IQR)	0.10 (0.06-0.17)	0.12 (0.07-0.18)	0.06 (0.05-0.10)	0.002
BNP, pg/mL (median, IQR)	592 (278-1216)	685 (302-1411)	494 (269-736)	0.123
Albumin, g/dL	4.1 ± 0.4	4.1 ± 0.3	4.3 ± 0.3	0.072
Creatinine, mg/dL	1.5 ± 0.5	1.5 ± 0.5	1.5 ± 0.5	0.750
eGFR, mL/min	56 ± 19	55 ± 19	56 ± 20	0.993
Modified BMI kg·g/dL·m2	110 ± 21	108 ± 20	118 ± 25	0.039
Echocardiography
LVEF, %	42 ± 16	40 ±16	49 ± 15	0.011
LA size, cm	4.6 ± 0.6	4.6 ± 0.6	4.8 ± 0.7	0.222
LVPW, mm	16 ± 4	16 ± 4	16 ± 4	0.531
IVS, mm	17 ± 4	17 ± 4	17 ± 4	0.819
EKG
Afib or flutter, N (%)	35 (30.4%)	27 (31.4%)	8 (27.6%)	0.817
Low voltage any lead, N (%)	38 (33.3%)	33 (38.8%)	5 (17.2%)	0.041
PRWP, N (%)	58 (54.7%)	44 (55.0%)	14 (53.9%)	0.918
Pseudoinfarct pattern, N (%)	8 (7.2%)	5 (6.0%)	3 (10.7%)	0.414
RHC
CO, L/min	3.6 ± 1.0	3.5 ± 1.0	3.8 ± 1.1	0.385
CI, L/min/m2	1.8 ± 0.5	1.8 ± 0.5	1.9 ± 0.5	0.617
PVR, woods units	3.6 ± 1.9	3.8 ± 1.9	3.0 ± 1.4	0.300

Abbreviations: NYHA, New York Heart Association; SBP, systolic blood pressure; DBP, diastolic blood pressure; BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; BMI, body mass index; LVEF, left ventricular ejection fraction; LA left atrial; LVPW, left ventricular posterior wall; IVS, interventricular septal thickness; PRWP, poor R wave progression; CO, cardiac output; CI, cardiac index; PVR, pulmonary vascular resistance

The study cohort included 120 patients (mean age 75 ± 8 years) with TTR-CA of which 84 (70%) were ATTRwt and 36 (30%) were ATTRm with the following mutations: Val122Ile (N=33), Thr60Ala (N=2), and Phe64Leu (N=1). The cohort was predominantly male (88%), with NYHA class II or III symptoms (89%). There were 29 (24%) patients who received stabilizer treatment (either tafamidis N=16, or diflunisal N=13) and 96 (76%) patients who did not receive stabilizer treatment. Compared to patients who were not on a stabilizer, patients who were treated with a stabilizer were more likely to be of White race (93% vs. 55%, p<0.001), classified as NYHA class I-II (79% vs. 39%, p=0.002), less likely to have a mutation (10% vs. 36%, p=0.010), had a lower troponin I (median 0.06 vs. 0.12 ng/mL, p=0.002), higher modified BMI (118 vs. 108 kg·g/dL·m², p=0.039), higher left ventricular ejection fraction (LVEF) (49% vs. 40%, p=0.0011), and less low voltage on ECG (17% vs 39%, p=0.041). There was no difference between the two groups with respect to baseline systolic or diastolic blood pressure, BNP, albumin, eGFR, left atrial size, interventricular septal wall thickness, poor R-wave progression or pseudoinfarct pattern on ECG, or cardiac output or pulmonary vascular resistance on right heart catheterization.

Predictors of Death or OHT (Table 2)

Table 2.

Cox Proportional Hazards Univariate and Multivariable Predictors of Death or OHT Among Patients with TTR Cardiac Amyloidosis

Predictor	N	Hazard Ratio	95% CI	P-value
Univariate
Age, per 1 year increase	120	1.00	0.97-1.03	0.843
Modified BMI, per 1 unit increase	116	0.99	0.98-1.01	0.331
Male sex	120	1.36	0.55-3.39	0.487
Black race	120	1.72	1.07-2.76	0.029
Presence of any mutation	120	1.67	1.05-2.64	0.033
NYHA class, I-IV, per 1 class increase	120	1.58	1.15-2.17	0.005
Era diagnosed	120	0.84	0.60-1.17	0.303
Stabilizer	120	0.32	0.18-0.58	<0.0001
Log10 Troponin I	115	1.20	0.98-1.47	0.084
Log10 BNP	118	1.30	1.02-1.65	0.037
eGFR<60 mL/min	119	0.96	0.60-1.52	0.850
LVEF, per 1% increase	116	0.98	0.97-0.99	0.041
LA size, per 1 cm increase	108	0.92	0.62-1.35	0.660
IVS, per 1 mm increase	115	1.01	0.96-1.07	0.691
Pseudoinfarct pattern, yes vs. no	111	1.01	0.44-2.35	0.978
Low voltage, yes vs. no	114	1.07	0.66-1.76	0.780
CO, per 1 L/min decrease	77	0.91	0.70-1.19	0.492
CI, per 1 L/min/m2 decrease	77	0.84	0.48-1.45	0.525
PVR, per 1 woods unit increase	55	1.02	0.87-1.20	0.836
Full Multivariable model with all univariate predictors with p<0.05
Stabilizer	112	0.37	0.19-0.75	0.003
Black race	112	1.20	0.56-2.54	0.643
NYHA class	112	1.19	081-1.74	0.387
Presence of any mutation	112	1.01	0.46-2.20	0.981
Log10 BNP	112	1.09	0.81-1.46	0.562
LVEF, by 1% increase	112	1.00	0.98-1.02	0.909

Abbreviations: BMI, body mass index; NYHA, New York Heart Association; BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; LA left atrial; IVS, interventricular septal thickness; CO, cardiac output; CI, cardiac index; PVR, pulmonary vascular resistance

Subjects were followed for median 1.9 (IQR 1.0-2.8) years from time of amyloid diagnosis. Over this time period, 78 subjects met the primary end-point of death (n=67) or OHT (n=11). Probability of death or OHT at 1 and 5 years comparing stabilizer vs not on stabilizer was 7% vs 21% and 41% vs 86% respectively. In Cox proportional hazards modeling, patients on a stabilizer had a lower risk of the combined endpoint of mortality or OHT (HR 0.32, 95% CI of 0.18-0.58, p<0.0001). Other univariate predictors of death or OHT included Black race (p=0.029), presence of any mutation (p=0.033), higher NYHA class (p=0.005), Log₁₀BNP (p=0.037) and lower LVEF (p=0.041). Modified BMI, male sex, Log₁₀ Troponin I, eGFR, era diagnosed, left atrial size, interventricular septal wall thickness, pseudoinfarct pattern or low voltage on ECG, and cardiac output did not significantly predict death or OHT in this cohort. In multivariable Cox analysis, the association between stabilizer and death or OHT persisted when adjusted by all non-colinear univariate predictors with p<0.05: Black race, presence of any mutation, NYHA class, Log₁₀ BNP, and lower EF (HR 0.37, 95% CI 0.19-0.75, p=0.003).

Propensity scores were constructed using logistic regression modeling with observed treatment groups as the outcome (Supplemental Figure 1). In Cox proportional hazards model for death or OHT adjusting for likelihood of treatment, stabilizer use alone remains significant (HR 0.355, 95% CI 0.19-0.66, p=0.0012). These results are consistent with our findings in the full multivariable model, suggesting an effect of treatment after accounting for non-random treatment assignment.

In Kaplan-Meier analysis (Figure 2) among patients with TTR-CA, median time to the combined outcome of death or OHT was significantly worse if patients were not on a stabilizer, 2.2 vs 5.4 years, log-rank P<0.0001. Similar results were found for the endpoint of survival (Supplemental Table 1 and Supplemental Figure 2). The association of type of stabilizer (diflunisal vs. tafamidis) with the composite outcome of death or OHT, did not differ (see Supplemental Figure 3).

Figure 2.

Kaplan-Meier Analysis Among 120 Patients with TTR-CA Over the 1.9 Year Median Follow-up for the Outcome of Death or OHT, Stratified by use of Stabilizer

Discussion

This retrospective review of a cohort of 120 patients with TTR-CA demonstrates that treatment with TTR stabilizers is associated with improved outcome of survival or OHT. Compared to patients who were not treated with a stabilizer, those treated with a stabilizer were more likely to have lower NYHA class, lower baseline troponin I, higher modified BMI, and higher LVEF, suggesting a less severe phenotype of disease at baseline. However, after accounting for univariate predictors of death or OHT, which include presence of mutation, higher NYHA class, higher troponin I and lower EF that account for severity of disease, the benefit of TTR stabilizers persists in multivariate modeling.

Mechanism of TTR Stabilization

TTR stabilizers block tetramer dissociation, the rate-limiting step in the amyloidogenic process and reduce de novo deposition of amyloid¹⁹. Negative cooperative binding of stabilizers to the two L-thyroxine binding sites in the TTR tetramer, stabilizes the native state over the dissociative state and inhibits amyloid fibril formation in vitro^{15, 32–34}. Phase I/II studies demonstrated effective binding of tafamidis and diflunisal to the thyroxine binding site, kinetically stabilizing circulating TTR tetramers and inhibiting release of TTR monomers required for amyloidogenesis^{13, 14, 23, 26, 35}. TTR tetramer stabilization is a critical first step to slowing cardiac disease progression but the impact of stabilizers on hard outcomes such as survival or OHT has not been previously delineated.

Stabilizers in TTR-familial amyloid polyneuropathy

TTR stabilizers inhibit progression of polyneuropathy and preserve quality of life in patients with transthyretin familial-amyloid-polyneuropathy (TTR-FAP). In a clinical trial of 130 patients with TTR-FAP, diflunisal reduced the rate of progression of neurologic impairment (NIS+7 score increased 25 points in placebo vs 8.7 in the diflunisal group, p<0.001) and preserved quality of life during two year follow up. At 2 years, the diflunisal group exhibited neurological stability compared to placebo, defined as <2-point increase in NIS+7 score (29.7% vs. 9.4%, p=0.007)²⁴. In a subsequent study of Japanese subjects with ATTRm with neuropathy, treatment with 12 months of diflunisal significantly increased serum TTR concentrations (21.8±5.2 mg/dL to 28.1±6.8 mg/dL, p<0.0001) and stabilized TTR structure. Clinical FAP score increased by 1.0/year (p=0.79) at 24 months compared to the previously described natural history of FAP at about 7.0/year, suggesting longitudinal benefits of diflunisal¹⁸.

Data from phase II trials in TTR-FAP have demonstrated long-term efficacy and safety of tafamidis^{20, 36}. In a randomized placebo-controlled trial in patients with TTR-FAP, those in the intent-to-treat population who received tafamidis demonstrated 52% less neurologic deterioration after 18 months of treatment than patients who received placebo, with a sustained response in a 12 month extension study^{23, 37}. The long term safety and efficacy of tafamidis for delaying disease progression in TTR-FAP for up to 5.5 years was also recently demonstrated²¹. Among ATTRm patients, those who continuously received tafamidis experienced numerically less deterioration of neurologic function from baseline compared with those initially treated with placebo and then switched to tafamidis 18 months later. In fact, patients who received tafamidis from the start of the pivotal study were less likely to progress to the next ambulatory stage by 6 years than patients who started tafamidis later^{21, 36}.

Phase II Studies of stabilizers in TTR-CA

Prior phase II studies of both tafamidis^{26, 27} and diflunisal^{13, 18, 25} have demonstrated adequate stabilization of the TTR homotetramer as well as potential clinical stabilization of disease progression in TTR-CA. Two recent open label studies evaluated the safety and preliminary effects of diflunisal demonstrated stabilization of cardiac structure and function in addition to preservation of renal function. In a single arm, small open label study (n=13, 85% male, 69 years old) in TTR-CA, Castano et al report no change in cardiac structure (LV mass) or function (EF), with only non-significant uptrend in Troponin I and BNP, suggestive of stabilization of disease²⁵. Likewise, in 40 subjects (70% male, 60.7 years old) with ATTRm with both cardiac and neurologic manifestations, Sekijma et al reported a non-significant decrease in cardiac wall thickness by 0.2mm/year and an increase in EF by 0.4%/year after 24 months of treatment accompanied by increased levels of serum TTR¹⁸. In our experience, diflunisal at a dose of 250 mg orally twice a day is generally well tolerated by patients with TTR-CA who are not acutely decompensated, have eGFR > 45 ml/min/m² and are not on high dose diuretics. Among such patients, approximately 10% of subjects do not tolerate diflunisal.

Two recent phase II, open label single treatment arm studies of tafamidis demonstrated safety and stabilization of TTR tetramer in TTR-CA. In a study of 21 patients with non-Val30Met TTR amyloid (13 with cardiac involvement) followed over 12 months, 18 patients demonstrated TTR stabilization at week 6. There was maintenance of quality of life, mBMI, echocardiographic parameters and stabilization of biomarkers over the follow up period²⁷. In a phase II, open label, multicenter 12-month study evaluating safety and efficacy on TTR stability of oral tafamidis with ATTRwt or ATTRm (V122I), 30 of 31 patients were stabilized at week 6 using a non-physiologic assay and no clinically relevant changes in echocardiographic parameters were observed during the follow up period. Cardiac biomarkers including NT-proBNP, troponin I, and troponin T increased from baseline to 12-month follow up, but there was minimal decrease of 8m in 6MWT and subjects had stable quality of life assessments. Tafamidis treatment was generally well tolerated although 7 of 31 patients had bouts of diarrhea²⁶. In a recent post hoc analysis, survival of patients from the TRACS natural history study was compared with those in the Fx1B-201 tafamidis open label study and was restricted to include only NYHA class I/II and ATTRwt in attempts to provide a comparable cohort of patients. In time-to-mortality analysis, there was a significant difference in survival for patients treated with tafamidis compared with untreated patients over the long-term follow-up (P = 0.0004 for NYHA I/II patients, P=0.0262 for ATTRm patients)²⁷.

Furthermore, a phase III trial evaluating the safety and efficacy of tafamidis on the primary outcome of a composite of all-cause mortality and cardiovascular disease related hospitalization is anticipated to report results in 2018²⁹.

Survival in TTR-CA

TTR-CA is progressive disorder, often undiagnosed until symptoms of heart failure are apparent in the presence of significant ventricular wall thickening and advanced diastolic dysfunction. In the largest and most recent databases of TTR-CA, median survival from the time of diagnosis usually ranges between 25 and 41 months, with longer survival in patients with ATTRwt^{8, 38, 39}. In this retrospective analysis, median survival in patients on stabilizers was 65 months whereas for those not on stabilizers, median survival was only 26 months. In a prospective longitudinal investigation of TTR-CA, the Transthyretin Amyloidosis Cardiac Study (TRACS) that followed both 29 ATTRwt and ATTRm (V122I) subjects for progression of disease⁴⁰, median survival following diagnosis was 26 and 43 months for ATTRm (V122I) and ATTRwt respectively. Similarly, Connors et al demonstrated a median survival of 27 months in 26 patients with ATTRm (V122I)³⁹ and separately report survival in 82 patients with ATTRwt to be 52 months⁴¹. Notably while stabilizer use was associated with longer survival in our data, long-term survival (e.g. > 8-10 years) was rare. Whether study of a larger cohort, higher doses of tafamidis (80 mg), the use of other agents such as TTR silencers, or treatment with multiple agents, will translate into long term survival benefit in early stages of disease requires further study.

Study limitations

Significant limitations to this investigation are worth noting. This was a retrospective case control cohort study and not a randomized control trial. As such, there were inherent baseline differences between patients who received treatment with TTR stabilizers and those who did not, including a potential selection bias for patients having a less severe disease phenotype. For example, patients without baseline renal dysfunction were more likely to receive TTR-stabilizers, especially diflunisal, an NSAID. Despite differences in baseline characteristics, in multivariable modeling, the hazard ratio for the association of TTR stabilizers with improved survival remained significant with a similar point estimate. Second, this study represents data from a single amyloid center with a relatively small cohort and a low percentage of African American patients with Val122Ile mutation treated with TTR stabilizers, which may limit the generalizability of these results. A prospective, multicenter, randomized, double-blinded, placebo-controlled study with sufficient power to detect differences in clinical outcomes among mutant and wild type forms of TTR-CA with tafamidis 20 or 80 mg compared to placebo is now fully recruited with results anticipated in 2018.²⁹ If effective, future trials might study whether multiple agents result in further reduced disease progression and survival in patients with TTR-CA. Finally, the comparison of type of stabilizer (tafamidis and diflunisal) association with the composite outcome of death or OHT is limited by a small sample size and should be considered hypothesis generating.

Conclusion

In conclusion, TTR stabilizers, tafamidis and diflunisal, are associated with a lower risk of the combined endpoint of death or OHT in a small, retrospective analysis among patients treated at a single center. Randomized, placebo controlled trials are required to confirm these findings.

Clinical Perspective.

What is new?

• The study shows an independent association of TTR stabilizer medications (tafamidis and diflunisal) in patients with transthyretin cardiac amyloidosis (TTR-CA) with the composite outcome of survival and orthotopic heart transplant (OHT).

What are the clinical implications?

• The results from this retrospective single-center study demonstrate for the first time an association between TTR tetrameter stabilizer use and improved survival.

• These data highlight the potential value of TTR stabilizers for patients with TTR-CA, which are being evaluated in an ongoing randomized, placebo controlled trials.

• If these findings are confirmed, future trials should be performed to determine if TTR stabilizers are equally effective.