Transthyretin cardiac amyloidosis: an update on diagnosis and treatment.

Transthyretin cardiac amyloidosis (ATTR-CA) demonstrates progressive, potentially fatal, and infiltrative cardiomyopathy caused by extracellular deposition of transthyretin-derived insoluble amyloid fibrils in the myocardium. Two distinct types of transthyretin (wild type or variant) become unstable, and misfolding forms aggregate, resulting in amyloid fibrils. ATTR-CA, which has previously been underrecognized and considered to be rare, has been increasingly recognized as a cause of heart failure with preserved ejection fraction among elderly persons. With the advanced technology, the diagnostic tools have been improving for cardiac amyloidosis. Recently, the efficacy of several disease-modifying agents focusing on the amyloidogenic process has been demonstrated. ATTR-CA has been changing from incurable to treatable. Nevertheless, there are still no prognostic improvements due to diagnostic delay or misdiagnosis because of phenotypic heterogeneity and co-morbidities. Thus, it is crucial for clinicians to be aware of this clinical entity for early diagnosis and proper treatment. In this mini-review, we focus on recent advances in diagnosis and treatment of ATTR-CA.

Introduction

Transthyretin cardiac amyloidosis (ATTR‐CA) demonstrates infiltrative cardiomyopathy caused by extracellular deposition of insoluble transthyretin (TTR) amyloid fibrils in the myocardium.1 TTR is a plasma protein mainly synthesized in the liver, recognized as a transporter of thyroxine and retinol‐binding protein. Unstable changes in two different types of TTR (wild type or variant) become misfolding, aggregate, and form ultimately amyloid fibrils. Cardiac amyloidosis (CA) has been recently highlighted as a cause of heart failure with preserved ejection fraction (HFpEF) among elderly persons, and its incidence has been constantly increasing because the population is aging.2, 3 CA is progressive and life‐threatening if left untreated, and thus, early diagnosis is critical. However, a definitive diagnosis of CA is considerably difficult, because left ventricular hypertrophy (LVH) shows a similar phenotype in hypertension, hypertrophic cardiomyopathy (HCM), and aortic stenosis (AS).4, 5, 6 Moreover, the development of an appropriate strategy to reach the correct diagnosis and treatment of CA has been long overdue. In this mini‐review, we focus on the current diagnosis and treatment of ATTR‐CA.

Epidemiology and clinical presentation

Wild type ATTR‐CA (ATTRwt‐CA), previously known as senile systemic amyloidosis, is caused by age‐related misfolding of TTR, but its mechanism remains to be fully elucidated.1 ATTRwt‐CA will become the most frequent form of amyloidosis in the USA. ATTRwt‐CA affects male‐predominant patients over 60 years of age and typically presents HFpEF.7 Dyspnoea, fatigue, and weakness are common symptoms. However, these symptoms are often misunderstood as non‐specific symptoms due to aging. Anginal chest pain caused by microvascular amyloid infiltration in patients without obstructive epicardial coronary stenosis can also occur.8, 9 Recently, ATTRwt‐CA is increasingly recognized as a cause of various common diseases. While ATTRwt‐CA was identified in 13% of elderly patients (>60 years old) with HFpEF and 16% of patients undergoing transcatheter aortic valve replacement, respectively,2, 4 bilateral carpal tunnel syndrome (CTS) and lumbar spinal stenosis frequently present many years before the development of cardiac symptoms..10, 11

Hereditary ATTR‐CA (ATTRv‐CA) is an autosomal‐dominant disease in which gene mutations lead to changes in the protein TTR.12 More than 120 mutations have been reported in the TTR gene with considerable phenotypic and geographical heterogeneity. Their clinical symptoms vary extensively from neurological‐predominant phenotype to cardiac‐predominant phenotype. This phenotypic heterogeneity depends on many factors, such as specific TTR mutation site, geographical distribution, inheritance pattern, timing of onset, and epidemic/non‐epidemic aggregation. While the Val30Met induces progressive peripheral sensory‐motor polyneuropathy, other mutations (Thr60Ala, Ile68Leu, Leu111Met, and Val122Ile) cause exclusively infiltrative cardiomyopathy.13 In particular, a sporadic case of ATTRv‐CA is extremely rare, and its diagnosis is considerably difficult.14

The Val122Ile variant is the most frequent form of TTR amyloid in the USA and prevalent in 3.4% of African Americans.15 The clinical phenotype is late‐onset restrictive cardiomyopathy, identical to ATTRwt‐CA and often mimics hypertensive cardiomyopathy due to high morbidity of hypertension. A prospective observational Atherosclerosis Risk in Communities study, which analysed 3856 African Americans, reported a low clinical penetrance of the disease. Echocardiogram revealed that only three (7%) carriers had typical echocardiographic findings characteristic of CA. However, surprisingly, Val122Ile carriers had a significantly increased risk of heart failure (HF) during the later years of the study compared with non‐carriers (age‐stratified and sex‐stratified hazard ratio, 1.47 and 95% confidence interval, 1.03 to 2.10),16 suggesting that Val122Ile carriers are predominantly at increased risk of HF with an age‐dependent clinical penetrance. The Thr60Ala variant, the second most frequent form of TTR amyloid in the USA, affects up to 1% of the population of Northwestern Ireland.17 The Thr60Ala variant causes both neurologic and cardiac phenotypes, and symptom onset mainly occurs between 50 and 60 years of age.18 The Ile68Leu variant is endemic in central‐northern Italy and shows exclusively cardiac phenotype with male predominance and age‐dependent penetrance.13, 19 In Denmark, the Leu111Met variant, which shows exclusively cardiac phenotype, has only been found in Danish families.20 Transthyretin Amyloidosis Outcomes Survey registry verified that patients with the Leu111Met variant were significantly younger (mean age at symptom onset: 47.6 years) and less likely to be male (63.6%) than those with other known cardiac variants (Val122Ile, Thr60Ala, and Ile68Leu).21 On the other hand, the Val30Met variant, known as transthyretin familial amyloid polyneuropathy (ATTR‐FAP), is the most frequent form of TTR amyloid in Europe and other countries.12 In endemic early‐onset Val30Met patients in Portugal, Brazil, and Japan, an exclusively neurologic phenotype is common, and its disease onset frequently occurs in the third and fourth decade. On the other hand, in non‐endemic late‐onset Val30Met patients in France, Sweden, and Japan, an exclusively cardiac phenotype is common, and its disease onset occurs relatively late (>50 years old), affecting predominantly male. Median survival was notably shorter in late‐onset Val30Met patients.22 There is a bimodal distribution of onset age among patients with the Val30Met variant in Japan.13, 23 Several studies demonstrated that the non‐Val30Met variants were associated with disease severity and worse outcomes compared with those with the Val30Met variant.24, 25

Implications of genetic diagnosis of transthyretin gene sequencing

Early and correct identification of ATTRv amyloidosis plays a vital role for the estimation of prognosis, treatment of choice, familial screening, and genetic counselling. In a prospective multicentre study in France, genetic screening for a TTR gene revealed that 5% of 298 consecutive elderly patients (>60 years old) clinically diagnosed with HCM have ATTRv‐CA with a predominance of the Val122Ile variant.26 Moreover, the Beta‐Blocker Evaluation of Survival Trial study showed a high prevalence of the Val122Ile variant in elderly African Americans with severe HF.27 Thus, TTR gene sequencing should be systematically carried out in even elderly patients of African descent having HCM phenotype of unknown cause or unexplained HF. Regular monitoring should start for asymptomatic carriers of a variant of interest 10 years before the predicted age of onset of symptomatic disease.28

Diagnosis

Lack of recognition of this clinical entity, non‐specific symptoms, and co‐morbidities often leads to delayed diagnosis, resulting in disease progression.29 One descriptive study revealed that 35% had been previously misdiagnosed with other cardiovascular diseases commonly recognized in clinical practice. Among them, hypertensive cardiomyopathy (35%) was the most frequent followed by HCM (23.5%), ischaemic heart disease (11.8%), HFpEF (8.8%), and AS (8.8%), respectively.30 Thus, it is critical to make an early diagnosis for proper treatment. ATTR‐CA should be highly suspected if left ventricular (LV) wall thickening is observed in combination with one or more of red‐flags shown in Figure . Newly diagnosed HFpEF patients with LVH (>12 mm) over 60 years old are good candidates for ATTR‐CA.2 Moreover, atrial fibrillation (AF) (38–67%) and symptomatic atrioventricular block requiring pacemaker (8–40%) are the most common in patients with ATTRwt‐CA (Table 1).7, 31, 32 Unexplained CTS (bilateral or intractable) should strongly suspect ATTRwt‐CA.10 Intriguingly, patients with ATTRv‐CA frequently had CTS and/or neuropathy (46% and 53%, respectively).26 In addition, a series of studies demonstrated that ATTR‐CA was prevalent in 14% to 16% of elderly patients with severe calcified AS undergoing transcatheter aortic valve replacement.4, 36 In particular, this is highly likely to be the case with low‐flow, low‐gradient severe AS. There are also warnings that should not be overlooked. Because blood pressure often falls as the disease progresses, symptomatic hypotension or resolution of hypertension in previously hypertensive patients might be other promising clues to suspect ATTR‐CA.

Figure 1

Red‐flags that should be highly suspected of transthyretin cardiac amyloidosis. ACE‐I, angiotensin‐converting enzyme inhibitor; ARB, angiotensin II receptor blocker; CCB, calcium‐channel blocker; CMR, cardiac magnetic resonance; ECG, electrocardiogram; Echo, echocardiogram; HF, heart failure; HFpEF, heart failure with preserved ejection fraction; LGE, late gadolinium enhancement; LV, left ventricle.

Table 1

Baseline characteristics and survival of patients with different types of cardiac amyloidosis

	AL31	ATTRwt7, 19, 31, 32	Val122Ile33, 34	Thr60Ala18	Ile68Leu19	Late Val30Met35
Age at diagnosis (year)	63	73‐76	69‐74	66	71	67.3
Male (%)	69.4	89‐98	76‐85	70	78	86
Common ethnicity	Variable	Caucasian	Afro‐Caribbean	Irish	Caucasian	Japanese
Family history (%)	NA	NA	NA	37	63	48
NYHA Class III–IV (%)	60	62‐85	47‐55	NA	27	14
CTS (%)	8	39‐48	29‐46	NA	43	NA
Peripheral neuropathy (%)	8	3‐9	38	54	19	80
Autonomic symptoms (%)	NA	NA	10	75	6	10
SBP/DBP (mmHg)	107/72	116/74	112/69	NA	120/80	NA
Atrial fibrillation (%)	11	38‐67	31‐52	NA	30	3.8
Pacemakers (%)	5	8‐40	8.7‐11	NA	9	12
IVSd/LVPWd (mm)	15/15	17/17	17/17	17/17	17/16	16/14
LVEF (%)	42	47‐51	39‐51	53	51	64
NT‐ proBNP (pg/mL)	6038	3361	2734	2528	3287	1116
	(3615–13 302)a	±845	(2307–4467)a	(42–18 148)b	(1745–5658)a	±1384
Diagnosis to death (year)	0.9	2.7‐3.9	2.62	3.4	5‐year survival (37%)	4.3

The numerical values of circulating NT‐proBNP levels are described in mean ± standard deviation,,7, 19, 31, 32, 35 median (range, min–max),18 or median (Q1–Q3 percentile),19, 31, 33, 34 according to the description in the original literature.

AL, light‐chain amyloidosis; ATTRwt, wild type transthyretin amyloidosis; CTS, carpal tunnel syndrome; DBP, diastolic blood pressure; IVSd, interventricular septum thickness at end‐diastole; LVEF, left ventricular ejection fraction; LVPWd, left ventricular posterior wall thickness at end‐diastole; NA, not available; NT‐proBNP, N‐terminal pro b‐type natriuretic peptide; NYHA class, New York Heart Association class; SBP, systolic blood pressure.

Median (Q1–Q3 percentile).

Median (range, min–max).

Electrocardiography

Patients may develop low QRS voltage on electrocardiogram or pseudoinfarction pattern even in the absence of epicardial coronary stenosis as the disease progresses. Unexplained LVH, characterized by low QRS voltage on electrocardiogram despite LVH on echocardiogram, provides valuable clues for the suspicion of CA. However, these electrocardiographic features are actually not sensitive enough to identify CA.37, 38

Echocardiography

Echocardiogram plays an important key role in diagnosing CA. Echocardiogram can recognize diastolic dysfunction at early stages and systolic dysfunction at later stages. The most common CA phenotype is the presence of increased LV wall thickness, small LV chamber size with systolic impairment, atrial enlargement, and signs of elevated filling pressures caused by restrictive diastolic filling. Left ventricular ejection fraction (LVEF) declines as the disease progresses, resulting in a transition from preserved LVEF to reduced LVEF at an advanced stage. Other clues include right ventricular hypertrophy, pericardial effusions, atrial septal or cardiac valve thickening, and intra‐cardiac thrombus,39 which are not pathognomonic of ATTR‐CA.

With recent advanced technology, strain echocardiography has become a new imaging modality to measure myocardial deformation. Relative apical‐sparing of longitudinal strain (LS) is considered as a hallmark for CA. A relative apical LS index, which is the ratio of average apical LS/(average basal LS + average mid‐LS), was useful for differentiating CA from other causes of LVH (93% sensitivity and 82% specificity).40 Another study evaluated the diagnostic accuracy of strain imaging. A cut‐off value of septal apical to basal longitudinal systolic strain ratio > 2.1 differentiated CA from other causes of LVH such as hypertension, Fabry disease, or Friedreich ataxia (88% sensitivity, 85% specificity, 67% positive predictive value, and 96% negative predictive value).41 In addition, a combination of the systolic strain gradient and deceleration time of early filling (<200 ms) could further improve the diagnostic accuracy for detecting CA (88% sensitivity, 100% specificity, 100% positive predictive value, 96% negative predictive value). The ratio of ejection fraction/global LS, which is another specific echo parameter, is effective for distinguishing CA from HCM.42

Cardiac magnetic resonance

Cardiac magnetic resonance can provide unique information on myocardial tissue properties. CMR shows various characteristic patterns of late gadolinium enhancement (LGE) in CA: global subendocardial LGE, global transmural LGE, atrial LGE, and suboptimal myocardial nulling. CMR with LGE was very useful for diagnosing CA (80% sensitivity and 94% specificity).43 Although speckle‐tracking imaging can reliably recognize and differentiate CA from the non‐amyloid cause of cardiomyopathy, echocardiogram alone cannot distinguish the CA subtypes. An original scoring system, the Query Amyloid Late Enhancement (QALE) score, has been designed to distinguish ATTR‐CA from immunoglobulin light‐chain cardiac amyloidosis (AL‐CA) with 82% sensitivity and 76% specificity.44 The QALE score is a semi‐quantitative index of measuring the extent of amyloid burden in both ventricles identified by LGE. However, the value of the QALE score remains inconsistent because another study has failed to prove its usefulness.45 Actually, LGE cannot detect diffuse myocardial fibrosis due to significant extracellular amyloid infiltration, and patients with renal dysfunction cannot utilize this technique. Recently, T1‐mapping technique can overcome these limitations, identify early disease, and quantitatively evaluate disease progression without contrast agents.46

Radionuclide bone scintigraphy

Nuclear imaging technique employing technetium pyrophosphate (99mTc‐PYP), once used as bone scintigraphy, has recently been reported as a reliable diagnostic method for ATTR‐CA, which is distinguished from AL‐CA, or other wall thickening disease with high specificity of 100%. Cardiac TTR deposition can be detected at an asymptomatic stage.47, 48 In addition, multicentre study validated that bone‐avid tracers provided excellent diagnostic performance for diagnosis of ATTR‐CA when used in combination without evidence of monoclonal protein (100% specificity and 100% positive predictive value). Moreover, marked myocardial uptake of 99mTc‐PYP was associated with poor prognosis in ATTR‐CA.49 Thus, cardiac radioisotope examination has established a valuable position as indispensable tools for the diagnosis, severity, and treatment planning.

Histology

Tissue diagnosis remains the golden standard for making the diagnosis of amyloidosis. Congo red or Direct Fast Scarlet 4BS staining binds to deposit amyloid fibrils and yields to characteristic apple‐green birefringence under polarized light microscopy. Note that apple‐green birefringence should be observed in the same amyloid deposits stained with Congo red dye. In addition, electron microscopy demonstrates randomly oriented and non‐branching fibrils with a diameter of approximately 7.5–10 nm. Subsequently, subtyping of amyloid fibril can be performed by immunohistochemistry or laser microdissection/mass spectroscopy. Laser microdissection/mass spectroscopy can allow accurate diagnosis, typing, and variants of amyloidosis.50, 51 Traditionally, a conventional endomyocardial biopsy was performed to confirm the diagnosis and typing of CA with high sensitivity. However, it was not always practical and limited by a risky and invasive procedure. Recently, a paradigm shift in diagnosing CA has occurred with technological advances in diagnostic imaging. Positive nuclear imaging with bone‐avid tracers (Grade 2 or 3 tracer uptake) in the absence of detectable monoclonal protein in serum or urine allows non‐histological diagnosis of ATTR‐CA reliably (Figure ).48 However, conventional histopathology and subsequent amyloid typing are still essential in all cases that do not meet such diagnostic criteria. Poor sensitivity of abdominal fat pad aspiration leads to delayed diagnosis (84% for AL, 15% for ATTRwt, and 45% for ATTRv, respectively).52 Moreover, the prevalence of monoclonal gammopathy of undetermined significance in patients with ATTR amyloidosis has been reported to be high.31, 53 Because treatment methods of each disease are totally different, endomyocardial biopsy should be carried out for definitive diagnosis.

Figure 2

Diagnostic algorithm for patients with suspected cardiac amyloidosis. −, negative test; +, positive test; 99mTc‐PYP, technetium pyrophosphate; AL‐CA, light‐chain cardiac amyloidosis; ATTR‐CA, transthyretin cardiac amyloidosis; ATTRv, hereditary transthyretin amyloidosis; ATTRwt, wild type transthyretin amyloidosis; CA, cardiac amyloidosis; CMR, cardiovascular magnetic resonance; EM, electron microscopy; IHC, immunohistochemistry; LMD/MS, Laser microdissection and mass spectrometry; sFLC, serum‐free light chain; sIFE, serum immunofixation electrophoresis; uIFE, urine immunofixation electrophoresis.

Treatment of transthyretin cardiac amyloidosis

A treatment strategy for CA includes both managing cardiovascular complications and treating the underlying disease process.

Management of cardiovascular complications

In CA, both the interstitial amyloid deposition and the subendocardial fibrosis due to ischaemia cause morphological and functional abnormalities. Amyloid infiltration, not myocyte hypertrophy, leads to the thickening of ventricular walls concomitant with normal to small cavity size with diastolic filling impairment due to reduced contractility. Moreover, atrial dysfunction due to amyloid deposit may further reduce diastolic filling. These pathological changes result in a reduced stroke volume concomitant with considerable elevated intra‐cardiac pressures.

Cardiac amyloidosis initially exhibits a clinical phenotype similar to HFpEF.54 Most of the well‐recognized anti‐HF drugs may be harmful due to its unique pathology. Angiotensin‐converting enzyme inhibitors, or angiotensin II receptor blockers, have the risk of producing profound hypotension because of activation of the renin–angiotensin–aldosterone system due to autonomic dysfunction. Moreover, beta‐blockers may be detrimental because they cause cardiac output reduction due to lowering heart rate and negative inotropic effect, resulting in profound hypotension. Calcium‐channel blocker or digitalis should be avoided because they bind irreversibly to amyloid fibrils and might cause serious side effects.55, 56 Diuretics remain the first‐line treatment for congestion in HF. A combination of loop diuretics and mineralocorticoid receptor antagonists is effective. In addition, the combined use of vasopressin V2 receptor antagonists (aquaretics) is useful to safely reach euvolaemia because the excessive use of loop diuretics (natriuretics) easily causes intravascular dehydration, resulting in pre‐renal azotaemia or symptomatic hypotension.57 Nevertheless, uncontrollable pleural effusion requiring frequent fluid drainage may suggest amyloid infiltration, which requires pleurodesis.58

Atrial fibrillation, which is very common in CA, may worsen HF, and its treatment is challenging. Rate and rhythm control in AF is very important because patients with CA have fixed stroke volumes and the cardiac output extremely depends on the heart rate. However, rate control options including beta‐blocker and calcium‐channel blocker are limited as the aforementioned reason. Amiodarone can be used to restore and maintain sinus rhythm safely. Catheter ablation for AF is not effective due to high recurrence rate.59

In CA, increased LV filling pressure, atrial contractile dysfunction caused by atrial amyloid infiltration, and AF highly cause intra‐cardiac thrombus, leading to an increased risk of thromboembolism.39, 60 Spontaneous echo contrast or decreased atrial appendage Doppler velocities (<40 cm/s) on transesophageal echocardiography is strongly suggestive of atrial contractile dysfunction.61 Thus, anticoagulation should be warranted in cases of AF or even sinus rhythm with evidence of atrial contractile dysfunction.

Pacemaker implantation

Many ATTR‐CA patients are facing at high risk of conduction system disorders requiring cardiac pacing for symptomatic atrioventricular block or bradycardia, and its indication follows current standard guidelines. However, there are no guidelines for the indication or optimal timing of prophylactic pacemaker implantation for asymptomatic ATTR‐CA patients. Although cardiac pacing can be helpful in symptomatic relief, there are no significant beneficial effects on patient survival.62, 63, 64, 65 Moreover, ventricular dyssynchrony or lead‐induced tricuspid regurgitation caused by right ventricular pacing might have deleterious impacts on haemodynamics.

Implantable cardioverter‐defibrillator

Sudden cardiac death (SCD) is a major problem in CA patients accounting for one quarter of all‐cause death.66 Although implantable cardioverter‐defibrillator (ICD) therapy can be a promising therapeutic option for high‐risk CA patients, the efficacy of ICD is controversial because electromechanical dissociation seems to be a dominant cause of SCD. Most clinical studies targeted AL‐CA patients with poor prognosis and a high frequency of SCD. Several cases reported that ICD has successfully terminated sustained ventricular arrhythmia in selected patients with AL‐CA.67 Based on such limited data, 2015 European Society of Cardiology guidelines recommend that ICDs should be considered in patients with AL‐CA or ATTR‐CA with a history of sustained ventricular arrhythmia and an estimated life expectancy of >1 year (Class IIa and Level C).68 On the other hand, ICD therapy is not recommended for primary prevention of SCD in ATTR‐CA patients because its effectiveness is questionable.69

Disease‐modifying treatments

Variants and aging cause dissociation of the TTR tetramer into non‐native monomers with low conformational stability, which misfolds into an amyloidogenic form and aggregates, leading to insoluble amyloid fibril formation.70 Subsequently, the extracellular deposition of amyloid fibrils in the myocardium causes ATTR‐CA. Thus, treatment options of the underlying misfolding protein disease include organ transplantation and investigational agents focusing on the TTR amyloidogenic pathway.

Organ transplantation

Liver transplantation (LT), which can remove the main source of circulating pathogenic TTR protein, has been considered a promising therapeutic choice to cure ATTR‐FAP.71 Unfortunately, there might be a serious risk of the development of progressive cardiomyopathy following LT among patients.72, 73 Autopsy findings revealed the preponderance of wild type TTR in cardiac amyloid deposits after LT,74 suggesting that paradoxical progressive deposition of wild type TTR on pre‐existing variant TTR amyloid deposits acting as templates. Thus, combined heart and liver transplant (CHLT) may be the most preferable alternative for advanced HF patients with ATTRv‐CA. According to the International Society for Heart and Lung Transplantation guideline, while young patients should be considered for CHLT to prevent systemic disease progression, elderly patients with cardiac‐dominant manifestations such as ATTR‐CA (wild type or Val122Ile variant) should be considered for isolated heart transplantation.75 Patients who have a definitive diagnosis of ATTR‐CA should perform an evaluation for CHLT or isolated heart transplantation as soon as possible to undergo organ transplantation in the earlier stage.

Transthyretin gene silencers

siRNA (patisiran)

A novel TTR‐targeted small‐interfering ribonucleic acid encapsulated lipid nanoparticle, namely, ALN‐TTR02 (patisiran), has developed. A Phase 3 APOLLO trial (NCT01960348), a randomized 18‐month trial, evaluated the efficacy and safety of the patisiran (0.3 mg/kg i.v., once every 3 weeks) in 225 ATTR‐FAP patients compared with the placebo group, demonstrated that patisiran significantly improved the quality of life (QOL) and clinical neuropathy scores, and ameliorated the disease progression in ATTR‐FAP patients.76 The drug was generally well tolerated, and the frequency of serious adverse effects (SAEs) was similar between the two groups. In a cardiac subgroup, patisiran significantly improved LV basal LS, lowered N‐terminal pro b‐type natriuretic peptide levels, and ameliorated abnormal LV geometric patterns including LVH compared with placebo groups.77, 78

Antisense oligonucleotides (inotersen)

Inotersen is a TTR‐directed antisense oligonucleotide, which interferes with hepatic TTR synthesis. A Phase 3 NEURO‐TTR trial (NCT01737398), a randomized 66‐week trial, evaluated the efficacy and safety of the subcutaneous inotersen (300 mg s.c., once weekly), in 172 ATTR‐FAP patients in the presence of cardiomyopathy and demonstrated that inotersen improved clinical manifestations and the neurological clinical scores in ATTR‐FAP patients. The most common SAEs were glomerulonephritis (3%) and life‐threatening thrombocytopaenia (3%). Thus, frequent laboratory monitoring is necessary to avoid these potential side effects.79

Transthyretin tetramer stabilizers

Tafamidis

Tafamidis is a novel molecule that can bind a thyroxine‐binding site of the TTR tetramer, resulting in inhibition of its dissociation into monomers, an important step in the TTR amyloid‐forming cascade.80 In a Phase 3 multicentre randomized ATTR‐ACT trial (NCT01994889), 441 patients with ATTR‐CA (wild type or variant) were randomly assigned in a 2:1:2 ratio to receive tafamidis 80 mg, 20 mg, or placebo orally every 24 h for 30 months.81 Patients with New York Heart Association (NYHA) class IV HF or an estimated glomerular filtration rate of <25 mL/min/1.73 m2 were excluded. Tafamidis reduced all‐cause mortality and cardiovascular‐related hospitalizations compared with placebo. Also, tafamidis significantly delayed decline in functional capacity assessed by the 6‐min walk distance and in QOL assessed by the Kansas City Cardiomyopathy Questionnaire‐Overall Summary/KCCQ‐OS. In sub‐analysis, tafamidis steadily improved all‐cause mortality and cardiovascular‐related hospitalizations in all subgroup (TTR genotype, NYHA class baseline, and tafamidis dosage) except for frequency of cardiovascular‐related hospitalizations in NYHA class III. This trial suggests that the severity of cardiac dysfunction at baseline drives the outcome yet leaves several questions to be solved. The preliminary experimental study demonstrated the effects of tafamidis on plasma TTR tetramer stability in a dose‐dependent manner.82 Thus, tailoring of the tafamidis dose to the patients in NYHA class III might need to achieve maximum kinetic stabilization. Because many randomized‐control trials for patients with systolic HF have suggested the need of months to recognize anti‐remodelling effects on the LV during the therapeutic intervention,83, 84 the observation period might not be sufficient to determine the drug treatment efficacy. The Phase 3 trial has further been extended up to 60 months and will end in 2021.

Diflunisal

Diflunisal, a non‐steroidal anti‐inflammatory drug, can stabilize TTR tetramers in vitro. In a Phase 3 randomized trial, 130 ATTR‐FAP patients with symptomatic neuropathy were randomly assigned to diflunisal 250 mg or placebo orally twice daily for 2 years.85 This trial demonstrated that diflunisal significantly reduced the progression of neurologic impairment and preserved QOL, compared with placebo. However, diflunisal had no beneficial effects on the improvement in cardiac status. Although the dosage of diflunisal generally were well tolerated in this trial, its cyclooxygenase inhibitory activity can cause renal and gastrointestinal damage, such as renal failure, gastric mucosal injury, volume overload, and hypertension. The potential for life‐threatening SAEs remains a major concern. Thus, the diflunisal remains to be used off‐label in the treatment of ATTR‐FAP.

AG10

Many pathologic TTR variants, such as Val30Met or Val122Ile, are of the loss‐of‐function type causing TTR destabilization and increase the risk of ATTR amyloidosis. However, a certain TTR variant, Thr119Met, is known as the gain‐of‐function type that super‐stabilizes a TTR tetramer, resulting in the prevention of ATTR amyloidosis. AG10 is a novel stabilizing compound that has a similar motif to the thyroxine‐binding site of the Thr119Met variant, specifically binding to TTR tetramer to inhibit TTR dissociation. In vitro study proved that AG10 had a stronger TTR tetrameric stability, compared with tafamidis and diflunisal.86 A randomized, double‐blind, placebo‐controlled, Phase 2 trial (NCT03458130) confirmed the safety and efficacy of AG10 in ATTR‐CA patients (wild type or variant).87 A Phase 3 trial (NCT03860935) in ATTR‐CA has been initiated.

Transthyretin amyloid disruptors

Doxycycline/tauroursodeoxycholic acid

Both the tetracycline antibiotic doxycycline and the antiapoptotic agent tauroursodeoxycholic acid (TUDCA) are effective in degrading non‐fibrillar TTR deposits.88 The concurrent treatment has been verified to attenuate disease progression in ATTR amyloidosis patients in limited clinical trials.89 An open‐label Phase 1/2 trial (NCT01855360) evaluated tolerability and efficacy of a combination of doxycycline and TUDCA on the disease progression in patients with ATTR‐CA (wild type or variant). Thirty‐eight ATTR‐CA patients were treated with the combination of TUDCA (250 mg orally three times daily) and doxycycline (100 mg orally twice daily) for 18 months. This trial has recently been completed, but the results have not been released yet.

Green tea

Epigallocatechin‐gallate, a well‐known major polyphenol in green tea, can inhibit TTR amyloid fibril formation and disaggregate amyloid deposits.90 Two observational studies revealed that 12 months of green tea consumption significantly reduced the LV mass by 6–13% evaluated by CMR in ATTRwt‐CA patients, suggesting that green tea consumption extracts have an inhibitory effect on the disease progression.91, 92 However, these studies are open‐label, observational studies and limited to small sample sizes.

Serum amyloid P component

Serum amyloid P component (SAP) is a normal plasma protein found in all types of amyloid deposits and stabilizes the amyloid deposit formation. The drug (R)‐1‐[6‐[(R)‐2‐carboxy‐pyrrolidin‐1‐yl]‐6‐oxo‐hexanoyl]pyrrolidine‐2‐carboxylic acid (CPHPC) has been developed to work as a competitive inhibitor of SAP binding to amyloid fibrils and efficiently depletes circulating SAP yet leaves some residual SAP in amyloid deposition.93 Subsequently, sequential treatment with CPHPC followed by anti‐SAP monoclonal antibodies efficiently elicited immunotherapeutic removal of amyloid deposits from key organs including liver in animal studies and in patients with systemic amyloidosis in early phase clinical trials (NCT01777243).94, 95, 96, 97

Conclusions

Both diagnosing and treating ATTR‐CA remain challenging. ATTR‐CA has been increasingly recognized as a cause of HFpEF. Large cohort study indicated that most patients diagnosed with ATTR‐CA were in NYHA class II (>70%).98 Also, with the rise of nuclear imaging techniques, the early non‐invasive diagnosis has become possible with high diagnostic accuracy.48 Moreover, the results of the ATTR‐ACT trial highlight the significance of high clinical suspicion and early diagnosis of ATTR‐CA given that tafamidis provides huge benefits to patients with early‐stage ATTR‐CA.81 For example, early screening for ATTR‐CA might be useful for initial evaluation of elderly patients with newly diagnosed HFpEF with LVH or elderly patients hospitalized for HF.2, 99 With the emergence of new efficient agents including tafamidis, tailor‐made medicine will be needed according to the individual disease stage and clinical phenotype. However, there are several problems to be solved: what should we do with treatment strategies for non‐responders? Or which is the most effective treatment for ATTR‐CA, a monotherapy or dual therapies (gene silencer and stabilizer)? Further studies are needed to address these unanswered questions. In addition, other novel innovative agents focused on the amyloidogenic process are ongoing. Thus, ATTR‐CA is becoming a treatable disease. However, misdiagnosis and delayed diagnosis still interfere with such benefits. Clinicians should be aware of the clinical entity and make much effort to identify ATTR‐CA as soon as possible.

Conflict of interest

None declared.

Author contributions

H.Y. and T.Y. drafted and revised the manuscript. Both authors discussed, read, and approved the submission of this manuscript to the journal.

Funding

No source of funding was declared for this study.