Article Text

Download PDFPDF

Transthyretin (ATTR) amyloidosis: clinical spectrum, molecular pathogenesis and disease-modifying treatments
  1. Yoshiki Sekijima1,2
  1. 1Department of Medicine (Neurology and Rheumatology), Shinshu University School of Medicine, Matsumoto, Japan
  2. 2Institute for Biomedical Sciences, Shinshu University, Matsumoto, Japan
  1. Correspondence to Dr Yoshiki Sekijima, Department of Medicine (Neurology and Rheumatology), Shinshu University School of Medicine, 3-1-1 Asahi, Matsumoto 390-8621, Japan; sekijima{at}


Transthyretin (ATTR) amyloidosis is a life-threatening, gain-of-toxic-function disease characterised by extracellular deposition of amyloid fibrils composed of transthyretin (TTR). TTR protein destabilised by TTR gene mutation is prone to dissociate from its native tetramer to monomer, and to then misfold and aggregate into amyloid fibrils, resulting in autosomal dominant hereditary amyloidosis, including familial amyloid polyneuropathy, familial amyloid cardiomyopathy and familial leptomeningeal amyloidosis. Analogous misfolding of wild-type TTR results in senile systemic amyloidosis, now termed wild-type ATTR amyloidosis, characterised by acquired amyloid disease in the elderly. With the availability of genetic, biochemical and immunohistochemical diagnostic tests, patients with ATTR amyloidosis have been found in many nations; however, misdiagnosis is still common and considerable time is required before correct diagnosis in many cases. The current standard first-line treatment for hereditary ATTR amyloidosis is liver transplantation, which allows suppression of the main source of variant TTR. However, large numbers of patients are not suitable transplant candidates. Recently, the clinical effects of TTR tetramer stabilisers, diflunisal and tafamidis, were demonstrated in randomised clinical trials, and tafamidis has been approved for treatment of hereditary ATTR amyloidosis in European countries and in Japan. Moreover, antisense oligonucleotides and small interfering RNAs for suppression of variant and wild-type TTR synthesis are promising therapeutic approaches to ameliorate ATTR amyloidosis and are currently in phase III clinical trials. These newly developed therapies are expected to be effective for not only hereditary ATTR amyloidosis but also wild-type ATTR amyloidosis.


Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


The amyloidoses are a group of gain-of-toxic-function diseases caused by the aggregation of a specific protein. According to the official International Society of Amyloidosis (ISA) Amyloid Fibril Protein Nomenclature List,1 31 different amyloid fibril proteins have been identified in humans, and the diseases are now classified according to the nature of the amyloid precursor protein. Transthyretin (TTR) is a representative amyloidogenic protein in humans. Variant TTR deposition causes autosomal dominant hereditary ATTR amyloidosis. To date, more than 120 TTR gene mutations have been reported and considerable genotype–phenotype correlations have been identified.2 The three main phenotypes of hereditary ATTR amyloidosis are familial amyloid polyneuropathy (FAP), familial amyloid cardiomyopathy (FAC) and familial leptomeningeal amyloidosis (table 1). On the other hand, wild-type ATTR deposition leads to an acquired amyloid disease, senile systemic amyloidosis (now termed wild-type ATTR (ATTRwt) amyloidosis), which typically presents later than hereditary ATTR amyloidosis.3 ,4 ATTR amyloidosis was historically considered to be a rare endemic disease. However, recent progress in diagnosis indicates that there are many patients with hereditary ATTR amyloidosis worldwide. Furthermore, ATTRwt amyloidosis is thought to be a common ageing-related disorder, as more than 10% of people over the age of 80 have wild-type TTR deposition in postmortem studies.3 ,5

Table 1

Clinical spectrum of transthyretin (ATTR) amyloidosis

ATTR amyloidosis used to be an incurable disease, but liver transplantation has been shown to be effective for FAP.6 While liver transplantation markedly improves the prognosis of patients with FAP, large numbers of patients are not suitable transplant candidates because of their age and/or advanced disease status. In addition, transplantation is not a viable option for FAC, familial leptomeningeal amyloidosis and ATTRwt amyloidosis. Therefore, it is desirable to develop more general alternative therapeutic strategies to ameliorate ATTR amyloidosis. Recently, the molecular, biological and chemical pathogeneses of ATTR amyloidosis have been clarified,7 ,8 and several novel disease-modifying treatments have been developed.9 ,10 Among these, the TTR tetramer stabilisers, diflunisal11 and tafamidis,12 have been shown to inhibit progression of polyneuropathy and preserve quality of life in patients with FAP in randomised controlled trials. This review focuses on recent progress in our understanding of the clinical spectrum, molecular pathogenesis and development of novel therapeutic approaches for ATTR amyloidosis.

Clinical spectrum of ATTR amyloidosis

Hereditary ATTR amyloidosis

Hereditary ATTR amyloidosis is the most common form of hereditary amyloidosis caused by TTR gene mutation. Hereditary ATTR amyloidosis is a life-threatening, gain-of-toxic-function disease that may present with peripheral neuropathy, autonomic neuropathy, cardiomyopathy, ophthalmopathy and/or leptomeningeal amyloidosis. FAP, FAC and familial leptomeningeal amyloidosis are the three main phenotypes of hereditary ATTR amyloidosis (table 1). However, the phenotype is not always uniform, even in patients with the same mutation. For example, most patients with Val30Met (p.Val50Met) mutation develop both neuropathy and cardiomyopathy during the course of the disease. Cardiomyopathy could be an initial and/or main symptom in patients with Val30Met mutation, although FAP is the most common phenotype in such cases.

Familial amyloid polyneuropathy

FAP is the most common clinical phenotype of hereditary ATTR amyloidosis. The Val30Met mutation, found worldwide, is the most common TTR mutation and is responsible for the well-known large foci of patients with FAP in Portugal, Sweden and Japan.2 Patients with Val30Met mutation (ATTR Val30Met) are classified into two groups, early-onset and late-onset. Early-onset ATTR Val30Met patients are seen in the endemic foci in Portugal and Japan, and are characterised by onset before the age of 50 years, high penetrance rate, predominant loss of superficial sensation including nociception and thermal sensation, and marked autonomic dysfunction, including orthostatic hypotension, sexual impotence, neurogenic bladder and disturbed bowel movement. In addition, various types of cardiac conduction blocks frequently appear, requiring the implantation of a pacemaker.2

In contrast to the early-onset patients, ATTR Val30Met patients in non-endemic areas and the endemic focus in Sweden typically have onset at a late age, usually after the age of 60 (late-onset ATTR Val30Met). The clinical characteristics of late-onset ATTR Val30Met include a low penetrance rate, often with lack of a family history, loss of all sensory modalities, relatively mild autonomic dysfunction and male predominance.13 Amyloid cardiomyopathy is also common in late-onset ATTR Val30Met.13 The basis of this variability is uncertain and may be attributable to other genetic factors, epigenetic factors, epistasis or environmental factors.

In patients with specific TTR mutations, such as Leu58His (p.Leu78His), Ile84Ser (p.Ile104Ser) and Tyr114His (p.Tyr134His), neuropathy starts in the upper extremities as carpal tunnel syndrome.2

Familial amyloid cardiomyopathy

FAC is another common clinical phenotype of hereditary ATTR amyloidosis, and Val122Ile (p.Val142Ile) is the most common mutation responsible for this disease. It was reported that ≥3% of African-Americans are heterozygous for Val122Ile mutation and develop late-onset cardiac amyloidosis,14 ,15 although the exact penetrance is unknown. The frequencies of Val122Ile in Caucasian, Hispanic and Asian populations are very low.15 Other representative TTR mutations responsible for FAC include Ser50Ile (p.Ser70Ile), Thr60Ala (p.Thr80Ala), Ile68Leu (p.Ile88Leu) and Leu111Met (p.Leu131Met).2

Patients with FAC show congestive heart failure, intractable arrhythmia and conduction blocks and occasionally require implantation of a pacemaker and/or implantable cardioverter defibrillator. Typical ECG findings of patients with FAC include low voltage in the standard limb leads and QS pattern in the right precordial leads with conduction blocks. Symmetrical thickening of interventricular septum and ventricular walls is seen on ultrasound and MRI. Global or patchy subendocardial late gadolinium enhancement of the myocardium is also a characteristic MRI finding in cases of cardiac amyloidosis. Left ventricular systolic function is usually preserved in the early stage of the disease. Technetium-99 m pyrophosphate (99mTc-PYP), 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (99mTc-DPD) and 99mTc-hydroxy methylene diphosphonate (99mTc-HPD) myocardial scintigraphy are also valuable for detecting cardiac ATTR amyloid deposition.

Familial leptomeningeal amyloidosis

Cerebral amyloid angiopathy accompanied by leptomeningeal amyloidosis are the main clinical features of patients with hereditary ATTR amyloidosis with several specific TTR gene mutations, such as Asp18Gly (p.Asp38Gly), Ala25Thr (p.Ala45Thr) and Tyr114Cys (p.Tyr134Cys).2 In this disease, the main source of variant TTR is the choroid plexus and amyloid deposition is observed in the media and adventitia of medium-sized and small arteries, arterioles and veins of the cortex and leptomeninges. These pathological changes induce cerebral infarction, cerebral haemorrhage, subarachnoid haemorrhage and/or hydrocephalus, and cause various central nervous dysfunctions, such as spastic paralysis, ataxia, convulsions and dementia. Leptomeningeal amyloidosis develops only rarely in ATTR Val30Met patients, but can be a serious complication in patients with post-transplant, as choroid plexus continues to produce variant TTR after transplantation.16

Other organ involvement in hereditary ATTR amyloidosis

Ocular involvement, including vitreous opacity, glaucoma, dry eye and ocular amyloid angiopathy, is common and occurs in most hereditary patients with ATTR amyloidosis during the course of the disease.17

Gastrointestinal symptoms, including recurrent vomiting, constipation and/or severe watery diarrhoea are frequently observed in hereditary patients with ATTR amyloidosis, especially in early-onset ATTR Val30Met patients in endemic foci.18

Renal involvement, including nephritic syndrome and progressive renal failure, occurs in about one-third of patients in Portugal;19 however, severe renal dysfunction rarely occurs in patients with late-onset disease.

Wild-type ATTR (ATTRwt) amyloidosis

Wild-type ATTR deposition in systemic organs is thought to be a common ageing-related phenomenon similar to Aβ deposition in the brain, as post-mortem studies showed that more than 10% of people over the age of 80 have ATTR amyloid deposition.3 ,5 However, only a limited number of ATTRwt amyloidosis patients have been reported to date, and the exact prevalence of the disease is unknown, mainly because (1) a substantial amount of wild-type TTR deposition is thought to be necessary to develop clinical symptoms or signs, and (2) ATTRwt amyloidosis is markedly underdiagnosed. Typically, patients with ATTRwt amyloidosis show cardiac manifestations, such as congestive heart failure, atrial fibrillation and intractable arrhythmia, and occasionally require implantation of an implantable cardioverter defibrillator. Carpal tunnel syndrome is another common clinical manifestation of ATTRwt amyloidosis and develops as an initial symptom of the disease in many patients.20 Cardiogenic embolism and mild to moderate renal dysfunction are also frequently seen in such cases. There are no effective disease-modifying therapies for ATTRwt amyloidosis, and mean survival period from the onset of congestive heart failure symptoms is 75 months.

Diagnosis of ATTR amyloidosis

Diagnosis of hereditary ATTR amyloidosis

In addition to the clinical symptoms described above, proven amyloid deposition in biopsy specimens and identification of disease-causing mutations in the TTR gene are necessary to establish the diagnosis. Deposition of amyloid in tissue can be demonstrated by Congo red staining of biopsy materials. Tissues suitable for biopsy include the subcutaneous fatty tissue of the abdominal wall and gastroduodenal mucosa. Amyloid deposition can also be detected in the sural nerve, rectal mucosa, salivary glands and endomyocardial biopsy specimens, as well as tenosynovial tissues obtained at carpal tunnel release surgery. Ideally, immunocytochemical study of the amyloid-positive tissue should be performed to confirm the presence of amyloid precursor protein. It should be noted that negative biopsy findings do not rule out amyloidosis because amyloid deposition is often patchy.2 ,21 ,22 Therefore, TTR gene analysis should be performed in parallel with tissue biopsy when hereditary ATTR amyloidosis is suspected. Hereditary ATTR amyloidosis can be excluded if genetic testing for TTR yields a negative result (no mutation).

Diagnosis of wild-type ATTR (ATTRwt) amyloidosis

Clinical manifestations, proven ATTR amyloid deposition in biopsy specimens and confirmation of wild-type TTR genotype are necessary to establish the diagnosis of ATTRwt amyloidosis. Immunocytochemical or proteomic analysis of the amyloid positive tissue is essential for the diagnosis of ATTRwt amyloidosis to confirm the type of amyloid protein involved in the disease. In most diagnosed cases, tissue samples are obtained by endomyocardial biopsy; however, this is rarely performed due to its high degree of invasiveness. Abnormal uptake of 99mTc in myocardial scintigraphy is highly suggestive of ATTR cardiac amyloidosis (see FAC), although histopathological confirmation is necessary. Surgical skin biopsy including the deep subcutaneous fat pad is a useful alternative histopathological tool for diagnosis of ATTRwt amyloidosis.23

If wild-type TTR-derived amyloid deposition is found in tenosynovial tissues obtained at carpal tunnel surgery, additional screening for amyloid deposition is necessary to confirm ‘systemic ATTRwt amyloidosis’, as wild-type TTR derived amyloid deposition localised to tenosynovial tissues, ‘localised ATTRwt amyloidosis’, is common in elderly men.20

Molecular pathogenesis of ATTR amyloidosis

Structure and function of TTR

TTR is a 55 kDa homotetrameric protein composed of 127-residue β-sheet-rich subunits synthesised in and secreted by the liver into the bloodstream.24 The main physiological functions of TTR are transport of thyroxine (T4) and retinol binding protein–vitamin A complex (holoRBP). The TTR tetramer contains two identical T4-binding sites located in a channel at the centre of the molecule.24 However, a very small proportion (<1%) of the TTR tetramers bind to T4 and most T4 binding sites are unoccupied in human serum due to the high affinity of T4 for thyroid-binding globulin, the major carrier of T4, and because the serum concentration of T4 is relatively low (0.1 μM) compared to that of TTR (3.6–7.2 μM). The TTR tetramer also contains four symmetry-related holoRBP binding sites; however, only two are concurrently accessible as a result of overlapping footprints25 and about 50% of TTR tetramer is actually bound to holoRBP in serum. TTR is also synthesised in and secreted by the choroid plexus into the cerebral spinal fluid. TTR protein in eye tissue is synthesised mainly in the retinal pigment epithelium of the eye.26 TTR is expressed during embryonic development and throughout life; however, TTR knockout mice showed a normal phenotype, suggesting that TTR may not be essential.27

Process of ATTR amyloid fibril formation

The conversion of the native folded TTR tetramer into insoluble amyloid fibrils is a dynamic, multistep process. It is widely accepted that TTR tetramer dissociation into monomers is the initial and rate-limiting step that allows subsequent partial misfolding and misassembly, leading to the formation of ATTR amyloid fibrils, as well as several other aggregate morphologies.28 ,29 Furthermore, energetic studies of a large number of recombinant TTR variants7 have suggested that amyloidogenic mutations destabilise the native quaternary and tertiary structures of TTR, thereby inducing conformational changes. Amyloidogenic mutations variably decrease protein thermodynamic stability, favouring the shift toward unfolded monomeric species, and kinetic instability, promoting tetramer dissociation and monomer unfolding. These destabilising effects also impact the efficiency of TTR secretion from the cell by endoplasmic reticulum-associated degradation, modulating disease severity.7 That is, the quality control system of the cell protects individuals from severe ATTR amyloidosis by lowering the amyloidogenic precursor concentration.

Amyloid formation is generally considered to be a nucleation-dependent process, where fibril growth requires the formation of an oligomeric high-energy quaternary structure before monomer addition can become energetically favourable; however, amyloid formation by the TTR monomer proceeds by spontaneous energetically downhill polymerisation.30 It is likely that C-terminal fragments of the TTR protein produced by trypsin proteolysis promote TTR aggregation,31 especially in ATTRwt amyloidosis,32 late-onset ATTR Val30Met and ATTR non-Val30Met patients.33 Extracellular factors also play a role in modulating ATTR amyloid fibril formation, as Bourgault et al34 reported that sulfated glycosaminoglycans, especially heparin, accelerate TTR amyloidogenesis by quaternary structural conversion.

Cytotoxicity of TTR and ATTR amyloid

Historically, it was assumed that accumulation of amyloid fibrils alone causes the tissue damage in various amyloidoses. However, recent studies indicated that low molecular weight oligomers and protofibrils are toxic to the cells in protein misfolding diseases, including Alzheimer's disease. In addition, free immunoglobulin light chain (soluble monomer of amyloid precursor protein) was shown to be responsible for organ damage in light chain (AL) amyloidosis.

In ATTR amyloidosis, it is obvious that amyloid fibrils themselves induce tissue damage by direct compression, obstruction and local blood circulation failure. Carpal tunnel syndrome is a representative disorder induced by massive deposition of amyloid fibrils, which is relieved by carpal tunnel release surgery. Vitreous opacity and glaucoma are other examples of tissue damage induced by amyloid fibrils, which can be relieved by ophthalmological surgery. On the other hand, cell biological studies35 ,36 have demonstrated that mature ATTR amyloid fibrils are not toxic, whereas monomeric or very low molecular weight TTR oligomers are cytotoxic. It was also reported that low molecular weight TTR oligomers effectively induced calcium influx via voltage-gated calcium channels, subsequently causing cytotoxicity.37 In addition, Saraiva and coworkers demonstrated that TTR-induced toxicity is mediated by the receptor for advanced glycation end (RAGE) products and that activation of the RAGE leads to endoplasmic reticulum stress, activation of ERK1/2 and caspase-dependent apoptosis.38 ,39 Moreover, Reixach and colleagues40 showed that the oxidised form of TTR is cytotoxic to a human cardiomyocyte cell line, suggesting that ageing-related oxidative modifications of TTR may contribute to the onset of the late-onset ATTR amyloidosis, including ATTRwt amyloidosis.

Disease-modifying therapy for ATTR amyloidosis

Liver transplantation

Until 1990, ATTR amyloidosis was considered an incurable disease; however, liver transplantation has been shown to be an effective therapeutic strategy for ameliorating FAP. Transplantation replaces the variant TTR gene by the wild-type gene in the liver, the main source of serum TTR protein (table 1, figure 1), and the serum concentration of variant TTR decreases rapidly after the surgery. Long-term observations of transplant recipients have clearly demonstrated the histopathological regression of amyloid deposits in abdominal fat tissue,41 with an overall patient survival rate at 5 years of >80%.42 Survival analyses also demonstrated that TTR genotype is one of the most influential factors, as the 5-year and 10-year patient survival rates in the Val30Met group (82% and 74%, respectively) were significantly better than those in the non-Val30Met group (59% and 44%, respectively).42 ,43 Although prognosis of transplanted early-onset ATTR Val30Met patients is excellent,44–46 survival of transplanted patients does not differ from that of non-transplanted patients in late-onset ATTR Val30Met.43 ,45 It was also reported that female transplanted late-onset ATTR Val30Met patients had significantly improved survival compared with corresponding male patients.45 Progression of cardiac amyloid deposition is of more significance in non-Val30Met and late-onset Val30Met transplant recipients, which largely explains the less favourable long-term survival rates in these groups.45 The effects of liver transplantation on neuropathy were evident as progression of autonomic and peripheral neuropathy was stopped or even slightly improved in most patients.46

Figure 1

Mechanism of ATTR amyloid fibril formation and sites of action of therapeutic strategies. *The clinical effectiveness of antisense oligonucleotides, small interfering RNA, combined use of doxycycline and tauroursodeoxycholic acid, and immune therapy has not yet been proven. TTR, transthyretin.

While liver transplantation is the current standard therapeutic strategy for FAP, it has several limitations, including the requirement for surgery, long-term post-transplantation immunosuppressive therapy and progression of eye, cardiac and leptomeningeal amyloidosis after transplantation. Eye and leptomeningeal deposition of TTR are not relieved by liver transplantation due to TTR synthesis by the retinal pigment epithelium and choroid plexus. On the other hand, cardiac amyloidosis progresses in some patients even after liver transplantation, because wild-type TTR deposition often continues.47 Furthermore, large numbers of patients are not suitable transplant candidates because of their age and/or advanced disease status. Finally, transplantation is not a viable option for FAC, familial leptomeningeal amyloidosis and ATTRwt amyloidosis.3 ,4 Therefore, it is necessary to develop more general, convenient and non-invasive alternative therapies for ATTR amyloidosis.

TTR tetramer stabiliser

Another therapeutic option for ameliorating ATTR amyloidosis is to stabilise the native TTR tetramer structure, as tetramer dissociation is necessary for ATTR amyloid fibril formation (figure 1). In fact, a trans-suppressor mutation, Thr119Met (p.Thr139Met), was shown to prevent disease in ATTR Val30Met/Thr119Met compound heterozygotes by stabilising the tetrameric native state.48 On the other hand, it was reported that binding of T4, a natural TTR ligand, to two sites in the TTR tetramer, stabilises the native state over the dissociative transition state and thus inhibits amyloidogenesis in vitro. However, more than 99% of the T4 binding sites are unoccupied in human serum, as described above (see MOLECULAR PATHOGENESIS OF ATTR AMYLOIDOSIS). Based on these findings, Kelly and co-workers performed robust screening and structure-based drug design to find small molecules that bind tightly and selectively to the TTR tetramer.49 ,50 Typical TTR stabilisers have two aromatic substructures and a linker, and two small molecules, diflunisal and tafamidis, have been shown to slow the rate of disease progression of FAP in randomised clinical trials10–,12 (table 1, figure 1).


Diflunisal, a salicylic acid derivative, is a well-known oral non-steroidal anti-inflammatory drug that was developed more than 40 years ago. Previous in vitro analysis49 ,50 showed that diflunisal is able to bind to the T4 binding sites of TTR (Kd1=75 nM, Kd2=1.1 μM) and inhibit ATTR amyloid fibril formation at relatively low concentrations. Based on these findings, phase I51 and II52 clinical trials of diflunisal were conducted and showed that diflunisal at a dose of 250 mg twice daily successfully complexed to the T4 binding sites and stabilised circulating TTR tetramers. Subsequently, Berk and co-workers conducted an investigator-initiated randomised double-blind placebo-controlled multicentre phase III trial of diflunisal from 2006 through 2012.11 In this study, 130 patients with FAP (mean age 59.7 years; Val30Met 54.6%, non-Val30Met 45.4%) were enrolled and randomly assigned to receive diflunisal at 250 mg or placebo twice daily for 2 years. This phase III trial demonstrated the clinical effect of diflunisal on polyneuropathy progression, as the primary endpoint, change from baseline in the Neuropathy Impairment Score plus 7 nerve test (NIS+7), deteriorated by 8.7 points in the diflunisal group and 25.0 points in the placebo group (p<0.001). This study also demonstrated that diflunisal preserved quality of life in patients with FAP. Gastrointestinal, renal, cardiac and blood-related adverse events occurred at similar rates between diflunisal and placebo groups,11 although the adverse effect of cyclooxygenase-1 inhibitory activity of diflunisal (eg, gastric mucosal injury, renal dysfunction and congestive heart failure) should be monitored carefully.

A long-term (up to 116 months) open-label study of diflunisal53 indicated that the clinical effects of diflunisal were sustained even after 2 years of treatment. However, clinical symptoms deteriorated slowly in most patients, indicating that diflunisal cannot stop disease progression of FAP completely.


Tafamidis meglumine (Vyndaquel; Pfizer Inc) is a newly developed oral agent that occupies the T4 binding sites of TTR (Kd1=2 nM, Kd2=154 nM) and potently stabilises the tetrameric structure.54 Coelho et al12 conducted a randomised double-blind placebo-controlled multicentre phase II/III trial (Fx-005) from 2007 through 2009 and showed the efficiency of tafamidis on polyneuropathy progression in patients with FAP. In this 18-month study, 128 ATTR Val30Met FAP patients (mean age, 39.1 years) with stage I (disease limited to the lower limbs, walking without any help) were randomly assigned to receive tafamidis meglumine at a dose of 20 mg (tafamidis 12.2 mg) or placebo once daily. Coprimary end points of the Fx-005 study were treatment group differences in the Neuropathy Impairment Score–Lower Limbs (NIS-LL) responder analysis (<2-point worsening) and change from baseline in Norfolk Quality of Life–Diabetic Neuropathy total score (TQOL) in the intent-to-treat (ITT) population. These end points were also evaluated in the efficacy-evaluable (EE) population. No significant differences were observed between the tafamidis and placebo groups for the coprimary end points, NIS-LL responder analysis (45.3% vs 29.5%, respectively; p=0.068) or change in TQOL (2.0 vs 7.2, respectively; p=0.116) in the ITT population. However, in the EE population, significantly more tafamidis patients than placebo controls were NIS-LL responders (60.0% vs 38.1%, respectively; p=0.041), and tafamidis patients had better preserved TQOL (0.1 vs 8.9, respectively; p=0.045). In addition, tafamidis was shown to improve nutritional status of patients with FAP, as reflected by an increase in the modified body mass index.

In a 12-month open-label extension study of tafamidis (Fx-006),55 patients who continued on tafamidis had stable rates of change in NIS-LL (from 0.08 to 0.11/month; p=0.60) and TQOL (from −0.03 to 0.25; p=0.16), while the monthly rate of change in NIS-LL declined (from 0.34 to 0.16/month; p=0.01), as did TQOL score (from 0.61 to −0.16/month; p<0.001) in patients switched from placebo. Patients treated with tafamidis for 30 months had 55.9% greater preservation of neurological function as measured by the NIS-LL than patients in whom tafamidis was initiated later, showing that treatment benefits were greater when tafamidis was begun earlier. The TTR tetramer stabilisation effect of tafamidis in non-Val30Met patients was also confirmed in a phase II open-label study.56

On the basis of these results, tafamidis was approved by the European Medicines Agency (EMA) in 2011 for the treatment of early-stage (stage I) FAP and by the Japanese Pharmaceuticals and Medical Devices Agency (PMDA) in 2013 for the treatment of FAP (any stage). In December 2013, Pfizer Inc initiated a phase III randomised double-blind placebo-controlled study to evaluate the efficacy, safety and tolerability of an oral dose of 20 or 80 mg of tafamidis meglumine in patients with ATTR cardiomyopathy, including ATTRwt amyloidosis and FAC.

Gene therapy

Gene silencing approaches using antisense oligonucleotides (ASOs)57–59 or small interfering RNAs (siRNAs)60 are promising therapeutic strategies for amelioration of ATTR amyloidosis, as reduction of amyloid fibril precursor protein level has proven to be effective in AL amyloidosis (chemotherapy to reduce amyloidogenic immunoglobulin light chain) and amyloid A protein (AA) amyloidosis (biological agents and immunosuppressive agents to reduce serum amyloid A protein). In addition, liver transplantation has proven to be effective in FAP by reduction of serum variant TTR concentration. Furthermore, TTR knockout mice did not show any abnormal phenotype,27 suggesting that it may be possible to suppress TTR protein without serious adverse effects. Based on the theoretical rationale described above, several TTR-specific ASOs and siRNAs have been developed and clinical trials are currently underway (table 1 and figure 1).

Antisense oligonucleotides (ASOs)

ASOs are short chemically modified oligonucleotides designed to prevent the expression of a target protein by selectively binding to the complementary RNA in cells, thereby preventing translation. Benson et al57 first identified TTR-specific ASOs that can potently suppress hepatic TTR mRNA and serum TTR protein levels of transgenic mice carrying human Ile84Ser TTR. Subsequently, Isis Pharmaceuticals Inc developed ISIS-TTRRX that binds within the 3′ non-translated portion of the human TTR mRNA and results in degradation of all forms of TTR mRNA including wild-type. In a human TTR transgenic mouse model and cynomolgus monkeys, subcutaneous injection of ISIS-TTRRX decreased liver TTR mRNA as well as circulating plasma TTR protein levels in a dose-dependent manner (up to 80%).59 Based on these findings, Isis Pharmaceuticals Inc conducted phase I clinical trials of this agent in normal healthy volunteers and showed that subcutaneously administered ISIS-TTRRx (50–400 mg) produced dose-dependent reductions in plasma TTR protein levels. Mild adverse events, including injection site pain and somnolence, were reported in the phase I studies.59 In February 2013, a phase III randomised double-blind placebo-controlled study was initiated to evaluate the long term safety and efficacy of ISIS-TTRRX in patients with FAP. In this phase III study, patients have been randomly assigned to receive ISIS-TTRRx at 300 mg or placebo by subcutaneous injection three times on alternate days in the first week, then once weekly for 64 weeks. The estimated study completion date is November 2016.

ASOs may have the potential to ameliorate familial leptomeningeal amyloidosis, the most intractable phenotype of ATTR amyloidosis, as intraventricular administration of TTR-specific ASOs significantly reduced TTR mRNA and protein levels in choroid plexus in a human TTR transgenic mouse model.58

Small interfering RNA

RNA interference (RNAi) is a natural mechanism for controlling gene expression in which sequence-specific siRNAs (21–23 nucleotides long double-stranded RNA molecules) are able to target and cleave complementary mRNA. In practice, siRNA can be produced synthetically and then be directly introduced into the cell, making it possible to knock down target genes in various diseases.

Alnylam Pharmaceuticals Inc developed several TTR-specific siRNA formulations and ALN-TTR02, an anti-TTR siRNA encapsulated in a second-generation formulation of lipid nanoparticles, showed the most potent pharmacological action. The siRNA in ALN-TTR02 targets a conserved sequence in the 3′ untranslated region of TTR mRNA, thereby suppressing all forms of TTR, including the wild-type. In cynomolgus monkeys that received ALN-TTR02 in a single dose of 0.3 mg/kg, the mean %TTR knock-down at the nadir level was approximately 90%, with >70% suppression persisting on day 28. In a single-dose phase I clinical trial, intravenous administration of ALN-TTR02 (0.15–0.3 mg/kg) reduced serum TTR protein levels by 82.3–86.8%, with reductions of 56.6–67.1% on day 28. Moderate infusion-related reactions occurred in 7.7% of participants receiving ALN-TTR02.60 Based on these results, a phase III randomised double-blind placebo-controlled study (APOLLO Phase III study) was initiated in November 2013, designed to evaluate the efficacy and safety of ALN-TTR02 in patients with FAP. In the APOLLO Phase III study, patients are randomly assigned to receive ALN-TTR02 0.3 mg/kg or placebo by intravenous injection every 3 weeks for 78 weeks. The estimated study completion date of the APOLLO Phase III study is May 2017. In addition, an open-label phase II study of ALN-TTRsc, a subcutaneous formulation of siRNA, is underway to evaluate the safety, pharmacokinetics, pharmacodynamics and exploratory clinical activity of this agent in patients with ATTR cardiac amyloidosis.

Other therapeutic strategies

Doxycycline/tauroursodeoxycholic acid

Combination therapy with doxycycline, an antibiotic that disrupts TTR amyloid fibrils, and tauroursodeoxycholic acid (TUDCA), a biliary acid that reduces non-fibrillar TTR aggregates, was demonstrated to have a synergistic effect on removal of tissue TTR deposits in mouse models carrying Val30Met TTR (table 1, figure 1). A phase II open-label study evaluating the pharmacokinetics, efficacy, safety and tolerability of doxycycline (200 mg/day) and TUDCA (750 mg/day) is currently underway at the University of Pavia, Italy. Preliminary data demonstrated that the combined treatment has an acceptable toxicity profile.61

Immune therapy

Active or passive immunisation with misfolded TTR proteins (figure 1) may have therapeutic benefit in ATTR amyloidosis, as suggested by previous studies using similar immunisation approaches for other protein misfolding diseases. Saraiva and coworkers62 immunised transgenic mice carrying Val30Met TTR with the highly destabilised Tyr78Phe (p.Tyr98Phe) TTR variant, and found that immunised mice showed significant reductions in non-fibrillar and fibrillar TTR deposition in the gastrointestinal tract compared to non-immunised controls. However, the immunisation also induced inflammatory cell infiltration and hyperplasia of lymphoid tissues at the site of TTR deposition. Immunisation-related inflammation is one of the most important issues to be resolved, as the first clinical trial of Aβ vaccination for Alzheimer's disease was halted because of the occurrence of subacute meningoencephalitis.

On the other hand, Ando and coworkers developed a monoclonal anti-TTR antibody that binds specifically to ATTR amyloid fibrils.63 This antibody may have potential to suppress ATTR amyloid deposition, although further investigations are necessary to determine its efficacy and safety profile.

In addition, immune therapy with a monoclonal antibody against human serum amyloid P component (SAP) was reported to successfully remove amyloid deposition in a mouse model of AA amyloidosis.64 Anti-SAP antibody therapy may be applicable to ATTR amyloidosis, as SAP is a major component of amyloid deposit in all types of amyloidosis.


This review described recent advances in the understanding of the clinical diversity, molecular pathogenesis and novel therapeutic approaches for ATTR amyloidosis. The novel therapeutic approaches are disease-modifying therapies that target different steps of the amyloid fibril formation cascade and can stop the cardinal pathogenesis of amyloidosis. Based on their mechanisms of action, TTR tetramer stabilisers, ASOs and siRNAs are expected to be effective for not only hereditary ATTR amyloidosis but also ATTRwt amyloidosis. Early diagnosis and therapy will become increasingly important in ATTR amyloidosis, as none of the therapies available can recover the organ damage already established. Recent pathological and biochemical research demonstrated that most neurodegenerative disorders involve protein misfolding, in an analogous fashion to ATTR amyloidosis. It is expected that the newly developed disease-modifying therapies for ATTR amyloidosis will open a new era of protein misfolding disease treatment.


The author would like to thank Professor Shu-ichi Ikeda and Prof Jeffery W Kelly for their helpful advice. This study was supported by a grant from Amyloidosis Research Committee, the Ministry of Health, Labour and Welfare, Japan, and a Group Research Grant for the Pathogenesis and Therapy for Intractable Neuropathy in Japan.



  • Correction notice This article has been corrected since it was published Online First. The provenance and peer review statement has been corrected.

  • Competing interests YS has received royalties from Pfizer related to tafamidis patents. YS has received speaker honoraria from Pfizer.

  • Provenance and peer review Commissioned; externally peer reviewed.