Advances in the treatment of hereditary transthyretin amyloidosis: A review

Abstract Introduction Amyloid transthyretin amyloidosis (ATTR) is a progressive and often fatal disease caused by the buildup of mutated (hereditary ATTR [hATTR]; also known as ATTR variant [ATTRv]) or normal transthyretin (wild‐type ATTR) throughout the body. Two new therapies—inotersen, an antisense oligonucleotide therapy, and patisiran, an RNA interference therapy—received marketing authorization and represent a significant advance in the treatment of amyloidosis. Herein, we describe the clinical presentation of ATTR, commonly used procedures in its diagnosis, and current treatment landscape for ATTR, with a focus on hATTR. Methods A PubMed search from 2008 to September 2018 was conducted to review the literature on ATTR. Results Until recently, there have been few treatment options for polyneuropathy of hATTR. Inotersen and patisiran substantially reduce the amyloidogenic precursor protein transthyretin and have demonstrated efficacy in patients with early‐ and late‐stage disease and in slowing or improving neuropathy progression. In contrast, established therapies, such as liver transplantation, typically reserved for patients with early‐stage disease, and tafamidis, indicated for the treatment of early‐stage disease in Europe, or diflunisal, a nonsteroidal anti‐inflammatory drug that is used off‐label, are associated with side effects and/or unclear efficacy in certain patient populations. Thus, inotersen and patisiran are positioned to be the preferred therapeutic modalities. Conclusions Important differences between inotersen and patisiran, including formulation, dosing, requirements for premedications, and safety monitoring, require an understanding and knowledge of each treatment for informed decision making.


| INTRODUC TI ON
Amyloid transthyretin (TTR) amyloidosis (ATTR amyloidosis) is a rare, progressive, and fatal disease caused by the buildup of amyloid fibrils in organs and tissues (Connors et al., 2016;Koike et al., 2016;Koike, Yasuda, et al., 2018). It can be caused by the buildup of mutated TTR, referred to as hereditary ATTR (hATTR) or ATTR variant (ATTRv), or by the buildup of normal TTR, referred to as wildtype ATTR (wtATTR) (Connors et al., 2016;Koike et al., 2016;Koike, Yasuda, et al., 2018). hATTR is transmitted in an autosomal dominant manner, and more than 130 TTR gene mutations have been identified thus far (Rowczenio & Wechalekar, 2018). The mechanism by which wild-type TTR becomes amyloidogenic is poorly understood.
Until recently, therapeutic options have been limited for patients with hATTR. This review describes the clinical presentation and commonly used diagnostic procedures and provides an overview of the current treatment landscape for ATTR, with a focus on hATTR.

| Clinical presentation
Hereditary ATTR manifests as sensorimotor neuropathy, autonomic neuropathy, cardiomyopathy, and nephropathy, whereas wtATTR mainly affects the heart, especially in men ˃60 years old (Table 1) (Carr et al., 2016;Conceicao et al., 2016;Connors et al., 2016;Maurer et al., 2016). Of note, manifestations consistent with peripheral and autonomic neuropathy involvement have been observed in patients with wtATTR (Connors et al., 2016;Maurer et al., 2016). A classic feature of hATTR is length-dependent peripheral sensorimotor neuropathy (Cappellari et al., 2011;Carr et al., 2016). Symptoms typically progress in a distal to proximal direction with feet affected first followed by upper limb involvement (Carr et al., 2016). As the disease progresses, patients experience increasing lower limb muscle weakness, walking difficulty, imbalance, and sensory loss (Carr et al., 2016;Coelho et al., 2017;Maurer et al., 2016). Neurologic symptoms and the pattern of progression of neurologic manifestations may vary according to mutation type (Cappellari et al., 2011;Carr et al., 2016).
Recognizing hATTR as a cause of polyneuropathy may be challenging because of its similarity to more common causes of neuropathy.
Symptoms that may help to distinguish hATTR from other causes of neuropathy include neuropathic pain (often described as lightning pain), autonomic dysfunction, absence of ataxia, small fiber sensory loss above the wrist, and weakness in the upper limbs, and are more common in hATTR (Lozeron et al., 2018). Autonomic neuropathy can manifest as orthostatic hypotension and/or sexual dysfunction but may not be reported (Carr et al., 2016;Koike, Nakamura, et al., 2018;Lozeron et al., 2018;Maurer et al., 2016;Wixner, Tornblom, Karling, Anan, & Lindberg, 2018). Therefore, patients presenting with progressive length-dependent neuropathy of unknown origin, particularly those with concomitant autonomic dysfunction, should be tested for ATTR. Additional symptoms include progressive cardiomyopathy and gastrointestinal disturbances. Progressive cardiomyopathy results in a rapid decline in cardiac functional capacity (Castano et al., 2016;. Patients may experience rhythm disturbances, dyspnea, syncope, and palpitations . Gastrointestinal disturbances include gastroparesis leading to nausea and vomiting, alternating diarrhea and constipation, and unintentional weight loss (Coelho, Maurer, & Suhr, 2013;Maurer et al., 2016;Wixner et al., 2018).

| Assessment and diagnosis of ATTR
In patients with signs, symptoms, or manifestations suggestive of amyloidosis, diagnostic and genetic testing are of utmost importance.
Neurologic and cardiac symptoms can be evaluated with numerous tests and procedures, including tissue biopsy, Congo red staining to confirm the presence of amyloid, and mass spectrometry or immunohistochemistry to confirm the type of amyloid (Figure 1a,b) (Benson et al., 2011;Carvalho, Rocha, & Lobato, 2015;Coelho, Maurer, et al., 2013;Gilbertson et al., 2015;Linke, Oos, Wiegel, & Nathrath, 2006;Sperry et al., 2018). Genetic testing is required to differentiate wtATTR from hATTR and to allow for the detection of specific TTR gene mutations, which may help predict the clinical course of disease (Coelho, Maurer, et al., 2013). In patients with a family history, it may be appropriate to proceed directly to genetic testing.
Myocardial radiotracer uptake on bone scintigraphy is an alternative to tissue diagnosis for patients with cardiac ATTR (Gillmore et al., 2016). Myocardial uptake of bone scintigraphy agents (e.g., 99m technetium [Tc]-pyrophosphate scintigraphy and 99m Tc-3,3-diphosphono-1,2-propanodicarboxylic acid) is sensitive and specific for the diagnosis of TTR cardiac amyloid, if results of screening tests for light chain amyloid (serum and urine electrophoresis with immunofixation and serum free light chains) are negative (Bokhari et al., 2018;Gertz et al., 2015;Gillmore et al., 2016;Nativi-Nicolau & Maurer, 2018). The use of bone scintigraphy is more sensitive than echocardiography in detecting early cardiac ATTR. Patients with cardiac amyloid detected via scintigraphy should also be evaluated for neuropathy because patients with hATTR often have a mixed phenotype. This will ensure appropriate treatment.

| Liver transplantation
Liver transplantation is a standard treatment option for hATTR because replacing the primary source of mutant TTR substantially reduces its production. The efficacy of liver transplantation has been demonstrated in studies showing improvement in sensory and motor impairment (Okumura et al., 2016), as well as improvement in long-term overall survival (Yamashita et al., 2012). The 10-year posttransplantation survival rate exceeds 70% for some patients with hATTR (Suhr, Larsson, Ericzon, & Wilczek, 2016).
Despite reports of successful outcomes with liver transplantation, progression of peripheral and autonomic neuropathy and cardiomyopathy have occurred following liver transplantation (Banerjee et al., 2017). Additionally, numerous factors, such as genotype and pretransplantation (e.g., modified body mass index [mBMI]) characteristics, have been shown to affect posttransplantation survival (Banerjee et al., 2017;Suhr et al., 2016). The 10-year survival rate is as low as 23% for patients with Ser50Arg hATTR and as high as 85% for patients with Val71Ala hATTR (Suhr et al., 2016).
In multivariate analysis, the presence of Val30Met hATTR is associated with significantly reduced mortality (Ericzon et al., 2015).
Additional factors to consider when selecting patients for transplantation are availability of donor liver/graft, patient fitness, the need for long-term immunosuppression, and posttransplantation complications.
The efficacy and safety of tafamidis have been demonstrated in patients with early-stage hATTR polyneuropathy and most recently in patients with hATTR cardiomyopathy; however, the drug's efficacy in mid-to late-stage hATTR and across a broad range of hATTR genotypes is mixed. Two studies in patients with early-stage hATTR polyneuropathy-Fx-005 (clinicaltrials.gov identifier: NCT00409175) and Fx-006 (clinicaltrials.gov identifier: NCT00791492)-demonstrated less neurologic deterioration with tafamidis 20 mg once daily compared with placebo but did not reach statistical significance (Coelho, Maia, et al., 2013;Coelho et al., 2012). The efficacy and adverse event profile observed in these studies led to the approval of tafamidis in the European Union in 2011 (Pfizer Ltd, 2016). Sustained activity and tolerability of tafamidis for up to 6 years have been reported (clinicaltrials.  Findings from this study provided data for the approval of tafamidis meglumine (Vyndaqel, Pfizer, Inc.) and tafamidis (Vyndamax, Pfizer, Inc.) in the treatment of cardiomyopathy of wtATTR or hATTR in adults to reduce cardiovascular mortality and cardiovascular-related hospitalization (Pfizer, 2019).

Diflunisal
Diflunisal, a nonsteroidal anti-inflammatory drug (NSAID), has TTRstabilizing properties and has been used off-label for the treatment of hATTR. In a randomized, placebo-controlled, double-blind, investigator-initiated study (clinicaltrials.gov identifier: NCT00294671), the difference in Neuropathy Impairment Score + 7 neurophysiologic tests composite score (NIS + 7) between diflunisal 250 mg twice daily and placebo at 2 years indicated decreased neuropathy progression (Berk et al., 2013). Additionally, physical and mental scores on the 36-Item Short-Form Health Survey indicated improvement in QOL.
Despite the efficacy of diflunisal, adverse events associated with NSAIDs-specifically gastrointestinal, renal, cardiac, and blood-related events-are a concern for patients with hATTR and may preclude use of diflunisal in specific patients. NSAID-related adverse events, including acute renal failure (Azorin, Cabib, & Campistol, 2017), deterioration in renal function, and thrombocytopenia (Sekijima, Tojo, Morita, Koyama, & Ikeda, 2015), have been shown in 2 recent studies conducted in Spain and Japan.  (Ackermann et al., 2016;Adams, Gonzalez-Duarte, O'Riordan, Yang, Ueda, et al., 2018;Akcea Therapeutics, Inc., 2018a;Alnylam Netherlands B.V., 2018;Alnylam Pharmaceuticals Inc., 2018;Ionis USA Ltd, 2018)  Collectively, tafamidis and diflunisal have demonstrated benefit in the treatment of hATTR, but their benefit may be limited by potential variable efficacy in specific patient populations and by an increased potential for adverse events (diflunisal).

Inotersen
Inotersen, a once-weekly subcutaneously administered antisense oligonucleotide therapy, is approved in the United States for the treat- In NEURO-TTR, 173 patients with hATTR polyneuropathy were randomly assigned 2:1 to receive subcutaneous inotersen 300 mg once weekly (n = 113; n = 112 treated) or placebo (n = 60) stratified by mutation status, disease stage, and prior stabilizer treatment . Baseline demographic and disease characteristics were generally balanced across both treatment groups; however, patients receiving inotersen had a longer mean duration of hATTR polyneuropathy from diagnosis compared with patients receiving placebo (mean, 42 vs. 39 months); this was particularly evident among patients with stage 2 hATTR (mean, 41 vs. 25 months) Gertz et al., 2018). Additionally, a greater proportion of patients receiving inotersen than placebo had cardiomyopathy (67% vs. 55%) and had a longer duration from the onset of cardiomyopathy symptoms (mean, 48 vs. 34 months) Gertz et al., 2018).
In NEURO-TTR, patients treated with inotersen showed substantial reductions in TTR levels that were sustained over time Dyck et al., 2018). Inotersen treatment resulted in a median TTR reduction from baseline (measured predose) of 75%-79% between months 3 and 15   . Improvement in neuropathy-related QOL, as indicated by the reduction in Norfolk QOL-DN scores, was Common adverse events, defined as adverse events occurring in ˃10% and twice as frequently in patients receiving inotersen than placebo, included nausea, pyrexia, chills, vomiting, anemia, thrombocytopenia, and lowered platelet counts . The rate of injection-site reactions was 1.1% of all injections in patients who received inotersen. Most (97%) were mild in severity, and no patient discontinued because of injection-site reactions. Five (4.5%) deaths occurred, and all were in inotersen-treated patients; however, all but one were because of disease progression or underlying disease. A single death occurred because of fatal intracranial hemorrhage associated with serious thrombocytopenia in a patient whose platelets had not been monitored, as the event occurred before routine monitoring was implemented; similar events did not occur after monitoring was implemented . Safety concerns include thrombocytopenia and glomerulonephritis, which were monitorable and manageable following implementation of more frequent platelet (weekly) and renal (every 2-3 weeks) monitoring in the NEURO-TTR study . In order to manage and minimize the potential risk of serious bleeding and glomerulonephritis, inotersen is available through a restricted distribution program under a Risk Evaluation and Mitigation Strategy (Akcea Therapeutics, Inc., 2018a). Enhanced monitoring for thrombocytopenia, glomerulonephritis/renal function, and liver function is recommended (Table 2) (Akcea Therapeutics, Inc., 2018a; Ionis USA Ltd, 2018).
The sustained benefit and tolerability of inotersen were demonstrated in the ongoing phase 3 open-label extension study of NEURO-TTR (clinicaltrials.gov identifier: NCT02175004) (Brannagan et al., 2018). In NEURO-TTR, 81% of patients completed the 15-month treatment period and ˃95% enrolled in the open-label extension study . Extended dosing with inotersen, reaching 5.2 years for the longest time any patient received treatment, demonstrated continued slowing or improvement of neuropathy progression after 2 years of follow-up (Brannagan et al., 2018). Additionally, initiation of inotersen in the extension study in patients who previously received placebo resulted in disease stabilization (Brannagan et al., 2018). Although patients who previously received placebo experienced rapid benefit, the delay in receiving inotersen resulted in less improvement in mNIS + 7 and Norfolk QOL-DN than those of patients who started inotersen earlier and remained on inotersen, suggesting that treatment with inotersen altered the natural history of the disease. Importantly, no new safety concerns were identified, and there was no evidence of increased risk for grade 4 thrombocytopenia or glomerulonephritis with increased duration of inotersen exposure. Taken together, these data confirm the urgency to treat patients as early as possible.

Patisiran
Patisiran, a triweekly intravenously administered TTR-directed, and were stratified by NIS, disease onset/genotype, and previous stabilizer use. Baseline demographics and disease characteristics were generally balanced between treatment groups, although a slightly higher proportion of patients in the overall population had disease stage 2 (53%) than disease stage 1 (46%). Additionally, a greater proportion of patients receiving patisiran had cardiomyopathy compared with patients receiving placebo (61% vs. 47%). Patisiran treatment resulted in a median TTR reduction from baseline (measured postdose) of 81% over 18 months of treatment (Adams, Gonzalez-Duarte, O'Riordan, Yang, Ueda, et al., 2018).
In patients receiving patisiran, a statistically significant improvement in mNIS + 7 (primary end point) was achieved at 18 months (p ˂ .001) compared with placebo, with effects seen as early as 9 months (Adams, Gonzalez-Duarte, O'Riordan, Yang, Ueda, et al., 2018). A similar effect was observed for mNIS + 7 across all subgroups based on study stratification factors. Significant improvement in neuropathy-related QOL, as indicated by Norfolk QOL-DN (secondary end point), was also observed at 18 months (p ˂ .001) with patisiran compared with placebo (Adams, Gonzalez-Duarte, O'Riordan, Yang, Ueda, et al., 2018). Similar to mNIS + 7, a benefit in favor of patisiran was observed for Norfolk QOL-DN across all subgroups, indicating a broad clinical benefit. In addition to improvement in neuropathy and QOL, improvement from baseline in favor of patisiran was observed at 18 months for gait speed (measured by 10-m walk test), nutritional status (measured by mBMI), and autonomic symptoms (measured by Composite Autonomic Symptom Score 31) (Adams, Gonzalez-Duarte, O'Riordan, Yang, Ueda, et al., 2018). In the subset of patients with cardiomyopathy, patisiran was associated with significant improvements for the exploratory end points, mean left ventricular wall thickness (p = .02) and longitudinal strain (p = .02), compared with placebo at 18 months (Adams, Gonzalez-Duarte, O'Riordan, Yang, Ueda, et al., 2018).

| Emerging/investigational therapies
Several investigational therapies are being evaluated for the treatment of hATTR amyloidosis. These therapies target specific aspects of the amyloidogenic cascade and, in some cases, may offer advantages over existing agents. Several agents that stabilize TTR are being investigated, including epigallocatechin-3-gallate (EGCG), AG-10, and CHF5074. EGCG, a catechin in green tea, has demonstrated increased stabilization of TTR tetramers and a reduction in TTR deposition (Ferreira, Saraiva, & Almeida, 2012). In a mouse model of hATTR amyloidosis with peripheral nervous system involvement, EGCG 100 mg kg −1 day −1 for 6 weeks produced a significant reduction in TTR deposition in dorsal root ganglia and sciatic nerve (Ferreira et al., 2012). AG-10, a kinetic stabilizer, binds with high affinity and negative cooperativity to TTR (Penchala et al., 2013). In preclinical studies, AG-10 stabilizes tetramers composed of wild-type and mutated TTR equally well, whereas tafamidis showed greater stabilization of wild-type than mutated TTR (Penchala et al., 2013). CHF5074, an NSAID derivative without cyclooxygenase inhibitory properties, is another molecule that has demonstrated stabilization of wild-type and mutant TTR tetramers (Mu et al., 2015). The absence of cyclooxygenase inhibitory activity offers an advantage over other NSAID-based TTR stabilizers in that it may prevent unwanted adverse effects associated with NSAIDs. Additional testing is needed to confirm whether that is the case for CHF5074. Monoclonal antibodies directed at misfolded TTR are another therapeutic strategy being investigated for ATTR amyloidosis. Preclinical activity demonstrated the ability of monoclonal antibodies to prevent amyloid fibril formation (Higaki et al., 2016). Additionally, antibody-dependent phagocytic uptake of misfolded TTR was observed (Higaki et al., 2016).  Similarly, direct comparison of inotersen and patisiran with tafamidis for patients with cardiac disease is not possible given differences in the NEURO-TTR, APOLLO, and ATTR-ACT studies. Additional research is necessary to determine the optimal treatment for a given patient with ATTR. It is possible that gene silencing therapy in combination with TTR-stabilizing therapy or an agent that removes fibril deposition may be beneficial. As additional evidence for current and emerging therapies expands, the outlook for patients with ATTR is becoming more promising, offering much-needed hope for a debilitating and life-threatening disease.

AUTH O R CO NTR I B UTI O N S
All authors contributed equally to this article, take responsibility for its contents, and were involved in the planning, preparation, and review/revision of this article. All authors approved the final draft and agreed to submit the article for consideration of publication. Medical writing support was provided by ApotheCom and funded by Akcea Therapeutics. Akcea Therapeutics had no role in determining the content of the article.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study were derived from the following resource available in the public domain: PubMed at https ://www.ncbi.nlm.nih.gov/pubmed.