Variant and wild type transthyretin amyloidosis: two sides of the same coin or different currencies in different pockets?

Abstract

This article refers to ‘Prevalence, characteristics and outcomes of older patients with hereditary versus wild-type transthyretin amyloid cardiomyopathy’ by A. Porcari et al., published in this issue on pages xxx. In 2002, the group from the National Amyloidosis Centre in the UK published data showing that 3.7% of patients referred to them with a presumptive diagnosis of light-chain (AL) amyloidosis actually had a mutation in the transthyretin (TTR) molecule as a cause of their amyloidosis.1 That important paper underscored the need for comprehensive evaluation of all patients with suspected AL amyloidosis, in order to provide accurate diagnosis and to avoid unhelpful and potentially toxic therapy. In the intervening years it has become apparent that, contrary to the belief 20 years ago, the prevalence of TTR amyloidosis (ATTR) is considerably higher than that of AL amyloidosis. This surge in diagnosis has been driven by the widespread availability of imaging agents used previously for bone imaging, most specifically 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (DPD) scintigraphy in Europe and 99mTc-pyrophosphate scintigraphy (PYP) in the United States. These agents are avidly taken up in the heart of patients with ATTR and their use has obviated the necessity for endomyocardial biopsy in the majority of patients, previously required in almost all patients with this condition. With accurate single photon emission computed tomography imaging (ideally with concomitant cardiac computed tomography), ATTR can be diagnosed with a high degree of confidence, provided that a plasma cell dyscrasia is excluded by simple blood tests. Transthyretin amyloidosis may be hereditary, transmitted in an autosomal dominant fashion, or due to amyloid deposition derived from wild-type TTR. In this issue of the Journal, Porcari and colleagues retrospectively report an overall prevalence of 20.7% of a pathogenic TTR variant (with a prevalence of 7.2% in those with non-African or Caribbean ancestry) among 2029 patients with ATTR aged 70 years or older studied between 2010 and 2022.2 The likelihood of a pathogenic TTR variant in this cohort was higher in relatively younger patients, females, patients with Afro-Caribbean ethnicity and in those with orthostatic hypotension or polyneuropathy. Importantly, patients diagnosed with the variant form of the disease (variant TTR [ATTRv]) compared with the wild-type form (wild-type TTR amyloidosis [ATTRwt]), showed higher all-cause mortality. The authors note that many of these patients had not undergone genetic testing by their referring physicians and thus the precise diagnosis had been missed. This is despite the European Society of Cardiology position statement on the diagnosis of cardiac amyloidosis, which strongly recommends genetic testing regardless of patient's age.3 The authors are to be congratulated on this large series of ATTR cardiomyopathy patients and their documentation that ‘real-life’ work-up falls short of expert recommendations despite the ease with which genetic testing for a TTR variant can nowadays be obtained. Their study highlights two particularly important elements of ATTRv that differ from patients with ATTRwt, namely the frequent presence of neuropathy and the more rapid disease progression in those with a variant TTR protein. Moreover, as they point out, knowledge of the presence of a TTR mutation confers considerable benefits in terms of the initiation of family screening and therapeutic options otherwise unavailable for patients with ATTRwt. However, the implications of these results can be taken one step further to pose the following three questions: ‘Why do ATTRv and ATTRwt differ?’, ‘What direction might it point us in understanding the mechanisms of organ dysfunction in cardiomyopathy and neuropathy of ATTR?’, and ‘How can understanding of possible mechanisms of organ damage help to predict outcome of ongoing studies of TTR gene silencing drugs for amyloid cardiomyopathy?’ Here, we briefly propose a hypothesis based on their data and these questions. One obvious difference that might account for the more rapid cardiac progression of patients with TTR variants is the instability of the mutant TTR molecule, leading to a greater degree of breakdown and to a more amyloidogenic product. This hypothesis is supported by the lower levels of circulating TTR protein observed in patients with the familial TTR gene mutation, with a reversion to normal levels following therapy with a TTR stabilizer.4 Nevertheless, while more rapid TTR breakdown may account for a more rapid disease progression in ATTRv compared to ATTRwt, it does not account for the fact that ATTRv is often associated with neuropathy, whereas proven amyloid neuropathy in ATTRwt is very uncommon. There is no doubt that the clinical development of heart failure in ATTR cardiomyopathy of either form is due to massive myocardial amyloid infiltration, but neuronal amyloid infiltration is probably not the major underlying pathophysiologic mechanism for early symptomatic ATTRv neuropathy. This is supported by prior studies which have shown that amyloid fibres are not always found in nerves in the early stages of amyloid neuropathy,5 and rare variants may have neuropathy in the apparent absence of nerve infiltration.6 The exact mode of neuronal damage in neuropathic ATTRv is not fully elucidated, but it is well-established that pre-amyloid oligomers are neurotoxic in vitro,7 and there is a reasonable consensus that neurotoxicity plays a significant role in patients with the disease.8 On the other hand, the absence of neuropathy in ATTRwt or the presence of only mild neuropathy in certain TTR variants such as V122Ile may well reflect the non-neurotoxic nature of wild-type TTR oligomers, and suggests a spectrum of oligomer toxicity in ATTRv dependent on the specific mutation. We therefore postulate that there are at least two mechanisms of major organ involvement in ATTR: (i) an infiltrative, non-toxic cardiomyopathy, which manifests in both ATTRv and ATTRwt; and (ii) a predominant mutation-specific toxic neuropathy, with neural amyloid infiltration occurring only later in the disease course, often after patients have developed clinically significant symptoms (Figure 1). If correct, then we believe that this hypothesis can serve as the basis to understand the reasons for the clinical differences between ATTRwt and ATTRv, and may affect the way in which we look at future outcomes in trials of TTR gene silencers for amyloid cardiomyopathy. Two TTR gene silencers phase 3 trials are currently ongoing—Alnylam's Helios B trial with vutrisiran and Ionis's CArdioTTRansform trial with eplontersen. Each company's earlier generation drug, patisiran and inotersen, respectively, was highly successful in mitigating the progression of neuropathy in ATTRv polyneuropathy and both have been in clinical use for several years now. An indirect comparison of outcomes between 20 mg of tafamidis and patisiran in patients with familial amyloid polyneuropathy suggested that the latter is superior in slowing the progression of neuropathy,9 triggering some amyloid experts to predict that the same superiority will be observed in the amyloid cardiomyopathy trials, where the majority of patients have ATTRwt. However, since the symptomatic heart failure phenotype of amyloid cardiomyopathy is primarily driven by myocardial TTR amyloid infiltration, as opposed to the potentially toxic effects of pre-amyloid oligomers in ATTRv neuropathy, we believe that extrapolating relative efficacy data from neuropathy trials to likely results of cardiomyopathy trials is flawed reasoning. In line with the hypothesis suggested above, a relatively rapid amelioration of neurologic symptoms was seen following the use of TTR gene silencers and its associated profound drop in circulating TTR levels, in contrast to the more modest amelioration of neurologic symptoms that was found with TTR stabilizers, a treatment which does not decrease TTR production, and thus might not have as great an effect as silencers on the levels of residual TTR oligomers. A hint at potential lack of superiority of silencers over stabilizers for amyloid cardiomyopathy comes from recently-released preliminary data in patients with amyloid cardiomyopathy treated with the TTR-gene silencer patisiran.10 While patisiran demonstrated superiority over placebo in the 6-min walk test in preventing deterioration over a 12-month period, this benefit that was not observed in the subgroup of patients who were already on tafamidis and there was no improvement in N-terminal pro-B-type natriuretic peptide or troponin levels compared to baseline values in the treated group. Overall this suggests disease stabilization but not disease improvement, findings that are similar to those previously seen with the TTR stabilizer, tafamidis. In our view of the available data and based on the presented hypothesis, we predict that TTR gene silencers will be found to be a highly-effective therapy for ATTR cardiomyopathy, yet will share a similar treatment outcome profile with tafamidis, currently the only Food and Drug Administration-approved TTR stabilizer. Therefore, we feel that the trend among some amyloid specialists to already prescribe combination therapy of a TTR-stabilizer and a TTR gene silencer (or the latter alone) cannot be justified, particularly in light of the enormous cost of both of these agents. As always, the paper from the UK National Amyloidosis Centre is like a culinary delight; providing plentiful food for thought, derived from a large helping of patients. Their data has whetted our cerebral taste buds and we hope that our ruminations on the findings will stimulate a fruitful debate. As all in the field do, we eagerly await the results of the TTR gene silencer trials, not only for their findings on the value of TTR gene silencers in the management of amyloid cardiomyopathy, but also for further insight into the mechanisms of ATTR organ damage. Whether we are proved right or wrong, these thoughts could not have been stimulated without the careful clinical descriptions in a large cohort of patients, such as the cohort studied by Pocari et al. Conflict of interest: none declared.