Research report: Hereditary transthyretin amyloidosis in peripheral neuropathy: prevalence, clinical features, and neurology-based screening in non-endemic Japan

A prospective, single-center study in a non-endemic Japanese region found pathogenic transthyretin variants in 3 of 130 consecutive patients evaluated for suspected peripheral neuropathy, a prevalence of 2.31% with an exact 95% confidence interval of 0.5% to 6.6%. The estimate was comparable to those reported in prospective European cohorts. All three affected patients had late-onset disease without a family history of amyloidosis or neuropathy, and cardiac amyloid involvement was demonstrated in every case.

The findings indicate that hereditary transthyretin amyloidosis with variant amyloid, or ATTRv, can occur among patients whose neuropathy is evaluated outside recognized Japanese endemic foci. The study examined ATTRv, a severe adult-onset systemic disorder caused by pathogenic variants in the transthyretin gene and transmitted in an autosomal dominant pattern with incomplete penetrance. Amyloid deposits can affect peripheral and autonomic nerves, the heart, kidneys, gastrointestinal tract and eyes.

Some transthyretin variants are associated predominantly with progressive sensorimotor polyneuropathy, others with restrictive cardiomyopathy, and many with a mixed phenotype. The neurological presentation commonly begins with small-fiber involvement, including distal sensory disturbance, dysesthesia and pain, before more evident motor involvement develops years later. Because these features can resemble diabetic polyneuropathy or chronic inflammatory demyelinating polyradiculoneuropathy, diagnostic delay is an important concern.

The cohort included 73 men and 57 women, with a mean age at symptom onset of 61.1 years. Common clinical features included polyneuritis in 89 patients, pain and muscle weakness in 44 each, constipation in 26 and dizziness in 23. Nerve conduction studies were performed in 128 of the 130 participants.

Of these, 123 had an axonal, demyelinating or mixed abnormality, while five had normal conventional results. The five patients with normal results remained in the cohort because small-fiber or autonomic neuropathy may be present before abnormalities appear on conventional nerve conduction testing. This broad definition means the reported prevalence applies to patients clinically suspected of having peripheral neuropathy rather than only those with a confirmed, specifically defined polyneuropathy.

As a sensitivity analysis, the researchers excluded 18 patients with amyotrophic lateral sclerosis; the prevalence then became 3 of 112, or 2.68%, with an exact 95% confidence interval of 0.6% to 7.6%. Three heterozygous pathogenic variants were identified: ATTRV122I, also designated p.(Val142Ile); ATTRI127V, or p.(Ile147Val); and ATTRV30M, or p.(Val50Met). Two of these variants were not associated with the principal Japanese endemic foci and were unrelated to the best-known V30M founder pattern.

The three patients developed symptoms between ages 59 and 79 and received genetic diagnoses between ages 62 and 83, four to seven years after the first symptom attributed to ATTRv. The affected patients had markedly different neurological presentations. The first was an 82-year-old woman with ATTRV122I whose initial manifestation was dropped head caused by neck-extensor weakness.

Ascending limb numbness, grip difficulty, gait instability, mild asymmetric weakness, glove-and-stocking dysesthesia and urinary frequency followed. Although she had diabetes, the investigators considered the neuropathy predominantly attributable to ATTRv because motor abnormalities preceded limb numbness, axial weakness is not characteristic of diabetic polyneuropathy, and endomyocardial biopsy demonstrated cardiac amyloid. A contribution from diabetic neuropathy could not be excluded.

The second patient, a 62-year-old man with ATTRI127V, had progressive distal-dominant weakness, sensory disturbance, tandem-gait difficulty and severe orthostatic hypotension with syncope. Sural-nerve biopsy showed severe loss of myelinated and unmyelinated fibers without demyelinating change. Cardiac evaluation produced clinically informative results even though cardiac symptoms were not overt at neurological presentation.

In the man with ATTRI127V, echocardiography was structurally normal, with an interventricular septal thickness of 9 mm and an ejection fraction of 71%, but 99mTc-pyrophosphate scintigraphy showed grade 2 uptake indicating cardiac amyloid involvement. The third patient, an 83-year-old man with ATTRV30M, had mild neurological disease marked by reduced vibration sensation in the left lower limb and a bilateral carpal-tunnel electrophysiological pattern, without clear reduction in compound muscle action potentials. He also had orthostatic hypotension, vertigo and gait instability.

His echocardiogram showed diffuse left-ventricular hypertrophy and reduced longitudinal strain, and cardiac amyloid was confirmed by biopsy. This patient entered the neurology pathway after cardiology-led scintigraphy rather than through the screening route examined by the investigators. All three variant-positive patients had a low-voltage electrocardiogram.

Among the 98 patients for whom ECG data were available, low voltage occurred in 3 of 3 variant-positive patients and in 5 of 95 variant-negative patients. Calculated sensitivity for ATTRv was 100%, with an exact 95% confidence interval of 29.2% to 100%, and specificity was 94.7%, with a confidence interval of 88.1% to 98.3%. Positive predictive value was 37.5%, because eight patients had low voltage but only three carried a pathogenic variant, while negative predictive value was 100%.

These estimates are highly imprecise because they rest on only three positive cases. The researchers therefore described the ECG finding as preliminary and hypothesis-generating, not as evidence for a validated screening strategy. Low QRS voltage is reported in only a minority of patients with ATTRv cardiomyopathy in cardiology series, so a normal ECG cannot be used to exclude ATTRv when neurological red flags are present.

The researchers also calculated a retrospective compound clinical-electrophysiological score derived from motor symptoms, a history of carpal tunnel syndrome and nerve-conduction amplitudes in the median, ulnar and tibial nerves. Complete data were available for 125 patients. A score of at least 5 was present in two of the three variant-positive patients, compared with 37 of 122 variant-negative patients.

The resulting sensitivity was 66.7%, specificity was 69.7%, positive predictive value was 5.1% and negative predictive value was 98.8%. The patient below the threshold had a score of 4 and early, mild neuropathy with a bilateral carpal-tunnel pattern. The score was therefore less sensitive in this least advanced case, but it was calculated retrospectively in a cohort broader than the population in which it was originally derived.

The findings describe its behavior in this clinic and neither validate nor refute its original performance. Several limitations restrict the conclusions. The study covered one neurology center and included patients with inflammatory neuropathies, motor neuron disease and other diagnoses, potentially diluting the prevalence estimate.

ECG and echocardiography were performed at the attending neurologist’s discretion rather than by protocol; 32 patients lacked ECG data and 33 lacked echocardiograms. Because patients considered more likely to have systemic disease were more likely to undergo cardiac testing, selection could have affected the apparent specificity of low voltage in an unknown direction. Only three patients were variant-positive, so comparisons between groups were descriptive and no hypothesis tests were performed.

The study also enrolled symptomatic patients rather than unselected carriers or families and therefore could not estimate age-dependent penetrance. Despite these constraints, the results identify a meaningful frequency of ATTRv among patients whose unexplained neuropathy led to neurological assessment in non-endemic Japan. Progressive neuropathy accompanied by autonomic dysfunction, gait instability, low body weight or weight loss, distal weakness, bilateral carpal tunnel syndrome or an inadequate response to immunotherapy can raise clinical suspicion.

The authors support placing TTR genetic testing, with appropriate genetic counseling, early in the diagnostic evaluation when such features are present rather than waiting for cardiac investigations to be completed. Their data further suggest that a routinely available ECG may complement neurological assessment and prompt cardiac evaluation, but that hypothesis requires testing in larger, independent prospective cohorts; the ECG should complement genetic testing rather than serve as a prerequisite or stand-alone diagnostic rule.

Subject of Research: Prevalence, neurological features and cardiac involvement of hereditary transthyretin amyloidosis in a non-endemic Japanese neurology clinic

Article Title: Hereditary transthyretin amyloidosis in peripheral neuropathy: prevalence, clinical features, and neurology-based screening in non-endemic Japan

Article References: Okuma, Y., Isahaya, K., Ohori, S., Mitsuhashi, S., Sato, T., & Yamano, Y. (2026). Hereditary transthyretin amyloidosis in peripheral neuropathy: prevalence, clinical features, and neurology-based screening in non-endemic Japan. Frontiers in Neurology, 17, Article 1955913. https://doi.org/10.3389/fneur.2026.1955913

Image Credits: AI Generated

DOI: 10.3389/fneur.2026.1955913

Keywords: hereditary transthyretin amyloidosis, ATTRv, peripheral neuropathy, genetic screening, Japan, cardiac amyloidosis, autonomic dysfunction, low-voltage ECG, nerve conduction studies, TTR variants, Hereditary, transthyretin