Detecting Polyneuropathy in Patients with Hereditary Transthyretin Amyloid Cardiomyopathy

Abstract

Background

Patients with hereditary transthyretin amyloid cardiomyopathy (hATTR-CM), can show neurologic signs that lead to the eventual onset of polyneuropathy (PN). Detection of PN among patients with hATTR-CM can identify candidates for disease-modifying therapies that could significantly reduce the incidence of mortality and morbidity for these complex patients. The purpose of this study is to determine the prevalence of PN among patients with hATTR-CM.

Methods

Sixty patients with hATTR-CM were enrolled in this prospective study, from the Amyloidosis Clinic at the University Health Network in Toronto, Ontario, Canada. All patients (average age: 69.12 ± 11.82 years; 51.7% male; 48.3% female) were referred to neurology programs specialized in hATTR PN. Comprehensive neurologic assessments were conducted, including a physical examination and nerve conduction studies (NCSs). NCSs were performed on the upper and lower extremities.

Results

Thirty patients (50.0%) displayed clinical and NCS abnormalities. Six patients (10.0%) did not display any neurologic abnormalities on the physical examination, but did have abnormalities on the NCSs, suggesting the presence of transthyretin-PN. The abnormalities on the NCSs were more indicative of sensory PN than of motor polyneuropathy. Twenty-four patients (40.0%) did not display any clinical or NCS abnormalities. Based on the systematic neurologic assessment, 36 patients in our study were identified as having a mixed hATTR phenotype, and they are to receive specific disease-modifying therapies, either patisiran or inotersen.

Conclusion

Patients with hATTR-CM should be systematically referred to a specialized neurologist for routine assessment, especially patients who report neurologic symptoms (ie, numbness and burning sensation in extremities).

Transthyretin amyloidosis (ATTR) is a rare disease caused by misfolded transthyretin (TTR) proteins that are produced in the liver and deposited in another organ or tissue, typically the heart and nerves.¹ The organs and tissues containing these TTR amyloid deposits lead to the manifestation of symptoms. Patients who present predominantly with cardiomyopathy (CM) are known to have TTR amyloid CM (ATTR-CM), as the misfolded TTR proteins are deposited in the myocardium.² Patients with ATTR-CM typically develop cardiovascular-related symptoms, including fatigue, dyspnea, and abdominal swelling.³ Those who present with only polyneuropathy (PN) are known to have TTR amyloid PN (ATTR-PN), as the deposits of the misfolded TTR proteins are in the nerves.² Patients with ATTR-PN develop neurologic symptoms, including loss of stability, a burning sensation in the extremities, and numbness in the hands.⁴ Sometimes, patients have deposits in the myocardium and nerves— a mixed phenotype, as they present with both cardiac and neurologic symptoms.^2,4

Two types of ATTR amyloidosis can occur.⁵ One is wildtype TTR amyloidosis (wtATTR), which is seen commonly among older individuals, and the other is hereditary ATTR (hATTR), which is acquired by a genetic mutation in the TTR gene that is passed down in families.⁵ A variety of phenotypes can be observed among each subtype. The most common phenotypes include the following: asymptomatic, only cardiac symptoms, only neurologic symptoms, and mixed. The most prevalent phenotypes observed among wtATTR patients are asymptomatic and only CM; patients with hATTR tend to display any of the 4 phenotypes.^4,5

Although this disease has no cure, the treatments that are provided for these patients help slow its progression, and these include TTR silencers.⁶ Inotersen and patisiran are TTR silencers that are effective in alleviating symptoms among patients with hATTR-PN and patients with hATTR who have a mixed phenotype (CM and PN).6, 7, 8 These drugs silence the TTR messenger RNA, which prevents it from translating into TTR proteins that would be released from the liver and deposited in the nerves and/or heart.⁹ As a result, patients would have a better quality of life, as the amyloid in the nerves and/or heart would be reduced, thereby reducing mortality and morbidity.

Without treatment, the disease progresses and eventually leads to death. Thus, patients with hATTR-PN may also develop hereditary ATTR-CM (hATTR-CM) over time, and patients with hATTR-CM also may develop hATTR-PN over time. Patients with hATTR-PN are eligible to receive a TTR silencer, but currently, patients with hATTR-CM are not eligible for a TTR silencer unless they also present with signs and symptoms of PN. Moreover, patients with hATTR-CM tend to present signs and symptoms of PN over time.¹⁰ As a result, this study aimed to determine the prevalence of PN in patients with hATTR-CM, to highlight the need for regular screening for PN to provide the optimal treatment for patients.

Methods

Study design and population

This single-centre prospective study included patients (aged > 18 years) diagnosed with hATTR-CM who were referred to the Amyloidosis Clinic at the Peter Munk Cardiac Centre (PMCC), from 2021 to 2023. Patients who were aged < 18 years who presented other phenotypes (ie, only PN, mixed phenotype, or asymptomatic), and had diabetes—as they may also present length-dependent PN—were excluded.¹¹ Patients were referred to a neurologist who specializes in hATTR-PN. The neurologist conducted a physical examination and nerve conduction studies (NCSs) on these patients to determine any neurologic deficits and/or abnormalities of nerve functioning, respectively, that are signs of PN.

Diagnostic methods for CM

To determine if they had ATTR-CM, patients underwent either technetium pyrophosphate scintigraphy (PYP) or an endomyocardial biopsy on either the left ventricular wall or the right ventricular septal wall. Some patients may have had additional testing, such as echocardiography (ECHO) or cardiac magnetic resonance imaging (cMRI). Patients had a PYP scan to confirm the presence of ATTR-CM; a semiquantitative scoring of 2/3, and a heart-to-contralateral-lung ratio uptake via quantitative analysis > 1.5 were considered positive tests.^12,13

Genetic test

Patients diagnosed with ATTR-CM have undergone a genetic test that confirms whether patients have hATTR-CM by determining whether a mutation is present on the TTR gene.¹⁴ The genetic result also indicates the type of variant the patients have, as well as how the gene was altered, by specifying the exact nucleotide that was replaced. Patients with confirmed hATTR-CM were referred to a neurologist who specializes in PN for amyloidosis.

Physical examination by neurologist

Before any assessments were conducted, the neurologist reviewed patients’ family history. Then, a physical examination was conducted to determine whether patients with hATTR-CM displayed signs of PN. The physical examination assesses the following: speech, language, cranial nerves (for pupillary light responses), motor ability, muscle strength, ankle reflexes, knee reflexes, plantar responses, pinprick, coordination, and gait. Any abnormalities during this examination indicate signs of PN among these patients. A few parameters were not included, as many patients were not able to be examined for them—facial symmetry, vibration sense, and vibration perception.

NCSs–

The NCSs were also conducted by the neurologist. The NCSs measure the speed of electrical impulses in nerves.^15,16 Temperature was controlled for this test, as maintaining skin temperature relative to body temperature is optimal for conducting NCSs.¹⁷ A motor NCS was conducted orthodromically (distal stimulation with proximal recording, in the physiological direction of flow for normal sensory impulses), and sensory studies were performed antidromically (proximal stimulation with distal recording, opposite to the physiological direction of impulse flow).¹⁸ The speed of the electrical impulse was measured by calculating the distance between electrodes and the time it takes the electrical impulse to travel between electrodes. The NCS examines the function of nerves in the lower and upper extremities. Differential changes indicate the presence of length-dependent PN, which occurs when the peripheral nerves in the longest nerve fibres have been damaged, affecting the hands and feet first by causing tingling and numbness.¹⁹ An NCS is a reliable indicator of PN, including length-dependent PN. An abnormality in an NCS may be the first quantitative indicator of PN and is needed for the diagnosis of PN. Abnormality in an NCS is determined by comparing the values obtained to reference values.²⁰

Statistical analysis

Demographic data were collected and analyzed with descriptive statistics, using means and standard deviations for continuous variables, and frequencies and proportions for dichotomous and/or polytomous variables. Neurologic data for the physical examination and the NCS were collected and analyzed using frequencies and proportions to determine the prevalence of abnormal findings. Comparisons between variants presented among patients with hATTR-CM who showed signs of PN and patients with hATTR-CM who did not show signs of PN were conducted using Fisher’s exact test. The analysis was conducted using a 2-tailed test, with a P-value < 0.05 considered to be statistically significant.

Results

Study population

Sixty patients with hATTR-CM were included in this study, with an average age of 69.12 ± 11.82 years (Table 1). The most frequent genetic variant was Val142Ile (n = 31; 51.7%). The predominant ethnicity among this patient population was Afro-Caribbean descent, and all of those with this ethnicity had the Val142Ile variant.

Table 1.

Baseline characteristics in patients with hereditary transthyretin amyloid cardiomyopathy

Variable	n = 60
Age, y	69.12 ± 11.82
Sex
Male	31 (51.7)
Female	29 (48.3)
Ethnicity
Canadian	7 (11.7)
African	35 (58.3)
Hispanic	7 (11.7)
European	9 (15.0)
South-Asian	2 (3.3)
CM diagnostic methods
PYP scan	49 (81.7)
Imaging (ie, ECHO, cMRI)	7 (11.7)
Endomyocardial biopsy	4 (6.7)
Variant
Val142Ile	31 (51.7)
Val50Met	7 (11.7)
Thr80Ala	5 (8.3)
Val142Del	2 (3.3)
Val150Leu	2 (3.3)
Other	9 (15.0)
Unknown (missing)	4 (6.7)

Values are reported as mean ± standard deviation, or frequencies and proportions (n [%]). CM, cardiomyopathy; cMRI, cardiac magnetic resonance imaging; ECHO, echocardiogram; PYP, pyrophosphate.

Neurologic assessments

Assessments included a history of neurologic symptoms, neurologic examination, and an NCS. The overall findings of the neurology assessments included whether patients had normal or abnormal results on the physical examination and NCS, and whether patients reported any neurologic symptoms. Most patients who did not report any neurologic symptoms showed normal results on both the physical examination and the NCS (n = 24; 40.0%). These patients had an average age of 69.05 ± 11.91 years, with an equal number of male (n = 12; 50.0%) and female (n = 12; 50.0% patients; Table 2). However, patients with abnormal results in both assessments reported having neurologic symptoms (n = 30; 50.0%). These 30 patients had an average age of 68.62 ± 11.97 years, and most were female (n = 18; 60.0%). Asymptomatic patients with a normal physical examination had an abnormal NCS in 10% of cases (n = 6). These 6 patients had an average age of 66.11 ± 13.39 years, and most were female (n = 5; 83.3%). In all 3 sets of patients, the predominant ethnicity was Afro-Carribean descent, and the most frequent genetic variant was Val142Ile. Regarding the cardiac parameters, in all 3 sets of patients, the predominant New York Heart Association class was II, and the predominant PYP grade was 3. The PYP ratios were 1.88 ± 0.38, 1.88 ± 0.45, 1.45 ± 0.30 for the 24 patients, 30 patients, and 6 patients, respectively.

Table 2.

Overall results of patient post-neurologic assessment

n = 60	24 patients with clinical exam & NCSs negative	30 patients with clinical exam & NCSs positive	6 patients with clinical exam negative & NCSs positive
Age, y	69.05 ± 11.91	68.62 ± 11.97	66.11 ± 13.39
Sex	Male: 50.0 Female: 50.0	Male: 40.0 Female: 60.0	Male: 16.7 Female: 83.3
Ethnicity	African: 79.2 Hispanic: 2.5 Canadian: 4.2 European: 4.2	African: 43.3 European: 23.3 Canadian: 13.3 Hispanic: 13.3 South Asian: 6.7	African: 50.0 Canadian: 33.3 European: 16.7
Variant	Val142Ile: 70.8 Val50Met: 12.5 Val142Del: 4.2 Unknown: 12.5	Val142Ile: 33.3 Val50Met: 13.3 Thr80Ala: 10.0 Val150Leu: 6.7 Val142Del: 3.3 Other: 30.0 Unknown: 3.3	Val142Ile: 66.7 Thr80Ala: 33.3
NYHA class	I: 25.0 II: 62.5 III: 12.5	I: 20.0 II: 40.0 III: 33.3 IV:6.7	II: 100
PYP grade	1: 8.3 2: 16.7 3: 75.0	1: 6.7 2: 33.3 3: 60.0	1: 33.3 2: 33.3 3: 33.3
PYP ratio	1.88 ± 0.38	1.88 ± 0.45	1.45 ± 0.30

Values are reported as frequencies (%), or mean ± standard deviation. Negative indicates normal results, and positive indicates abnormal results. Exam, examination; NCS, nerve conduction study; NYHA, New York Heart Association; PYP, pyrophosphate.

Among the subgroup of patients with hATTR-CM who showed signs of PN in both assessments, the following components of the physical examination were normal: language (n = 27; 90.0%), cranial nerves (n = 25; 83.3%); motor ability (n = 20; 66.7%), muscle strength (n = 17; 56.7%), plantar responses (n = 15; 50.0%), and coordination (n = 20; 66.7%; Table 3). However, within this subgroup, the results for the following parameters predominantly showed signs of PN: ankle reflexes (n = 19; 63.3%)’ knee reflexes (n = 15; 50.0%); and pinprick (n = 24; 80.0%). Unfortunately, some reports within this subgroup were incomplete (n = 3), as patients’ conditions limited their ability to participate in this examination.

Table 3.

Results of the physical examination among patients who showed signs of polyneuropathy in both neurologic assessments

Variable n = 30	Language	Cranial nerves	Motor ability	Muscle Strength	Ankle reflex	Knee reflexes	Plantar response	Pinprick	Coordination	Gait
Normal	27 (90.0)	25 (83.3)	20 (66.7)	17 (56.7)	5 (16.7)	6 (20.0)	15 (50.0)	0 (0.0)	20 (66.7)	13 (43.3)
Abnormal	1 (3.3)	0 (0.0)	8 (26.7)	9 (30.0)	19 (63.3)	15 (50.0)	12 (40.0)	24 (80.0)	1 (3.3)	11 (36.7)
Missing	2 (6.7)	5 (16.7)	2 (6.7)	4 (13.3)	6 (20.0)	9 (30.0)	3 (10.0)	6 (20.0)	9 (30.0)	6 (20.0)

Values are reported as frequencies (%).

In addition, within this subgroup of patients with hATTR-CM, NCS results also varied (Table 4). An abnormal result on an NCS is determined by the presence of nerve damage in the axons (axonal PN) in at least 2 nerves and the onset of carpal tunnel syndrome (CTS). CTS is indicated by a reduction in the electrical signals (compound motor action potential) and a nerve conduction speed that is < 50 metres per second across the carpal tunnel (narrow passageway along the wrist).21, 22, 23 A couple of patients presented neither axonal PN nor CTS (n = 2; 6.7%). Alternatively, patients predominantly presented both axonal PN and CTS (n = 11; 36.7%). A smaller subgroup of patients showed only axonal PN, but not CTS (n = 9; 30.0%). NCS results in a few cases (n = 8; 26.7%) were not reported.

Table 4.

Results of nerve conduction studies among patients who showed signs of polyneuropathy (PN)in both neurologic assessments

Presence of axonal PN	Presence of CTS	Number of patients (n = 30)
No	No	2 (6.7)
Yes	Yes	11 (36.7)
Yes	No	9 (30.0)
Missing	No	8 (26.7)

Values are reported as frequencies (%).CTS, carpal tunnel syndrome. Eight patients with missing NCSs were diagnosed clinically.

Comparison between variants for patients with hATTR-CM who presented with vs without PN

Patients with hATTR-CM who presented signs of PN in both the physical examination and the NCS had one of the following variants: Val142Ile (n = 10; 30.0%); Val50Met (n = 6; 20.0%); other variants (n = 13; 43.3%); and unknown variants (n = 1; 3.3%; Table 5). Alternatively, patients with hATTR-CM who did not present any signs of PN in either neurologic assessment had one of the following variants: Val142Ile (n = 17; 70.8%); Val50Met (n = 1; 4.2%); other variants (n = 3; 12.5%); and unknown variants (n = 3; 12.5%). The differences between these 2 subgroups of patients were statistically significant (P = < 0.001).

Table 5.

Comparisons between the variants observed among patients with signs of polyneuropathy (PN) vs patients without signs of PN

Variant	Patients with signs of PN (n = 30)	Patients without signs of PN (n = 24)	P-value
Val142Ile	10 (33.3)	17 (70.8)	< 0.001
Val50Met	6 (20.0)	1 (4.2)
Others	13 (43.3)	3 (12.5)
Unknown (missing)	1 (3.3)	3 (12.5)

Values are reported as frequencies (%), unless otherwise indicated.

Discussion

This study is the first and largest in Canada to determine the prevalence of patients with hATTR-CM presenting signs of PN, by conducting neurologic assessments. This study included 60 patients with confirmed hATTR-CM.

Our results demonstrate that 50.0% of patients with hATTR-CM showed signs of PN on physical examination and on the NCS. These findings suggest that 50.0% of patients may have had TTR amyloid deposits in their nerves. In addition to this group of patients, 10.0% of patients did not report any symptoms of PN, did not show any deficits on the physical examination, but did show deficits during the NCS. This finding suggests that 10.0% of patients had TTR amyloid deposits in their nerves. However, these deposits were not significant enough to cause manifestation of PN symptoms, indicating that PN was detected at an early stage among these patients. The abnormal NCS results could also reflect involvement of specific never fibres that do not necessarily lead to clinical manifestations of PN. As these patients did not exhibit motor or sensory symptoms, they may not have had large nerve fibers involvement.^24,25 These results are in accordance with a study conducted by Williams et al., in which they described the tendency of patients with hATTR-CM to present with PN as well.²⁶

In addition to evaluating the overall results of the neurologic assessments, we also analyzed the specific physical examination parameters used to identify the signs of PN in the 50.0% of patients who showed abnormal results. The predominant parameters were ankle reflexes, knee reflexes, plantar responses, pinprick, and gait. These findings are corroborated by other studies, such as the one by Inês et al., which explained the stages of PN. In the early stages (stages 1 and 2), sensory symptoms typically begin in the lower limbs and feet.²⁷ Although these parameters showed the most frequent abnormalities, some patients also exhibited abnormal results in the following parameters: language; motor examination; muscle strength; and coordination. These results could be due to age, as this study involved an aging population.

The results of the NCS of patients with hATTR-CM who showed signs of PN in both neurologic assessments also were analyzed, as the NCS provides information on the function of nerves. The NCS examined whether axonal PN was present, by assessing the farthest nerve endings in the feet, as well as normal vs abnormal functioning of nerves in the lower and upper extremities.^15,28 The data demonstrate that two-thirds of the patients showed axonal PN in their NCS (n = 20); however, only 36.7% (n = 11) of these patients also had CTS. This finding demonstrates that a small subgroup of patients had more-diffuse axonal PN, whereas the remaining 30.0% of patients had less-diffuse axonal PN (n = 9). The presence of CTS may explain why some patients showed signs of PN during the physical examination upon assessment of the upper extremities. Amyloid deposits may be present in the upper extremities resulting in the presentation of CTS. Patients who presented only axonal PN may have amyloid deposits in the lower extremities, resulting in the presentation of signs of PN in the lower extremities in both neurologic assessments.

In addition, 6.7% of patients (n = 2) did not show axonal PN or CTS, an interesting finding, as they did show signs of PN during the physical examination. This result may be due to the inability of the NCS to measure the nerve conduction of certain types of nerve fibres, especially small fibres.²⁸ Small-fibre nerve testing was conducted for only 12 of the 30 patients. Eight of the 12 patients displayed PN in their small fibres. Determining what factors are unique to the asymptomatic TTR patients who had an abnormal NCS is difficult because this group was small (n = 6). The abnormal NCS could reflect early, subclinical neuropathy.²⁸ Lastly, 26.7% of patients (n = 8) did not complete their assessments for the NCS but had received a clinical diagnosis from a neurologist.

This study assessed only patients with hATTR-CM; patients with wtATTR-CM were not included. Other studies, such as those by Wajnsztajn et al., and Russell et al., evaluated the prevalence of PN in patients with wtATTR-CM; this work has shown that 30%-50% of these patients exhibit signs and/or symptoms of PN.^29,30 The prevalence among this subgroup of patients may be higher; however, PN tends to be milder and slower among patients with wtATTR-CM.³¹ Another important point to consider is that patients with wtATTR-CM tend to be older, so the prevalence of neuropathy may be higher for other reasons, such as spinal stenosis and musculoskeletal conditions.³²

Comparisons between the variants in patients with hATTR-CM who showed signs of PN in both neurologic assessments vs those in patients with hATTR-CM who did not show signs of PN in the 2 neurologic assessments were analyzed. In both groups of patients, a relatively significant number had the Val142Ile variant. This variant tends to lead to the onset of hATTR-CM, which would explain the large number of patients expressing this mutation.³³ In addition, a few patients had the Val50Met variant, which tends to result in the onset of PN, thus explaining this finding.³⁴ In addition, most patients who had signs of PN had other rare variants, such as Val150Leu, and whether these rare variants can lead to the manifestation of one or both phenotypes is currently unknown. These findings were statistically significant (P < 0.001), which indicates that these 2 groups of patients differ. Identifying the type of variant patients have could help determine which phenotype they are more prone to presenting.

The prognosis for hATTR-CM is poor without treatment; its course is progressive and ultimately fatal. The prognosis is influenced by the stage at which the disease is diagnosed, with earlier detection associated with better outcomes. Among patients with predominant cardiac involvement, the median length of survival is approximately 3.4 years after diagnosis. The presence of concomitant neurologic involvement is associated with an even poorer prognosis.^35,36 These findings highlight the critical importance of early screening to enable timely diagnosis and intervention.

Overall, these findings demonstrate that neurologic screening among patients with hATTR-CM is necessary in determining the presence of early signs/symptoms of PN among this patient population. Such a reclassification from hATTR-CM to a mixed phenotype may render these patients eligible for TTR silencer therapy (eg, inotersen or patisiran), which addresses both CM and PN.⁶

Study limitations

Although this study may include the largest cohort of Canadians with amyloidosis that has been studied, the sample size is small. In addition, data for 11 patients were not collected post-screening (ie, they were not included in the final sample size of 60 patients), due to the following reasons: (i) patients refused to do a genetic test; (ii) patients refused neurology referral; (iii) patients were deceased before being able to obtain or attend a neurology appointment; and (iv) patients were not able to undergo a complete assessment of the neurologic testing in which data from a few parameters were not included in the study. In addition, a few patients were included who had incomplete neurology assessments due to the severity of their PN, which limited their ability to undergo the physical examination and/or the NCSs. Moreover, a referral bias may have occurred—that is, patients who presented symptoms and/or had a family history of hATTR-CM may have been referred to the amyloidosis clinic at University Health Network and then included in this study. Lastly, various neurologists have consulted these patients, and they may not have used the same assessments to examine patients.

Conclusions

Patients with hATTR-CM may often have underlying PN that goes unrecognized. In our study, 50% of hATTR-CM patients exhibited signs and symptoms of PN. An additional 10% were asymptomatic but demonstrated signs of PN on the NCS. This finding effectively reclassifies them from a pure CM phenotype to a mixed phenotype. The distinction of a mixed phenotype is clinically important because medications are available that target both CM and PN, offering a more comprehensive approach to treatment. Without PN being recognized, patients may miss the opportunity to access these therapies. Our research supports the systematic referral of all patients presenting with hATTR-CM (especially those with a Val142Ile or Val50Met mutation) to a neurologist who specializes in TTR, for regular screening.