1. Introduction
Hereditary amyloid transthyretin (ATTRv) amyloidosis with polyneuropathy is a systemic adult-onset disease that affects the sensorimotor and autonomic functions along with other organs (especially the heart, gastrointestinal tract, eyes, and kidneys) [
1]. The disease has an autosomal dominant pattern of inheritance and is considered a worldwide disease with an estimated global prevalence of 38,000 persons [
2,
3,
4,
5]. Patients with the most frequent neuropathic mutation, V30M (p.V50M), have characteristically early onset (<50 years) in endemic areas, whereas late onset (>50 years) is prevalent in non-endemic areas. Other variants, such as T49A (p.T69A) and E89Q (p.E109Q), often present with an early onset even in non-endemic areas [
6,
7,
8,
9,
10]. The most common hereditary cardiac transthyretin amyloidosis is that associated with V122I (p.V142I), carried by 3.9% of individuals of African descent [
11].
TTR is a transport protein in the serum and cerebrospinal fluid that carries the thyroid hormone thyroxine and retinol-binding protein bound to retinol. It is encoded by the
TTR gene located in the 18th chromosome. The protein is a 55kDa homotetramer synthesized in the liver, choroid plexus, and retinal pigment epithelium for secretion into the bloodstream, cerebrospinal fluid, and eye, respectively. Each monomer is a 127-residue polypeptide rich in beta sheet structure [
12,
13,
14]. The presence of missense mutations in altering the amino acid sequence can make the tetramer less stable, favoring its dissociation. The misfolded monomers aggregate, generating amyloid fibrils, which precipitate into tissues [
15,
16]. In peripheral nerves, fibrils cause axon and Schwann cell damage, resulting in the predominant loss of small-fiber axons characteristic of early-onset cases, while vasculopathy can also determine the pathogenesis of neuropathy in late-onset cases [
17]. Once polyneuropathy begins, it progresses rapidly, evolving in three defined stages: in the first, patients have a sensory polyneuropathy that leads to difficulty walking without assistance; in the second, assistance with walking is required; and in the last stage, patients are wheelchair-bound or bedridden. Death occurs on average 12 years after the onset of the disease in the absence of therapy, but survival can vary with different mutations [
18]. The presence of amyloidotic heart disease can influence this classical progression, and the survival for variants with prevalent cardiological involvement is generally shorter [
19,
20]. The clinical phenotype based on the mutations is not always predictable, as there is considerable variability in clinical expression even within the same family [
21].
In the last three decades, there has been an overall improvement in the management of this disease, and various disease-modifying therapies have been developed [
15,
16]. Moreover, awareness of being ahead of a multisystem pathology requiring a multidisciplinary approach has also improved the management of symptoms [
18,
22]. In parallel, increased diagnostic tools and awareness of the disease have reduced the diagnostic delay, favoring an earlier start of treatment [
4,
23].
To date, some studies on the efficacy of new drugs in modifying the outcomes of patients over short periods have been performed, but information on survival is lacking, mainly in non-V30M patients [
24,
25,
26,
27,
28,
29,
30,
31,
32]. A recent paper showed that in non-endemic late-onset ATTRv amyloidosis, long-term treatment with tafamidis is safe and efficacious, and the severity of polyneuropathy at the start of treatment is the main predictor for disease progression on tafamidis [
30].
The aim of this work was to analyze whether current therapies have prolonged survival in Sicilian patients affected by ATTRv amyloidosis, representing an almost exclusively non-V30M patient population. Another purpose was to update the clinical–epidemiological data in Sicily [
10], where diagnosed cases have increased dramatically in the last five years.
2. Materials and Methods
We conducted a retrospective study, including all patients and carriers who had a pathogenic missense mutation on the TTR gene followed in the two Sicilian referral centers for the disease (Amyloidosis Centre of Messina University Hospital and Amyloidosis Centre of Palermo University Hospital) from January 1991 until December 2020. The study was approved by the Ethical Committee of Messina Province. All methods were performed in accordance with the relevant guidelines and regulations. The diagnosis was based on clinical and neurophysiological features and confirmed by positive genetic testing for TTR gene variants. In some patients, usually the probands, especially in the first half of the observed period, the diagnosis was also corroborated by biopsy, in most cases of the sural nerve. We also performed TTR gene analysis in asymptomatic relatives close to the typical age of onset, with the aim of following them serially to reduce the diagnostic delay and, consequently, to start therapy in a timely manner.
The clinical data collected from available medical records consisted of:
The age at onset, diagnosis, and death, or age of living patients as of 31 December 2020,
The family history of neuropathy or ATTR amyloidosis,
The type of TTR mutation,
The clinical and neurological examination at diagnosis,
Electromyography and nerve conduction studies at diagnosis,
The presence of autonomic nervous system involvement at diagnosis, when available and based on autonomic cardiovascular tests, or, in their absence, based on clinical evaluation,
The presence of cardiac involvement at diagnosis based on ECG, echocardiogram, atrial natriuretic hormone dosage (nt-pro-BNP), and, in some cases, 99 m Tc-3, 3-diphosphono-1, 2-propanodicarboxylic acid scintigraphy or cardiac magnetic resonance (CMR) with T2-weighted imaging and late gadolinium enhancement,
Sural nerve biopsy in selected cases,
Therapeutic drugs (or procedures) and the overall duration of their administration,
Symptomatic treatment.
Statistical Analysis
The characteristics of the patients were described by using standard descriptive statistics. All the results are documented as mean values, ranges, or percentages when appropriate. To assess the relationships between treatment and survival, between clinical characteristics in treated patients at the onset and survival, and between mutation (E89Q vs. F64L) and survival, curves were estimated by Kaplan–Meier analysis and compared using the log rank test. The effect size is expressed as the hazard ratio, and statistical uncertainty is expressed as 95% confidence intervals. The comparison of the age of onset and disease duration for deceased patients between treated and untreated was performed using chi squared tests.
3. Results
3.1. General Aspects
This study examined 159 patients with pathogenic mutations in the
TTR gene at the two reference centers from 1991 to 2020, 135 in Messina and 24 in Palermo. The most frequently encountered mutation was F64L (p.F84L) in 81 patients, followed by E89Q in 56, V122I (p.V142I) in 11, T49A in 8, and V30M in 3. Among them, 40 subjects (20 men and 20 women, mean age 46.6, range 28–66) were still asymptomatic as of 31 December 2020. In addition, 14 patients (4 men and 10 women, mean age 58.3 years, range 41–78) were affected by carpal tunnel syndrome, without other symptoms or signs of ATTRv amyloidosis. The clinical characteristics of the other 105 patients with ATTRv amyloidosis are described in
Table 1. Of the 105 patients, 71 received disease-modifying therapy and 34 did not receive specific drugs for the disease.
Three mutations usually have late onset (F64L, V122I, and V30M), T49A has early onset, and the age of onset of patients with the E89Q variant is usually around 50. Clinical evaluations at diagnosis show frequent cardiac involvement in E89Q and V122I patients, while dysautonomia is frequently present in T49A patients. New diagnoses progressively increased in the 30-year period, reaching 45 cases in the last five years (
Figure 1). The most frequently identified variant over the past five years was F64L, in 29 of 45 patients.
Regarding the type of polyneuropathy, in the 105 patients who developed polyneuropathy, a nerve conduction study at diagnosis showed an axonal pattern in 94, demyelinating in 5, and mixed in 6. Sural nerve biopsy was performed in 16 patients, among which 10 were positive to Congo Red staining; in all of these cases there was a marked loss of myelinated fibers.
3.2. Actual Scenario
As of 31 December 2020, 73 living patients and 40 asymptomatic carriers underwent regular follow-up, and 44 patients received disease-modifying treatment: 23 patisiran (3 also received liver transplants), 17 tafamidis, and 4 inotersen. Only one patient was liver-transplanted and did not receive other therapies. No other disease-modifying treatments are currently in use in Sicily. In light of these results, the current prevalence of ATTRv amyloidosis in Sicily is 15.0/1,000,000 (12.1/1,000,000 excluding patients with carpal tunnel syndrome only).
3.3. Comparison of Treated and Untreated Patients
Table 2 shows the characteristics of patients who received a disease-modifying treatment, compared to those patients who did not.
The two groups did not show a significant difference in age of onset (
p = 0.24) and were quite similar, considering patient
TTR mutations and symptoms at diagnosis. The type of treatment administered was measured as years of therapy received, and in the case of patients who underwent liver transplantation (six in this cohort), the years of therapy were considered as all years of life after the transplant. The Kaplan–Meyer curves of survival in the two groups are represented in
Figure 2. The survival was longer for treated patients (
p = 0.0004).
The median survival of treated and untreated patients was 12 years vs. 8 years, and the hazard ratio for untreated/treated was 2.55. Furthermore, considering only deceased patients, the disease duration was longer for treated patients than untreated patients (
p = 0.04) (
Table 2).
3.4. Relation between Clinical Characteristics at the Onset and Survival in Treated Patients
In the 71 treated patients, survival was not significantly different as it related to the presence of cardiac involvement (
p = 0.79), weight loss (
p = 0.1), or autonomic dysfunction (
p = 0.05) at diagnosis. Furthermore, survival was not significantly different comparing patients with E89Q vs. those with F64L mutation (
p = 0.19). A very significant difference (
p = 0.0002) was found between patients who had a polyneuropathy disability score (PND) [
34] of 1 (sensory disturbance but preserved walking capability) at diagnosis vs. patients with a more advanced polyneuropathy score (
Figure 3).
3.5. Symptomatic Drugs
In 51 of the 105 patients with polyneuropathy, a specific treatment for neuropathic pain had been administered. However, these findings are likely to have been underreported in the clinical notes, especially in the first half of the period under review, because the general practitioner sometimes managed these drugs. The most used drugs for neuropathic pain were pregabalin (38 cases), gabapentin (4), amitriptyline (21), duloxetine (12), and tramadol (6). The most used symptomatic treatments for autonomic symptoms were fludrocortisone, midodrine, loperamide, and octreotide. Cardiac dysfunction was generally treated with diuretics, ace inhibitors, beta blockers, and implantable cardioverter defibrillators. No patients underwent a heart transplant.