Abstract
Background
Amyloid transthyretin (ATTR) amyloidosis is a rare, life-threatening disease frequently manifesting with cardiomyopathy (ATTR-CM), polyneuropathy (ATTR-PN) or both (ATTR-mixed). We retrospectively analysed US electronic health records and claims data to provide up-to-date estimates of ATTR amyloidosis epidemiology (overall and by phenotype).
Methods
Data were extracted from the Clarivate Real-World Data repository (2016−2023). Given the lack of established coding for ATTR amyloidosis, we used different combinations of diagnostic codes to obtain narrow and broad estimates of incident and prevalent cases in the USA in 2022. Temporal trends (2019–2022) were also assessed.
Results
Using narrow definitions, the 2022 estimated incidence of ATTR amyloidosis overall, ATTR-CM, ATTR-PN and ATTR-mixed was 16.6, 12.7, 3.5 and 1.9 cases per million people, respectively; the corresponding prevalence estimates were 59.8, 41.1, 15.1 and 9.8 cases per million people. Estimates were consistently lower with the narrow (vs broad) definitions. Over time, the incidence and prevalence of ATTR amyloidosis overall increased, driven by ATTR-CM cases. No major changes were reported for the other phenotypes.
Conclusion
This study provides comprehensive and up-to-date epidemiological data for ATTR amyloidosis in the USA. Our findings corroborate the need for appropriate differential diagnostic coding and standardised criteria.
Keywords: Amyloidosis, Cardiomyopathies, Epidemiology, Polyneuropathy, Rare Diseases
WHAT IS ALREADY KNOWN ON THIS TOPIC
Amyloid transthyretin (ATTR) amyloidosis is a progressive and life-threatening disease that can present with different phenotypes. ATTR amyloidosis is currently classified as a rare disease, but its prevalence is likely to be underestimated. In particular, epidemiological estimates for the USA are outdated and limited to the cardiomyopathy phenotype.
WHAT THIS STUDY ADDS
This study used different combinations of diagnostic and drug treatment codes to provide comprehensive and up-to-date estimates of incident and prevalent cases of ATTR amyloidosis in the USA in 2022 as well as temporal trends over the 2019–2022 period.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Our findings corroborate the need for appropriate differential diagnostic coding and standardised criteria in ATTR amyloidosis.
Introduction
Amyloid transthyretin (ATTR) amyloidosis is a progressive, systemic, life-threatening disease caused by the tissue deposition of misfolded transthyretin (TTR), a carrier protein mainly produced in the liver. Accumulation of insoluble amyloid fibres causes tissue damage and eventual failure of the affected organs, most commonly the heart, peripheral nerves and the renal and gastrointestinal systems.1
The three main phenotypes of ATTR amyloidosis are cardiomyopathy (ATTR-CM), polyneuropathy (ATTR-PN) and a mixed form (ATTR-mixed) with both cardiac and neurological manifestations. Moreover, many patients present with musculoskeletal conditions (eg, carpal tunnel syndrome and lumbar spinal stenosis) years before the appearance of cardiac signs.2 Most patients with ATTR-CM have a wild-type TTR gene, whereas the majority of those with ATTR-PN or a mixed phenotype have a TTR mutation (variant ATTR, often referred to as hereditary ATTR).3,5 ATTR-CM is more common in men and elderly (≥65 years) patients4 5 and is characterised by accumulation of the TTR amyloid fibrils in the myocardium, causing stiffening and cardiac function impairment.6 Most patients with ATTR-CM develop heart failure (HF),3 and 2.9–15.0% of those with HF have been reported to have ATTR-CM,7 8 making HF a marker that should prompt consideration of an ATTR-CM diagnosis.9 ATTR-PN is characterised by extracellular deposition of TTR fibrils and progressive destruction of somatic and autonomic peripheral nerves leading to loss of autonomy and death.1 10
ATTR amyloidosis is currently classified as a rare disease, but its prevalence is likely to be underestimated. Despite the considerable variability in epidemiology data across studies and patient populations,3,57 11 12 the incidence and prevalence of diagnosed ATTR amyloidosis have risen over the past decade, primarily due to advances in diagnostic tools (ie, cardiac imaging and genetic testing) and greater disease awareness.13,19 Major contributors to under-recognition and misdiagnosis of the disease, which can lead to delayed diagnosis and care and inadequate treatment,20 21 include lack of standardised coding algorithms for the different subtypes of ATTR amyloidosis,10 22 23 poor disease awareness and high symptom heterogeneity.24 25 In the Transthyretin Amyloidosis Outcomes Survey (THAOS) study, the largest global observational study of ATTR amyloidosis with >6000 patients, the phenotype distribution for symptomatic patients (n=4428) at enrolment was 40.7% predominantly cardiac, 40.1% predominantly neurological, 16.6% mixed and 2.5% no phenotype.26 In the North American patient cohort (n=1594), the most common phenotype at enrolment was predominantly cardiac (63.9%).3 A recent analysis of the THAOS study showed that about one in three patients initially diagnosed with either ATTR-CM or ATTR-PN were reclassified as having a mixed phenotype within a median of 1–2 years of follow-up.27 Given the poor prognosis of untreated ATTR amyloidosis, especially of ATTR-CM (life expectancy estimated at 2–5 years after diagnosis28), and the disease-modifying agents under development,9 10 early and accurate identification of affected patients and prompt treatment should be a priority of clinical management.9
Most of the available observational studies that provide ATTR amyloidosis epidemiological estimates in the US report data for the 2000–2020 period and for the CM phenotype only.4 5 16 29 Gilstrap et al16 assessed the incidence and prevalence of ATTR-CM among Medicare beneficiaries aged ≥65 years from 2000 to 2012 and Brown et al29 used a large US commercial claims database to identify adult patients diagnosed with ATTR-CM between 2014 and 2018. The latter was the only study, out of the 26 US studies identified in the systematic literature review (SLR) of epidemiological studies by Delgado et al,4 5 that reported incidence and prevalence data in the general population. In another US study identified in the SLR, Akers and Lynch30 reported the prevalence of ATTR-CM among patients aged ≥65 years included in a health plan between 2019 and 2020.
To fill this evidence gap, we conducted a large real-world study to provide up-to-date estimates of ATTR amyloidosis epidemiology (overall and by phenotype) in the USA, using electronic health records (EHRs) and claims data to identify patients diagnosed with ATTR amyloidosis between 2017 and 2023 and describe patient demographics and comorbidities.
Methods
Study design and data sources
This retrospective study used data from the commercially available Clarivate Real-World Data repository, a database of US EHRs and open claims obtained from clearinghouses (digital hubs for the transmission of electronic claims from healthcare providers to government and commercial payers).31 32 More details on the type of data included in the Clarivate Real-World Data repository are provided in the online supplemental methods.
Because data were deidentified (in accordance with the US Health Insurance Portability and Accountability Act regulations) and retrospectively analysed, no institutional review board approval was required.
Data for this study were collected from records dated between 1 January 2016 and 31 December 2023.
Patient identification
Given the lack of established coding for ATTR amyloidosis, we developed different algorithms using a combination of diagnostic codes (International Classification of Diseases (ICD)−10 and Systemized Nomenclature of Medicine (SNOMED)), drug treatment codes (national drug code (NDC) assigned to a Healthcare Common Procedure Coding System) and physician’s mention of ATTR amyloidosis (hereinafter referred to as physician note) to obtain a range of narrow to broad estimates. Therefore, the main difference between broad and narrow definitions relies on the codes used, particularly with respect to ICD-10 and SNOMED codes (see Online supplemental methods, online supplemental tables 1 and 2). These algorithms were based on the literature and expert opinion (epidemiologists, neurologists and cardiologists). Narrow and broad definitions were used to identify the following cohorts: ATTR amyloidosis overall, ATTR-CM, ATTR-PN, ATTR-CM-only, ATTR-PN-only and ATTR-mixed. The date of the first ATTR amyloidosis-qualifying event during the patient identification period was considered as the index date and had to occur between 1 January 2017 and 31 December 2023 (online supplemental figure 1). The phenotype diagnosis had to occur within 1 year preindex or postindex date. Due to incomplete postindex data for patients with index dates in 2023, estimates for this year were not reported. For each phenotype, the index date definition, phenotype time frame and inclusion and exclusion criteria, with the ICD-10 codes, are specified in online supplemental table 1; the SNOMED codes used for the narrow and broad definitions, and the drug codes are reported in online supplemental tables 2 and 3), respectively. Patients were required to be ≥18 years old on the index date.
Using the broad definition, patients were identified as having ATTR-CM based on the methods from Lauppe et al14: a SNOMED broad code for ATTR-CM, physician notes for ATTR-CM or a drug treatment code for ATTR-CM. However, while Lauppe et al14 used the date of the phenotype diagnosis as the index date, we used the date of the first incidence of amyloidosis diagnosis to align with all the cohort definitions. Moreover, they included patients with a 2-year interval between the diagnosis of amyloidosis and of the phenotype, whereas we considered a 1-year interval. Further details on the identification of patients with ATTR amyloidosis overall, ATTR-CM and ATTR-PN are provided in the online supplemental methods.
All patients who were ATTR-CM and not ATTR-PN during the study period by either narrow or broad definitions were included in two additional cohorts: ATTR-CM only narrow and ATTR-CM only broad, respectively. Similarly, all patients who were ATTR-PN and not ATTR-CM during the study period by narrow or broad definitions were included in the ATTR-PN only narrow and ATTR-PN only broad cohorts, respectively. A similar methodology as above was also created for mixed cohorts where patients had both ATTR-CM and ATTR-PN during the study period. The index date for the mixed cohort is the earliest of either ATTR-CM or ATTR-PN index dates.
Key exclusion criteria included a diagnosis of light-chain amyloidosis (ICD-10 code E85.81, physician notes or SNOMED code), a diagnosis of multiple myeloma and treatment for light-chain amyloidosis or multiple myeloma during the study period. For the ATTR-CM cohort, the following additional exclusion criteria were considered based on the study by Lauppe et al14: patients who had ≥3 haematologist visits during the study time period and patients who had liver, heart or haematopoietic stem cell transplants prior to the amyloidosis diagnosis.
Incident and prevalent cohorts
Patients with no ATTR amyloidosis events during the baseline period (ie, the period between study start (1 January 2016) and the first possible index date (1 January 2017) with ≥12 months of continuous clinical activity) were considered to have incident ATTR amyloidosis. A minimum of 3 years of wash-out prior to the index date was implemented for incident cases. Patients with ATTR amyloidosis-qualifying events during the baseline period were considered to have prevalent disease. Due to the chronic nature of the disease, incident and prevalent cases were considered to have prevalent disease for the remainder of the study period unless they died or became unobservable in the data, irrespective of any subsequent ATTR amyloidosis-qualifying events.
Patient data
Demographics (age, sex) on the index date and the Charlson Comorbidity Index (CCI) 1 year prior to the index date were extracted for the 2022 ATTR amyloidosis incident population. The CCI score allows categorisation of comorbidity severity as mild (CCI=1–2), moderate (CCI=3–4) or severe (CCI ≥5).33 A score of 0 indicates absence of any comorbidities among AIDS/HIV, cancer, cerebrovascular disease, chronic pulmonary disease, congestive HF, connective tissue disease, dementia, diabetes with and without complications, metastatic carcinoma, mild liver disease, moderate or severe liver disease, myocardial infarction, paraplegia and hemiplegia, peptic ulcer disease, peripheral vascular disease and renal disease.
Outcomes
The primary outcomes were incidence and prevalence of ATTR amyloidosis (overall and by phenotype) over 1 year (2022). Sex and age distribution of incident and prevalent cases was also evaluated, together with temporal trends for the 2019–2022 period.
Statistical analysis
Annual prevalence and incidence were calculated using population-based data to provide the most up-to-date epidemiological estimates available. Sample means of prevalence and incidence were calculated from bootstrapped samples. The denominator for the sample mean consisted of all individuals within the Clarivate Real-World data repository that had an encounter during each individual year. From these data, 10 000 samples with replacement were created in order to calculate incidence and prevalence estimates. The sample means were averaged to obtain the yearly incidence and prevalence, and then adjusted by year, state of residence at index date, gender and age, using US census data. This method has previously been used to provide population estimates.32 Data for the overall ATTR amyloidosis population and for the phenotype cohorts were summarised using mean, SD and frequencies. Annual prevalence and incidence estimates were expressed as cases per million people. The compound annual growth rate was used to estimate the change in incidence and prevalence over time. Structured Query Language and Python were used to conduct all statistical analyses.
Results
Study population
The attrition flow chart is presented in online supplemental figure 2. Overall, after applying the criteria for the broad definition specified in the methods, 88 680 patients were included in the ATTR amyloidosis overall cohort, 81 935 in the ATTR-CM cohort and 13 095 in the ATTR-PN cohort. With the narrow definition, the corresponding figures were 17 775, 13 400 and 3811. In general, ICD-10 codes and NDCs were the data sources leading to the identification of most of the cases across the different ATTR cohorts, whereas physician notes and SNOMED led to the identification of a minority of cases (online supplemental table 4).
Incidence of ATTR amyloidosis
The 2022 incidence estimates for ATTR overall and by phenotype are reported in figure 1. Estimates were consistently higher for all cohorts when the broad definitions were used. The number of incident cases per million people was 81.1 for ATTR overall with the broad definition and 16.6 using the narrow definition. Regardless of the definition used, the highest estimates were reported for the ATTR-CM cohort (75.1 cases per million people with the broad definition and 12.7 with the narrow definition) and ATTR-CM only cohort (66.9 and 10.8 cases per million people, respectively), followed by the ATTR-PN (12.0 and 3.5 cases per million people, respectively), ATTR-mixed (8.0 and 1.9 cases per million people, respectively) and ATTR-PN only (4.7 and 1.7 cases per million people, respectively) cohorts.
Figure 1. Incidence estimates of ATTR (overall and by phenotype) by definition, in 2022 in the USA. Bars represent the 95% CIs. ATTR, amyloid transthyretin; CM, cardiomyopathy; PN, polyneuropathy.
The main characteristics of patients with incident disease in 2022 are reported in table 1 and figure 2.
Table 1. Characteristics of patients with incident disease in 2022 in the USA.
| Phenotype | Broad definition | Narrow definition |
|---|---|---|
| ATTR, n | 15 164 | 3243 |
| Age (years), mean (SD) | 75.3 (11.8) | 75.0 (12.6) |
| Age groups, n (%) | ||
| 18–34 years | 121 (0.8) | 56 (1.7) |
| 35–54 years | 877 (5.8) | 194 (6.0) |
| 55–64 years | 1394 (9.2) | 251 (7.7) |
| ≥65 years | 12 772 (84.2) | 2742 (84.6) |
| Males, n (%) | 9503 (62.7) | 2291 (70.6) |
| CCI, n (%) | ||
| Mild (1–2) | 4139 (27.3) | 978 (30.2) |
| Moderate (3–4) | 3247 (21.4) | 614 (18.9) |
| Severe (5+) | 4996 (32.9) | 692 (21.3) |
| CCI score of 0 | 2782 (18.3) | 959 (29.6) |
| ATTR-CM, n | 14 190 | 2632 |
| Age (years), mean (SD) | 75.8 (11.5) | 77.9 (9.8) |
| Age groups, n (%) | ||
| 18–34 years | 91 (0.6) | 15 (0.6) |
| 35–54 years | 750 (5.3) | 70 (2.7) |
| 55–64 years | 1218 (8.6) | 122 (4.6) |
| ≥65 years | 12 131 (85.5) | 2425 (92.1) |
| Males, n (%) | 9053 (63.8) | 2011 (76.4) |
| CCI, n (%) | ||
| Mild (1–2) | 3841 (27.1) | 816 (31.0) |
| Moderate (3–4) | 3106 (21.9) | 562 (21.4) |
| Severe (5+) | 4852 (34.2) | 631 (24.0) |
| CCI score of 0 | 2391 (16.8) | 623 (23.7) |
| ATTR-CM only, n | 12 637 | 2265 |
| Age (years), mean (SD) | 75.7 (11.6) | 78.2 (9.8) |
| Age groups, n (%) | ||
| 18–34 years | 82 (0.6) | 13 (0.6) |
| 35–54 years | 704 (5.6) | 59 (2.6) |
| 55–64 years | 1118 (8.8) | 103 (4.5) |
| ≥65 years | 10 733 (84.9) | 2090 (92.3) |
| Males, n (%) | 8000 (63.3) | 1745 (77.0) |
| CCI, n (%) | ||
| Mild (1–2) | 3446 (27.3) | 719 (31.7) |
| Moderate (3–4) | 2746 (21.7) | 482 (21.3) |
| Severe (5+) | 4279 (33.9) | 522 (23.0) |
| CCI score of 0 | 2166 (17.1) | 542 (23.9) |
| ATTR-PN, n | 2250 | 644 |
| Age (years), mean (SD) | 74.4 (11.4) | 72.0 (12.0) |
| Age groups, n (%) | ||
| 18–34 years | 16 (0.7) | 7 (1.1) |
| 35–54 years | 120 (5.3) | 50 (7.8) |
| 55–64 years | 220 (9.8) | 74 (11.5) |
| ≥65 years | 1894 (84.2) | 513 (79.7) |
| Males, n (%) | 1371 (60.9) | 412 (64.0) |
| CCI, n (%) | ||
| Mild (1–2) | 636 (28.3) | 175 (27.2) |
| Moderate (3–4) | 475 (21.1) | 113 (17.5) |
| Severe (5+) | 669 (29.7) | 152 (23.6) |
| CCI score of 0 | 470 (20.9) | 204 (31.7) |
| ATTR-PN only, n | 812 | 294 |
| Age (years), mean (SD) | 70.3 (12.4) | 67.4 (12.7) |
| Age groups, n (%) | ||
| 18–34 years | 10 (1.2) | 5 (1.7) |
| 35–54 years | 77 (9.5) | 39 (13.3) |
| 55–64 years | 124 (15.3) | 55 (18.7) |
| ≥65 years | 601 (74.0) | 195 (66.3) |
| Males, n (%) | 389 (47.9) | 154 (52.4) |
| CCI, n (%) | ||
| Mild (1–2) | 272 (33.5) | 81 (27.6) |
| Moderate (3–4) | 139 (17.1) | 35 (11.9) |
| Severe (5+) | 142 (17.5) | 48 (16.3) |
| CCI score of 0 | 259 (31.9) | 130 (44.2) |
| ATTR-mixed, n | 1551 | 361 |
| Age (years), mean (SD) | 76.7 (10.1) | 75.8 (10.0) |
| Age groups, n (%) | ||
| 18–34 years | 8 (0.5) | 3 (0.8) |
| 35–54 years | 45 (2.9) | 11 (3.0) |
| 55–64 years | 102 (6.6) | 18 (5.0) |
| ≥65 years | 1396 (90.0) | 329 (91.1) |
| Males, n (%) | 1053 (67.9) | 264 (73.1) |
| CCI, n (%) | ||
| Mild (1–2) | 397 (25.6) | 97 (26.9) |
| Moderate (3–4) | 354 (22.8) | 78 (21.6) |
| Severe (5+) | 571 (36.8) | 108 (29.9) |
| CCI score of 0 | 229 (14.8) | 78 (21.6) |
ATTR, amyloid transthyretin; CCI, Charlson Comorbidity Index; CM, cardiomyopathy; PN, polyneuropathy.
Figure 2. Sex and age distribution of the ATTR amyloidosis incident population (overall and by phenotype) by definition, in 2022 in the USA. ATTR, amyloid transthyretin; CM, cardiomyopathy; PN, polyneuropathy.
Across all cohorts, most patients were aged ≥65 years (66.3–92.3%) and predominantly male (60.9–77.0%), except for the ATTR-PN only cohort, where males accounted for 47.9% of the population when using the broad definition and 52.4% with the narrow definition. The proportion of males was consistently lower using the broad definition, compared with the narrow definition. Regardless of the definition used, the mean age of patients was ≥72.0 years in all cohorts, except for the ATTR-PN only cohort, where it was slightly lower (70.3 years and 67.4 years with the broad and narrow definitions, respectively). The proportion of patients with no comorbidities (CCI score=0) was 18.3% with the broad definition and 29.6% with the narrow definition in the ATTR amyloidosis overall cohort and was lowest in the ATTR-mixed cohort (14.8% and 21.6%, respectively) and highest in the ATTR-PN only cohort (31.9% and 44.2%, respectively). The corresponding figures for patients with a CCI score ≥5, indicating severe comorbidities, were 32.9% and 21.3% in the ATTR amyloidosis overall cohort, with the lowest proportions reported for the ATTR-PN only cohort (17.5% and 16.3%) and the highest for the ATTR-mixed cohort (36.8% and 29.9%).
Prevalence of ATTR amyloidosis
The 2022 prevalence data for ATTR amyloidosis overall and by phenotype are reported in figure 3. Estimates were consistently higher with the broad definition compared with the narrow definition. There were 250.8 prevalent cases of ATTR amyloidosis per million people with the broad definition and 59.8 with the narrow definition. Similarly to incidence, the highest prevalence estimates were reported for the ATTR-CM (224.0 and 41.1 cases per million people with the broad and narrow definitions, respectively) and ATTR-CM only cohorts (193.3 and 31.9 cases per million people, respectively) followed by the ATTR-PN (47.3 and 15.1 cases per million people, respectively), ATTR-mixed (30.7 and 9.8 cases per million people, respectively) and ATTR-PN only (19.1 and 6.0 cases per million people, respectively) cohorts.
Figure 3. Prevalence estimates of ATTR (overall and by phenotype) by definition, in 2022 in the USA. Bars represent the 95% CIs. ATTR, amyloid transthyretin; CM, cardiomyopathy; PN, polyneuropathy.
As shown in figure 4, patients were predominantly male in all cohorts except for the ATTR-PN only cohort (48% with the broad definition and 54% with the narrow definition), and the proportion of males was consistently higher with the narrow definitions compared with the broad definitions. Moreover, over 80% of patients were aged ≥65 years in all cohorts except for the ATTR-PN only cohort (75% with the broad definition and 66% with the narrow definition).
Figure 4. Sex and age distribution of the ATTR amyloidosis prevalent population (overall and by phenotype) by definition, in 2022 in the USA. ATTR, amyloid transthyretin; CM, cardiomyopathy; PN, polyneuropathy.
ATTR amyloidosis incidence and prevalence trends from 2019 to 2022
Results for the incidence and prevalence trends over the 2019–2022 period are shown in figures5 6, respectively. With both the narrow and broad definitions, the incidence of ATTR amyloidosis overall increased by 6.9% and 9.2%, respectively, driven by an increase in ATTR-CM (by 10.5% and 10.0%) and ATTR-CM only (by 13.9% and 10.2%) incidence. The changes observed for the ATTR-PN, ATTR-PN only and ATTR-mixed cohorts over time were 3.0%, 15.6% and −4.8% with the narrow definitions and 7.7%, 10.3% and 4.6% with the broad definitions.
Figure 5. Trends for ATTR incidence by broad (A) and narrow (B) definitions over the 2019–2022 period. Bars represent the 95% CIs. In 2020 and 2021, which were the years when the COVID-19 pandemic occurred, there was a significant drop in the denominator of the data. While adjustments were made, the trend during 2020 and 2021 needs to be interpreted with caution as results may reflect the healthcare situation during the pandemic, rather than an actual decrease in ATTR amyloidosis incidence. ATTR, amyloid transthyretin; CM, cardiomyopathy; PN, polyneuropathy.
Figure 6. Trends for ATTR prevalence by broad (A) and narrow definitions (B) over the 2019–2022 period. Bars represent the 95% CIs. ATTR, amyloid transthyretin; CM, cardiomyopathy; PN, polyneuropathy.
A similar pattern was observed for prevalence (ATTR amyloidosis overall: 26.1% and 22.0% increase with the narrow and broad definitions, respectively), which increased over time primarily due to ATTR-CM (by 32.7% and 23.3%) and ATTR-CM only (by 38.1% and 23.5%) regardless of the definition used. A 15.2–23.3% increase in the prevalence of ATTR-PN, ATTR-PN only and ATTR-mixed was also observed.
Discussion
This is a comprehensive and up-to-date analysis of epidemiological data for ATTR amyloidosis in the USA, providing incidence and prevalence estimates of the disease overall and by the different phenotypes using various definitions.
For 2022, the present study identified 81.1 incident cases of ATTR amyloidosis overall per million people with the broad definition and 16.6 with the narrow definition, 75.1 and 12.7, respectively, of ATTR-CM and 12.0 and 3.5 of ATTR-PN. Prevalent cases per million people were 250.8 and 59.8 for ATTR amyloidosis overall, 224.0 and 41.1 for ATTR-CM and 47.3 and 15.1 for ATTR-PN. Moreover, we examined the trends in the incidence and prevalence of ATTR amyloidosis, overall and by phenotype, over the 2019–2022 period, using both the broad and narrow definitions. Regardless of the definition used, the incidence and prevalence of ATTR amyloidosis overall tended to increase over time, driven by a marked increase in the number of cases of ATTR-CM. An increase in the incidence and prevalence of ATTR was also found by Gilstrap et al,16 who included only patients with ATTR-CM aged ≥65 years and reported trends for the period 2000–2012. The increase in the incidence and prevalence of ATTR amyloidosis may be attributed to several factors, including improvements in diagnostic tools and disease management, that have translated into improved patient outcomes.13,1934 Traditionally, the diagnosis of ATTR amyloidosis has relied on histology, which is invasive and time-consuming. The adoption of non-invasive techniques, such as bone scintigraphy for ATTR-CM, has enabled accurate diagnosis while eliminating the need for confirmatory endomyocardial biopsies.35 Advancements in ATTR treatment have led to benefits in patient function as well as survival, particularly when the disease is diagnosed and treated at an early stage, resulting in increased disease prevalence.36 37
A recent SLR of studies published between January 2018 and April 2023 assessed the epidemiology of ATTR amyloidosis.4 5 Of the 26 included studies conducted in North America, only one analysis of US commercial claims reported estimates in the general population, but these were limited to ATTR-CM.29 In this study, Brown et al29 included adult patients diagnosed with ATTR based on ≥1 inpatient or ≥2 outpatient claims with an ICD-10-CM code for hereditary (E85.1, E85.2) or wild-type (E85.82) form in 2018 or another amyloidosis form in 2018; those with evidence of congestive HF or CM (restrictive or hypertrophic) between 2014 and 2018 were identified as ATTR-CM cases. Patients with ≥1 claim for chemotherapy, stem cell transplant, light-chain amyloidosis or dementia were excluded. In 2018, the incidence of ATTR-CM was 4.0 per million people-year and the prevalence was 6.1 cases per million people, which is far less than the estimates in our analysis, even when the narrow definition was used (2019 incidence: 9.4 cases per million people; 2019 prevalence: 17.6 cases per million people).29 While the ICD-10-CM codes used by Brown et al29 to identify patients with ATTR-CM are the same as those used in our study to develop the narrow definition of this same phenotype, they did not consider SNOMED codes for ATTR-CM, physician notes for ATTR-CM or any medications for ATTR-CM, and excluded patients with dementia. Moreover, Brown et al29 extracted data for the 2014–2018 period from the IBM MarketScan Commercial database, which included data from 250 million patients collected between 2007 and 2022,38 and the Medicare Supplement databases, which, in 2024, included 0.6 million people.39 These differences in patient identification criteria, population size and period of data collection may at least partly explain the higher estimates reported in our study. The different period of time is particularly relevant considering the rise in disease awareness and improvements in diagnostic technologies over the past few years.13,19 Similarly, methodological differences likely contributed to the variation in estimates between our study and that of Gilstrap et al,16 where data were extracted from the Medicare database (which, in 2024, included 66.4 million people, 88.8% aged ≥65 years) for the period 2000–2012. In this study, the incidence and prevalence of ATTR-CM ranged from 18.0 to 55.2 cases per 100 000 person-years and from 8 to 16.6 cases per 100 000 person-years, with the highest estimates reported for black men (incidence: 36 per 100 000 person-years; prevalence: 174 cases per 100 000 person-years).16
In the current literature, the lack of specific diagnostic codes for all ATTR amyloidosis phenotypes has prompted the development of various strategies for identifying potential ATTR cases in existing databases. These rely on combinations of ICD diagnostic codes for amyloidosis, presence of related cardiovascular conditions (eg, CM or HF, or abnormal ECG results),15 16 18 29 40 or medical procedures related to HF, arrhythmias, conduction disorders or cardiomyopathies.15 Differences in the case definition used across studies likely contribute to the inconsistent epidemiology data reported for ATTR amyloidosis.3,57 11 12 In our study, we used a combination of parameters to provide both broad and narrow disease definitions. As expected, using narrow definitions, we observed consistently lower prevalence and incidence estimates in the overall population with ATTR amyloidosis and across all cohorts, compared with the broad definitions, likely due to the identification of fewer false positive cases. However, the rather broad ranges of incidence and prevalence estimates found in the present study and in previous analyses highlight the need for a more standardised approach to accurately identify patients with ATTR in existing databases for research purposes.
In 2022, patients with incident ATTR amyloidosis were predominantly males, aged ≥65 years and with a CM phenotype, in line with previous studies of US patients.3 29 Regardless of the definition used (broad or narrow), the ATTR-CM and ATTR-CM-only phenotypes had the highest incidence and prevalence, followed by ATTR-PN, ATTR-mixed and ATTR-PN only. Differences in phenotype distribution have been observed across geographies. In the large multinational THAOS study, which included patients from Asia, Europe and South and North America, the predominantly cardiac phenotype accounted for the majority of cases in North America (63.9% vs 7.3–20.8% in the other geographies), whereas in the other continents most patients had the predominantly neurological or mixed phenotype.3 The study was performed after tafamidis was first approved for the treatment of early-stage ATTR-PN by the European Medicines Agency in 2011,41 and, therefore, the relative proportions of CM and PN reported for Europe do not reflect the real-world share of the two phenotypes. In the THAOS study, patients with the predominantly neurological or mixed phenotype accounted for 12.9% and 18.7%, respectively, of all symptomatic patients in North America.3 In our study, however, the ATTR-PN phenotype was more prevalent than the ATTR-mixed phenotype, with 47.3 and 30.7 cases per million people, respectively.
Here, we included patients who exhibited a consistent phenotype (either CM or PN) throughout the study in the ATTR-CM only and ATTR-PN only cohorts, and those with mixed characteristics into the ATTR-mixed cohort. Patients with a CM-only phenotype accounted for 85.1–89.1% (based on the definition used) of the total CM incident cases, whereas those with a mixed phenotype represented <15% of cases. In contrast, in the ATTR-PN cohort, patients with the PN-only phenotype accounted for 39.1–48.6% of the total incident cases, with mixed phenotype patients being the majority. The greater proportion of patients with a mixed phenotype within the ATTR-PN versus ATTR-CM cohorts is in line with previous findings indicating that a considerable proportion of patients presenting with ATTR-PN exhibit cardiac involvement (35–63% in ATTR-PN trials).27 42 43 In the THAOS study, among the patients initially diagnosed with ATTR-CM or ATTR-PN who were reclassified within 1–2 years as having a mixed phenotype, 69.2% were initially from a predominantly neurological phenotype and 31.8% were from a predominantly cardiac phenotype.27 Overall, these findings suggest that mixed phenotypes may be more common than previously thought and emphasise the need for multidisciplinary evaluation at both the initial assessment and the follow-up for all patients with ATTR amyloidosis.
While real-world claims and EHR data are a valuable source of information, they are limited by missing and erroneous data, coding imperfections, a lack of standardisation of clinical measures, variations between clinical testing centres and measurements that are taken with varying periodicity. Two additional intrinsic limitations of the database used in the present analysis may have led to underestimation or misclassification of the actual cases of ATTR. First, it includes only data from patients who access the healthcare system, excluding uninsured individuals (eg, low-income populations). Indeed, evidence from real-world studies of patients with ATTR demonstrates the impact of disparities in social determinants of health on disease care, including non-equitable access to genetic screening for people living in high-deprivation areas.14 15 Second, claims data are primarily used for reimbursement and not for clinical diagnosis, meaning diagnosis codes may reflect suspected conditions being considered during a diagnostic workup, rather than a confirmed diagnosis. However, real-world data sets with similar limitations remain one of the best available sources of incidence/prevalence data and have been used in previously published studies.16 29 Moreover, analysis of data relative to wild-type versus mutated TTR was not performed. While the collection of these data would have been of interest in the context of diagnosis, it was out of the scope of this analysis. As the study provides US data, results may not be generalisable to other geographies.
Strengths of this analysis include the large sample size (>300 million people), the data coverage and the varied data sources, coding systems and case definitions that were used to capture ATTR amyloidosis cases. Indeed, the comprehensive methodology we employed to identify ATTR amyloidosis diagnoses optimises the identification of cases, especially given the lack of standardised coding systems and of nationwide ATTR registry data.
Conclusion
Using a large sample of the US population and various combinations of diagnostic codes, this study provides up-to-date epidemiology estimates of ATTR amyloidosis, both overall and by phenotype. Given the increasing incidence and prevalence of the disease, as well as the heterogeneous approaches to diagnosis reported in the literature, there is a pressing need for a widely accepted and validated definition of ATTR amyloidosis cases, not only to get a better picture of the actual burden, but, more importantly, to facilitate appropriate patient management. Key steps could include improving the coding systems in use (eg, ICD-10, which currently lacks a code specific for ATTR-CM), the methodologies for coding allocation in clinical practice and the standardising of coding algorithms for research.
Supplementary material
Acknowledgements
Clara Ricci (Clarivate) provided medical writing support.
Footnotes
Funding: This work was supported by Alexion and AstraZeneca Rare Disease.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: Not applicable.
Data availability free text: All data are included in the article.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
References
- 1.Gertz MA, Benson MD, Dyck PJ, et al. Diagnosis, Prognosis, and Therapy of Transthyretin Amyloidosis. J Am Coll Cardiol. 2015;66:2451–66. doi: 10.1016/j.jacc.2015.09.075. [DOI] [PubMed] [Google Scholar]
- 2.Wininger AE, Phelps BM, Le JT, et al. Musculoskeletal pathology as an early warning sign of systemic amyloidosis: a systematic review of amyloid deposition and orthopedic surgery. BMC Musculoskelet Disord. 2021;22:51. doi: 10.1186/s12891-020-03912-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gentile L, Coelho T, Dispenzieri A, et al. A 15-year consolidated overview of data in over 6000 patients from the Transthyretin Amyloidosis Outcomes Survey (THAOS) Orphanet J Rare Dis. 2023;18:350. doi: 10.1186/s13023-023-02962-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Delgado DS, Pao C, Dabbous F, editors. Systematic Literature Review of the Epidemiology and Characteristics of Transthyretin Amyloidosis (ATTR). Poster. 4th International ATTR Amyloidosis meeting; Madrid. 2023. In. [Google Scholar]
- 5.Delgado D, Dabbous F, Shivappa N, et al. Epidemiology of transthyretin (ATTR) amyloidosis: a systematic literature review. Orphanet J Rare Dis. 2025;20:29. doi: 10.1186/s13023-025-03547-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Shams PAI. Cardiac amyloidosis. Treasure Island, FL: StatPearls, StatPearls Publishing; 2024. [Google Scholar]
- 7.Aimo A, Merlo M, Porcari A, et al. Redefining the epidemiology of cardiac amyloidosis. A systematic review and meta-analysis of screening studies. Eur J Heart Fail. 2022;24:2342–51. doi: 10.1002/ejhf.2532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.See ASY, Ho JS-Y, Chan MY, et al. Prevalence and Risk Factors of Cardiac Amyloidosis in Heart Failure: A Systematic Review and Meta-Analysis. Heart Lung Circ. 2022;31:1450–62. doi: 10.1016/j.hlc.2022.08.002. [DOI] [PubMed] [Google Scholar]
- 9.Brito D, Albrecht FC, de Arenaza DP, et al. World Heart Federation Consensus on Transthyretin Amyloidosis Cardiomyopathy (ATTR-CM) Glob Heart. 2023;18:59. doi: 10.5334/gh.1262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Adams D, Ando Y, Beirão JM, et al. Expert consensus recommendations to improve diagnosis of ATTR amyloidosis with polyneuropathy. J Neurol. 2021;268:2109–22. doi: 10.1007/s00415-019-09688-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Schmidt HH, Waddington-Cruz M, Botteman MF, et al. Estimating the global prevalence of transthyretin familial amyloid polyneuropathy. Muscle Nerve. 2018;57:829–37. doi: 10.1002/mus.26034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Antonopoulos AS, Panagiotopoulos I, Kouroutzoglou A, et al. Prevalence and clinical outcomes of transthyretin amyloidosis: a systematic review and meta-analysis. Eur J Heart Fail. 2022;24:1677–96. doi: 10.1002/ejhf.2589. [DOI] [PubMed] [Google Scholar]
- 13.Porcari A, Allegro V, Saro R, et al. Evolving trends in epidemiology and natural history of cardiac amyloidosis: 30-year experience from a tertiary referral center for cardiomyopathies. Front Cardiovasc Med. 2022;9:1026440. doi: 10.3389/fcvm.2022.1026440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lauppe RE, Liseth Hansen J, Gerdesköld C, et al. Nationwide prevalence and characteristics of transthyretin amyloid cardiomyopathy in Sweden. Open Heart. 2021;8:e001755. doi: 10.1136/openhrt-2021-001755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Damy T, Bourel G, Slama M, et al. Incidence and survival of transthyretin amyloid cardiomyopathy from a French nationwide study of in- and out-patient databases. Orphanet J Rare Dis. 2023;18:345. doi: 10.1186/s13023-023-02933-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gilstrap LG, Dominici F, Wang Y, et al. Epidemiology of Cardiac Amyloidosis-Associated Heart Failure Hospitalizations Among Fee-for-Service Medicare Beneficiaries in the United States. Circ Heart Fail. 2019;12:e005407. doi: 10.1161/CIRCHEARTFAILURE.118.005407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Prasad M, Kim M, Chandrashekar P, et al. Evolving epidemiology of transthyretin amyloid cardiomyopathy due to increased recognition in women. Int J Cardiol. 2023;374:116–9. doi: 10.1016/j.ijcard.2022.12.048. [DOI] [PubMed] [Google Scholar]
- 18.Sepehrvand N, Youngson E, Fine N, et al. The Incidence and Prevalence of Cardiac Amyloidosis in a Large Community-Based Cohort in Alberta, Canada. J Card Fail. 2022;28:237–46. doi: 10.1016/j.cardfail.2021.08.016. [DOI] [PubMed] [Google Scholar]
- 19.Westin O, Butt JH, Gustafsson F, et al. Two Decades of Cardiac Amyloidosis: A Danish Nationwide Study. JACC CardioOncol. 2021;3:522–33. doi: 10.1016/j.jaccao.2021.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Cortese A, Vegezzi E, Lozza A, et al. Diagnostic challenges in hereditary transthyretin amyloidosis with polyneuropathy: avoiding misdiagnosis of a treatable hereditary neuropathy. J Neurol Neurosurg Psychiatry. 2017;88:457–8. doi: 10.1136/jnnp-2016-315262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lousada I, Maurer M, Warner M, et al. AMYLOIDOSIS RESEARCH CONSORTIUM CARDIAC AMYLOIDOSIS SURVEY: RESULTS FROM PATIENTS WITH AL AND ATTR AMYLOIDOSIS AND THEIR CAREGIVERS. J Am Coll Cardiol. 2018;71:A890. doi: 10.1016/S0735-1097(18)31431-1. [DOI] [Google Scholar]
- 22.Huda A, Castaño A, Niyogi A, et al. A machine learning model for identifying patients at risk for wild-type transthyretin amyloid cardiomyopathy. Nat Commun. 2021;12:2725. doi: 10.1038/s41467-021-22876-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mitchell JD, Lenihan DJ, Reed C, et al. Implementing a Machine-Learning-Adapted Algorithm to Identify Possible Transthyretin Amyloid Cardiomyopathy at an Academic Medical Center. Clin Med Insights Cardiol. 2022;16:11795468221133608. doi: 10.1177/11795468221133608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Porcari A, Merlo M, Rapezzi C, et al. Transthyretin amyloid cardiomyopathy: An uncharted territory awaiting discovery. Eur J Intern Med. 2020;82:7–15. doi: 10.1016/j.ejim.2020.09.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Obi CA, Mostertz WC, Griffin JM, et al. ATTR Epidemiology, Genetics, and Prognostic Factors. Methodist Debakey Cardiovasc J. 2022;18:17–26. doi: 10.14797/mdcvj.1066. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Dispenzieri A, Coelho T, Conceição I, et al. Clinical and genetic profile of patients enrolled in the Transthyretin Amyloidosis Outcomes Survey (THAOS): 14-year update. Orphanet J Rare Dis. 2022;17:236. doi: 10.1186/s13023-022-02359-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.González-Moreno J, Dispenzieri A, Grogan M, et al. Clinical and Genotype Characteristics and Symptom Migration in Patients With Mixed Phenotype Transthyretin Amyloidosis from the Transthyretin Amyloidosis Outcomes Survey. Cardiol Ther. 2024;13:117–35. doi: 10.1007/s40119-023-00344-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Nativi-Nicolau JN, Karam C, Khella S, et al. Screening for ATTR amyloidosis in the clinic: overlapping disorders, misdiagnosis, and multiorgan awareness. Heart Fail Rev. 2022;27:785–93. doi: 10.1007/s10741-021-10080-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Brown D, Vera-Llonch M, Perez JTO, et al. USE OF COMMERCIAL CLAIMS DATA TO ESTIMATE TRANSTHYRETIN-AMYLOID CARDIOMYOPATHY PREVALENCE AND INCIDENCE IN THE US. J Am Coll Cardiol. 2021;77:884. doi: 10.1016/S0735-1097(21)02243-9. [DOI] [Google Scholar]
- 30.Akers TE, Lynch K. Retrospective Evaluation Of Suspected Vs. Confirmed Transthyretin Amyloid Cardiomyopathy (ATTR-CM) Cases. J Card Fail. 2022;28:S110. doi: 10.1016/j.cardfail.2022.03.281. [DOI] [Google Scholar]
- 31.Clarivate Real-world data repository. https://clarivate.com/products/real-world-data/#dynamics n.d. Available.
- 32.Rodrigues E, Umeh E, Navaratnarajah N, et al. Incidence and prevalence of myasthenia gravis in the United States: A claims-based analysis. Muscle Nerve. 2024;69:166–71. doi: 10.1002/mus.28006. [DOI] [PubMed] [Google Scholar]
- 33.Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–83. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 34.Gillmore JD, Maurer MS, Falk RH, et al. Nonbiopsy Diagnosis of Cardiac Transthyretin Amyloidosis. Circulation. 2016;133:2404–12. doi: 10.1161/CIRCULATIONAHA.116.021612. [DOI] [PubMed] [Google Scholar]
- 35.Kittleson MM, Maurer MS, Ambardekar AV, et al. Cardiac Amyloidosis: Evolving Diagnosis and Management: A Scientific Statement From the American Heart Association. Circulation. 2020;142:e7–22. doi: 10.1161/CIR.0000000000000792. [DOI] [PubMed] [Google Scholar]
- 36.Quarta CC, Fontana M, Damy T, et al. Changing paradigm in the treatment of amyloidosis: From disease-modifying drugs to anti-fibril therapy. Front Cardiovasc Med. 2022;9:1073503. doi: 10.3389/fcvm.2022.1073503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Girard AA, Asif T, Sperry BW. Advances in the screening, diagnosis, and treatment of transthyretin amyloid cardiomyopathy: New insights and future directions. Trends Cardiovasc Med. 2025;35:354–62. doi: 10.1016/j.tcm.2025.03.002. [DOI] [PubMed] [Google Scholar]
- 38.Stanford Medicine MarketScan data. https://med.stanford.edu/phs/data/marketscan-data.html n.d. Available.
- 39.Khan U. RetirementLiving.com; 2023. Medicare statistics.https://www.retirementliving.com/best-medicare-supplement-plans-companies/medicare-statistics Available. [Google Scholar]
- 40.Lauppe R, Liseth Hansen J, Fornwall A, et al. Prevalence, characteristics, and mortality of patients with transthyretin amyloid cardiomyopathy in the Nordic countries. ESC Heart Fail. 2022;9:2528–37. doi: 10.1002/ehf2.13961. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.European Medicines Agency Vyndaqel-epar-product-information. 2016 https://www.ema.europa.eu/en/documents/product-information/vyndaqel-epar-product-information_en.pdf Available.
- 42.Adams D, Gonzalez-Duarte A, O’Riordan WD, et al. Patisiran, an RNAi Therapeutic, for Hereditary Transthyretin Amyloidosis. N Engl J Med. 2018;379:11–21. doi: 10.1056/NEJMoa1716153. [DOI] [PubMed] [Google Scholar]
- 43.Benson MD, Waddington-Cruz M, Berk JL, et al. Inotersen Treatment for Patients with Hereditary Transthyretin Amyloidosis. N Engl J Med. 2018;379:22–31. doi: 10.1056/NEJMoa1716793. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All data relevant to the study are included in the article or uploaded as supplementary information.