Prevalence and Outcomes of p.Val142Ile Transthyretin Amyloidosis Cardiomyopathy: A Systematic Review

Abstract

Background:

The p.Val142Ile variant, predominantly found among people of African descent, is the most common cause of variant transthyretin amyloidosis (vATTR) and carriers predominantly develop a cardiomyopathy (vATTR-CM) phenotype. Yet, there are conflicting data on the prevalence and outcomes of p.Val142Ile variant carriers.

Methods:

We performed a systematic review of the prevalence and outcomes of p.Val142Ile vATTR-CM among subjects of African descent. We found 62 relevant articles after searching the MEDLINE databases from 1980 – 2020 that reported data for approximately 150,000 subjects.

Results:

The reported worldwide prevalence of the p.Val142Ile variant is 0.3% – 1.6% in the general population. Among people of African descent, the reported prevalence from all studies ranges from 1.1% to 9.8%, but for studies with >1000 subjects it is 3% −3.5%. The prevalence of the p.Val142Ile variant in a region is dependent on the reported percentage of subjects who are of African descent in that region. p.Val142Ile vATTR-CM typically presents in the 7–8^th decade of life and the majority of cases reported were male, with 25% – 38% diagnosed with atrial fibrillation. It was associated with a longitudinally worse quality of life and a lower adjusted survival compared to other types of ATTR-CM.

Conclusion:

The p.Val142Ile variant is the most common variant of the transthyretin gene with most carriers being of African descent. The true penetrance is unknown but the p.Val142Ile variant is associated with increased rates of incident heart failure and portends a lower overall survival. Increased awareness could lead to earlier diagnosis and improved heart failure outcomes among those of African descent, which is of increasing importance given the advent of novel therapeutics for this disease.

Introduction

Variants in the transthyretin (TTR) gene which destabilize the TTR tetramer are associated with the development of hereditary or variant transthyretin amyloidosis (vATTR). Some variants are predominantly associated with a polyneuropathy phenotype, some with a cardiomyopathy phenotype (vATTR-CM), and others show a mixed phenotype. The most commonly identified variant is caused by a single amino acid substitution of valine to isoleucine at position 142 (p.Val142Ile or p.V142I or V142I), historically reported as p.Val122Ile as a 20-amino acid signal peptide at the start of the TTR gene was previously not incorporated into the codon count^1,2. The geographic origin of this coding sequence is consistent with a founder variant in West Africa². This was determined based on analysis of de-identified DNA samples from 14 modern African countries. Higher prevalence of the variant among the non-Bantu-speaking Niger-Kordofanian-derived populations suggests that the variant arose prior to the so-called Bantu expansion from 1000 B.C. to 500 A.D. This is the ancestry both commonly found in countries from where 50–70% of Africans forced into slave labor were brought to North America during the Atlantic slave trade in the 16^th to 19^th century and that is mostly highly represented in African Americans, which further supports the worldwide distribution of this allele. As a result of this the prevalence of p.Val142Ile carrier status has been estimated to be approximately 3–4% among the United States African American population (Figure 1) ^1,3.

Figure 1.

Geographical Distribution of p.Val142Ile Variant Prevalence. World map that shows prevalence of p.Val142Ile among subjects of African descent is highest in West Africa, with next highest in USA, UK, and the Caribbean. Dark blue arrows represent 16^th century Atlantic slave trade routes providing an explanation for higher rates of p.Val142Ile in these countries. || Expanded map of USA and UK showing prevalence of p.Val142Ile variant from reported studies among those of African descent and among vATTR-CM. || Inset blue box: Prevalence of p.Val142Ile variant among those of African descent is 3% – 3.5% when restricted to studies with at least 1000 subjects. When all studies are considered, the prevalence ranges from 1.1% – 9.8%. | Prevalence of p.Val142Ile variant among vATTR-CM is 66% – 79% in the USA and UK. The prevalence ranges from 0% – 22% among studies outside of the USA and UK. | Demographics of p.Val142Ile vATTR-CM have been summarized, full details are available in Table 4. (Created with BioRender.com)

p.Val142Ile vATTR-CM remains an underdiagnosed cause of heart failure (HF) among those of African descent^4,5. Historically, the diagnosis required an invasive endomyocardial biopsy, and the treatment of ATTR-CM was limited to supportive care, providing little incentive to search for and diagnose these patients. The resurgence of bone scintigraphy and the increased utilization of cardiac magnetic resonance imaging has uncovered a significant prevalence of ATTR-CM^6–8. With the advent of tafamidis, a TTR tetramer stabilizer shown to improve quality of life and decrease mortality in ATTR-CM, as well as other TTR silencing drugs that decrease the progression of polyneuropathy in vATTR, early identification of patients with ATTR has become paramount^9–12. The purpose of this systematic review is to summarize the evidence of the prevalence of p.Val142Ile carriers and the vATTR-CM phenotype among those of African descent as well as the demographics and clinical outcomes of p.Val142Ile vATTR-CM.

Methods

The full methods are available in the Supplemental Material. Briefly, we conducted this systematic review of published literature with an inclusion criteria of original studies of adults, both interventional and non-interventional, with the p.Val142Ile variant or p.Val142Ile vATTR-CM that reported p.Val142Ile prevalence, its association with HF, demographics of vATTR-CM, and clinical outcomes of p.Val142Ile vATTR-CM. We identified 62 studies that were included in this systematic review (Figure 2, Supplemental Table I) ¹³. The authors declare that all supporting data are available within the article and its online supplemental material. This study was exempt from institutional review board approval.

Figure 2.

PRISMA flow diagram for search strategy and results.

Results

Prevalence of p.Val142Ile Variant in General Population

The best estimate of the p.Val142Ile variant prevalence in the general population is 0.3%, which comes from the large GnomAD Database with data from >138,000 subjects compiled from multiple large-scale sequencing projects across the world (Supplemental Table II)^14–16. Based on these data, in the general population the p.Val142Ile variant is likely the most common cause of vATTR in the world, representing approximately 88% of all variants that affect function¹⁴. The prevalence of the p.Val142Ile variant in a region is dependent on the reported percentage of subjects who are of African descent in that region. For example, in the GnomAD Database, 8.6% were reported as African descent and the p.Val142Ile variant prevalence was 0.30%¹⁴. In a smaller study in Indiana among 1973 consecutive newborns, half were of African descent, and the p.Val142Ile prevalence was 1.62%¹⁶.

Prevalence of p.Val142Ile Variant among those of African descent

Among people of African descent in a population, the prevalence of the p.Val142Ile variant ranges from 1.1% to 9.8% and the allele frequency 0.006 to 0.058 (Table 1)^1,2,14–23. This was determined by compiling data from several prospective cohort studies and genetic biobanks. When restricting to only studies with at least 1000 subjects, the prevalence ranges from 3.0% to 3.5%. There was not a large variation in the prevalence despite being from various regions in the USA, failing to reveal any obvious clustering (Figure 1).

Table 1.

Prevalence of p.Val142Ile Variant: African descent only

Data Source, Location	Time Period	Allele Frequency	Prevalence of p.Val142Ile variant
GnomAD Database, International14	2017–2018	0.016	3.2% (385/11,935)
BioMe BioBank, USA20	2007–2015	n/r	3.2% (211/6609)
Penn Medicine Bio Bank, USA20	2008–2017	n/r	3.3% (190/5737)
Yale-Penn Cohort, USA23	n/r	0.018	3.5% (152/4361)
ARIC Cohort, USA15	1987–2007	n/r	3.2% (124/3856)
Multiple databases, USA22	1949–1991	0.020	3.9% (65/1688)
Newborn Cord Blood Marion County, Indiana16	n/r	0.015	3.0% (30/1000)
CHS Cohort, USA18	1989–1993	0.011	2.1% (17/805)
Kaiser Health Appraisal Clinic, SD, USA19	n/r	0.015	3.0% (16/526)
New York University and NY Veterans Clinic, USA21	1995–2004	n/r	4.9% (23/471)
Random healthy South Africa Volunteers, Africa17	n/r	0.006	1.1% (1/89)
Indianapolis Veteran’s Administration Hospital, USA17	n/r	0.058	9.8% (5/51)
Newborns, New York, USA1	n/r	0.019	n/r
Jackson Heart Study, USA1	2000–2004	0.017	n/r
Dallas Heart Study, USA1	1999	0.018	n/r
San Diego Wellness, USA1	n/r	0.021	n/r
Southwest United States Thousand Genome, USA1	n/r	0.076	n/r
African DNA Samples + 1000Genomes Project, Africa2	n/r	0.020	n/r
		0.006 – 0.058	1.1% – 9.8%

Sorted by decreasing sample size.

ARIC: Atherosclerosis Risk in Communities; CHS: Cardiovascular Health Study; n/r: not reported; USA: United States of America

Only two small studies were found that evaluated subjects in Africa^2,17. In the largest one to date, Jacobson et al compiled several smaller studies in Africa and found that the allele prevalence varies from 0.0253 in high prevalence areas to 0.010 in the lowest prevalence areas². Strong inferences and conclusions from such studies are limited by the lack or a large systematic evaluation of cohorts in Africa^2,17. Analysis of the GnomAD database showed that 92.3% of all p.Val142Ile variant carriers were of African descent. Although rare, Gentile et al described a series of 12 Caucasian family members with the p.Val142Ile variant from two hospitals in Italy²⁴.

Prevalence of p.Val142Ile Variant among those with confirmed transthyretin amyloidosis phenotype

There is a wide geographic variation in the distribution of the underlying genetic variant found in patients with vATTR phenotype. The prevalence of p.Val142Ile appears to be higher in the USA and UK ranging from 11% – 84% among those with a vATTR phenotype. In all these studies greater than 75% of p.Val142Ile carriers were of African descent (Supplemental Table III)^25–28. In contrast, the best estimate of the prevalence in the rest of the world is 0.78% and this comes from the THAOS rest of world (ROW) registry. Smaller individual studies, except for one French cohort, show a similarly lower prevalence^25,29–40.

Autopsy studies show comparable results with Jacobson et al demonstrating that in Los Angeles County, USA among 45 viable myocardial blocks found to have transthyretin amyloid deposits (58% African descent), 13% were positive for the p.Val142Ile variant³. In Germany, Eriksson et al evaluated 30 ATTR pathology specimens (anti-ATTR immunostaining positive) from its amyloid pathology registry, a third were myocardial⁴¹. Among all 30, only one colon specimen was positive, showing a polymorphism for both p.Val142Ile and p.Gly26Ser.

Among those diagnosed with vATTR-CM phenotype in the USA and UK 66% – 79% carried the p.Val142Ile variant compared to 0% – 22% reported from the rest of the world (Table 2)^42–51. Among studies of clinically diagnosed vATTR polyneuropathy the p.Val142Ile variant frequency was expectedly low ranging from 0% – 1.7% (Supplemental Table IV)^52–61.

Table 2.

p.Val142Ile Variant Frequency among vATTR-CM

Data Source, Location	Time Period	Proportion of ATTR-CM that is of African Descent	Proportion of p.Val142Ile among vATTR-CM
Kaiser San Francisco, USA44	2001–2016	48% (32/67)	79% (23/29)	USA + UK
AG10 Trial; USA, Europe, South America45	2018	20% (10/49)	79% (11/14)
Johns Hopkins Hospital, USA42	2003–2016	41% (23/56)	75% (21/28)
Cleveland Clinic, USA46	2002–2014	34% (57/169)	73% (37/51)
National Amyloidosis Centre, UK43	2000–2019	n/r	66% (314/474)
		20% – 48%	66% – 79%
Peking Union Medical College, Beijing, China 48	n/r	0% (0/23)	0% (0/23)	Rest of World
Henri-Mondor Hospital, Creteil, France51	2009–2013	n/r	22% (8/36)
Aarhus University Hospital, Denmark47	2015–2016	0% (0/15)	0% (0/5)
Multiple cardiology clinics, Japan50	2001–2012	0% (0/32)	0% (0/10)
Heidelberg Amyloidosis Center, Germany49	n/r	n/r	0% (0/8)
		0%	0% – 22%

n/r: not reported; USA: United States of America; UK: United Kingdom; vATTR-CM: variant transthyretin amyloid cardiomyopathy

Estimated Prevalence of p.Val142Ile Homozygotes

The best estimate of homozygosity in the general population comes from the large GnomAD database, which showed that the majority of p.Val142Ile carriers are heterozygotes with only 0.72% of subjects estimated to be homozygous¹⁴. As expected, in clinical cohorts with an enriched population the percentage of homozygotes identified were found to be higher (Supplemental Table V)^{3,5,16–19,22–24,26,28,42,62–64}. For example, among endomyocardial biopsy confirmed p.Val142Ile vATTR-CM 6% – 10% were homozygotes^5,42. In comparison, 4% to 14% of subjects referred to an amyloidosis clinic who underwent TTR variant testing were homozygotes.

These should be interpreted with caution due to the low absolute number of subjects that will be identified and may not be applicable clinically in smaller practices. For example, among cohorts of African Americans in the general population in whom the TTR variant was screened, the prevalence of homozygotes ranges from 0% (0 of 17) – 20% (1 of 5). But if restricted to cohorts with at least 1000 subjects, it ranged from 1% to 3%.

Association of the p.Val142Ile Variant with Incident Heart Failure

The studies describing the increased rates of HF among p.Val142Ile variant carriers are summarized in Table 3^{5,15,18,20,21,63,65}. Prospective cohorts estimated that the risk of incident HF ranges from 1.47 to 1.75. Quarta et al analyzed the Atherosclerosis Risk in Communities (ARIC) cohort, a prospective observational study of subjects <65 years old with a median follow up time of 21.5 years where most subjects of African descent are from the Jackson, Mississippi area ¹⁵. Incident HF, defined by a first HF hospitalization, was observed in 29% of p.Val142Ile carriers (1.59 cases/100 person-years) vs. 20% of non-carriers (1.12 cases/100 person-years; p=0.04) with a stratified hazard ratio (HR) of 1.47 [95% Confidence Interval (CI): 1.03 – 2.10]. One important caveat to this study is that only 37% of all subjects with any genotyping information were male, likely an underrepresentation as the majority of p.Val142Ile vATTR-CM reported cases are males, and only a minority of these included had an echocardiogram limiting full phenotypic assessment. Buxbaum et al analyzed the prospective Cardiovascular Health Study (CHS) cohort from North Carolina and found that TTR p.Val142Ile carriers represented 2.8% of African American individuals with HF, but almost 7 percent in the similar age and gender analysis¹⁸. In the total population, the relative risk of developing HF in p.Val142Ile carriers is 2.62 [95% CI: 1.28 – 5.37]; however, in the post-hoc matched subset it was no longer statistically significant (1.75 [95% CI: 0.79 – 3.86]).

Table 3.

Effect of p.Val142Ile Variant on Development of Heart Failure

Data Source, Location	Sample Size	Effect size	Details
ARIC Cohort, USA15	3856	Stratified HR: 1.47 [95% CI: 1.03 to 2.10]	• Incident HF hospitalization was observed 29% of carriers (1.59 cases/100 person-years) vs. 20% of non-carriers (1.12 cases/100 person-years), p=0.04
CHS Cohort, USA18	5201	Relative Risk: Total population: 2.62 [95% CI: 1.28 – 5.37] Post-hoc matched subset: 1.75 [95% CI: 0.79 – 3.86]	• p.Val142Ile carriers represented 2.8% of African American individuals with HF, but almost 7 percent in the similar age and gender analysis.
Penn Medicine BioBank, USA20	3724	Adjusted odds ratio: 1.6 [95% CI: 1.1–2.3]	• p.Val142Ile variant carriers had higher rates of HF than non-carriers (44% vs. 30%; p=0.006)
Mt. Sinai BioMe BioBank, USA20	5970	Adjusted odds ratio: 1.8 [95% CI: 1.2–2.7]	• p.Val142Ile variant was more commonly found in cases of HF compared to controls (2.6% vs. 1.8%; p=0.008)
New York University & Veterans Administration Clinic, USA21	383	Relative risk: 2.0 [95% CI: 1.04 to 3.71]	• Among all subjects, 39% p.Val142Ile carriers had evidence of HF vs. 17% of matched controls; p=0.07) • Among the subjects referred for echocardiography, there was a statistically significantly higher incidence of HF in p.Val142Ile carriers vs. controls.

ARIC: Atherosclerosis Risk in Communities; CHS: Cardiovascular Health Study; CI: Confidence interval; HF: Heart failure; HR: hazard ratio; n/r: not reported; USA: United States of America

Penetrance of p.Val142Ile variant

It is difficult to determine the true penetrance of the p.Val142Ile variant that leads to the development of the vATTR-CM phenotype from current published studies. Due to historical under-recognition of this disease most cohorts lack a detailed phenotypic assessment, which can lead to under-diagnosis of ATTR-CM, particularly in the absence of the severe phenotype on echocardiography. Furthermore, there is an age-dependent expression of this phenotype. This can lead to a wide range of estimates of penetrance as demonstrated by the analysis of one cross-sectional study by Damrauer et al of the Penn Medicine BioBank²⁰. Among 116 p.Val142Ile carriers of African ancestry greater than 50 years old the calculated penetrance could be as low as 8% if only subjects who were formally diagnosed with vATTR-CM were included as positive cases. In contrast 46% of all carriers had heart failure, but all did not undergo assessment for vATTR-CM. Similarly, in the prospective ARIC cohort, nearly 50% of carriers developed incident HF by age 75 but a formal detailed phenotypic assessment of etiology of HF was also not feasible ¹⁵.

Prevalence of p.Val142Ile vATTR-CM among adults with Heart Failure

Among those with the p.Val142Ile variant in retrospective cohorts, the adjusted odds ratio for HF ranged from 1.6 to 1.8. Damrauer et al found that among African-Americans >50 years old who were p.Val142Ile carriers in the Penn Medicine BioBank the odds ratio for having an International Classification of Disease (ICD) code for HF was 1.6 [95% CI: 1.06 – 2.3] after adjusting for age, sex, confounding by ancestry, hypertension, and myocardial infarction/coronary revascularization²⁰. This was replicated in their case-control analysis of the Mt. Sinai BioMe BioBank (adjusted odds ratio: 1.8 [95% CI: 1.2 – 2.7]) where the p.Val142Ile variant was more common among cases compared to controls of African/Hispanic descent >65 years old without an ICD code for HF (2.6% vs. 1.8%; p=0.008). As this was based on administrative ICD codes, it lacked the granularity to identify if vATTR-CM was the true etiology of HF. Nonetheless, supportive evidence could be extrapolated from the multicenter observational DISCOVERY study aimed to characterize the frequency of TTR variants among 1001 patients with a clinical or echocardiographic phenotype of cardiac amyloidosis⁶². They found that p.Val142Ile carriers made up 92% of all 74 carriers of a pathological TTR variant, the majority of whom were African Americans with New York Heart Association (NYHA) functional class II or above.

In a cohort from routine clinical care, Jacobson et al evaluated consecutive elderly African Americans predominantly referred for echocardiography in the New York City area and noted that p.Val142Ile carriers are twice as likely to have HF (relative risk 2.0 [95% CI: 1.04 – 3.71]²¹. Similarly, a UK study evaluated 1392 patients that presented to their HF clinic in London of whom 211 were of Afro-Caribbean descent⁵. Among them 8.5% (18/211) were found to have the p.Val142Ile variant, and 100% were subsequently confirmed to have p.Val142Ile vATTR-CM.

The p.Val142Ile variant has even been identified within patients of African descent for whom ATTR-CM was believed to have been excluded. Arvanitis et al found that among 99 elderly African Americans with left ventricular (LV) wall thickness >12 mm in whom ATTR-CM was already thought to have been excluded as a cause of HF, five had the p.Val142Ile variant and two were subsequently diagnosed with p.Val142Ile vATTR-CM⁶³. In a sub-study of the BEST trial (Beta-Blocker Evaluation in Survival Trial) where a clinical diagnosis of ATTR-CM was also an exclusion criterion, 10% of subjects of African descent (13/207) >60 years of age with an ejection fraction <35% and NYHA Class III-IV were found to have the p.Val142Ile variant; but no further genotype-phenotype correlations were performed⁶⁵.

Demographics of p.Val142Ile vATTR-CM

The demographics of p.Val142Ile vATTR-CM subjects are summarized in Table 4^{5,24,25,27,28,43,64,66–70}. When excluding studies that enrolled all ethnicities, 75% to 100% of all p.Val142Ile vATTR-CM cases were of African descent. Swiecicki et al found median age of symptom onset was almost 6 years earlier at 64 years (10–90^th percentile: 50–75 years) than the age of diagnosis²⁷. Looking exclusively at cases in patients of African descent, Connors et al found a comparable mean age at diagnosis of 70.2 ± 7.3 years²⁸. Reddi et al found that the mean age of diagnosis in homozygotes was 64 ± 6 years, almost 10 years younger than heterozygotes⁶⁴. There is a predominance of males and unsurprisingly no studies were identified that evaluated the characteristics of p.Val142Ile vATTR-CM exclusively among females. Finally, the atrial fibrillation was detected in 25% – 38% of p.Val142Ile vATTR-CM subjects^{25,28,66–68}. This is lower than the reported rate of atrial fibrillation among wild type TTR amyloidosis cardiomyopathy (wtATTR-CM) of 40–60%⁹. There were no studies that specifically reported the burden of arrhythmias on ambulatory monitoring specifically to p.Val142Ile vATTR-CM.

Table 4.

Demographics of p.Val142Ile vATTR-CM

Study, Date	Sample Size	African descent	Age at diagnosis, years	Age at enrollment, years	Males	Atrial Fibrillation
Connors 2009 28	36	100% (36/36)§	70 Ɨ	n/r	78% (25/36)	28% (8/29*)
Ruberg 2012 69	11	100% (11/11)	n/r	71 Ɨ	82% (9/11)	27% (3/11)
Givens 2013 66	28	96% (27/28)	71 #	n/r	82% (23/28)	38% (10/28)
Reddi 2014 64	194	92% (44/48*)	68 #	n/r	83% (157/189*)	n/r
Swiecicki 2015 27	28	75% (21/28)	69 # 64 (s)#	n/r	96% (27/28)	n/r
Maurer 2015 67	4	100% (4/4)	n/r	73 #	75% (3/4)	25% (1/4)
Dungu 2016 5	18	100% (18/18)	n/r	71 #	n/r	n/r
Maurer 2016 25	91	n/r	n/r	69 Ɨ	76% (69/91)	30% (27/91)
Singh 2017 70	35	n/r	n/r	72 Ɨ	n/r	n/r
Gentile 2020 24	5	0% (0/5) §	67 Ɨ	n/r	n/r	n/r
Monfort 2020 68	18	100% (18/18) §	n/r	72 Ɨ	89% (16/18)	33% (6/18)
Chacko 2020 43	314	87% (272/314)	n/r	75 Ɨ	73% (229/314)	n/r
		75% – 100%	67 – 71	69 – 75	73% – 96%	25% – 38%

(s) - age of symptom onset; vATTR-CM - variant transthyretin amyloid cardiomyopathy

^Ɨaverage

^#median

^§Studies with single race/ancestry as inclusion criteria not included in range

^*incomplete data reported

Quality of Life for p.Val142Ile vATTR-CM

Given the progressive nature of ATTR-CM and the historical lack of any proven therapies, studies reporting on patients’ quality of life (QoL) demonstrated deterioration over time. Lane et al observed that the direction of change in Kansas City Cardiomyopathy Questionnaire (KCCQ) QoL scores was overwhelmingly negative in all ATTR-CM cohorts and in all QOL domains. p.Val142Ile vATTR-CM specifically showed clinically significant deterioration in seven out of ten domains, compared to five for non-p.Val142Ile vATTR-CM and three domains for wtATTR-CM⁷¹. Among the USA cohort of the THAOS (Transthyretin Amyloid Outcome Survey) worldwide registry, although the duration of disease approached that of other ATTR patients, p.Val142Ile carriers had worse NYHA functional class and worse QOL, as evidenced by lower Euro-Qual-5D index and lower Karnofsky performance²⁵.

Effect of p.Val142Ile Variant on Survival and Clinical Outcomes of p.Val142Ile vATTR-CM

The negative effect of p.Val142Ile variant on survival and the clinical outcomes of p.Val142Ile vATTR-CM are summarized in Table 5. Analysis of the CHS cohort found that the relative risk of death for cases with the p.Val142Ile variant is 2.59 [95% CI: 1.01 – 6.63] versus age matched controls after 15 years of follow up time ¹⁸. The distribution of the recorded causes of death did not differ between groups. In contrast p.Val142Ile carriers in the ARIC cohort did not have an increased risk of death after 21.5 years (age and sex-stratified HR: 0.99 [95% CI: 0.73 – 1.36]¹⁵. The differences between these studies are likely the result of variation in age and sex: CHS only enrolled subjects >65 years old with 43% of them male, whereas ARIC consisted of subjects <65 years old with 38% of them male. Although p.Val142Ile vATTR-CM subjects are most likely to be male, Lane et al did not find biological sex to be an independent predictor of mortality among their vATTR-CM cohort⁷¹.

Table 5:

Effect of p.Val142Ile Variant on Survival

Data Source, Location	Population Studied	Effect Size	Follow up time	Details
CHS Cohort, USA 18	General Population	Relative risk of death:  • Overall: 1.95 [95% CI: 0.92 – 4.14]  • Matched Cohort: 2.59 [95% CI: 1.01 – 6.63]	15 years	• The distribution of the recorded causes of death did not differ between groups.
ARIC Cohort, USA 15	General population	• Age and sex-stratified HR: 0.99 [95% CI: 0.73 – 1.36]	21.5 years	• p.V142I carriers: 1.76 deaths/100 person-years  • Non-carriers: 1.96 deaths/100 person-years
St. George Hospital, UK 5	HF Patients of Afro-Caribbean descent	• ATTR p.V142I had the poorest survival (2.33 years) versus dilated and hypertensive CM, but no significant difference vs. ischemic or valvular CM	8 years	• Afro-Caribbean patients with p.V142I vATTR-CM presented at a relatively old median age of 76 years (IQR: 74 – 78)
Boston Medical Center, USA 28	Cardiac Amyloidosis - African descent	Survival estimates for p.V142I vATTR-CM vs. AL-CM:  • At 1 year: 79% vs 26%  • At 3 years: 48% vs 10%	n/r
Kaiser San Francisco, USA 44	Cardiac amyloidosis – ICD codes	• In ATTR patients, African American race was associated with higher mortality	2.8 ± 2.9 years
Mayo Clinic, USA 27	vATTR	• Univariate RR: 1.76 (p<0.05)  • Multivariate RR: 2.83 (p<0.05)	46 months (7.2 – 164.4)	• The shortest survival was for p.V142I carriers as all had cardiac involvement at diagnosis
THAOS, USA cohort 25	ATTR	• Survival from enrollment in THAOS did not differ between p.V142I vATTR and wtATTR.	~24 months	• 48/66 (73%) died at 24 months
NAC, UK 26	ATTR	• 55% diagnosed with p.V142I ATTR died at end of follow up period	n/r	• Median age at death: 78 years (range 63–91)  • Median survival from diagnosis: 33 months (range 1–240)
TRACS, USA 69	ATTR-CM	• p.V142I were more likely to have died at end of follow up (73% vs 22%, p = 0.03)	n/r	Median survival  • wtATTR-CM: 43 months versus  • p.V142I vATTR-CM: 26 months (p = .04)
NAC, UK 71	ATTR-CM	• Among African ancestry, median survival for p.Val142Ile vATTR-CM was 31 months vs. 40 months for wtATTR-CM (P=0.85)  • p.Val142Ile vATTR-CM multivariate HR for death 2.07 [95% CI: 1.415–3.031, p<0.01]	n/r	Median survival from diagnosis: • p.Val142Ile vATTR-CM: 31 months vs. • wtATTR-CM: 57 months vs. • non–p.Val142Ile vATTR-CM: 69 months (p<0.01)
Columbia University, USA 66	ATTR-CM	• Among non-transplanted patients OR: 0.96 [95% CI: 0.31–3.01; p = 0.94]  • The median time to death alone was not significantly different.	n/r	Median time from diagnosis to death/OHT: • p.Val142Ile vATTR-CM: 36 months vs. • wtATTR-CM: 67 months (p = 0.09) Survival among those who did not undergo OHT: • p.Val142Ile 32% (7/22) vs. • wtATTR-CM 25% (8/36) (p = 0.4). Mean age at death among p.Val142Ile: 73 ± 6.1 years
Brigham and Women’s Hospital, Boston, USA 70	ATTR-CM	• Hazard ratio: 2.1 [95% CI: 1.2 – 3.6]		Median time from diagnosis to death: • p.Val142Ile vATTR-CM: 47 months vs. • wtATTR-CM: 59 months (p = 0.09)

ARIC: Atherosclerosis Risk in Communities; CHS: Cardiovascular Health Study; CI: Confidence interval; HF: Heart failure; HR: hazard ratio; n/r: not reported; NAC: National Amyloidosis Center; THAOS: Transthyretin Amyloid Outcome Survey; TRACS: Transthyretin Amyloidosis Cardiac Study; USA: United States of America; UK: United Kingdom; vATTR-CM: variant transthyretin cardiac amyloidosis

From the National Amyloidosis Center (NAC) in the UK, the median age at death of p.Val142Ile carriers was 78 years (range 63–91)²⁶. There and at the Mayo Clinic it was found that p.Val142Ile carriers had the shortest median survival from diagnosis among all vATTR subgroups^26,27. Conversely Maurer et al reported that survival two years from enrollment did not differ for p.Val142Ile vATTR cases in the USA cohort of the THAOS registry²⁵.

Patients of Afro-Caribbean descent with p.Val142Ile vATTR-CM at a HF clinic in London had significantly lower survival (2.33 years) compared to non-ischemic dilated cardiomyopathy (7.41 years, p<0.001) and hypertensive heart disease (7.32 years, p=0.002), but there was no significant difference compared to ischemic heart disease or valvular heart disease⁵. The conclusion was the difference may be driven by the median age of presentation, which was significantly higher for p.Val142Ile vATTR-CM (76 years; IQR 74–78) compared to dilated CM and hypertensive CM, but akin to ischemic CM and valvular CM. This study as well as other smaller single center studies showed a comparatively shorter median survival for p.Val142Ile vATTR-CM (Table 5)^66,69–71.

Czobor et al identified all patients with cardiac amyloidosis using ICD codes at Kaiser Permanente hospital systems in San Francisco, USA and found that among all ATTR-CM patients, African American race was associated with higher mortality⁴⁴.

Orthotopic heart transplantation for p.Val142Ile vATTR-CM

It is uncommon for patients with cardiac amyloidosis to undergo orthotopic heart transplantation (OHT), only comprising 0.14% of those performed in the USA, although p.Val142Ile vATTR-CM makes up a large proportion of these patients^72,73. In the USA cohort of the THAOS registry, OHT was more frequently performed in p.Val142Ile carriers compared to wtATTR among the 11 who underwent OHT out of 330 symptomatic ATTR-CM patients²⁵. Givens et al found that 28.6% (8/28) of p.Val142Ile vATTR-CM subjects received an OHT compared to 2.7% (1/37) wtATTR-CM subjects⁶⁶. This was likely driven by eligibility for transplant and lower average age of diagnosis (60.7 ± 4.7 years) for the eight who received an OHT versus those who did not (73.5 ± 5.5; p=0.0001).

Discussion

In the USA, HF is more prevalent among African Americans than other races⁷⁴. The Multi-Ethnic Study of Atherosclerosis (MESA) study revealed that African Americans also had the highest risk of incident HF⁷⁵. Although it has been found that the greater proportion of HF risk was attributable to more traditional modifiable risk factors compared to those identified as white (68% vs. 49%), it is also important to identify and be familiar with non-modifiable risk factors for development of HF that are unique to those of African descent. The p.Val142Ile variant is one such risk factor that can appreciably contribute to the burden of HF, particularly in those with atypical presentations⁷⁶. In HF clinics or among those at risk for HF, p.Val142Ile vATTR-CM was the etiology of 4–8% of the cases of HF^5,63,65. It is thus important to raise awareness of this among clinicians to help better care for this population that has higher rates of hospitalization, higher rates of death after HF hospitalization, and higher age- adjusted death rates^74,77,78.

The overall prevalence of the p.Val142Ile variant is low, but it appears to increase proportionally to the percentage of the population in the area studied who is of African descent. Thus, there appears to be a consistent linear relationship across studies with the p.Val142Ile prevalence increasing by 0.034% for every 1% increase in reported percentage of people of African descent in a region. This is expected given that the founder mutation likely originated in West Africa. The majority of eligible studies were in the USA or Western Europe, where most people of African descent were brought from Western Africa as a result of the African slave trade (Figure 1) ⁷⁹. This is the most plausible explanation for the geographical differences in prevalence, although increased healthcare utilization in more developed countries with increased use of diagnostic tools for amyloidosis may also play a role.

Based on the above prevalence and a Census Bureau estimate of the USA population, there are approximately 1.6 million African Americans who could be p.Val142Ile carriers, 160,000 of whom are >65 years old⁸⁰. Additionally, an increasing number of carriers will be diagnosed as more first-degree relatives undergo cascade genetic testing for the TTR variant. This should be an impetus to determine the true penetrance to develop vATTR-CM as well as study the utility of screening for the p.Val142Ile variant ⁸⁰. Although intuitively it would be assumed that the etiology of incident heart failure among p.Val142Ile carriers is vATTR-CM, this has not been confidently confirmed and thus the true penetrance has not yet been determined. Even in studies where the etiology was thought to have been determined, it is very plausible that the assumed etiology of HF could be incorrect as demonstrated by a sub-study of the BEST trial⁶⁵. Among African Americans with NYHA class III and IV, where the diagnosis of cardiac amyloidosis was an exclusion criterion, 10% of subjects >60 years old carried the p.Val142Ile variant. Factors that influence penetrance will also be valuable information, including epigenetic and gene-environment interactions that have been studied in other TTR variants^81,82.

Screening will require educational and community grassroots efforts. There have been historically lower rates of genetic testing among African Americans with concerns about the potential misuse of genetic testing likely playing a role and hence ^83–85. African Americans have been subject to systemic injustice leading to socio-economic barriers that contribute to decreased healthcare utilization and lower rates of HF diagnosis ^74,75,86. Additionally when LV hypertrophy (LVH) is noted in people of African descent this is more likely to be attributed to hypertension compared to other races⁸⁷. Educating physicians to carefully evaluate if there is an underlying etiology of LVH besides hypertension in patients of African descent will be a significant step in the right direction. There have not been well designed prospective studies testing strategies to screen for vATTR-CM, however the SCAN-MP (Screening for Cardiac Amyloidosis Using Nuclear Imaging for Minority Populations) trial represents a large effort aiming at answering this exact question ⁸⁸. With the advent of new imaging techniques to diagnose ATTR-CM, it will be easier to identify those with early vATTR-CM phenotype.

Earlier diagnosis is more important than ever given the introduction of therapeutics that appear to be more beneficial earlier in the disease course, as shown in the ATTR-ACT (Transthyretin Amyloidosis Cardiomyopathy Clinical Trial) phase III randomized controlled trial of the tetramer stabilizer tafamidis, ¹⁰. Recently, a prespecified analysis of ATTR-ACT showed that vATTR-CM patients (58% p.Val142Ile) had worse outcomes compared to wtATTR-CM ⁸⁹. In spite of the vATTR-CM patients being significantly younger, they had more severe disease at baseline with significantly fewer being NYHA Class I or II, lower 6-minute walk times, and higher NT-proBNP levels. This is particularly concerning as 52% of vATTR-CM patients were reported to be Black compared to 2% of wtATTR-CM. Although small and retrospective, an Italian study showed that p.Ile88Leu vATTR-CM patients (100% white) did not have a lower age-adjusted survival⁹⁰. Despite other studies attempting to control for African ancestry, these current data raise the question of whether the worse outcomes in vATTR-CM is related to socio-economic disparities or due to the underlying disease itself. Regrettably, participants of African descent represented only 2% of the patients enrolled in the pivotal clinical studies that led to the approval of inotersen and patisiran in vATTR polyneuropathy and 14% of the population in ATTR-ACT⁸⁰. While executing clinical trials in ATTR-CM is a difficult task, intentional effort and specific steps should be taken to recruit a more diverse clinical trial population.

The basis for male predilection in p.Val142Ile vATTR-CM is not well described, but it is thought to be due to milder phenotypes in females on presentation when compared to males of the same age. Sex differences in LV function of p.Val142Ile vATTR-CM patients (37% female; 96% Black) were recently evaluated using non-invasive pressure-volume analysis by Batra et al⁹¹. They found that despite females being diagnosed eight years later they had similar LV function and mortality, which further suggests that females have a less aggressive disease trajectory, making vATTR-CM diagnosis even more difficult. Likewise, Gonzalez-Lopez et al found that females made up 50% of those diagnosed wtATTR-CM, by prospectively performing ^99mTc-DPD (3,3-diphosphono-1,2-propanodicarboxylicacid) scintigraphy scans on consecutive elderly inpatients with LVH and unexplained HF with preserved ejection fraction⁸. This surprisingly high proportion of females may be explained by the older mean age of 86 years among all wtATTR-CM patients. There may be a pathophysiological basis for this observation. A mouse model showed that androgens increase hepatic TTR mRNA expression more than estrogen⁹². Rapezzi et al also showed that among vATTR-CM, females in the highest tertile of mean LV wall thickness were more often post-menopausal although none were p.Val142Ile carriers⁹³; this is consistent with the published evidence demonstrating thinner LV walls when females present with ATTR-CM⁹⁴. Finally, traditional sex disparities in cardiovascular outcomes, particularly for African American women may also contribute to this probable under-diagnosis^95,96.

The heterogeneity in the methodology and data required for extraction is a limitation of this systematic literature review. We included both observational studies and prospective therapeutic trials. Given the heterogeneity we did not perform a meta-analysis and limited summary statistics to ranges. There was additional heterogeneity in terms of the stage at which patients were recruited into a study as well as in the reporting of statistics. Furthermore, many studies were excluded due to the lack of information on specific type of ATTR-CM and these excluded studies may have altered the results if they could have been reported. We assumed that all those identified as Black race were of African descent, which can be associated with inaccuracies given that the genetic ancestry was not individually defined. Conversely, studies of p.Val142Ile were included even if ancestry or race was not mentioned. The consistency in the extracted data across studies that were included was reassuring that despite these limitations it was likely represented accurately.

Conclusion

The p.Val142Ile variant is the most common variant of the TTR gene with the majority of carriers being of African descent. The true penetrance to develop p.Val142Ile vATTR-CM is unknown, but it is associated with increased rates of incident heart failure and portends a lower overall survival. p.Val142Ile vATTR-CM is a frequent but under-recognized cause of HF among individuals of African descent and is associated with a worse overall survival and quality of life even compared to other causes of cardiac amyloidosis. It is paramount to increase awareness and diagnosis of p.Val142Ile ATTR-CM as the availability of newer therapeutics bring the promise of improving clinical outcomes in this historically disadvantaged population.

Nonstandard Abbreviations and Acronyms:

ARIC

Atherosclerosis Risk in Communities

ATTR-ACT

Transthyretin Amyloidosis Cardiomyopathy Clinical Trial

BEST

Beta-Blocker Evaluation in Survival Trial

CHS

Cardiovascular Health Study cohort

CI

Confidence Interval

CM

Cardiomyopathy

HF

heart failure

ICD

International Classification of Disease

KCCQ

Kansas City Cardiomyopathy Questionnaire

LV

left ventricle

LVH

left ventricular hypertrophy

MESA

Multi-Ethnic Study of Atherosclerosis

NAC

National Amyloidosis Centre, UK

NYHA

New York Heart Association

OHT

orthotopic heart transplantation

QoL

Quality of Life

ROW

Rest of World

THAOS

Transthyretin Amyloid Outcome Survey

TTR

Transthyretin

USA

United States of America

UK

United Kingdom

vATTR

variant transthyretin amyloidosis

vATTR-CM

variant transthyretin amyloidosis cardiomyopathy

wtATTR-CM

wild-type transthyretin amyloidosis cardiomyopathy