Population-based genomic screening is an emerging approach to uncover disease risk in unsuspecting individuals [1
]. There is increasing evidence that genomic screening may identify and improve outcomes for certain medically actionable conditions for which preventive measures and/or effective treatments are available [2
]. Genomic screening programs, typically performed in the setting of large health system clinics [4
], biobanks [2
], or public health initiatives [3
], may choose to return medically actionable findings with varying degrees of clinical utility. Recent studies have found that most individuals at genomic risk for medically actionable conditions are unaware of their risk [5
], and that the majority of individuals informed of such risk will undergo recommended surveillance or risk-reducing interventions [2
]. Therefore, genomic screening holds the promise of identifying preventable and treatable diseases that may be otherwise missed in routine clinical care.
Hereditary transthyretin amyloidosis (hATTR) is an under-recognized systemic disease caused by pathogenic variants in the transthyretin (TTR) gene [9
]. Such variants cause misfolding of the tetrameric transthyretin protein complex, resulting in the extracellular deposition and accumulation of insoluble amyloid fibrils. hATTR is transmitted in an autosomal dominant manner and with incomplete penetrance. Due to the nonspecific and progressive nature of its symptoms, hATTR diagnosis is frequently delayed or missed [11
]. V142I is the most common pathogenic TTR variant worldwide, with recent prevalence estimates of 1 in 330 individuals [14
]. Up to 4% of African Americans (AA) in the U.S. harbor the V142I variant [16
], and V142I was also recently shown to be prevalent (1%) in Hispanic/Latinx (HL) populations in New York City [17
V142I is classically associated with cardiac amyloidosis, resulting in an increased risk for cardiomyopathy and heart failure (HF) [18
]. However, there is wide variability in the clinical spectrum, with many patients experiencing overlapping cardiac, neurologic, and other phenotypes [20
]. Certain systemic “red flags”, including carpal tunnel syndrome (CTS) and spinal stenosis (SS), raise clinical suspicion for hATTR and have been found to precede TTR-associated cardiac amyloidosis by 5–10 years [22
]. Identifying patients with hATTR early in the disease course is key to realizing the benefits of newly approved targeted therapies to suppress and stabilize transthyretin [26
]. Tafamidis, an FDA approved medication for hATTR patients with cardiac amyloidosis [29
], prevents the accumulation of amyloid fibrils, but it does not reduce existing amyloid deposits. Given the high frequency of V142I among AAs and HLs, overlooked or delayed diagnoses may miss the optimal therapeutic window and potentiate health disparities impacting these patient populations.
We recently launched a pilot program to screen for medically actionable genomic conditions in a large, ancestrally diverse biobank in New York City [30
]. As part of this program, we screened eligible participants for pathogenic variants in TTR associated with hATTR. The aim of the present study was to assess whether genomic screening could effectively identify individuals at risk for hATTR and prompt their follow-up for risk management. We identified biobank participants who received TTR V142I results through the pilot genomic screening program, and we evaluated their medical records for hATTR-related systemic features, uptake of recommended follow-up, and short-term outcomes. This study highlights the potential for translating genomic risk information into the care of diverse patient populations.
In total, 33 BioMe
participants received a TTR V142I result through a pilot genomic screening program. One participant with a pre-existing neurological condition was excluded from the present study due to the inability to discern neuropathic features potentially associated with hATTR. Among the 32 remaining V142I-positive participants, self-reported race/ethnicity was 17 (53%) AA and 15 (47%) HL (Table 1
). Median age was 57 years (range 30–79), and 26 (81%) participants were female. None had a previous diagnosis of hATTR.
We queried participants’ medical records for the presence of cardiac, peripheral neuropathy, and autonomic neuropathy features of hATTR (Figure 2
and Supplemental Table S1
). At the time of result disclosure, 18 (56%) participants had at least one hATTR-related systemic feature. Peripheral neuropathy was most common, present in 15 (47%) individuals (Figure 2
). This was largely driven by the high prevalence of CTS (N
= 10, 31%) and SS (N
= 10, 31%). Five participants had lumbar stenosis, 6 had cervical stenosis, and 1 had both. Autonomic neuropathy features were noted in 7 (22%) individuals and included cardiac autonomic dysfunction (N
= 1, 3%), incontinence (N
= 4, 13%), and sexual dysfunction/impotence (N
= 3, 9%). Four (13%) individuals had HF. An additional 12 (38%) participants had at least one of seventeen related symptoms in the absence of a systemic feature of hATTR (Figure 2
and Supplemental Table S1
Fifteen (47%) V142I-positive individuals had a family history of any hATTR-related systemic feature in first- or second-degree relatives (Table 1
and Supplemental Table S1
). A family history of CTS was most common, reported in 13 first- or second-degree relatives of 8 (25%) participants. Six (19%) participants reported a family history of HF in a total of 8 relatives. No participants had a known family history of hATTR.
We evaluated the uptake of recommended follow-up among participants who received TTR V142I results. Eighteen (56%) participants followed up with a cardiologist/cardiovascular geneticist (Table 2
) after result disclosure. These visits occurred within 8 months of results disclosure (median 4.9 weeks). Recommended cardiac evaluation included baseline ECG and echocardiogram for all V142I-positive individuals (unless a test had been obtained in the past 12 months) and Tc-99m-PYP scintigraphy with SPECT imaging based on age and risk factors. Sixteen of the 18 individuals (89%) who followed up with a specialist underwent recommended cardiac evaluations, including ECG (N
= 10), echocardiogram (N
= 12), and/or Tc-99m-PYP scintigraphy (N
= 10). An AA man with HF had findings on echocardiogram and Tc-99m-PYP scintigraphy consistent with cardiac amyloidosis, and was subsequently diagnosed with hATTR. This is the only participant with a diagnosis of hATTR at the time of this study.
Three individuals had opportunistic histological evaluation of tissue samples for amyloid. Two HL women had histological analysis with Congo red staining of previous tissue specimens, which were negative for amyloid. The third, an AA woman with coronary artery disease, had an endomyocardial biopsy performed during a previously scheduled cardiac catheterization, which was negative for amyloid. None of these individuals had a diagnosis of hATTR at the time of this study.
Additional hATTR-related features were documented for three individuals who completed follow-up evaluations, including one individual with HF and atrial fibrillation, a second with sexual dysfunction, and a third with incontinence. With these, a total of 19 (59%) V142I-positive individuals were found to have an hATTR-related systemic feature (Supplemental Table S1
). All 10 individuals with CTS had evidence of bilateral involvement. Specialists also identified additional symptoms associated with hATTR in 12 participants. In total, following a specialist’s evaluation, 31 individuals (96%) had either a systemic feature of hATTR or a related symptom.
The presence of hATTR-related systemic features was compared across age, sex, and self-reported race/ethnicity groups (Supplemental Table S2
). The proportion of individuals with hATTR-related features was not significantly different in those ≥60 years vs. <60 years (P
= 0.29), in males vs. females (P
= 0.67), or in AA vs. HL participants (P
hATTR is an under-recognized yet treatable disease. Accurate diagnosis early in the disease course is critical, as novel therapeutics prevent new cardiac amyloid deposition but do not ameliorate existing amyloid [28
]. hATTR diagnosis can be improved by better understanding the natural history and phenotypic spectrum of disease. In November 2019, we initiated the return of pathogenic TTR variants to biobank participants through a pilot genomic screening program [30
], which offers an unprecedented opportunity to examine the presence of hATTR-related systemic features in an unselected population at high genomic risk. Here, we explored the characteristics of 32 biobank participants receiving a TTR V142I result who were previously unaware of their risk. All participants were either AA or HL by self-report, populations that have been underrepresented in genomic research in the U.S. and elsewhere [37
]. We found that the majority had at least one hATTR-related systemic feature at the time of their result disclosure (56%), and even more so after some underwent follow-up specialist evaluation (59%).
Penetrance of HF in V142I-positive individuals is not well defined [19
], but the risk has been shown to increase with age [39
]. Five individuals in our study had a diagnosis of HF, four of which were present at the time of result disclosure and one of which became clinically apparent directly following result disclosure. This is the only participant diagnosed with hATTR to date. Only one additional patient had sufficient features to prompt an opportunistic endomyocardial biopsy, which was ultimately negative. Although the penetrance of cardiac amyloid deposition in V142I has been estimated to be as high as 80% by age 70 [40
], when considered from a genomics-first approach, penetrance may be much lower [19
]. This suggests that routine inclusion of invasive procedures such as endomyocardial biopsies in the workup of younger and/or asymptomatic V142I-positive individuals may not be justified.
Studies of V142I have primarily centered on cardiac features, with limited focus on neuropathy [21
]. Bilateral CTS and SS emerged as common phenotypes in V142I-positive individuals in our study. Though often highlighted as a feature of wild-type ATTR or with other pathogenic TTR variants [23
], CTS has also been described in association with V142I [42
]. A recent study found CTS in 16 of 55 V142I-positive individuals, most of whom had a known diagnosis of hATTR [21
]. While SS—particularly lumbar SS—is well-recognized in wild-type ATTR [22
], its association with hATTR has only been appreciated more recently [23
]. Symptoms that could be attributable to an hATTR-related systemic feature were exceedingly common in V142I-positive individuals. Shortness of breath, numbness/tingling, back pain, joint pain, and gastrointestinal symptoms were all observed in over half of the participants. Many of these symptoms are non-specific and, thus, may not raise suspicion for hATTR, particularly if other associated comorbidities are present (e.g., paresthesia with diabetes) [12
]. As such, these symptoms may be most appropriately used to guide workup for amyloidosis in known V142I variant positive individuals, rather than to help identify candidates for TTR genetic testing.
The current approach to TTR V142I identification typically relies on the presence of cardiac amyloidosis in a symptomatic individual. This can be detected via tissue biopsy, or more recently, non-invasive nuclear imaging, such as Tc-99m-PYP scintigraphy with corroborative monoclonal protein testing [11
]. TTR genetic testing is then performed to differentiate hereditary from wild-type ATTR. Unaffected, at-risk individuals may be discovered through cascade testing of close relatives, and consensus guidelines for counseling have recently been proposed in this setting [43
]. In April 2019, a direct-to-consumer genetic testing company 23andMe added V142I and two other common TTR variants, V50M and T80A, to its Health Risk Reports [45
], allowing individuals to discover their genomic risk without the involvement of a healthcare provider. We therefore expect increasing numbers of asymptomatic individuals harboring pathogenic TTR variants to come to medical attention. There are currently no evidence-based guidelines for the management of TTR variant positive individuals. However, recommendations based on expert opinion underscore the call for early identification and multidisciplinary care of at-risk individuals so that surveillance and, if indicated, therapy would have the greatest potential for benefit [44
]. Strategies for managing both asymptomatic and symptomatic individuals will likely evolve over time. This study helps inform a process to initiate risk management in V142I-positive individuals identified by a genomics-first approach (Figure 1
V142I-positive individuals made medical management decisions based on their results. Over half sought follow-up care with a cardiologist within 8 months of result disclosure, and most of these individuals completed recommended cardiac evaluation. This is consistent with uptake observed in a genomic screening program for other actionable conditions, where 56% of participants with no prior knowledge of genomic risk followed up with a specialist referral and/or procedure [2
]. In our study, follow-up specialist evaluations led to the identification of additional cardiac and non-cardiac phenotypes related to hATTR. This underscores the value of multidisciplinary specialty care in the evaluation of individuals at high risk of hATTR.
There are limitations to this study. We retrospectively reviewed participants’ medical records, which may not fully capture all relevant medical and family history. However, targeted information was obtained during result disclosure visits, and during subsequent specialist visits, reducing the chance of missing relevant information. This study included a small sample of mostly female biobank participants, which limits the generalizability of our findings. The timeframe of the study may have been too short for participants to complete recommended medical management, and it coincided with the peak of the COVID-19 pandemic in New York City, which may have impacted uptake of recommendations. Concomitant comorbidities, such as diabetes and hypertension, may have contributed to some of the phenotypes captured in this study. That said, it is important to recognize that such comorbidities may obscure recognition of hATTR-related phenotypes in day-to-day clinical practice, leading to missed diagnosis and/or patients experiencing a diagnostic odyssey [47
This study helps to inform genomics-first approaches for an under-recognized condition that largely impacts AA and HL populations in the U.S. The presence of population-specific disease variants in underrepresented populations may cause health disparities [48
]. Approximately 4% of AA and 1% of HL in NYC harbor the TTR V142I variant and are, thus, at increased risk for hATTR. Despite having potentially related features of hATTR, none of our study participants had prior knowledge of their genomic risk. Genomic result disclosure prompted follow-up evaluation with a specialist in over half of participants, with initiation of surveillance in most. Findings in one individual were diagnostic for hATTR, facilitating appropriate therapy. Longitudinal follow up will be necessary to improve the early identification of hATTR cardiac amyloidosis in V142I-positive individuals and to determine optimal surveillance strategies and treatment once amyloidosis is present. Further questions remain regarding the appropriate implementation of genomic screening, cost-benefit of this approach, and equitable access to recommended follow-up care.