Discussion

In this study we demonstrated that cascade screening in ATTRv allows the early identification of a significant number of additional patients. Early detection of phenotype at the time of genetic diagnosis and early detection of phenotypic conversion during follow-up allowed early initiation of disease-modifying treatments improving the long-term outcome. To the best of our knowledge, this is the first comprehensive investigation of the effectiveness of cascade genetic screening in ATTRv in a nationwide cohort spanning over 20 years.

The main findings of our study can be summarized as follows: (i) cascade genetic screening identified 569 carriers, including 461 G+/P− and 108 G+/P+ family members, (ii) patients identified by screening received a diagnosis earlier than probands, allowing for potentially early referral for disease-modifying therapy, (iii) over a median follow-up of 5 years, 16.7% of G+/P− carriers converted to an overt clinical phenotype and became eligible for disease-modifying therapies: 11/62 converted before the PADO, (iv) disease-modifying treatment was significantly associated with improved clinical outcome in both index patients and G+/P+ relatives, with almost a 90% relative risk reduction in mortality, and (v) early treatment initiation in patients identified by cascade screening, either affected at baseline or developing the phenotype during follow-up, was associated with a more favourable risk profile compared to index patients (Structured Graphical Abstract).

Taken together these findings highlight the clinical relevance and reinforce the indication of cascade genetic screening in ATTRv. These findings further support the importance of a structured multidisciplinary care model—including neurology, cardiology, clinical genetics and, when appropriate, rehabilitation and ophthalmology—since neurological manifestations are common in variant carriers and may be under-recognized within cardiology-led pathways.

Impact of cascade genetic screening on ATTRv epidemiology

The demographic and clinical profiles of patients identified through cascade screening differed significantly from those of index patients. Over time, this practice showed the potential to reshape the epidemiology of ATTR-related disease and supports the need for genotype-, gender-, and age-specific screening approaches. This aligns with recent national data showing that, over the past two decades, improved awareness, advances in diagnostic tools, and wider genetic screening access have accelerated diagnosis, increased prevalence and incidence, and shifted the mutation spectrum towards mixed and cardiac phenotypes.17 Moreover, in our cohort, the yield of screening asymptomatic carriers increased in the later years of the study, likely reflecting this growing awareness, simplified diagnostic pathways, and earlier use of noninvasive imaging within structured screening programmes.

The G+/P+ cohort identified at baseline through family screening included both older siblings and younger offspring, reflecting the wide intrafamilial variability in disease expression. While this heterogeneity likely contributed to their intermediate clinical profile across cardiac and neurological assessments, differences in eGFR across groups should be interpreted with caution, as renal function may vary due to age and comorbidities and does not necessarily indicate amyloid-related renal involvement. Notably, the G+/P+ group was on average younger than probands, yet already showed clinical manifestations, and included a higher proportion of women—nearly 50%—a figure that exceeds what is typically reported in the literature.17,18 Given the autosomal dominant inheritance pattern of ATTRv, this finding raises the possibility that, still today, many women remain undiagnosed, not because of lower penetrance per se, but due to under-diagnosis.15,19 Although the underlying reasons remain uncertain, one possible explanation involves the growing prevalence of the Ile68Leu and other predominantly cardiac variants, for which diagnostic thresholds such as septal wall thickness may exhibit important sex-specific differences.17,20

In addition to gender-specific considerations, the high proportion of G+/P+ individuals among offspring—accounting for approximately one-third of cases—challenges the common practice of screening strategies based on the predicted age at symptom onset in the proband.1 This need is further supported by our finding that among the 16.7% of carriers converting to overt phenotype, a nonnegligible number of carriers developed symptoms more than 10 years earlier than the proband’s age at onset at both methods of PADO determination (determined either according to the age at disease onset in the family proband or according to average at disease diagnosis in the family pedigree). These early converters were clustered among carriers of the Ile68Leu, Glu89Gln, Phe64Leu, and Val30Met variants—which, although uncommon, highlights the substantial variability in age at onset associated with these variants.

Overall, while these findings are consistent with recent data supporting the feasibility and effectiveness of European Society of Cardiology (ESC)-recommended cascade screening protocols to identify specific ATTRv phenotypes,18 they also explore and extend the indication to testing and to routine multidisciplinary assessments even in the presence of mild symptoms (especially when specific variant carriers) in order to prioritize early diagnosis and determine eligibility to disease modifiers.

Impact of cascade genetic screening on treatment and survival

Beyond early diagnosis, at survival analysis, cascade family screening was also associated with a meaningful survival benefit. Over a median follow-up of 3.8 years, overall mortality exceeded 25%, with index patients experiencing higher risk and a significantly higher cumulative incidence of death compared to G+/P+. In time-dependent multivariable Cox regression, patients diagnosed via cascade screening (G+/P+) had a 57% lower risk of mortality relative to probands (HR 0.433, 95% CI 0.238–0.788, P = .006), even after adjustment for age, genotype, phenotype, and treatment exposure, indicating that earlier identification through cascade screening is associated with improved prognosis: further confirming this hypothesis, although initiation of disease-modifying therapy was associated with an approximate 90% reduction in mortality risk (HR 0.109, 95% CI 0.005–0.173, P < .001), its overall benefit declined with each year of delay, indicating a gradual attenuation of the initial protective effect during follow-up. This trend could really be driven by the probands who started treatment later in their life and have higher event rates despite DMD therapy. Of note, genotype and phenotype were both retained in the multivariable model, as phenotype expression in ATTRv is increasingly recognized to be heterogeneous across variants. Recent multicentre data have shown that TTR mutations once considered phenotype-specific (e.g. Val30Met, Ile68Leu, Phe64Leu, Val122Ile) frequently display mixed cardiac-neurological presentations, thereby reducing the risk of major genotype–phenotype collinearity in multivariable analyses.17

These observations should also be interpreted in the context of the evolving epidemiology of ATTRv and the historical nature of our cohort. Many index cases were diagnosed before disease-modifying therapies became available or reimbursed, whereas broader cascade screening in more recent years has enabled earlier detection, earlier treatment, and the identification of milder phenotypes—factors that directly influence survival estimates. For these reasons, some degree of residual confounding cannot be excluded. Nevertheless, these data reinforce the clinical value of structured cascade screening as a strategy to improve long-term outcomes in ATTRv, particularly when paired with prompt referral to treatment in the early stages of disease, while also underscoring the importance of considering potential reimbursement and access barriers.

Notably, cascade screening and periodic carrier follow-up can be time- and resource-consuming processes that may result in increased medicalization and anxiety—key modulators of quality of life, especially as it is often more complex given the challenges of communicating age-dependent penetrance.18 Recent evidence in related contexts, such as hypertrophic cardiomyopathy, has shown that the yield of genetic diagnosis can be as high as 1-in-4 screened individuals, with similar clinically meaningful long-term consequences.21 While these strategies may determine higher upfront costs in the short-term, they are later associated with a favourable cost-effectiveness profile.22–24 Importantly, at a time when DMDs are becoming increasingly available, the issue of a proper timing of treatment in mildly symptomatic carriers to prevent further disease burden is not only ethically very relevant but also cannot any longer be postponed. Obviously, ATTRv could benefit from early screening to guide disease-specific therapy and prevent disease progression–related hospitalizations, functional capacity deterioration and death.15

Another key finding from our cohort is that 16.7% of G+/P− carriers converted to an overt clinical phenotype over a median follow-up of 5 years. Our findings reinforce the concept that ATTRv is a dynamic continuum, beginning with early TTR amyloid deposition that may remain asymptomatic or subclinical for many years. Progression to overt clinical disease occurs once amyloid burden exceeds a threshold sufficient to impair organ function, a process that can vary greatly according to genotype, age, and other host factors. The observation that a proportion of carriers converted significantly earlier than predicted underscores this heterogeneity, and highlights the importance of genotype-specific, tailored surveillance to enable intervention at the earliest signs of phenotypic transition. Specifically, the rate of conversion, varied significantly by mutation. During a median follow-up of 5.3 [1.7–9.8] years, Glu89Gln carriers showed the highest conversion rate (42.2%), with a much earlier age of conversion compared to other more common pathogenic variants like Ile68Leu and Val122Ile. While we cannot exclude that the duration of follow-up could be in part responsible for this, this heterogeneity suggests a patient-tailored approach once genetic testing is confirmed positive.

Overall, these results align with the recommendations of the International Society of Amyloidosis, the American Heart Association, and the American College of Cardiology, which advocate genetic testing in all patients with ATTR.25–27 Our findings extend this message by demonstrating how such testing should be systematically integrated into clinical practice, coupled with genetic counselling, to enable early identification of at-risk relatives—not only in offspring (often carriers) but also in siblings (often presenting with overt phenotype). This approach supports timely access to both disease-modifying therapies and comprehensive supportive care (e.g. pharmacological, rehabilitative, physical, and nutritional interventions). Moreover, our data reinforce the rationale underpinning the ACT-EARLY trial (NCT06563895), emphasizing the strategic importance of early detection and surveillance of asymptomatic carriers to facilitate prompt intervention and improve long-term outcomes.

Study limitations

The retrospective design over 20 years may introduce bias, particularly in the referral and selection of index patients. However, the multicentre nature of the study including tertiary referral centres specialized in the care of rare diseases with similar screening and monitoring protocols, may have mitigated such potential bias. Follow-up intervals varied across centres (ranging from 6 to 24 months), particularly for asymptomatic carriers, which may have introduced bias in estimating the exact timing of phenotypic conversion, as those on longer surveillance intervals could have had delayed recognition of disease onset. Early subclinical patients may have been missed, leading to potential underestimation of ATTRv disease. Furthermore, given the long study period and the multicenter nature of the cohort, troponins were measured inconsistently and with different assays (including high-sensitivity assays). As such, data on troponins were not collected. Similarly, data on left atrial size, ejection fraction, and global longitudinal strain were not collected. No data on individuals who refused to undergo testing could be retrieved. From a clinical standpoint, data on PND score and NYHA class were recorded only when considered clinically relevant (i.e. when neurological or cardiac involvement was suspected or present); similarly, NAC scores or other potential residual confounding or modulating factors for survival analysis may have not been recorded systematically due to the retrospective nature of the study. Furthermore, only associations between covariates and mortality can be determined. Biopsies, like fat pads, have a reasonable sensitivity for ATTRv, but far from perfect: this may induce underestimation bias.

In addition, the study is focused on a predominantly Italian cohort with regional-specific variant distribution and lower prevalence of some genotypes (like Val122Ile or Thr60Ala). This may limit the generalizability of the findings to other populations, where genetic variants, mutation portfolio, and healthcare infrastructure may differ. On the other hand, this approach highlights the importance of a nationwide network, with shared protocols and collaboration in the setting of a uniform national healthcare system.

Future prospective international collaborative studies with standardized follow-up protocols across diverse populations are needed to validate these findings in different national cohorts, and identify variant-specific surveillance strategies.