Women Remain a Small Minority in Tafamidis-Treated Amyloid Heart Disease Cohort

Women made up just over one in ten patients in a large pooled analysis of people with transthyretin amyloid cardiomyopathy treated with tafamidis, yet they differed from men in striking ways at the moment they entered care. The analysis, published in Cardiology and Therapy, pooled 1924 tafamidis-treated patients from two contemporary sources: 1476 from an early-access arm of a long-term extension study and 448 enrolled between 2019 and 2023 in the Transthyretin Amyloidosis Outcomes Survey, a global longitudinal registry. Of the total, 214 patients, or 11.1 percent, were women. The patients came from 77 study sites in 19 countries, and the analysis set out to examine how sex relates to clinical presentation and to all-cause mortality in a treated, real-world population rather than in a randomized trial.

Transthyretin amyloid cardiomyopathy arises when misfolded transthyretin protein deposits as amyloid fibrils in the heart muscle, driving progressive heart failure, arrhythmias and reduced survival. It occurs in a wild-type form, ATTRwt-CM, and in hereditary variant forms, ATTRv-CM, both associated with substantial morbidity and mortality. Tafamidis, a once-daily TTR stabilizer, has shown a statistically significant reduction in mortality at 30 months compared with placebo in the phase 3 ATTR-ACT trial, and real-world outcomes have since been described through the THAOS registry. Despite this accumulating experience, the disease has historically been viewed as one that predominantly affects older men, and women remain underrepresented in prior studies, leaving sex-specific data in contemporary treated patients limited.

The genotype split between the sexes was one of the clearest findings. A smaller proportion of women than men had the wild-type form of the disease, 59.8 percent versus 88.9 percent, while a greater proportion of women carried hereditary variants, 40.2 percent versus 11.1 percent, a difference that was highly statistically significant. Women were also more frequently of African descent, 23.9 percent versus 7.4 percent, whereas men were more often white, 88.0 percent versus 66.2 percent. The authors note that this pattern fits the known epidemiology of variant disease, which is more prevalent among people of African ancestry, particularly in association with the V122I variant, and they argue that genotype and ancestry must be considered when interpreting sex-related differences in amyloid cardiomyopathy cohorts.

Timing, by contrast, looked remarkably similar across sexes. Mean age at enrollment was 77.2 years for women and 76.7 years for men; mean age at symptom onset was 74.3 versus 73.7 years; and mean time from symptom onset to diagnosis was 48.3 versus 47.1 months, with no significant differences on any of these measures. Medical histories were broadly comparable, although women reported carpal tunnel syndrome less often, 4.7 percent versus 10.6 percent, and tendon rupture not at all, versus 2.7 percent in men. Mean follow-up was 1.5 years in both groups, a duration the authors themselves flag as modest in their discussion of the study’s limitations.

At presentation, women appeared more symptomatic. A higher proportion of women than men were in New York Heart Association class III or IV, 37.7 percent versus 29.4 percent, and a higher proportion were classified as National Amyloidosis Centre stage II or III, 68.1 percent versus 47.7 percent, both differences reaching statistical significance. Yet mean NT-proBNP levels, a key biomarker of cardiac strain, were similar between the sexes, at 3459.8 pg/ml in women and 3227.3 pg/ml in men. Women also had a higher mean heart rate, 74.8 versus 71.6 beats per minute, and shorter mean PR, QRS, QT and QTcF intervals on electrocardiography. The authors observe that most of these conduction differences match established sex differences in healthy populations, with the notable exception that women showed shorter QT and QTcF intervals, which runs counter to what is seen in healthy people.

Genotype-stratified analyses sharpened the picture. Among patients with the wild-type form, women were older than men both at enrollment, with a mean age of 80.7 versus 77.6 years, and at symptom onset, 77.5 versus 74.5 years, and a higher proportion of women were in NYHA class III/IV, 42.5 percent versus 28.8 percent, and in NAC stage II/III, 69.7 percent versus 48.0 percent. Among patients with variant disease, women were older at enrollment, 72.1 versus 69.6 years, but age at symptom onset did not differ significantly, and neither NYHA class nor NAC stage differed significantly by sex. In both genotype subgroups, women had significantly lower mean estimated glomerular filtration rates: 54.5 versus 65.9 ml/min/1.73 m2 in variant disease and 45.0 versus 57.5 ml/min/1.73 m2 in wild-type disease.

Regression modeling, adjusted for age, cohort, baseline NYHA class and genotype, confirmed two consistent differences. Women had significantly higher eGFR than men once these factors were accounted for, with a coefficient for men versus women of −10.61, and women had lower heart rate, with a coefficient of 2.60 for men versus women. The eGFR pattern held in both genotype subgroups. The authors emphasize that most apparent sex-related differences in the pooled cohort were attenuated after stratification by genotype, leaving men and women with broadly similar symptom burden, biomarker profiles and clinical characteristics; the only consistent differences were older age and lower eGFR in women.

Mortality told a more nuanced story. During the study, 34 women, or 15.9 percent, and 346 men, or 20.2 percent, died, with cardiovascular causes recorded in 2 and 26 of these deaths respectively. Unadjusted Kaplan–Meier analysis showed no statistically significant sex-based difference in all-cause mortality, with a log-rank p value of 0.178. But after adjustment for age, study, baseline NYHA class and TTR genotype, men had a significantly higher risk of all-cause death than women, with a hazard ratio of 1.60 and a 95 percent confidence interval of 1.12 to 2.30. The authors attribute the gap between the unadjusted and adjusted findings to confounding by baseline factors that differed between the sexes, and they caution that adjusted mortality findings should be interpreted with care because adjusting for genotype, which was unevenly distributed between men and women, may have over-controlled in the multivariable models.

Within genotype subgroups, no statistically significant sex-based differences in all-cause mortality emerged, either in Kaplan–Meier analyses or after adjustment for baseline variables, suggesting that differences in genotype distribution may have contributed to the overall mortality result. Cause-of-death patterns were broadly comparable between the sexes and do not suggest that a split between cardiovascular and non-cardiovascular deaths explains the survival findings, although the authors note that a substantial proportion of deaths had unknown cause, limiting that interpretation. They also situate the symptom pattern in a wider context: women reporting greater functional limitation despite similar biomarker levels has been described in broader heart failure populations, where women often report greater symptom burden and poorer quality of life despite comparable biomarkers and similar or more favorable survival, with possible mechanisms including body size, comorbidity burden, exercise capacity, symptom perception and referral patterns.

The study’s limitations are substantial and stated plainly. The analysis was observational and pooled two distinct cohorts with different enrollment criteria, follow-up procedures and definitions of baseline, the latter potentially introducing heterogeneity in time zero and survival follow-up. The small number of women limited power for subgroup analyses, residual confounding cannot be excluded, and baseline NT-proBNP and NAC stage data were missing for a substantial proportion of patients, which also constrained the ability to control for disease severity in the mortality analysis. Inclusion was restricted to tafamidis-treated patients and may not reflect the broader untreated population, and without a comparison group the data could not assess differential tafamidis efficacy by sex. Within those bounds, the authors’ conclusion is measured: women remain underrepresented among contemporary treated patients, a greater proportion carry hereditary variants, women appear more symptomatic at presentation despite similar NT-proBNP, and men show higher adjusted mortality risk, with further studies needed to understand sex-related differences in diagnosis, referral patterns and disease expression.

Subject of Research: Sex differences in clinical characteristics and all-cause mortality among patients with transthyretin amyloid cardiomyopathy treated with tafamidis

Article Title: Sex Differences in a Contemporary, Real-World Cohort of Patients with Transthyretin Amyloid Cardiomyopathy Treated with Tafamidis

Article References: Fontana, M., Aimo, A., Grogan, M., Crespo-Leiro, M. G., Maurer, M. S., Siepen, F. A. D., Hanna, M., Delgado, D., Carlsson, M., Damy, T., & Davis, M. K. (2026). Sex Differences in a Contemporary, Real-World Cohort of Patients with Transthyretin Amyloid Cardiomyopathy Treated with Tafamidis. Cardiology and Therapy, 15(S1), 81-93. https://doi.org/10.1007/s40119-026-00475-3

Image Credits: AI Generated

DOI: 10.1007/s40119-026-00475-3

Keywords: transthyretin amyloid cardiomyopathy, tafamidis, sex differences, ATTRwt-CM, ATTRv-CM, THAOS registry, NT-proBNP, all-cause mortality, NYHA class, eGFR, genotype, cardiac amyloidosis