Objective:To explore the clinical characteristics and prognostic value in hereditary transthyretin amyloidosis cardiomyopathy (hATTR-CM) patients based on cluster analysis, and to explore the risk factors for cardiovascular composite events.Methods:This retrospective cohort study included hATTR-CM patients who were admitted to Peking Union Medical College Hospital from January 2000 to January 2024. These patients were divided into two clusters using cluster analysis, based on genetic information, demographic information and clinical information. During the follow-up period, cardiovascular composite events were defined as all-cause death and hospitalization for heart failure. Both cardiovascular composite events and all-cause death were the endpoints. Kaplan-Meier survival curves and log-rank method were used to compare the prognostic significance of cluster analysis subgroups. Univariate and multivariate Cox proportional hazard regression models were used to analyze the risk factors affecting the incidence of cardiovascular composite events.Results:A total of 43 patients were included in this study, 30 were male (69.8%). In cluster 1 ( n=27), whose age of onset was (49.9±13.9) years old, 24(88.9%) of them started with neuropathy or gastrointestinal symptoms, and all clinical phenotypes were mixed type (neurological and cardiac). In cluster 2 ( n=16), whose age of onset was (59.0±10.6) years old, 15(93.8%) of them started with heart failure symptoms, and 13(81.3%) were pure cardiomyopathy. During the median follow-up time of 2.6 years, a total of 16 patients (37.2%) experienced composite cardiovascular events, and a total of 12 patients (27.9%) died. Kaplan-Meier survival curves showed a significantly lower cumulative survival rate for cardiovascular composite endpoint events (log-rank P=0.04) and all-cause death (log-rank P=0.04) in cluster 2 than in cluster 1. Univariate Cox proportional hazard regression model analysis showed that hATTR-CM patients with reduced estimated glomerular filtration rate, left ventricular ejection fraction≤40%, and moderate to severe mitral regurgitation were risk factors for vascular composite events (all P<0.05). Multivariate Cox proportional hazard regression analysis showed that left ventricular ejection fraction≤40% was an independent risk factor ( P<0.01). Conclusions:Cluster analysis is a valuable prediction tool on the prognostic stratification of hATTR-CM. Cluster 2, which is late-onset with onset of heart failure symptoms has a worse prognosis during follow-up period. The occurrence of composite cardiovascular events in hATTR-CM is related to left ventricular ejection fraction≤40%. Cluster analysis is helpful for clinical identification of high-risk groups.