Multimodal phenotypic clustering predicts cardiac outcomes in Fabry disease

Fabry disease (FD) is a rare lysosomal storage disorder with cardiac involvement. The efficacy of cardiac assessments in predicting disease progression is uncertain and few long-term studies have evaluated a clinically comprehensive approach in tracking cardiac outcomes. We developed a multi-modal framework integrating sex, genetics, clinical assessments, electrocardiography (ECG), imaging, and biomarkers to characterize the progression of cardiac involvement in FD and assess the predictive value of each modality for adverse cardiovascular outcomes.

Our cohort encompassed 525 "visit years" across 98 participants (60 females; 60 with classic mutations; mean age at presentation: 42.5 ± 17). We defined four cardiac phenotypic clusters (CPC): abnormal ECG, elevated cardiac biomarkers, left ventricular hypertrophy (LVH) (by echocardiography or cardiac magnetic resonance [CMR]), and late gadolinium enhancement (LGE) (by CMR). The primary outcome was a composite of cardiac hospitalization, device implantation, or cardiovascular/cerebrovascular death. A CPC progression map was plotted, outlining the course of events in FD cardiomyopathy, where changes in ECG and the occurrence of hypertrophy precede increase in cardiac biomarkers and the formation of LGE. Mutation type (classic vs. late-onset) and sex (female vs. male) showed a significant link with a composite outcome of cardiac-related hospitalizations, device implantations (defibrillator/pacemaker), and cardiovascular or cerebrovascular mortality [Hazard ratio, HR (95% confidence interval, CI, p-value): 12.8 (3.5,46.7), p = 0.0001 and 0.09 (0.03,0.33), p = 0.0002, respectively]. The presence of LGE CPC was associated with an increased likelihood of encountering the composite outcome.

We propose a readily-applied "4 domain CPC" model, that can be routinely used in clinical practice, as a prospective "guide-map" for FD cardiomyopathy progression. Its utility is demonstrated in showing the expected role of sex, mutation and CMR findings in the cardiac trajectory of FD.