Long-Term Monitoring of Cardiac Involvement under Migalastat Treatment Using Magnetic Resonance Tomography in Fabry Disease

**Background:** Fabry disease (FD) is an X-linked lysosomal storage disorder caused by pathogenic GLA mutations leading to deficient α-galactosidase A activity and glycosphingolipid accumulation. Cardiac involvement—characterized by left ventricular hypertrophy (LVH) and myocardial fibrosis—is a major cause of morbidity and reduced life expectancy (approximately 20 years in males, 15 years in females if untreated). Migalastat, an oral pharmacological chaperone approved in the EU in 2016 and Switzerland in 2017, has shown promise in stabilizing or improving LVH in phase III trials (ATTRACT, FACETS) and real-world studies, but those studies primarily used 2D echocardiography, which has high inter- and intra-observer variability. Cardiac magnetic resonance (CMR) is the gold standard for quantifying left ventricular mass and detecting fibrosis via late gadolinium enhancement (LGE). This study aimed to evaluate the evolution of Fabry cardiomyopathy under migalastat using CMR in a real-world cohort including both ERT pre-treated and treatment-naïve patients.

**Methods:** This retrospective multinational study included 15 consecutive adult patients (11 females, 4 males) with amenable GLA mutations from centers in Zurich, Lausanne, and Vienna. Patients started migalastat between April 2017 and December 2020; last follow-up was censored in February 2023. Median follow-up was 34 months (range 25–47). CMR examinations were performed on 1.5T scanners (one on 3T) using Medis Suite MR or Syngovia software. Parameters measured included LVMi (LVH defined as >75 g/m² for males, >59 g/m² for females), interventricular septal thickness (IVS), posterior wall thickness (PWT), left ventricular end-diastolic volume (LVEDV), ejection fraction (EF), LGE extent (% of LV mass using full width at half-maximum), and septal T1 relaxation times. α-Gal A activity (as % of lower reference limit) and lyso-Gb3 in dried blood spots were measured. eGFR was calculated using CKD-EPI. Wilcoxon tests and t-tests were used for paired comparisons; p < 0.05 was considered significant.

**Key Results:** Median α-Gal A activity significantly increased from 37.3% (IQR 5.88–89.3) at baseline to 105% (IQR 37–177) at follow-up (p = 0.005). Lyso-Gb3 showed a non-significant trend toward reduction, mainly in males. Overall, there were no significant changes in LVMi (baseline median 64.2 g/m² vs. follow-up 66.8 g/m², p = 0.294), IVS (13 vs. 13 mm, p = 0.470), PWT (9 vs. 10 mm, p = 0.577), LVEDV (96 vs. 95 mL, p = 0.096), or LVEF (67% vs. 61%, p = 0.152). However, LGE extent increased significantly from 0% (IQR 0–4.1) to 0% (IQR 0–5.5) overall (p = 0.043). Among the 7 patients with pre-existing LGE, fibrosis increased from a median of 3% (IQR 2.3–14.3) to 6% (IQR 4.24–21.6) of LV mass. No patient without LGE at baseline developed new-onset LGE. T1 relaxation times varied without a clear trend (baseline median 916 ms vs. follow-up 927 ms, p = 0.859). eGFR remained stable (baseline median 93.5 vs. follow-up 90.48 mL/min/1.73 m², p = 0.074). Individual analysis showed LVMi progression in two male patients with the late-onset mutation c.713G>A p.(S238N) and one 70-year-old female with classic mutation c.581C>T p.(T194I) who also had progressive LGE despite being the only patient on acetylsalicylic acid. No correlation was found between change in α-Gal A activity and LVMi course.

**Clinical Implications:** This study confirms an overall stable course of CMR-measured LVMi and kidney function under migalastat, consistent with the recent MAIORA study. However, the significant progression of LGE in patients with pre-existing fibrosis—despite therapy—highlights that migalastat may not fully halt fibrotic processes once initiated. The absence of new-onset LGE suggests a potential preventive effect on initial fibrosis formation. The discrepancy with prior echocardiography-based studies that reported LVMi reduction underscores CMR's superior sensitivity and lower variability, making it the preferred modality for monitoring Fabry cardiomyopathy, especially in small cohorts. Regular CMR surveillance is recommended, with shorter intervals for patients with LGE, who may be at risk for malignant arrhythmias and may benefit from optimized device therapy. The potential enhancing effect of ASA co-medication on migalastat (reported in cell models) warrants further clinical investigation, as the single ASA-treated patient in this cohort showed enzyme activity increase but also disease progression. Limitations include the small sample size, predominance of females and late-onset variants, retrospective multicenter design with different MRI scanners/software, and missing T1 data in some cases.