Enhancing FSHD Diagnosis: a one-year follow-up study on the Efficacy of a Combined Methylation Assay and Machine Learning Pipeline

Domenica Megalizzi

Enhancing FSHD Diagnosis: a one-year follow-up study on the Efficacy of a Combined Methylation Assay and Machine Learning Pipeline

Autori

DOMENICA MEGALIZZI (GENOMIC MEDICINE LABORATORY-UILDM, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY; DEPARTMENT OF BIOMEDICINE AND PREVENTION, TOR VERGATA UNIVERSITY, 00133 ROME, ITALY. – PHD STUDENT)
GIULIA TRASTULLI (GENOMIC MEDICINE LABORATORY-UILDM, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY. – )
EMMA PROIETTI PIORGO (GENOMIC MEDICINE LABORATORY-UILDM, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY. – )
LUCA COLANTONI (GENOMIC MEDICINE LABORATORY-UILDM, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY. – )
RAFFAELLA CASCELLA (GENOMIC MEDICINE LABORATORY-UILDM, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY; DEPARTMENT OF BIOMEDICAL SCIENCES, CATHOLIC UNIVERSITY OUR LADY OF GOOD COUNSEL, TIRANA, ALBANIA. – )
CLAUDIA STRAFELLA (GENOMIC MEDICINE LABORATORY-UILDM, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY. – )
ELEONORA TORCHIA (UNITÀ OPERATIVA COMPLESSA DI NEUROLOGIA, FONDAZIONE POLICLINICO UNIVERSITARIO A. GEMELLI IRCCS, 00168 ROME, ITALY. – )
SARA BORTOLANI (UNITÀ OPERATIVA COMPLESSA DI NEUROLOGIA, FONDAZIONE POLICLINICO UNIVERSITARIO A. GEMELLI IRCCS, 00168 ROME, ITALY. – )
MAURO MONFORTE (UNITÀ OPERATIVA COMPLESSA DI NEUROLOGIA, FONDAZIONE POLICLINICO UNIVERSITARIO A. GEMELLI IRCCS, 00168 ROME, ITALY. – )
GUIDO PRIMIANO (NEUROFISIOPATHOLOGY UNIT, FONDAZIONE POLICLINICO UNIVERSITARIO AGOSTINO GEMELLI IRCCS, ROME, ITALY. – )
CRISTINA SANCRICCA (NEUROFISIOPATHOLOGY UNIT, FONDAZIONE POLICLINICO UNIVERSITARIO AGOSTINO GEMELLI IRCCS, ROME, ITALY. – )
CARLO CALTAGIRONE (DEPARTMENT OF CLINICAL AND BEHAVORIAL NEUROLOGY, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY. – )
ENZO RICCI (UNITÀ OPERATIVA COMPLESSA DI NEUROLOGIA, FONDAZIONE POLICLINICO UNIVERSITARIO A. GEMELLI IRCCS, 00168 ROME, ITALY; ISTITUTO DI NEUROLOGIA, UNIVERSITÀ CATTOLICA DEL SACRO CUORE, 00168 ROME, ITALY. – )
EMILIANO GIARDINA (GENOMIC MEDICINE LABORATORY-UILDM, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY; DEPARTMENT OF BIOMEDICINE AND PREVENTION, TOR VERGATA UNIVERSITY, 00133 ROME, ITALY. -)

Presentatore

DOMENICA MEGALIZZI (GENOMIC MEDICINE LABORATORY-UILDM, SANTA LUCIA FOUNDATION IRCCS, 00179 ROME, ITALY; DEPARTMENT OF BIOMEDICINE AND PREVENTION, TOR VERGATA UNIVERSITY, 00133 ROME, ITALY.)

Modalità

Oral Communication

Abstract

FacioScapuloHumeral Dystrophy (FSHD) is a myopathy characterized by the loss of repressive epigenetic features affecting the D4Z4 locus (4q35). The study aimed at validating the molecular test for discriminating FSHD subjects according to the DNA methylation profile of D4Z4. To this purpose, 218 subjects with clinical suspect of FSHD collected in 2022-2023 were analyzed. The recruited subjects were tested for the 4q subtelomeric variant and analysis of methylation levels related to DUX4-PAS and DR1 regions. Subsequently, a Machine Learning (ML) pipeline was employed to classify FSHD subjects, which showed reduced methylation levels compatible with the disease. Among the 218 subjects, the 4q variant type distribution was 54% 4qA/4qA, 43% 4qA/4qB and 3% 4qB/4qB. Among patients carrying at least a 4qA allele, the ML model correctly predicted 112 subjects as FSHD, consistently with the presence of FSHD genetic signatures. The remaining 100 subjects were predicted as non-FSHD. Among these subjects, 76 were concordant with the absence of FSHD genetic alterations and 24 patients displayed borderline methylation levels and included patients with 4qA/4qA genotype, asymptomatic patients and peculiar cases requiring a clinical evaluation. As a result, the test showed an excellent performance in rapidly discriminating FSHD patients and providing clinicians with a powerful tool for supporting the clinical diagnosis. In conclusion, this study proved the feasibility of the workflow based on methylation levels assessment and ML approach in the FSHD diagnostic practice.