A Muscle Biosignature Differentiating Between Limb-Girdle Muscular Dystrophy and Idiopathic Inflammatory Myopathy on Magnetic Resonance Imaging

Machine learning-based radiomics to differentiate immune-mediated necrotizing myopathy from limb-girdle muscular dystrophy R2 using MRI

Objectives

This study aimed to assess the feasibility of a machine learning-based radiomics tools to discriminate between Limb-girdle muscular dystrophy R2 (LGMDR2) and immune-mediated necrotizing myopathy (IMNM) using lower-limb muscle magnetic resonance imaging (MRI) examination.

Methods

After institutional review board approval, 30 patients with genetically proven LGMDR2 (12 females; age, 34.0 ± 11.3) and 45 patients with IMNM (28 females; age, 49.2 ± 16.6) who underwent lower-limb MRI examination including T1-weighted and interactive decomposition water and fat with echos asymmetric and least-squares estimation (IDEAL) sequences between July 2014 and August 2022 were included. Radiomics features of muscles were obtained, and four machine learning algorithms were conducted to select the optimal radiomics classifier for differential diagnosis. This selected algorithm was performed to construct the T1-weighted (TM), water-only (WM), or the combined model (CM) for calf-only, thigh-only, or the calf and thigh MR images, respectively. And their diagnostic performance was studied using area under the curve (AUC) and compared to the semi-quantitative model constructed by the modified Mercuri scale of calf and thigh muscles scored by two radiologists specialized in musculoskeletal imaging.

Results

The logistic regression (LR) model was the optimal radiomics model. The performance of the WM and CM for thigh-only images (AUC 0.893, 0.913) was better than those for calf-only images (AUC 0.846, 0.880) except the TM. For "calf + thigh" images, the TM, WM, and CM models always performed best (AUC 0.953, 0.907, 0.953) with excellent accuracy (92.0, 84.0, 88.0%). The AUCs of the Mercuri model of the calf, thigh, and "calf + thigh" images were 0.847, 0.900, and 0.953 with accuracy (84.0, 84.0, 88.0%).

Conclusion

Machine learning-based radiomics models can differentiate LGMDR2 from IMNM, performing better than visual assessment. The model built by combining calf and thigh images presents excellent diagnostic efficiency.

Novel mutations in the SGCA gene in unrelated Vietnamese patients with limb-girdle muscular dystrophies disease

Background: Limb-girdle muscular dystrophy (LGMD) is a group of inherited neuromuscular disorders characterized by atrophy and weakness in the shoulders and hips. Over 30 subtypes have been described in five dominant (LGMD type 1 or LGMDD) and 27 recessive (LGMD type 2 or LGMDR). Each subtype involves a mutation in a single gene and has high heterogeneity in age of onset, expression, progression, and prognosis. In addition, the lack of understanding of the disease and the vague, nonspecific symptoms of LGMD subtypes make diagnosis difficult. Even as next-generation sequencing (NGS) genetic testing has become commonplace, some patients remain undiagnosed for many years. Methods: To identify LGMD-associated mutations, Targeted sequencing was performed in the patients and Sanger sequencing was performed in patients and family members. The in silico analysis tools such as Fathmm, M-CAP, Mutation Taster, PolyPhen 2, PROVEAN, REVEL, SIFT, MaxEntScan, Spliceailookup, Human Splicing Finder, NetGene2, and Fruitfly were used to predict the influence of the novel mutations. The pathogenicity of the mutation was interpreted according to the ACMG guidelines. Results: In this study, six patients from four different Vietnamese families were collected for genetic analysis at The Center for Gene and Protein Research and The Department of Molecular Pathology Faculty of Medical Technology, Hanoi Medical University, Hanoi, Vietnam. Based on clinical symptoms and serum creatine kinase (CK) levels, the patients were diagnosed with limb-girdle muscular dystrophies. Five mutations, including four (c.229C>T, p.Arg77Cys; exon one to three deletion; c.983 + 5G>C; and c.257_258insTGGCT, p.Phe88Leufs*125) in the SGCA gene and one (c.946-4_946-1delACAG) in the CAPN3 gene, were detected in six LGMD patients from four unrelated Vietnamese families. Two homozygous mutations (c.983 + 5G>C and c.257_258insTGGCT) in the SGCA gene were novel. These mutations were identified as the cause of the disease in the patients. Conclusion: Our results contribute to the general understanding of the etiology of the disease and provide the basis for definitive diagnosis and support genetic counseling and prenatal screening.

The Clinicopathological Distinction between Immune-Mediated Necrotizing Myopathy and Limb-Girdle Muscular Dystrophy R2: Key Points to Prevent Misdiagnosis

Background: Limb−girdle muscular dystrophy R2 (LGMD R2) is most frequently misdiagnosed as immune-mediated necrotizing myopathy (IMNM). This study aimed to compare the clinicopathological data of IMNM and LGMD R2 to find distinguishing features. Methods: We retrospectively reassessed the medical data of patients with IMNM (n = 41) and LGMD R2 (n = 8) treated at Tongji Hospital from January 2017 to December 2021. Results: In our cohort, patients with LGMD R2 had a longer interval of onset to first visit, mild muscle weakness with late upper limb involvement, less myalgia, no cervical muscle weakness or dysphagia, no extramuscular organs affected except cardiac involvement, and lack of various autoantibodies, such as antinuclear antibodies. These features were completely reversed in IMNM. Moreover, thigh MRIs showed that muscle edema prominently affecting the adductor magnus was a characteristic of IMNM, while extensive fatty replacement was more common in LGMD R2 (p = 0.0086). Necrotic myofibers presented in both entities (p = 0.1693), while features such as ring/whorled and splitting myofibers were more often found in LGMD R2 (p = 0.0112 and p < 0.0001, respectively). Conversely, sarcoplasmic p62 expression was more pronounced in IMNM (p < 0.05). There were 4 of 8 (50%) patients with LGMD R2 initially considered as seronegative IMNM, and therefore unnecessarily treated with immunosuppressive drugs. Insufficient recognition of the early clinical, imaging, and histopathological features of LGMD R2 is the main reason for misdiagnosis. Conclusions: These findings may help clinicians differentiate seronegative IMNM and LGMD R2, reducing early misdiagnosis and mismanagement. Particularly, prominent adductor magnus edema on MRI and abundant p62 staining seem to be good markers for IMNM, while the presence of splitting myofibers is a crucial clue to early hereditary myopathy, including LGMD R2.