1161: Proteasome Inhibitor Repurposed for Dermatomyositis: Results of a Drug Repurposing Analysis Based on the Transcriptomic Signature of Patients' Perifascicular Fibers Validated in Pre-clinical Models

Léa Debrut¹, Margherita Giannini², Céline Keime¹, Daniela Rovito¹, Béatrice Lannes³, Anne-Laure Charles⁴, Daniel Metzger¹, Bernard Geny⁵, Gilles Laverny¹ and Alain Meyer⁶, ¹Institut de Génétique et de Biologie Moléculaire et Cellulaire (IGBMC), Illkirch, France, ²Nouvel Hopital Civil, Strasbourg, Strasbourg, France, ³Hôpital de Hautepierre - Hôpitaux Universitaires de Strasbourg, Strasbourg, France, ⁴UR3072, Centre de Recherche en Biomédecine de Strasbourg (CRBS), Strasbourg, France, ⁵Nouvel Hôpital Civil, Strasbourg, France, ⁶Hôpitaux Universitaires de Strasbourg, Strasbourg, France

Background/Purpose: Dermatomyositis (DM) is an autoimmune myopathy responsible for muscle weakness associated with decreased quality of life and increased mortality. DM muscular histology is characterized by specific lesions of perifascicular fibers consisting of atrophy, type I interferon (IFN-I) signature, expression of major histocompatibility complex class I (MHC-I) and mitochondrial dysfunctions (1). The origin of these lesions is not understood. Current treatments are empirical, partially effective, expose to a risk of side effects and present a high rate of relapse upon discontinuation. The objective of the study was to identify innovative therapeutic strategies for DM, based on the identification of the molecular pathways that underlie perifascicular fibers lesions and the repositioning of drugs already approved in humans.

Methods: To reveal the molecular pathways underlying DM perifascicular fibers lesions, perifascicular and endofascicular fibers from muscle biopsy of 19 patients with recent and untreated myositis (DM, other myositis) or without myopathy were microdissected by laser, their transcriptome was established by RNA sequencing and analyzed by bioinformatic methods. To identify innovative therapeutic strategies based both on DM pathophysiological mechanisms and on existing drugs, the transcriptomic signature specific to DM perifascicular fibers obtained by microdissection experiments was used for a drug repositioning analysis as described by Karatzas et al. (2). To validate the predications obtained, the effect of drug candidates already used in humans was tested in in vitro and in vivo preclinical models: in 1) cultured human muscle cells treated with IFN-β and in 2) a mouse model of myositis experimentally induced by immunization against skeletal muscle fast-type C protein.

Results: Transcriptomic analysis of patient's muscle fibers combined with topographic information (perifascicular VS endofascicular localization) revealed that a proteasome deregulation predominant in the perifascicular fibers is a hallmark of DM. The integration of 3 computer databases of drug repositioning allowed the identification of 9 molecules predicted to reverse the pathological signature of the DM perifascicular fibers. The drug with the highest therapeutic potential was a proteasome inhibitor (Ixazomib). A second proteasome inhibitor (MG-132) was also identified. 2 drugs already used for DM (prednisolone and a JAK inhibitor) was identified with a lower therapeutic score. In the cellular model (human myotubes treated with IFN-I), ixazomib reversed the atrophy, IFN-I signature, MHC-I expression and mitochondrial dysfunctions induced by IFN-β treatment. In the mouse model of myositis, ixazomib restored muscle strength and decreased blood creatine kinase level.

Conclusion: Proteasome inhibition could be a new effective therapeutic strategy for DM.

(1) Meyer et al. 2017. IFN-β-Induced Reactive Oxygen Species and Mitochondrial Damage Contribute to Muscle Impairment and Inflammation Maintenance in Dermatomyositis. Acta Neuropathol. 134 (4): 655-66. < ! (2) Karatzas et al. 2019. An Application of Computational Drug Repurposing Based on Transcriptomic Signatures. Methods Mol Biol 1903, 149-177.


L. Debrut: None; M. Giannini: None; C. Keime: None; D. Rovito: None; B. Lannes: None; A. Charles: None; D. Metzger: None; B. Geny: None; G. Laverny: None; A. Meyer: None.