Objective: To characterize morphological and molecular underpinnings of polymyositis with mitochondrial pathology (PM-Mito) in comparison to sporadic inclusion body myositis (IBM) and to define common and distinct pathophysiological features with a focus on interferon-associated inflammation and T-cell-response.

Methods: In this cross-sectional study, skeletal muscle biopsy samples as well as clinical and laboratory data from PM-Mito and IBM patients were analyzed at Charité university hospital in Berlin, Germany. All available PM-Mito biopsy samples and an equal number of randomly selected IBM biopsy samples were included in the study as well as randomly selected non-diseased controls (NDC). Biopsy samples were studied by histopathology, immunohistochemistry, and quantitative PCR and compared to biopsies derived from NDC. Primary outcomes included cell counts for immunohistochemistry, and gene expression (fold-change values compared to NDCs) for quantitative PCR.

Results: Twenty-five skeletal muscle biopsy samples of patients with PM-Mito and IBM were included in the study and compared to five biopsy samples from non-diseased controls. PM-Mito and IBM qualitatively harbored a strikingly similar molecular signature and shared important histopathological features. Expression of interferon-induced GBP6 and T-cell function-related KLRG1 distinguished IBM from PM-Mito biopsies with IBM patients showing significantly higher expression of GBP6 and KLRG1. Cryptic exon expression was detected in both patient groups with IBM patients showing higher expression levels. Skeletal muscle biopsies from IBM patients showed significantly more GBP6+ cells and KLRG1+ lymphocytes in comparison to biopsies from PM-Mito patients. CD45+, CD68+, CD57+, PD1+ and CD8+ cytotoxic T-cells were also significantly more abundant in IBM. Clinically, PM-Mito patients presented with a spectrum of muscle-related symptoms including myalgia, proximal paraparesis, proximal tetraparesis and incomplete IBM-like patterns. 13 out of 14 (93%) PM-Mito patients for whom clinical follow-up was available later developed clinically defined IBM. Notably, two follow-up biopsies obtained 5 and 7 years after the first ones were available in this cohort, both showing histopathological progress to net IBM including GBP6 and KLRG1 upregulation.

Conclusions: Our combined data suggest that specific interferon-mediated inflammation plays a key role in both IBM and PM-Mito. GBP6 was identified as a new molecule of type II interferon-induced inflammation distinguishing IBM from PM-Mito. Skeletal muscles from both groups harbor dysfunctional T-cells of similar type, albeit in different quantity. T-cell senescence exemplified by KLRG1-positivity does not play a significant role in PM-Mito. Based on these findings, we propose to include PM-Mito in the spectrum of IBM (IBM-spectrum disease, IBM-SD) as a possible early form of this disease. The establishment of IBM-SD as a larger entity could potentially have a significant impact on the design of trials and therapeutic interventions.