Background: Myositis is a heterogeneous group of muscular auto-immune diseases with clinical and pathological criteria that allow the classification of patients into different sub-groups. Inclusion body myositis is the most frequent myositis above fifty years of age. Diagnosing inclusion body myositis requires expertise and is challenging. Little is known concerning the pathogenic mechanisms of this disease in which conventional suppressive-immune therapies are inefficacious.Objectives: Our aim was to deepen our understanding of the immune mechanisms involved in inclusion body myositis and identify specific biomarkers.Methods: Using a panel of thirty-six markers and mass cytometry, we performed deep immune profiling of peripheral blood cells from inclusion body myositis patients and healthy donors, divided into two cohorts: test and validation cohorts. Potential biomarkers were compared to myositis controls (anti-Jo1-, anti-3-hydroxyl-3-methylglutaryl CoA reductase-, and anti-signal recognition particle-positive patients).Results: Unsupervised analyses revealed substantial changes only within CD8+ cells. We observed an increase in the frequency of CD8+ cells that expressed high levels of T-bet, and containing mainly both effector and terminally differentiated memory cells. The senescent marker CD57 was overexpressed in CD8+T-bet+ cells of inclusion body myositis patients. As expected, senescent CD8+T-bet+ CD57+ cells of both patients and healthy donors were CD28nullCD27nullCD127null. Surprisingly, non-senescent CD8+T-bet+ CD57- cells in inclusion body myositis patients expressed lower levels of CD28, CD27, and CD127, and expressed higher levels of CD38 and HLA-DR compared to healthy donors. Using classification and regression trees alongside receiver operating characteristics curves, we identified and validated a frequency of CD8+T-bet+ cells >51.5% as a diagnostic biomarker specific to inclusion body myositis, compared to myositis control patients, with a sensitivity of 94.4%, a specificity of 88.5%, and an area under the curve of 0.97.Conclusion: Using a panel of thirty-six markers by mass cytometry, we identify an activated cell population (CD8+T-bet+ CD57- CD28lowCD27lowCD127low CD38+ HLA-DR+) which could play a role in the physiopathology of inclusion body myositis, and identify CD8+T-bet+ cells as a predominant biomarker of this disease.