Peripheral lymphocyte count defines the clinical phenotypes and prognosis in patients with anti‐MDA5‐positive dermatomyositis

Abstract Objective To explore the role of peripheral lymphocyte count in phenotyping and prognosis prediction in dermatomyositis (DM) patients with anti‐MDA5 antibodies. Methods In total, 1669 patients with idiopathic inflammatory myopathy (IIM) were retrospectively enrolled. Clinical characteristics and prognosis of patients with anti‐MDA5+ DM were analyzed in association with peripheral lymphocyte counts and clusters determined by unsupervised machine learning. Results The peripheral lymphocyte count was significantly lower in the anti‐MDA5+ DM group ( N = 421) than in the other IIM serotype groups. The anti‐MDA5+ DM patients were divided into three groups; the severe lymphopenia group had skin ulcers and rapidly progressive interstitial lung disease (RP‐ILD); patients with a normal lymphocyte count had a younger age of onset, more frequent arthritis, and normal serum ferritin levels, whereas mild lymphopenia group showed a moderate increase of serum ferritin and intermediate incidence of RP‐ILD. Survival analysis revealed that the 3‐ and 6‐month mortality rates were significantly higher in the severe lymphopenia group (29.0% and 42.1%, respectively) than in the mild lymphopenia group and normal lymphocyte count group ( p value <0.001). Consistently, unsupervised machine learning identified three similar groups; the arthritis cluster shows the highest lymphocyte counts and best prognosis; the RP‐ILD cluster presents the lowest peripheral lymphocyte, high incidence of RP‐ILD, and poor prognosis; the typical DM rash cluster had a moderate peripheral lymphocyte count and an intermediate prognosis. Conclusions Lymphopenia is a unique manifestation of anti‐MDA5+ DM. Peripheral lymphocyte count can define clinical phenotypes and predict prognosis in anti‐MDA5+ DM.