Nomogram to predict dermatomyositis prognosis: a population-based study of 457 cases

Dermatomyositis (DM) is a systemic autoimmune disease, which typically affects the striated muscle with a variable involvement of the skin and other organs. Clinically amyopathic DM (CADM) is a combination of hypomyopathic DM (HDM) and amyopathic DM (ADM), with a characteristic of skin-predominant lesions. To date, large-scale studies on the prognostic factors of DM/CADM have been limited. The aim of this study is to evaluate the prognostic values of clinical manifestations in DM/CADM and to develop a prognostic nomogram for DM/CADM. A development cohort (n=239), an internal validation cohort (n=128) and an external validation cohort (n=90) were included in this study. Overall survival (OS) was estimated by the Kaplan-Meier method. Univariate and multivariate Cox proportional hazards regression analyses were performed. Cox proportional hazards model and forward stepwise selection with the Akaike information criterion were used for multivariate analysis of prognostic factors. The concordance index (C-index) and calibration curve were calculated to evaluate the predictive accuracy of the proposed nomogram. Rapidly progressive interstitial lung disease (RP-ILD) and erythrocyte sedimentation rate (ESR) were identified as risk independent prognostic factors, with antinuclear antibodies (ANA) was identified as protective independent prognostic factors, for DM/CADM. A prognostic nomogram was formulated based on these three predictors. The C-index of the proposed nomogram in the development cohort was 0.874 (95%CI, 0.819-0.929). The predictive accuracy of the proposed nomogram was further validated in the internal validation cohort, with a C-index of 0.799 (95%CI, 0.681-0.917). Furthermore, the C-index was 0.864 (95%CI, 0.699-1.000) in the external validation cohort, indicating a good calibration ability. This proposed nomogram showed a promising predictive accuracy on the prognosis of DM/CADM. RP-ILD, ANA and ESR are prognostic factors for DM/CADM. The proposed nomogram based on these three factors could accurately predict the 10-year OS probabilities of patients with DM/CADM.