Prognostic factors for long-term survival in anti-MDA5-positive dermatomyositis: a 10-year cohort study

Anti-melanoma differentiation-associated gene 5-positive dermatomyositis (anti-MDA5+ DM) is a severe subtype frequently complicated by rapidly progressive interstitial lung disease (RP-ILD), leading to high early mortality. However, factors associated with long-term survival beyond 10 years remain largely uncharacterized. This exploratory analysis aimed to identify baseline factors associated with long-term all-cause mortality in a 10-year follow-up cohort using LASSO-Cox regression.

This retrospective longitudinal study enrolled 84 inpatients diagnosed with anti-MDA5+ DM between 2010 and June 2014; surviving patients had more than 10 years of potential follow-up until March 2026. The primary endpoint was all-cause mortality. All candidate covariates, including clinical features, laboratory parameters, HRCT score, and initial treatment variables, were entered simultaneously into LASSO-penalized Cox regression with 10-fold cross-validation, and retained variables were directly entered into a multivariable Cox model. The proportional hazards assumption was tested, and bootstrap resampling was performed to assess selection stability. Survival was illustrated using Kaplan-Meier curves with log-rank tests.

Among the 84 patients (27 males and 57 females), 41 (48.8%) were alive at the final follow-up and 43 (51.2%) had died. As a descriptive observation, 25 surviving patients (29.8% of the total cohort) had maintained sustained drug-free remission for over 2 years. Four baseline factors were independently associated with long-term mortality. Older age (HR = 1.05 per 1-year increase, P < 0.001), fever on admission (HR = 2.08, P = 0.036), and higher HRCT score (HR = 1.94 per 1-point increase, P < 0.001) were associated with higher mortality, whereas concurrent arthritis was associated with lower mortality (HR = 0.45, P = 0.022). Male sex showed a trend toward higher mortality that was not statistically significant after adjustment (P = 0.086).

Older age, admission fever, higher HRCT score, and absence of arthritis - all assessable at admission - were independently associated with all-cause mortality over long-term follow-up in the full anti-MDA5+ DM cohort; landmark analysis showed only arthritis and HRCT score remained associated with mortality after year 1. Given the limited sample size and lack of external validation, these exploratory findings are hypothesis-generating and require confirmation in independent multicenter cohorts.