D23-10 Anti-MDA-5-positive Interstitial Lung Disease Without Skin Manifestations Is Associated With Worse Outcomes: A Multicenter Cohort Study

Abstract Rationale Anti-melanoma differentiation-associated protein-5 (MDA-5)-positive interstitial lung disease (ILD) often presents with skin manifestations and rapidly progressive ILD, with high short-term mortality. However, the prognosis of MDA-5-positive ILD without a characteristic rash remains uncertain. We aimed to evaluate the association between the presence of rash and clinical outcomes in MDA-5-positive ILD. Methods We conducted a retrospective multicenter cohort study using ILD registries from three tertiary centers and included patients with MDA-5-positive ILD diagnosed between January 2015 and March 2025. The exposure of interest was a typical dermatomyositis rash at baseline, defined as any of the following: heliotrope rash, Gottron’s papules/sign, shawl or V-neck sign, or mechanic’s hands, as described in any doctor’s visit (dermatology, rheumatology, etc). Patients were classified as “rash present” if any feature was observed and “rash absent” otherwise. The outcome was a composite of first ILD hospitalization, lung transplantation, or all-cause mortality. Multivariable Cox proportional hazards models were fit to estimate hazard ratios (HRs), adjusted for age, sex, anti-Ro52 antibody, and baseline forced vital capacity. Results A total of 82 patients with MDA-5-positive ILD were included. Median (Q1, Q3) age was 65 (55, 72) years and 54% were female. A typical dermatomyositis rash was present in 31 (38%) and absent in 51 (62%) cases. Over a median follow-up of 48.9 months, component outcomes occurred as follows (rash present vs absent): ILD hospitalization in 6/31 (19%) vs 13/51 (25%); lung transplantation in 1/31 (3%) vs 2/51 (4%); and death in 0/31 (0%) vs 18/51 (35%). In multivariable analysis, absence of rash was associated with a higher risk of the composite outcome (adjusted HR 3.58, 95% confidence interval 1.09-11.8, P=0.037). Our findings were consistent in sensitivity analyses with further covariate adjustment (smoking history, DLCO, UIP pattern, baseline immunosuppressant and antifibrotic use: adjusted HR 4.61, 95% CI 1.12-19.0; P=0.036) and when redefining the presence of rash using an expanded rash definition to include ulcerations and other (adjusted HR 8.64, 95% CI 0.98-76.4; P=0.052). Conclusion Among patients with MDA-5-positive ILD, the absence of a typical dermatomyositis rash may be associated with a worse prognosis. This abstract is funded by: Sidney and Deanna Wolk Family Foundation