Abstract Introduction Quantitative CT analysis is an emerging diagnostic and prognostic tool in ILD that can aid therapeutic decision-making. This study explored its utility in predicting outcomes in myositis-ILD patients. Methods Myositis-ILD patients were identified through retrospective chart review with ILD findings and myositis antibody positivity (MSA and/or MAA). Lung texture analysis using AVIEW LCS + (v1.1.46.28, Coreline Soft, Co. ltd, Seoul, Korea) quantified % normal lung, honeycombing, reticulation, groundglass opacities (GGOs), consolidation, and emphysema in non-contrast CT chest images (n = 187). Three scores were derived: total ILD score (sum of all ILD patterns), fibrosis ratio (honeycombing + reticulation/total ILD), and groundglass ratio (GGO/total ILD). ROC analysis assessed predictive value for death, transplant, or ≥ 10% FVC decline. Youden’s index identified optimal cut-offs; logistic and cox regression tested associations. Results Total ILD score and fibrosis ratio predicted death or transplant (AUC 0.75, p 0.001; AUC 0.63, p = 0.02). Groundglass ratio predicted transplant-free survival (AUC 0.68, p = 0.001). ILD score 41.5% (HR 3.94, 95% CI 1.96-7.93, p 0.001), fibrosis ratio 0.52 (HR 2.40, 95% CI 1.21 -4.75, p = 0.01), and groundglass ratio 0.59 (HR 5.09, 95% CI 2.20-11.78, p 0.001) were associated with increased risk of death or transplant. Only fibrosis ratio predicted FVC decline (AUC 0.62, p = 0.04); values 0.05 were significantly associated (HR 8.72, 95% CI 1.18-68.14, p = 0.04). Conclusion Quantitative CT lung texture analysis was able to prognosticate outcomes in myositis-ILD patients. Fibrosis ratios 5% were linked to lung function decline; slowing progression may reduce mortality. Higher groundglass ratios may call for increased anti-inflammatory therapy. This abstract is funded by: None
Iancu et al. (Fri,) studied this question.