Risk Factors and Predictive Model for Dermatomyositis Associated with Rapidly Progressive Interstitial Lung Disease

Abstract

Background Rapidly progressive interstitial lung disease (RP-ILD) is a significant complication that determines the prognosis of dermatomyositis (DM). Early RP-ILD diagnosis can improve screening and diagnostic efficiency and provide meaningful guidance to carry out early and aggressive treatment. Methods A retrospective screening of 284 patients with DM was performed. Clinical and laboratory characteristics of the patients were recorded. The risk factors of RP-ILD in DM patients were screened by logistic regression (LR) and machine learning methods, and the prediction models of RP-ILD were developed by machine learning methods, namely least absolute shrinkage and selection operator (LASSO), random forest (RF), and extreme gradient boosting (XGBoost). Results According to the result of univariate LR, disease duration is a protective factor for RP-ILD, and ESR, CRP, anti-Ro-52 antibody and anti-MDA5 antibody are risk factors for RP-ILD. The top 10 important variables of the 3 machine learning models were intersected to obtain common important variables, and there were 5 common important variables, namely disease duration, LDH, CRP, anti-Ro-52 antibody and anti-MDA5 antibody. The AUC of LASSO, RF and XGBoost test set were 0.661, 0.667 and 0.867, respectively. We further validated the performance of these three models on validation set, and the results showed that, the AUC of LASSO, RF and XGBoost were 0.764, 0.727 and 0.909, respectively. Based on the results of the models, XGBoost is the optimal model in this study. Conclusion Disease duration, LDH, CRP, anti-Ro-52 antibody and anti-MDA5 antibody are vital risk factors for RP-ILD in DM. The prediction model constructed using XGBoost can be used for risk identification and early intervention in DM patients with RP-ILD and practical application.

Association of anti-Ro52 autoantibody with interstitial lung disease in autoimmune diseases: a systematic review and meta-analysis

“…In addition to its association with ILD, several studies have linked anti-Ro52 with RP-ILD, a progressive subtype of ILD with a high mortality rate and resistance to treatment 12. We showed that anti-Ro52-positive patients had 2.69 times higher odds of having RP-ILD than anti-Ro52-negative subjects.…”

“…A higher OR shows that patients with positive anti-Ro52 were more likely to have simultaneous RP-ILD. 12 13 16 27 34 35 40 47 77–79…”

“…Beyond acting as a diagnostic marker for pSS, IIM patients with anti-Ro52 positivity might have a higher prevalence of ILD 11. Additionally, it has been suggested that anti-Ro52 may be associated with a more severe phenotype of ILD in IIM patients called rapidly progressive ILD (RP-ILD), but conclusive evidence is lacking 12…”

Risk Prediction for Rapidly Progressive Interstitial Lung Disease in Anti-MDA5-Positive Dermatomyositis: The CRAFT Model

“…A single indicator fails to accurately predict a particular outcome, owing to limited predictive value; however, a prediction model incorporating several relevant risk factors may serve as a useful tool to enhance predictive power. To date, many prediction models associated with MDA5+ DM or RP-ILD have been established [ 6 , 17 – 21 ]. In their study, So et al proposed a prediction model in MDA5 + DM patients consisting of fever, lactate dehydrogenase, age, and neutrophil-to-lymphocyte ratio to stratify the risk of the RP-ILD complication [ 6 ].…”

Three cases of Anti-MDA5 positive dermatomyositis with interstitial lung disease and pneumomediastium

“…Cutaneous ulcers (ulcerated or inverted Gottron papules) are also associated with an increased risk of this complication [3]. Patients with positive anti-MDA5 and anti-Ro52 antibodies, especially in high titers as seen in case 1, have a higher risk of rapidly progressive-interstitial lung disease and a lower survival rate compared to patients with only anti-MDA5 [4][5][6][7][8]. Coexistence of these antibodies may help identify a subgroup of patients with severe disease, predict prognosis, and possibly guide risk-based therapy as previously shown in very few cases.…”