The following is a summary of “Radiomics based on HRCT can predict RP-ILD and mortality in anti-MDA5 + dermatomyositis patients: a multi-center retrospective study,” published in the June 2024 issue of Pulmonology by He et al.

This study aimed to evaluate the utility of high-resolution computed tomography (HRCT)-based radiomics in predicting rapidly progressive interstitial lung disease (RP-ILD) and mortality in patients with anti-MDA5 positive dermatomyositis-related interstitial lung disease (anti-MDA5+ DM-ILD).

Between August 2014 and March 2022, researchers retrospectively and consecutively enrolled 160 patients from Institution 1, randomly divided into a training dataset (n = 119) and an internal validation dataset (n = 41). Additionally, 29 patients from Institution 2 were enrolled as an external validation dataset. Radiomics features were extracted from four areas of HRCT to generate four distinct risk scores. The study group developed a nomogram integrating selected clinico-radiologic variables with the most discriminative radiomics model’s risk score. The models’ predictive performance for RP-ILD were assessed using the area under the receiver operating characteristic curves (AUC), calibration curves, and decision curves. Survival analysis was conducted using Kaplan-Meier curves, the Mantel-Haenszel test, and Cox regression.

Over a median follow-up period of 31.6 months (interquartile range: 12.9–49.1 months), 24 patients were lost to follow-up, and 46 patients (27.9%, 46/165) died. The risk score based on bilateral lung radiomics features demonstrated the best performance, achieving AUCs of 0.869 and 0.905 in the internal and external validation datasets, respectively. The nomogram, which integrated clinico-radiologic variables and the Risk-score, outperformed the clinico-radiologic model and the standalone Risk-score, with AUCs of 0.882 and 0.916 in the internal and external validation datasets, respectively. Patients were classified into low- and high-risk groups based on the nomogram, with the high-risk group showing a significantly higher risk of mortality compared to the low-risk group in both Institution 1 (hazard ratio [HR] = 4.117) and Institution 2 cohorts (HR = 7.515).

In patients with anti-MDA5+ DM-ILD, the nomogram, which primarily relies on HRCT-based radiomics, effectively predicts RP-ILD and is an independent predictor of mortality. This tool offers a robust approach for risk stratification and prognosis in this patient population, highlighting the potential of radiomics in enhancing clinical decision-making and patient management.

Source: respiratory-research.biomedcentral.com/articles/10.1186/s12931-024-02843-w