Integrated phenotyping identifies reproducible prognostic subgroups in idiopathic pleuroparenchymal fibroelastosis

BACKGROUND: Idiopathic pleuroparenchymal fibroelastosis (PPFE) is a rare interstitial lung disease characterized by upper-lobe-predominant fibrosis and heterogeneous prognosis. An integrated phenotype classification incorporating clinical, radiological and biological markers has not been established. METHODS: We conducted a retrospective observational study of 32 patients with idiopathic PPFE in an exploratory cohort and 37 patients in a separate validation cohort. Unsupervised k-means clustering was performed using % forced vital capacity (%FVC), residual volume/total lung capacity (RV/TLC), serum KL-6 levels, and extent of fibrotic lesions on quantitative computed tomography. Leukocyte telomere length (LTL) was measured using quantitative PCR and age-adjusted. Survival was assessed using Kaplan-Meier and Cox regression models. A simplified prognostic score was constructed using predefined clinical thresholds, and its discriminative performance was evaluated using the concordance index (C-index) with bootstrap-derived confidence intervals. Sensitivity analyses using multiple imputation were performed to assess robustness. RESULTS: Two distinct phenotypic clusters were identified in the exploratory cohort (Cluster 1, n = 15; Cluster 2, n = 17). Cluster 2 showed a severe fibrotic phenotype characterized by lower %FVC, higher RV/TLC, higher KL-6 levels, and greater fibrotic extent, and was associated with significantly worse survival (adjusted hazard ratio, 8.63; 95% CI, 1.06-70.4). Similar phenotypic separation and prognostic differences were reproduced in the validation cohort. The four-variable scoring system reproducibly stratified patients into three risk categories with stepwise prognostic separation in both cohorts. The model demonstrated moderate-to-good discriminative ability, with a C-index of 0.74 (95% CI, 0.59-0.88) in the exploratory cohort and 0.84 (95% CI, 0.74-0.92) in the validation cohort. In contrast, age-adjusted LTL in PPFE was not significantly shorter than in healthy controls and was not associated with survival. CONCLUSIONS: Integrated phenotyping using quantitative CT and clinical indicators identifies reproducible prognostic subgroups in idiopathic PPFE, whereas telomere shortening does not appear to play a major prognostic role. A simple scoring system may provide clinically applicable risk stratification.