Truace
Subscribe

Plasma proteomics and machine learning deliver non-invasive distinction between fibrotic hypersensitivity pneumonitis and idiopathic pulmonary fibrosis

Hypersensitivity pneumonitis (HP) manifests as fibrotic (FHP) and non-fibrotic (NFHP) phenotypes. Clinically distinguishing FHP from idiopathic pulmonary fibrosis (IPF) remains challenging owing to phenotypic overlap, despite divergent management protocols. This investigation sought to develop a plasma proteomics-based framework for differential diagnosis between these entities. A total of 119 subjects were enrolled from the Chinese Interstitial Lung Disease (ILD) National Cohort and the PORTRAY IPF Cohort betwe…

By the Truvace desk2 min read1 sourcePeer-reviewedTRV-2026-0264· cite & verify
Plasma proteomics and machine learning deliver non-invasive distinction between fibrotic hypersensitivity pneumonitis and idiopathic pulmonary fibrosis
"OhioHealth - Doctors Hospital 1" by Sixflashphoto is licensed under CC BY-SA 4.0. To view a copy of this license, visit https://creativecommons.org/licenses/by-sa/4.0/.

In brief

Between July 2018 and June 2022, investigators enrolled 119 subjects across healthy controls, non-fibrotic HP, fibrotic HP, and IPF cohorts and performed plasma proteomic profiling with WGCNA. They identified 813 proteins, noted enrichment of glycolysis/gluconeogenesis and pyruvate metabolism in FHP, and distilled nine differential proteins to a six-protein signature that was used to train seven machine learning classifiers.

The work matters because FHP and IPF overlap clinically but require divergent management, and a blood-based classifier could reduce reliance on invasive or ambiguous workups. By the July 2026 publication date, the result was a validated proof-of-concept with 71.4% accuracy on an independent test set, not yet a deployed clinical test, with generalizability beyond the two Chinese cohorts still unproven.

Main points

  1. Study enrolled 119 subjects: 32 healthy controls, 31 non-fibrotic HP, 28 fibrotic HP, and 28 IPF from Chinese ILD National Cohort and PORTRAY IPF Cohort.
  2. Quantitative plasma proteomics identified 813 proteins and WGCNA highlighted glycolysis/gluconeogenesis and pyruvate metabolism pathways.
  3. LASSO and recursive feature elimination reduced nine differentially expressed proteins to six-protein signature: H2BC12, SHBG, APCS, PTPRG, IGHV1-58, and GAPDH.
  4. Seven machine learning algorithms were compared to construct diagnostic models, with SVM selected as optimal on independent test set.

The gain

A six-protein plasma signature tested with SVM distinguished fibrotic hypersensitivity pneumonitis from idiopathic pulmonary fibrosis on an independent test set with 71.4% accuracy, offering a non-invasive diagnostic aid.

The rundown

Researchers profiled plasma from 119 participants enrolled between July 2018 and June 2022 and performed quantitative proteomics, WGCNA, and bioinformatics to map disease-associated proteins and modules.

From nine differentially expressed proteins, a six-protein panel was derived and fed into seven machine learning algorithms; the SVM model delivered the best independent test performance at 71.4% accuracy for FHP versus IPF discrimination.

Sources

  1. Peer-reviewedJournal of Translational Medicine2026-07-17

The debate