Faculty, Staff and Student Publications

Language

English

Publication Date

12-8-2025

Journal

Nature Communications

DOI

10.1038/s41467-025-65930-6

PMID

41360981

PMCID

PMC12686440

PubMedCentral® Posted Date

12-8-2025

PubMedCentral® Full Text Version

Post-print

Abstract

Early detection of lung cancer is crucial for improving patient outcomes. However, accurately diagnosing invasive pulmonary nodules and predicting tumor invasiveness remain major clinical challenges. Given the established role of immune dysfunction in cancer development, we hypothesize that peripheral immune profiling could provide a strategy for managing pulmonary nodules. In this multi-center, prospective study, we combine peripheral immune profiling via mass cytometry with machine learning algorithms to develop an integrated pulmonary nodule management platform. This platform accurately distinguishes invasive from non-invasive pulmonary nodules (AUC = 0.952), outperforming established clinical and radiomics-based models. Furthermore, it effectively predicts tumor invasiveness, differentiating minimally invasive from invasive adenocarcinoma (AUC = 0.949), thereby offering valuable guidance for surgical decision-making. In conclusion, the platform demonstrates substantial clinical utility and holds significant promise as a precision tool for future management of pulmonary nodules.

Keywords

Humans, Lung Neoplasms, Prospective Studies, Male, Female, Single-Cell Analysis, Multiple Pulmonary Nodules, Middle Aged, Aged, Solitary Pulmonary Nodule, Machine Learning, Neoplasm Invasiveness, Lung cancer, Cancer screening, Predictive markers

Published Open-Access

yes

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.