What question did this study set out to answer?

To create and validate a diagnostic model using ProGRP and inflammatory biomarkers for early lung cancer detection.

April 8, 2026BMC Cancer0 citationsOpen Access

A novel diagnostic model combining ProGRP and inflammatory biomarkers for early detection of lung cancer: development and validation in treatment-naive cohorts

JMJin MaHebei Medical University SZShumin ZhuHebei Medical University XLXinyuan LiHospital of Hebei Province

Key Points

To create and validate a diagnostic model using ProGRP and inflammatory biomarkers for early lung cancer detection.
Developed a diagnostic model using serum tumor markers and inflammation-related biomarkers.
Utilized a training cohort of 110 lung cancer patients and 107 healthy controls, and a validation cohort of 44 lung cancer patients and 52 controls.
Conducted logistic regression analysis to identify independent risk factors and construct the clinical prediction model.
Assessed model discrimination and calibration using ROC curves, the Hosmer-Lemeshow test, and decision curve analysis.
Achieved an AUC of 0.972 in the training cohort with 89.2% sensitivity and 99.1% specificity.
Demonstrated an AUC of 0.897 in external validation with 72.7% sensitivity and 98.1% specificity.
Confirmed ProGRP and inflammatory markers as independently associated risk factors for lung cancer.
Indicated significant clinical utility for the model across various probability thresholds (0-80%).

Abstract

Lung cancer remains the leading cause of cancer-related mortality globally, with late-stage diagnosis contributing to poor outcomes. Non-invasive biomarkers capable of complementing existing screening modalities are urgently needed to improve early detection. This retrospective study developed and validated a diagnostic model using serum tumor markers and inflammation-related biomarkers. The training cohort included 110 treatment-naive lung cancer patients and 107 healthy controls, while an external validation cohort comprised 44 lung cancer patients and 52 controls. Compare the differences in tumor markers, cytokines, and indices derived from whole blood cell counts between the lung cancer group and the healthy control group. Logistic regression analysis was used to identify independent risk factors for lung cancer and to construct a clinical prediction model. The model’s discrimination, calibration, and clinical applicability were assessed using receiver operating characteristic (ROC) curves, the Hosmer-Lemeshow test, and decision curve analysis (DCA), among other methods. The predictive model demonstrated high diagnostic accuracy in the training cohort, with an AUC of 0.972 (sensitivity 89.2%, specificity 99.1%). External validation showed an AUC of 0.897 (sensitivity 72.7%, specificity 98.1%), confirming its good generalizability. Notably, Pro-Gastrin Releasing Peptide(ProGRP)and inflammatory markers were independently associated with the risk of lung cancer (P < 0.05). DCA analysis indicated that the model has significant clinical utility across a wide range of probability thresholds (0–80%). This study presents a high-performance, non-invasive diagnostic model integrating ProGRP and systemic inflammatory biomarkers for lung cancer detection. The combined model outperforms individual biomarkers and may enhance early lung cancer screening, particularly in settings where imaging modalities are limited. Further multi-center prospective studies are warranted to validate its clinical applicability.

Ask AI

Helpful

Bookmark

View Full Paper