Proteomic and serological markers for diagnosing cardia gastric cancer and precursor lesions in a Chinese population

Q: What diagnostic indicators were evaluated for cardia gastric cancer in this study?

The study assessed serum pepsinogen, gastrin-17, Helicobacter pylori antibodies, and plasma proteomic profiling to evaluate diagnostic value for CGC and precursor lesions.

Q: How was the study cohort structured?

It used a case-control design with biopsy-confirmed CGC cases, low-grade and high-grade intraepithelial neoplasia cases, and healthy controls, matched for age and sex from high-incidence areas in China.

Q: Which machine-learning approach performed best and what biomarkers were included in the final model?

An explainable XGBoost model showed the best discriminative ability. After feature reduction, it incorporated five protein biomarkers: CDHR2, ICAM4, PTPRM, CDC27, and FLT1.

Sep 22, 2026

Upload by Skyler

Cardia gastric cancer (CGC) is common in East Asia, requiring noninvasive and cost-effective screening. This case-control study evaluated serum pepsinogen, gastrin-17, Helicobacter pylori antibodies, and plasma proteomic profiling for CGC and precursor lesions. Biopsy-confirmed cases (CGC, n=60; low-grade, n=60; high-grade, n=64) and matched healthy controls (n=120) were analyzed using ELISA, Western blot, and Olink technology. Machine-learning models including XGBoost identified biomarkers and achieved strong discrimination versus traditional serology.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Background

Need for noninvasive screening for CGC

Role of serological markers

Proteomic profiling and assay platforms

Study design and methods

Case-control cohorts and matching

Serological measurements

Proteomic profiling using PEA/Olink

Feature selection and machine-learning modeling

Results

Associations of serological markers with risk

Best-performing predictive model

Final explainable model and biomarker panel

Conclusion

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN Proteomic and serological markers for diagnosing cardia gastric cancer and precursor lesions in a Chinese population Jiqing Li1,7, Wei Zhao1,7, Jia Yang2, Peipei Lu3,7, Heming Sun4, Zhenhong Zhang5 & Jianhua Gu1,6􀀍 Cardia gastric cancer (CGC) is prevalent in East Asia, and noninvasive, cost-effective screening methods are needed. This study investigated the diagnostic value of serum pepsinogen (PG), gastrin-17 (G-17), Helicobacter pylori (H. pylori) antibodies, and proteomic profiling for CGC and precancerous lesions. We conducted a case-control study involving biopsy-confirmed patients with CGC (n = 60), low-grade intraepithelial neoplasia (CLGD, n =60), high-grade intraepithelial neoplasia (CHGD, n = 64), and healthy controls (n = 120) matched for age and sex from high-incidence areas in China. Serological markers including PGI, PGII, G-17, and H. pylori were measured using ELISA and Western blot, while plasma protein markers were assessed using Olink® technology. The VSOLassoBag algorithm and nine machine learning (ML) algorithms were employed to identify crucial features and construct predictive models. Various evaluation metrics, including the area under the receiver-operatingcharacteristic curve (AUC), were utilized to compare predictive performance. Elevated PGII levels, decreased PGR, and H. pylori infection were significantly associated with an increased risk ofCGC and precancerous lesions (Pfor trend < 0.05). The eXtreme Gradient Boosting (XGBoost) model performed best in discriminative ability among the 9 ML models. Following feature reduction based on predictive performance, a final explainable XGBoost model was developed, incorporating five protein biomarkers (CDHR2, ICAM4, PTPRM, CDC27, and FLT1). This model exhibited excellent performance in distinguishing individuals with CGC and precancerous lesions from healthy controls (AUC = 0.931 for CGC, 0.867 for CHGD, and 0.763 for CLGD), surpassing the traditional serological marker-based model. This study underscores the diagnostic potential of serological markers and proteomic profiling in the detection ofCGC. Further validation and exploration of combined biomarker approaches are warranted to enhance early diagnosis and improve outcomes in high-risk populations. Keywords Cardia gastric cancer, Precancerous lesions, Serological markers, Proteomics, Machine learning Abbreviations CGC Cardia gastric cancer CHGD High-grade intraepithelial neoplasia CLGD Low-grade intraepithelial neoplasia t-SNE t-distributed stochastic neighbor embedding UMAP Uniform manifold approximation and projection Cardia gastric cancer (CGC) accounted for 18% of global gastric cancer (GC) cases in 2018, with approximately 70% of these cases occurring in East Asia, particularly China1. While endoscopy remains the primary 1Qilu Hospital of Shandong University, 107 Wenhua West Road, Lixia District, Jinan, Shandong Province, China. 2Liaocheng People’s Hospital, Liaocheng, China. 3Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, China. 4Department of Computer and Information Science and Engineering, University of Florida, Gainesville, USA. 5Medical Imaging Center, Qilu Hospital of Shandong University Dezhou Hospital, Jinan, China. 6Office of National Central Cancer Registry, National Clinical Research Center for Cancer/Cancer Hospital, National Cancer Center, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. 7Jiqing Li, Wei Zhao and Peipei Lu contributed equally to this work. 􀀍 [email: jianhuagu@yeah.net](email: jianhuagu@yeah.net) [www. nature.com/scientificreports/](www. nature.com/scientificreports/) diagnostic method in high-incidence regions, its invasive nature and high cost limit its broader application2. This underscores the urgent need for noninvasive and cost-effective alternativ

FAQs

What diagnostic indicators were evaluated for cardia gastric cancer in this study?

The study assessed serum pepsinogen, gastrin-17, Helicobacter pylori antibodies, and plasma proteomic profiling to evaluate diagnostic value for CGC and precursor lesions.

How was the study cohort structured?

It used a case-control design with biopsy-confirmed CGC cases, low-grade and high-grade intraepithelial neoplasia cases, and healthy controls, matched for age and sex from high-incidence areas in China.

Which machine-learning approach performed best and what biomarkers were included in the final model?

An explainable XGBoost model showed the best discriminative ability. After feature reduction, it incorporated five protein biomarkers: CDHR2, ICAM4, PTPRM, CDC27, and FLT1.

Sep 22, 2026

Upload by Skyler