1. Using Proteome Microarray and Gene Expression Omnibus Database to Screen Tumour-Associated Antigens to Construct the Optimal Diagnostic Model of Oesophageal Squamous Cell Carcinoma.
- Author
-
Sun, G., Ye, H., Yang, Q., Zhu, J., Qiu, C., Shi, J., Dai, L., Wang, K., Zhang, J., and Wang, P.
- Subjects
- *
AUTOANTIBODIES , *MICROARRAY technology , *PROTEOMICS , *BIOINFORMATICS , *GENE expression profiling , *ENZYME-linked immunosorbent assay , *CANCER genes , *DESCRIPTIVE statistics , *TUMOR antigens , *LOGISTIC regression analysis , *SENSITIVITY & specificity (Statistics) , *RECEIVER operating characteristic curves , *SQUAMOUS cell carcinoma , *ESOPHAGEAL cancer - Abstract
Autoantibodies against tumour-associated antigens (TAAs) are promising biomarkers for early immunodiagnosis of cancers. This study was designed to screen and verify autoantibodies against TAAs in sera as diagnostic biomarkers for oesophageal squamous cell carcinoma (ESCC). The customised proteome microarray based on cancer driver genes and the Gene Expression Omnibus database were used to identify potential TAAs. The expression levels of the corresponding autoantibodies in serum samples obtained from 243 ESCC patients and 243 healthy controls were investigated by enzyme-linked immunosorbent assay (ELISA). In total, 486 serum samples were randomly divided into the training set and the validation set in the ratio of 2:1. Logistic regression analysis, recursive partition analysis and support vector machine were performed to establish different diagnostic models. Five and nine candidate TAAs were screened out by proteome microarray and bioinformatics analysis, respectively. Among these 14 anti-TAAs autoantibodies, the expression level of nine (p53, PTEN, GNA11, SRSF2, CXCL8, MMP1, MSH6, LAMC2 and SLC2A1) anti-TAAs autoantibodies in the cancer patient group was higher than that in the healthy control group based on the results from ELISA. In the three constructed models, a logistic regression model including four anti-TAA autoantibodies (p53, SLC2A1, GNA11 and MMP1) was considered to be the optimal diagnosis model. The sensitivity and specificity of the model in the training set and the validation set were 70.4%, 72.8% and 67.9%, 67.9%, respectively. The area under the receiver operating characteristic curve for detecting early patients in the training set and the validation set were 0.84 and 0.85, respectively. This approach to screen novel TAAs is feasible, and the model including four autoantibodies could pave the way for the diagnosis of ESCC. • Proteome microarray and GEO were used to identify potential TAAs. • We identified nine autoantibodies against TAAs with diagnostic value for patients with ESCC. • Logistic regression model including four autoantibodies against TAAs could be used as an optimal diagnosis model for ESCC. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF