Cargando…

Evaluation of the diagnostic value of serum-based proteomics for colorectal cancer

BACKGROUND: Colorectal cancer (CRC) is a highly malignant cancer with a high incidence and mortality in China. It is urgent to find a diagnostic marker with higher sensitivity and specificity than the traditional approaches for CRC diagnosis. AIM: To provide new ideas for the diagnosis of CRC based...

Descripción completa

Detalles Bibliográficos
Autores principales: Wang, Hui-Juan, Xie, Yi-Bin, Zhang, Peng-Jun, Jiang, Tao
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Baishideng Publishing Group Inc 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9412932/
https://www.ncbi.nlm.nih.gov/pubmed/36160749
http://dx.doi.org/10.4251/wjgo.v14.i8.1562
Descripción
Sumario:BACKGROUND: Colorectal cancer (CRC) is a highly malignant cancer with a high incidence and mortality in China. It is urgent to find a diagnostic marker with higher sensitivity and specificity than the traditional approaches for CRC diagnosis. AIM: To provide new ideas for the diagnosis of CRC based on serum proteomics. METHODS: Specimens from 83 healthy people, 62 colon polyp (CRP) patients, and 101 CRC patients were analyzed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. The diagnostic value of the profiles of differentially expressed proteins was then analyzed. RESULTS: Compared with the healthy control group, CRC patients had elevated expression of 5 proteins and reduced expression of 14 proteins. The area under the curve (AUC) for a differentially expressed protein with a mass-to-charge ratio of 2022.34 was the largest; the AUC was 0.843, which was higher than the AUC of 0.717 observed with carcinoembryonic antigen (CEA), and the sensitivity and specificity of this identified marker were 75.3% and 79.5%, respectively. After cross-validation, the accuracy of diagnosis using levels of this differentially expressed protein was 82.37%. Compared with the CRP group, the expression of 3 proteins in the serum of CRC patients was elevated and 11 proteins were expressed at reduced levels. Proteins possessing mass-to-charge ratio values of 2899.38 and 877.3 were selected to establish a classification tree model. The results showed that the accuracy of CRC diagnosis was 89.5%, the accuracy of CRP diagnosis was 81.6%, and the overall accuracy of this approach was 86.3%. The overall sensitivity and specificity of diagnosis using the proteomics approach were 81.8% and 66.75%, respectively. The sensitivities and specificities of diagnoses based on CEA and carbohydrate antigen 19-9 expression were 55.6% and 91.3% and 65.4% and 65.2%, respectively. CONCLUSION: We demonstrated that serum proteomics may be helpful for the detection of CRC, and it may assist clinical practice for CRC diagnosis.