New potential biomarkers in the diagnosis of esophageal squamous cell carcinoma

Abstract

Objective: To analyse the alterations of serum proteins in cases of esophageal squamous cell carcinoma (ESCC) in order to screen and validate serum marker patterns for the diagnosis of ESCC in the high-risk populations of Xinjiang, China. Methods: The serum proteomic patterns of 188 cases, including 139 patients with ESCC (54 Uygur, 45 Kazakh and 40 Han subjects) and 49 sex- and age-matched healthy controls, were detected using the SELDI-TOF-MS (surface-enhanced laser desorption/ionization–time of flight–mass spectrometry) technology with the CM10 ProteinChip. Differences in protein peaks between patients with ESCC and controls were analysed using the Biomarker Pattern Software, and a primary diagnosis model of ESCC was developed and validated with SVM (support vector machines). This model was further evaluated by a large-scale blind test. Results: Two hundred and eighty-three protein peaks were detected within the molecular range of 0–20 kDa, among which, 140 peaks were significantly different between ESCC cases and controls (p < 0.05). A diagnostic pattern consisting of six protein peaks (m/z 5667, 5709, 5876, 5979, 6043 and 6102) was established with a sensitivity of 97.12% and a specificity of 83.87%. The large-scale blind test generated a sensitivity of 91.43% and a specificity of 88.89%. Conclusions: The differential protein peaks analysed by SELDI-TOF-MS may contain promising serum biomarkers for screening ESCC. The diagnostic model which combined only six protein peaks had a satisfactory discriminatory power. The model should be further evaluated in other populations of ESCC patients and tested against controls. The nature and function of the discriminating proteins have yet to be elucidated.

Keywords::

• ESCC
• SELDI-TOF-MS
• bioinformatics
• SVM

Acknowledgements

The project was supported by the National Natural Science Foundation (no. 30760223, 30860097), Xinjiang Scientific-technique Bureau Grant (no. XJKJT200511113), and Urumqi Scientific-technique Bureau Grant (no. Y05331002). The authors thank Dr Patrick Ducoroy, from the Clinical and Innovation Proteomic Platform, University of Burgundy and University of Franche-Comte, France, and Professor Dominique Angele Vuitton, from Clinical Immunology WHO Collaborating Center CHU Besançon, University of Franche-Comté, France for their precious advice and revision of the manuscript.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.