Abstract
Finding new peptide biomarkers for stomach cancer in human sera that can be implemented into a clinically practicable prediction method for monitoring of stomach cancer. We studied the serum peptidome from two different biorepositories. We first employed a C8-reverse phase liquid chromatography approach for sample purification, followed by mass-spectrometry analysis. These were applied onto serum samples from cancer-free controls and stomach cancer patients at various clinical stages. We then created a bioinformatics analysis pipeline and identified peptide signature discriminating stomach adenocarcinoma patients from cancer-free controls. Matrix Assisted Laser Desorption/Ionization–Time of Flight (MALDI-TOF) results from 103 samples revealed 9 signature peptides; with prediction accuracy of 89% in the training set and 88% in the validation set. Three of the discriminating peptides discovered were fragments of Apolipoproteins C-I and C-III (apoC-I and C-III); we further quantified their serum levels, as well as CA19-9 and CRP, employing quantitative commercial-clinical assays in 142 samples. ApoC-I and apoC-III quantitative results correlated with the MS results. We then employed apoB-100-normalized apoC-I and apoC-III, CA19-9 and CRP levels to generate rules set for stomach cancer prediction. For training, we used sera from one repository, and for validation, we used sera from the second repository. Prediction accuracies of 88.4% and 74.4% were obtained in the training and validation sets, respectively. Serum levels of apoC-I and apoC-III combined with other clinical parameters can serve as a basis for the formulation of a diagnostic score for stomach cancer patients.
Highlights
Mortality rates of many cancers have not changed dramatically in the last 20 years [1]
Training stomach cancer patients identified by model
Upon removal of bias-contributing factors, it was shown that SELDI-TOF MS whole serum proteomic profiling with IMAC surface did not reliably detect prostate cancer [23]
Summary
Mortality rates of many cancers have not changed dramatically in the last 20 years [1]. Detection was shown to greatly improve the efficacy of cancer treatment, yet detection is often only possible after the appearance of the first clinical symptoms, which in some cancers occurs too late for successful intervention. This is largely due to the absence of specific and sensitive tests that allow early screening and monitoring of cancerous states. Despite immense efforts, only a very small number of plasma proteins have been proven to have diagnostic value [2,3,4,5] These biomarkers do not stand alone and are accompanied by other tests for monitoring and diagnosis. Most of them are not specific and sensitive enough for wide screen diagnosis [6,7]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
