الفهرس 
COLORECTAL CANCEROnly 14 pages are availabe for public view
 |  | from | 149 |  |  |
| | | from | 149 | | |
Abstract
Colorectal cancer (CRC) is among the most common malignancies and remains a leading cause of cancer-related morbidity and mortality. It is characterized by high recurrence rate and most of the recurrent cases may progress to a higher stage. So, early diagnosis is the most appropriate tool to reduce disease-related mortality. The gold standard for diagnosing CRC is colonoscopy. It provides direct visualization and localization of the tumor and the mean for biopsying abnormal areas. However, this procedure is invasive, relatively expensive, and often a source of patient anxiety.
Serum tumor markers as carcinoembryonic antigen (CEA) has been used for diagnosis. However, it has shown poor sensitivity and specificity, in addition to failure in judging the effectiveness in surgical resection of the tumor. These facts have driven the search for other non-invasive biomarkers that can improve the diagnosis.
Proteomics is an advancing technique in CRC. Studies have shown that this technique may provide a novel non-invasive means of diagnosing CRC, and it may have additional value as a prognostic tool. There is a great interest in the low molecular weight protein. The panels of peptidome markers might be more sensitive and specific than conventionally biomarker approaches, which may provide a novel means of diagnosing cancer and other diseases.
The aim of the present study was to determine serum proteomic pattern(s) that is (are) specific for primary colorectal cancer and to test its(their) performance as diagnostic utility in these patients.
The present study is a case control comparative analysis between CRC patients and healthy controls where WCX-MBs, Ultraflextreme MALDI-TOF, and ClinProt Tools software 3.0 were used to study the differential expression of peptides and proteins in the serum of both groups. This study included 70 patients with primary colorectal cancer and 70 apparently healthy subjects, serving as a control group. CRC patients were classified according to TNM staging system into 2 subgroups; subgroup 1a including CRC patients with stages 0 and 1 and subgroup 1b including CRC patients with stage II. For proteomic analysis, all patients and controls were randomly divided into model generation and model validation groups. Each group included 35 CRC patients and 35 healthy control subjects.
On comparing the serum proteome profile of CRC patients to that of healthy controls, our results revealed 98 peaks were common between the two groups in the mass range from 900 to 20000 Da. Out of these 98 peaks, 65 peaks were identified by the ClinProt software to show statistically different intensities among the model generation group. Forty-seven peaks of them were up-regulated and 18 peaks were down-regulated peptides
By using the classification models; GA, SNN and QC; to discriminate CRC patients from healthy controls. Our results revealed five discriminating peaks by GA model analysis, four of them were down-regulated, and the last one was up-regulated. As regard SNN analysis, our results revealed three discriminating peaks. All of them were up-regulated. By QC analysis, four discriminating peaks were detected. Two of them were up-regulated, the others were down-regulated. The three models revealed similar recognition capability (100%).While, SNN and QC models revealed higher internal validation than GA model (97.6%, respectively) Moreover, the QC model revealed the best combination of sensitivity and specificity (100% &94%, respectively).
To the best of our knowledge, our study was the first one to apply proteomic profiling for CRC tumor stages. 96 signals have been detected common between the two subgroups in the mass range (900-20000 Da and 55 were identified with statistically different area. The GA classification model was the only one applied to discriminate CRC patients with stages 0&I from those with stage II. It revealed four discriminating peaks with a recognition capacity of 100% and an internal-validation of 93.42%.
In conclusion, the magnetic beads-based separation technique with MALDI-TOF proteomic profiling followed by data analysis by classification models represents a new promising tool for early diagnosis of CRC patients with high sensitivity and specificity exceeding that of CEA. In addition to its usefulness in applying proteomic profile for CRC tumor stages.