DESCRIPTION:

List of 40 discriminating genes between the UC-Ca and UC-NonCa groups.

LABEL:

UC-Ca vs UC-NonCa groups

SCORE TYPE:

P-Value

THRESHOLD:

<= 0.5

GENES IN THRESHOLD:

31

DATE ADDED:

2010-06-24

DATE UPDATED:

2024-10-22

SPECIES:

AUTHORS:

Watanabe T, Kobunai T, Toda E, Kanazawa T, Kazama Y, Tanaka J, Tanaka T, Yamamoto Y, Hata K, Kojima T, Yokoyama T, Konishi T, Okayama Y, Sugimoto Y, Oka T, Sasaki S, Ajioka Y, Muto T, Nagawa H

TITLE:

Gene expression signature and the prediction of ulcerative colitis-associated colorectal cancer by DNA microarray.

JOURNAL:

Clinical cancer research : an official journal of the American Association for Cancer Research Jan 2007, Vol 13, pp. 415-20

ABSTRACT:

PURPOSE: Ulcerative colitis (UC) is associated with a high risk of colorectal cancer. To identify genes that could predict the development of cancer in UC, we conducted a DNA microarray analysis using nonneoplastic rectal mucosa of UC patients. EXPERIMENTAL DESIGN: Gene expression in nonneoplastic mucosa of 53 UC patients were examined. Gene expression profiles were examined using human Genome U133 Plus 2.0 gene chip array (Affymetrix). Among 53 UC patients, 10 had UC-associated cancer (UC-Ca group) whereas 43 did not (UC-NonCa group). RESULTS: By comparing gene expression profiles of nonneoplastic rectal mucosae between the UC-Ca and UC-NonCa groups, we could identify 40 genes that were differentially expressed between two groups. The list of discriminating genes included low-density lipoprotein receptor-related protein (LRP5 and LRP6). Previous studies suggested that LRP5 and LRP6 expression promotes cancer cell proliferation and tumorigenesis and are considered as candidate oncogenes. In the present study, both LRP5 and LRP6 showed significantly higher expression in the UC-Ca group, which suggests the importance of these genes in the development of UC-associated colorectal cancers. With the 40 selected discriminating genes, we did class prediction of the development of colorectal neoplasms in UC patients. Using the k-nearest neighbor method and the support vector machine, we could predict the development of UC-associated neoplasms with an accuracy of 86.8% and 98.1%, respectively. CONCLUSIONS: These findings have important implications for the early detection of malignant lesions in UC and may provide directions for future research into the molecular mechanisms of UC-associated cancer. PUBMED: 17255260
Find other GeneSets from this publication