Skip to main content
. 2009 Jun 23;50(3):309–321. doi: 10.3349/ymj.2009.50.3.309

Fig. 1.

Fig. 1

The relationships among various forms of genomic instability in CRC. (A) The pie chart illustrates the characterization of 209 sporadic CRCs for MSI and CIN.2 As shown, 48% of tumors had evidence of LOH, while 14% of tumors had MSI, including a minority of tumors (3%) that overlapped with LOH. We were most interested in the observation that nearly one third of the tumors (38%) did not have either MSI or LOH. (B) The bars demonstrate the mean methylation index based upon the number of markers methylated in the three subsets of CRCs.80 Methylation alterations were analyzed at 12 markers including 6 canonical CIMP markers (MINT-1, -2, -31, p16, p14 and MLH1) and 6 additional tumors suppressor genes (PTEN, TIMP3, RUNX3, HIC1, APC, and RARβ2). The three vertical columns represent the mean methylation ratios in MSI (blue), MSI-/LOH-(red) and LOH (white) CRCs. The error bars denote the S.D. The filled circles (color matched with vertical columns) represent the 95% confidence interval (CI) of the mean methylation ratios. The rectangular boxes in the upper panels represent the pair-wise correlation between the mean methylation ratios in each subset of CRC; p-values were calculated by the Wilcoxon test. As shown in the three panels, analysis of the methylation ratios using only four CIMP-related markers (versus using 12 or 6 CIMP-related markers) demonstrates a significant positive association for MSI and MSI-/LOH-tumors, but an inverse correlation for LOH CRCs. (C) The relationship between methylation ratios calculated using all 12 methylation markers was compared with LOH ratios in the total cohort of 126 CRCs. As shown, an inverse correlation was observed (p = -0.3690; p < 0.0001) between the methylation ratio and LOH.

CRC, colorectal cancers; MSI, microsatellite instability; LOH, loss of heterozygosity.

HHS Vulnerability Disclosure