Research Winner
Molecular Markers of Rapidly Growing Tumors: Another Piece of the Puzzle
By Shehla Sheikh, M.D., Mustafa Arain, M.D., Bharat Thaygarajan, M.D., and Aasma Shaukat, M.D., Minneapolis Veterans Affairs Medical Center and the University of Minnesota
Sporadic colorectal cancers that develop after complete colonoscopy (interval cancers) are likely to be rapidly growing tumors, and comparison of these tumors with noninterval cancers offers a unique means to identify the genetic pathways responsible for rapid tumor growth. The aim of this study was to compare the CpG island methylator phenotype (CIMP) status of interval versus noninterval colorectal cancers in the Minneapolis Veterans Affairs population and to determine the relationship between CIMP and microsatellite instability (MSI) in interval and noninterval colon cancers.
Methods: We searched our institution’s cancer registry for interval cancers (colorectal cancers that developed within 5 years of a complete colonoscopy). These were frequency matched in a 1:2 ratio by age and sex to patients with noninterval cancers. Over a 17-year period, we identified 194 cancers that met the study criteria. MSI testing had been performed in 163 of these cancers in a previous study. Tumor DNA was extracted and tested for MSI and CIMP gene markers (MINT1, MINT2, MINT31, p16INK4, MGMT, hMLH1). CIMP was defined as methylation in 3 or more genes.
Results: Of the 1,323 colorectal cancers diagnosed during the study period, 63 (4.7%) were identified as interval cancers; the 131 subjects with noninterval cancers served as controls. Study subjects were almost all Caucasian men. Interval cancers were significantly more likely to be CIMP-positive than noninterval cancers (56% vs. 31%, P=0.004), to be right-sided, and, consistent with our previous study, and to have MSI than noninterval cancers (29% vs. 11%, P=0.004). In a multivariate logistic regression model, the location of the cancer (OR 1.85; 95% CI 1.01-3.8), MSI (OR 2.6; 95% 1.08-6.7), and CIMP-positive status (OR 2.41; 95% CI 1.1-4.9) were associated with interval cancers. The McNemar’s test for discordance between MSI and CIMP was significant, suggesting that MSI and CIMP were independently associated with interval cancers. There was no difference in 5-year survival between the 2 groups. Interval cancers were more likely to have the BRAF mutation (28% versus 18%) and less likely to have the KRAS mutation (12% versus 21%); however, these differences were not statistically significant. Interval cancers were more likely to have MSI in combination with BRAF mutation (MSI+/BRAF+) than noninterval cancers (27% versus 8%, P=0.008).
Conclusion: Interval colon cancers are 3 times more likely to have MSI alone, and 4 times more likely to have MSI in association with BRAF mutation. Thus, this study indicates that MSI and CIMP-positive status are independently associated with interval cancers. In addition to MSI, a CIMP-positive status adds information in explaining rapid tumor growth and biology. Whether CIMP and MSI lead to rapid tumor growth or occur in rapidly growing tumors remains unknown.