Epigenomic diversity of colorectal cancer indicated by LINE-1 methylation in a database of 869 tumors

Abstract

Background: Genome-wide DNA hypomethylation plays a role in genomic instability and carcinogenesis. LINE-1 (L1 retrotransposon) constitutes a substantial portion of the human genome, and LINE-1 methylation correlates with global DNA methylation status. LINE-1 hypomethylation in colon cancer has been strongly associated with poor prognosis. However, whether LINE-1 hypomethylators constitute a distinct cancer subtype remains uncertain. Recent evidence for concordant LINE-1 hypomethylation within synchronous colorectal cancer pairs suggests the presence of a non-stochastic mechanism influencing tumor LINE-1 methylation level. Thus, it is of particular interest to examine whether its wide variation can be attributed to clinical, pathologic or molecular features.

Design: Utilizing a database of 869 colorectal cancers in two prospective cohort studies, we constructed multivariate linear and logistic regression models for LINE-1 methylation (quantified by Pyrosequencing). Variables included age, sex, body mass index, family history of colorectal cancer, smoking status, tumor location, stage, grade, mucinous component, signet ring cells, tumor infiltrating lymphocytes, CpG island methylator phenotype (CIMP), microsatellite instability, expression of TP53 (p53), CDKN1A (p21), CTNNB1 (beta-catenin), PTGS2 (cyclooxygenase-2), and FASN, and mutations in KRAS, BRAF, and PIK3CA.

Results: Tumoral LINE-1 methylation ranged from 23.1 to 90.3 of 0-100 scale (mean 61.4; median 62.3; standard deviation 9.6), and distributed approximately normally except for extreme hypomethylators [LINE-1 methylation < 40; N = 22 (2.5%), which were far more than what could be expected by normal distribution]. LINE-1 extreme hypomethylators were significantly associated with younger patients (p = 0.0058). Residual plot by multivariate linear regression showed that LINE-1 extreme hypomethylators clustered as one distinct group, separate from the main tumor group. The multivariate linear regression model could explain 8.4% of the total variability of LINE-1 methylation (R-square = 0.084). Multivariate logistic regression models for binary LINE-1 hypomethylation outcomes (cutoffs of 40, 50 and 60) showed at most fair predictive ability (area under receiver operator characteristics curve < 0.63).

Conclusions: LINE-1 extreme hypomethylators appear to constitute a previously-unrecognized, distinct subtype of colorectal cancers, which needs to be confirmed by additional studies. Our tumor LINE-1 methylation data indicate enormous epigenomic diversity of individual colorectal cancers.

Figure 1

Pyrosequencing to measure LINE-1 methylation. A. LINE-1 hypermethylated tumor. B. LINE-1 hypomethylated tumor. The % numbers (in blue shade) are proportions of C and T at each CpG site after bisulfite conversion, and the methylation level of each CpG site is estimated by the proportion of C (%). An overall LINE-1 methylation level is calculated as the average of the proportions of C (%) at the 4 CpG sites. The first, third and fourth CpG sites follow mononucleotide T repeats, resulting in higher T peaks (in yellow shade) than the second CpG site, and the proportion of C (%) has been adjusted accordingly. The arrows indicate no residual C at the non-CpG site, ensuring complete bisulfite conversion.

Figure 2

Distribution of LINE-1 methylation levels in 869 colorectal cancers. LINE-1 methylation levels distributed approximately normally, except for the presence of extreme hypomethylators (LINE-1 methylation value < 40), which appear to be separate from the remaining normal distribution.

Figure 3

Residuals plot of multivariate linear regression model to predict LINE-1 methylation in 869 colorectal cancers. Each dot represents each case. The x axis represents LINE-1 methylation level predicted by the regression model, and the y axis represents the residuals (observed LINE-1 methylation level minus predicted LINE-1 methylation level by the regression model). A distribution of the residuals is approximately normal across predicted LINE-1 methylation levels, except for the presence of outliers (arrow). The arrow suggests the presence of a distinct group of tumors, separate from the normal and homoscedastic distribution around the 0 residual line.

Figure 4

ROC (receiver operator characteristics) curves for the diagnosis of LINE-1 hypomethylation as a binary outcome using different cutoffs (40, 50 and 60 of 0-100 scale). Multivariate logistic regression models for the binary LINE-1 methylation outcomes were constructed using the same set of covariates as in the linear regression model. ROC curves for each cutoff was averaged over 10 folds each using 25% randomly heldout test data for cross validation. Area under the curve (AUC) represents performance characteristics of the multivariate model as a clinical test for the specific binary outcome.