LRP1 Methylation and Colon Cancer

Colorectal cancer (CRC) is a major public health problem in France and worldwide. CRC is the third most common cancer in incidence and mortality in France. The vast majority of these cancers are adenocarcinomas that arise sporadically and develop from precursor lesions: adenoma. All CRC with the same disease stage do not have the same prognosis. Various parameters have been identified as factors influencing the prognosis and allows adjustment of the treatment. The poor histoprognostic factors are vessels and nerves invasion by the tumor or the mucinous adenocarcinoma subtype. At the molecular level, the presence of microsatellite instability (MSI) improves the prognosis, while the presence of a BRAF mutation is an independent poor prognostic factor.

The different molecular pathways of colonic carcinogenesis are the chromosomal instability pathway, the microsatellite instability pathway inducing errors in DNA mismatch repair and the CpG Island Methylator Phenotype (CIMP). The hypermethylation of CpG islands of genes promoters leads to an over or most frequently under gene expression. CIMP is observed in near 15% of CRC and is associated with specific clinical and pathological features: older patients, female predominance, right colonic involvement, poorly differentiated or mucinous adenocarcinomas. From a molecular point of view, the high CIMP phenotype is strongly associated with the presence of BRAFV600E mutation, the absence of RAS mutation and the presence of microsatellite instability. The prognostic value of CIMP is actually controversial. A recent meta-analysis found that the CIMP phenotype was associated with a poor prognosis. Methylation of some genes promoters as CDKN2A is associated with a poor prognosis.

LRP-1 (low density lipoprotein receptor-related protein 1) is a multifunctional endocytic receptor that belongs to the LDL receptors the family. It mediates the clearance of many extracellular enzymes involved in the spread of cancer cells: metalloproteinases and serine proteinases. Decrease of LRP-1 activity or loss of LRP-1 expression correlates with increased aggressiveness of cancer cells in certain types of cancer. The expression of LRP-1 has almost never been studied in CRC. Only one immunohistochemical study of LRP-1 protein expression in colonic adenocarcinoma has been published to date. This study shows that tumor cells express LRP-1, but in nearly half the cases, weaker than in normal cells colic. The mechanisms involved in the decrease of expression are not known.

An epigenetic mechanism might be involved as hypermethylation of the of LRP-1 gene promoter, especially as the promoter of this gene is rich in CpG islands (methylation targets).

The clinical impact of the level of LRP-1 expression on overall survival of cancer patients has been assessed in two studies: one on patients suffering from hepatocellular carcinoma and one on patients suffering from primitive lung adenocarcinomas. In both studies, the decrease of immunohistochemical and gene expression of LRP-1 correlated with overall survival decrease. The clinical and prognostic impact of LRP-1 expression in CRC and its association with a particular molecular or morphological profile has not been studied to date.

In this work, the investigators will study the methylation pattern of the LRP-1 gene promoter in a well-characterized series of CRC.

The main objective of this study is to study by methylation analysis the role LRP-1 promoter gene methylation in the prognosis of CRC.

The secondary objective of this study is to evaluate the prognostic impact of CIMP.

This study will be the first to explore the LRP-1 gene promoter methylation in in CRC. This study will help to increase knowledge of the prognostic role of LRP-1 gene promoter methylation in CRC and its possible association with clinical, morphological or molecular parameters.

This study is a single center retrospective cohort study.

The investigators will first design the list of all eligible patients by using a request in the software DIAMIC of th pathology department with component code "colon", damage code "adenocarcinoma " and period "between 2006 and 2012". Patients suffering from Lynch syndrome or other familial cancer syndrome will be excluded of the study. Informed consent will be send to all the selected patients.

Once the patients list will be completed:

• The patients' medical records will be reviewed for: sociodemographic and clinical data colonic tumor location, TNM stage, MSI, RAS and BRAF status, presence of synchronous metastases, treatment type and duration, patient follow up to the date of 01/06/2016 point (metastases occurrence, disease recurrence, patient death, ...).
• A central review of all pathological slides will be performed for these criteria: adenocarcinoma subtype according to WHO 2010 classification, differentiation, association with polyps and their types, pTNM stage, invasion front and budding, stroma type, presence and amount of tumor necrosis).
• Methylation analysis:
• LRP1 gene promoter methylation analysis by pyrosequencing.
• CIMP status determination by High Resolution Melting analysis of MLH1, MINT1, MINT2, MINT31 and CDKN2A genes promoters.

Statistical analysis:

• Data description: mean and standard deviation for quantitative variables; number and percentage for categorical variables.
• Comparison of tumors having a high methylation of the LRP-1 gene promoter and tumors with not or little methylation of the LRP-1 gene promoter by univariate analysis (Student's test and the Wilcoxon test, Chi2 or Fisher's exact) and multivariate (logistic regression).
• Search factors associated with patient outcomes, including the methylation of LRP1 promoter and CIMP status, by univariate analysis (log-rank test) and multivariate (Cox model).