Deciphering Promoter Hypermethylation of Genes Encoding for RASSF/Hippo Pathway Reveals the Poor Prognostic Factor of RASSF2 Gene Silencing in Colon Cancers

Abstract

The aims of this study were to assess the frequency of promoter hypermethylation of the genes encoding the Ras associated domain family (RASSF)/Hippo pathway, as well as the impact on overall (OS) and progression-free survival (PFS) in a single-center retrospective cohort of 229 patients operated on for colon cancers. Hypermethylation status was investigated by methylation-specific PCR on the promoters of the <i>RASSF1/2</i>, <i>STK4/3</i> (encoding Mammalian Ste20-like protein 1 and 2 (MST1 and 2), respectively), and <i>LATS1/2</i> genes. Clinicopathological characteristics, recurrence-free survival, and overall survival were analysed. We found the RASSF/Hippo pathway to be highly silenced in colon cancer, and particularly RASSF2 (86%). The other promoters were hypermethylated with a lesser frequency of 16, 3, 1, 10 and 6%, respectively for <i>RASSF1</i>, <i>STK4</i>, <i>STK3</i>, <i>LATS1</i>, and <i>LATS2</i> genes. As the hypermethylation of one RASSF/Hippo family member was by no means exclusive from the others, 27% of colon cancers displayed the hypermethylation of at least two RASSF/Hippo member promotors. The median overall survival of the cohort was 60.2 months, and the median recurrence-free survival was 46.9 months. Survival analyses showed a significantly poorer overall survival of patients when the <i>RASSF2</i> promoter was hypermethylated (<i>p</i> = 0.03). The median OS was 53.5 months for patients with colon cancer with a hypermethylated <i>RASSF2</i> promoter versus still not reached after 80 months follow-up for other patients, upon univariate analysis (HR = 1.86, [95% CI: 1.05-3.3], <i>p</i> < 0.03). Such difference was not significant for relapse-free survival as in multivariate analysis. A logistic regression model showed that <i>RASSF2</i> hypermethylation was an independent factor. In conclusion, <i>RASSF2</i> hypermethylation is a frequent event and an independent poor prognostic factor in colon cancer. This biomarker could be investigated in clinical practice.