Abstract
Recent studies suggest that long-interspersed nucleotide element-1 (LINE-1) hypomethylation is commonly found in colorectal cancer (CRC), and is associated with worse prognosis. However, the utility of LINE-1 methylation on the prognosis of CRC is still controversial, and may be due to the fact that some clinical and pathological features may affect LINE-1 methylation. Thus, the aim of this study was to assess the prognostic value of tumor LINE-1 methylation in CRC, through their association with the CRC clinical and pathological characteristics. Survival of sixty-seven CRC patients was evaluated according to the median of tumor LINE-1 methylation, as well as pathological and oncological variables. We also studied the association between LINE-1 methylation and pathological features, and finally, we assessed the overall and disease-free survival of LINE1 methylation, stratified by neoadjuvant treatment and further checked by multivariate Cox regression to assess the statistical interactions. LINE-1 was hypomethylated in the CRC tumor with respect to the tumor adjacent-free area (p < 0.05), without association with any other clinical and oncological features, nor with overall and disease-free survival rates for CRC. Relevantly, in neoadjuvant treatment, LINE-1 methylation was associated with survival rates. Thus, disease-free and overall survival rates of treated CRC patients were worse in the hypomethylated LINE-1 tumors than those with normal LINE-1 methylation (p = 0.004 and 0.0049, respectively). Indeed, LINE-1 was hypermethylated more in the treated patients than in the non-treated patients (p < 0.05). The present study showed that tumor LINE-1 hypomethylation was associated with worse survival rates in only treated patients. Our data suggest an interactive effect of neoadjuvant treatment and tumor LINE-1 methylation, which could be a specific-tissue biomarker to predict survival of the treated patients, and help to personalize treatment in CRC.