Abstract
BACKGROUND: Hypomethylation of Long Interspersed Nucleotide Element-1 (LINE-1) is associated with worse prognosis in colorectal cancer (CRC). However, little is known about the relevance of this marker for the prognosis and response to chemotherapy of metastatic and recurrent (advanced-stage) CRC. Our aim was therefore to investigate whether tumor LINE-1 hypomethylation correlates with patient survival and with response to 5-fluorouracil (5-FU)/ oxaliplatin (FOLFOX) chemotherapy in advanced-stage CRC. METHODS: The study included 40 CRC patients who developed metastasis or local recurrence after surgery and subsequently underwent FOLFOX therapy. Progression-free and overall survival were estimated using the Kaplan-Meier method. LINE-1 methylation levels in formalin-fixed and paraffin-embedded primary tumor tissues were measured by MethyLight assay and correlated with patient survival. In vitro analyses were also conducted with human colon cancer cell lines having different LINE-1 methylation levels to examine the effects of 5-FU and oxaliplatin on LINE-1 activity and DNA double-strand-breaks. RESULTS: Patients with LINE-1 hypomethylation showed significantly worse progression-free (median: 6.6 vs 9.4 months; P = 0.02) and overall (median: 16.6 vs 23.2 months; P = 0.01) survival following chemotherapy compared to patients with high methylation. LINE-1 hypomethylation was an independent factor for poor prognosis (P = 0.018) and was associated with a trend for non-response to FOLFOX chemotherapy. In vitro analysis showed that oxaliplatin increased the LINE-1 score in LINE-1-expressing (hypomethylated) cancer cells, thereby enhancing and prolonging the effect of 5-FU against these cells. This finding supports the observed correlation between tumor LINE-1 methylation and response to chemotherapy in CRC patients. CONCLUSIONS: Tumor LINE-1 hypomethylation is an independent marker of poor prognosis in advanced-stage CRC and may also predict non-response to combination FOLFOX chemotherapy. Prospective studies are needed to optimize the measurement of tumor LINE-1 methylation and to confirm its clinical impact, particularly as a predictive marker.