Discovering pathways in the tumor microenvironment important for recurrence-free survival in patients with colorectal liver metastasis.

Abstract

480 Background: Colorectal cancer (CRC) is the third leading cause of cancer-related deaths in the United States. Curative treatment for CRC liver metastasis can be achieved by surgical resection, but 5 year survival ranges between 20-40%. Current treatment strategies to prevent recurrences after resection of CRC liver metastases have focused on the use of chemotherapy cytotoxic to malignant epithelial cells of the tumor. However, stromal components within the tumor microenvironment have recently been implicated as contributors to the aggressiveness of the malignancy. We hypothesized that genomic based signatures associated with tumor stroma would be prognostic for recurrence after CRC hepatic metastasectomy. Methods: Thirteen independent CRC liver metastatic samples were subjected to laser-capture microdissection to isolate the tumor epithelium and tumor-associated stroma separately. Microarray analysis using class neighbors and gene set enrichment analysis (GSEA) was applied to the stromal mRNA to identify differentially expressed genes and pathways between samples from patients with early recurrence (<1yr) after resection (Group R) and those who remain disease free (>2yrs) (Group NoR). Results: Class neighbor analysis revealed a greater number of genes associated with immunity and cell to cell adhesion in Group R compared to Group NoR (p<0.05). Using GSEA, gene sets associated with T-cell immunity, Wnt pathway, and homeostasis were enriched in stroma Group R when compared to stroma Group NoR (p<0.05). Conclusions: These findings suggest several pathways are prognostic for recurrence of CRC liver metastasis. Development of stroma-directed therapy combined with tumor-directed therapy in preclinical models could hold the promise of major advances in the treatment of colorectal cancer liver metastasis in order to improve the clinical outcomes of these patients.