Background: Colorectal cancer (CRC) is the third leading cause of cancer-related death worldwide. About 30% of patients present with metastatic disease involving predominantly the liver and a similar percentage will develop distant metastases later after removal of the primary tumor. In metastatic CRC, chemotherapies and biological drugs have prolonged survival for up to 30 months. However, there is a great need for biomarkers predictive of response and prognosis to optimize treatments. CXC chemokine receptor 4 (CXCR4) is a chemokine receptor; it binds to CXCL12 and plays a central role in colon cancer cells' growth and dissemination.
Materials and methods: CXCR4 was evaluated in CRC primary tissues by immunohistochemistry. Formalin-fixed, paraffin-embedded 4-μm tissue sections were immunostained using a biotin-streptavidin-peroxidase method and categorized into 2 semiquantitative classes: (i) absence of staining, ≤50% positive cells (negative/low) and (ii) >50% positive cells (high). Associations between clinic-pathologic variables and CXCR4 expression were evaluated using the χ test. The Kaplan-Meier product-limit method was applied to graph overall survival (OS). OS was defined as the time elapsed from diagnosis to death from any cause. Univariate analysis was carried out using the log-rank test. Cox proportional hazards regression was used to analyze the effect of several risk factors on OS.
Results: Seventy-eight primary adenocarcinomas were analyzed; 26 were categorized as negative/low and 52 as high. Age, sex, performance status, site of metastases, KRAS mutational status, type of first-line therapy, and a number of therapy lines did not correlate with CXCR4 expression. Although not significant (P=0.0533), high CXCR4 expression was more frequently localized on the right side of the colon. Significant correlations were detected with grading (P=0.0041) and response to first-line anti-epidermal growth factor receptors agents (P<0.0001), bevacizumab (P=0.0029), and chemotherapy alone (P=0.0260). At a median follow-up of 53 months, 77 deaths have been registered. Grading [hazard ratio (HR): 1.42; confidence interval (CI): 0.89-2.28; P<0.0001], KRAS mutational status (HR: 1.73; CI: 1.03-290; P=0.0133), response to first-line chemotherapy (HR: 3.39; CI: 2.10-5.48; P<0.0001), and CXCR4 expression (HR: 3.18; CI: 2.01-5.02; P<0.0001) showed prognostic power at univariate and multivariate analyses.
Conclusion: In the present report, we show that CXCR4 expression on the primary tumor is an independent prognostic factor and correlates with response to first-line chemotherapy in metastatic CRC patients.