A prognostic model comprising pT stage, N status, and the chemokine receptors CXCR4 and CXCR7 powerfully predicts outcome in neoadjuvant resistant rectal cancer patients.

Despite the optimization of the local treatment of advanced rectal cancer (LARC), combination of preoperative chemoradiotherapy (CRT) and surgery, approximately one third of patients will develop distant metastases. Since the chemokine receptor CXCR4 has been implicated in metastasis development and prognosis in colorectal cancer, the role of the entire axis CXCR4-CXCL12-CXCR7 was evaluated to identify high relapse risk rectal cancer patients. Tumor specimens of 68 LARC patients undergoing surgery after neoadjuvant-CRT were evaluated for CXCR4, CXCR7, and CXCL12 expression through immunohistochemistry. Multivariable prognostic model was developed using classical prognostic factors along with chemokine receptor expression profiles. High CXCR4 correlated with a shorter relapse-free survival (RFS) (p = 0.0006) and cancer specific survival (CSS) (p = 0.0004). Concomitant high CXCR4-negative/low CXCR7 or high CXCR4-negative/low CXCL12 significantly impaired RFS (p = 0.0003 and p = 0.0043) and CSS (p = 0.0485 and p = 0.0026). High CXCR4/N+ identified the worst prognostic category for RFS (p < 0.0001) and CSS (p = 0.0003). The optimal multivariable predictive model for RFS was a five-variable model consisting of gender, pT stage, N status, CXCR4, and CXCR7 (AUC = 0.92, 95% CI = 0.77-0.98). The model is informative and supportive for adjuvant treatment and identifies CXCR4 as a new therapeutic target in rectal cancer.