Pretreatment expression of 13 molecular markers as a predictor of tumor responses after neoadjuvant chemoradiation in rectal cancer

OBJECTIVE:: This study evaluated the predictive value of a number of tissue biomarkers, including proliferating cell nuclear antigen, survivin, thymidine phosphorylase, thymidylate synthase, bax, p53, nuclear factor-kappa B, vascular endothelial growth factor, matrix metalloproteinase-2, matrix metalloproteinase-9, CD133, CD44, and cyclooxygenase-2 with regard to preoperative chemoradiation in rectal cancer. BACKGROUND:: The ability to predict tumor response before treatment may significantly impact the selection of patients for preoperative chemoradiation therapy for rectal cancer. However, no definite predictive marker is known. METHODS:: Pretreatment biopsies from 123 patients who underwent preoperative chemoradiation were included. The mRNA levels of 13 biomarkers were analyzed by reverse transcriptase-polymerase chain reaction, with normalization relative to glyceraldehydes 3-phosphate dehydrogenase. Response to treatment was assessed by a 4-point tumor regression grade scale based on the ratio of fibrosis to residual cancer. RESULTS:: Among the 13 markers, no significant correlations in terms of T downstaging, N downstaging, and tumor-node-metastasis downstaging were observed. On multiple logistic regression analysis, only CD44 expression was found to be significant independent predictive factors for tumor regression grade response [odds ratio, 4.694 (1.155, 17.741), P = 0.030]. CD44 mRNA expression was significantly associated with expressions of the remaining 12 markers (all P < 0.05). Among the 118 patients receiving radical resection, proliferating cell nuclear antigen was the only independent factor to predict pathologic node negative status [odds ratio, 4.328 (1.078, 12.536), P = 0.037]. CONCLUSIONS:: Elevated CD44 mRNA levels in pretreatment biopsies might be predictive of poor tumor regression after preoperative chemoradiation in rectal cancer. Moreover, the proliferating cell nuclear antigen mRNA level might be predictive of nodal regression.