Predicting the pathologic response of locally advanced rectal cancer to neoadjuvant concurrent chemoradiation using enzyme linked immunosorbent assays (ELISAs) for biomarkers

PURPOSE: To investigate the role of biomarkers including serum tissue inhibitor of metalloproteinases-1 (TIMP-1), urokinase plasminogen activator receptor, vascular endothelial growth factor, and epidermal growth factor receptor in predicting pathologic response to neoadjuvant chemoradiation (NACRT) for rectal cancer. METHODS: Between 2007 and 2009, 50 clinical TNM stage II or III patients were analyzed in this prospective study. Pre- and post-NACRT serum levels of biomarkers were assessed using ELISAs. The primary and secondary endpoints were pathologic complete response (pCR) and Mandard regression grade (MRG). RESULTS: The pCR was reported in 5 patients (10.0%). According to the MRG, fifteen patients (30.0%) were divided into group A (Grade I-II), the others in group B (Grade III-V). On univariate analysis, post-NACRT TIMP-1 showed notable significance with pCR (P = 0.092) and significant correlation with MRG group A (P = 0.003). Post-NACRT TIMP-1 ≤ 204.5 ng/mL as cut-off value by ROC curve was associated with more pCR and MRG group A patients (P = 0.016 and 0.002). Interval between NACRT and surgery was related to pCR with approached trend levels of significance (P = 0.05) and to MRG group A significantly on univariate analysis of clinical factors (P = 0.031). On multivariate analysis, post-NACRT TIMP-1 was not significantly related to pCR (P = 0.187), while it was significantly associated with MRG (P = 0.009). Among clinical responders, post-NACRT TIMP-1 level ≤ 204.5 ng/mL was significantly associated with pCR (P = 0.021) and MRG group A (P = 0.003). CONCLUSIONS: Post-NACRT serum TIMP-1 could be used as a predictive marker of pathologic response to NACRT in rectal cancer, even in patients with clinical response.ope