Predictors of Pathologic Complete Response After Neoadjuvant Treatment for Rectal Cancer: A Multicenter Study.

ABSTRACT

BACKGROUND: Pathologic complete response (pCR) to neoadjuvant chemoradiation (CRT) for rectal cancer is associated with better long-term outcomes, and is used as an early indicator of response to novel agents. To assess the rate and predictors of pCR, we performed a retrospective multicenter study involving 5 Canadian cancer centers. PATIENTS AND METHODS: Cancer registries identified consecutive patients with locally advanced rectal adenocarcinoma from the Tom Baker Cancer Centre, Cross Cancer Institute, British Columbia Cancer Agency, Ottawa Hospital Cancer Centre, and the Dr H. Bliss Murphy Cancer Centre who received fluoropyrimidine-based CRT and had curative intent surgery from 2005 to 2012. Patient, tumor, and therapy characteristics were correlated with response. RESULTS: Of the 891 patients included, 885 patients had pCR data available. Of the included patients, 161 (18.2%) had a pCR to CRT, and 724 (81.8%) did not. Patients with a pCR had a lower pretreatment carcinoembryonic antigen (CEA) level, and higher hemoglobin level in univariate analysis. In multivariable analysis, statin use at baseline (odds ratio [OR], 1.7; 95% confidence interval [CI], 1.02-2.92; P = .04), lower pretreatment CEA level (OR, 1.03; 95% CI, 1.01-1.06; P = .03), and distance closer to anal verge (OR, 1.07; 95% CI, 1.01-1.15; P = .04) were significant predictors of pCR. The 3-year disease-free survival was 86% in those with a pCR versus 62.5% in those without a pCR (P < .0001) and pCR was associated with improved overall survival (hazard ratio, 0.29; 95% CI, 0.17-0.51; P < .0001). CONCLUSION: Lower pretreatment CEA level, proximity to anal verge, and statin use are predictors of pCR in our large retrospective cohort. Clinical trials to investigate statins combined with neoadjuvant CRT might be warranted.