CEA Clearance Pattern As a Predictor for Pathologic Complete Response after Neoadjuvant Chemoradiation for Rectal Cancer: Results of the FOWARC Trial.

Huabin Hu,Jianwei Zhang,Yue Cai,Ping Lan,Lei Wang,Meijin Huang,Liang Kang,Jianping Wang,Yanhong Deng
DOI: https://doi.org/10.1200/jco.2016.34.15_suppl.e15112
IF: 45.3
2016-01-01
Journal of Clinical Oncology
Abstract:e15112 Background: Approximately 10% to 30% locally advanced rectal cancer patients receiving neoadjuvant chemoradiation (CRT) will have a pathologic complete response (pCR). Predicting pCR can have a significant effect on patients in terms of helping confer prognosis, possibly enabling organ preservation, and potentially avoiding unnecessary surgery. The aim of this study was to investigate the predictive value of carcinoembryonic antigen (CEA) clearance pattern for the tumor response to preoperative CRT in rectal cancer patients with elevated CEA levels. Methods: A total of 71 patients with elevated CEA levels ( > 5ng/ml) were randomly assigned to receive neoadjuvant 5-FU-based chemotherapy with pelvic radiation in FOWARC trial. Serum CEA were measured at six different time points, including base line, and the 2, 4, 6, 8 and 14 weeks during CRT. An exponential trend line was drawn using the six CEA values. Patients were categorized into two groups based on R2 values calculated through trend line, which indicates the correlation coefficient between exponential graph and measured CEA values: exponential decrease group (0.9 < R2≦1.0), and non-exponential decrease group (R2≦0.9). Results: 31 patients were classified as CEA exponential decrease group, and 40 patients were in non-exponential decrease group. 11 (15%) of 71 patients were achieved pCR. CEA decrease with exponential pattern was a significant independent predictor for TRG (0-1) (OR: 3.04, CI 1.10-8.43), good down-staging (ypTNM stage 0-I) (OR 7.26, CI 1.79-29.37) and pCR (OR 10.15, CI 1.50-68.62) in multivariate analysis, and the AUC for pCR was 0.726, which was the strongest factor among clinicopathologic indexes (clinical T stage,tumor differentiation, tumor length, circumferential extent, and normalization of post-CRT CEA level). Conclusions: The pattern of CEA clearance, a simple clinical parameter, normally available in everyday practice, was a independent predictor of pCR for rectal cancer patients. These results would be beneficial to guide individualized treatment strategies for this patients with elevated CEA levels. Clinical trial information: NCT01211210.
What problem does this paper attempt to address?