An Integrated Model of Pathological Complete Response and Postoperative CEA Predicts Survival and Benefit of Adjuvant Chemotherapy for Locally Advanced Rectal Cancer

Shu-Biao Ye,Dong-Wen Chen,Pei-Si Li,Zhen-Sen Lin,Jia-Hui Long,Yi-Kan Cheng,Ping Lan
DOI: https://doi.org/10.21203/rs.3.rs-546136/v1
2021-01-01
Abstract:Abstract Background: Pathological complete response (pCR) and postoperative carcinoembryonic antigen (CEA) respectively represent pathological and biological response to treatment and prognostic indictors for locally advanced rectal cancer. However, the prognostic significance of integrated pCR and postoperative CEA model is rarely investigated. Methods: This big-data intelligence platform-based cohort biomarker study extracted clinicopathological characteristics and postoperative CEA of 919 locally advanced (clinical stage II-III) rectal cancer patients receiving neoadjuvant treatment from 6125 cases between April 2011 through September 2018 at Sun Yat-sen University, the Sixth Affiliated Hospital. An integrated model was constructed using pCR combined with the cut-off value of postoperative CEA generated from the receiver operating characteristic (ROC) curve. The model stratified locally advanced rectal cancer patients into four groups. Results: Both pathological response-pCR and biological response-postoperative CEA were independent prognostic indicators. In all patients, the 3-year distant metastasis-free survival (DMFS) (70.0%) and disease-free survival (DFS) (64.8%) rates were significantly lower in patients without pCR and postoperative CEA > 2.78 ng/ml (all p values < 0.05). Among patients treated with < 3 months of postoperative adjuvant chemotherapy (ACT), all others had higher 3-year DMFS and DFS rates. Among patients receiving ≥ 3 months of ACT, those without pCR had similar 3-year DMFS rates regardless of postoperative CEA level. Conclusions: The response integrated model with pCR and postoperative CEA not just effectively predict DMFS and DFS in locally advanced rectal cancer patients, but also was a predictive tool to potentially modify the optimal duration of ACT for locally advanced rectal cancer patients.
What problem does this paper attempt to address?