The Causes Behind the Survival Paradox in Non-Metastatic Colorectal Cancer Patients and its Impact on the Treatment Regimen: A Retrospective Cohort Study

Pengwen Zheng; Guohui Xu; Zhaofu Qin; Shiqi Zhou; Hai Huang; Dening Ma; Xinyi Gao

doi:10.1016/j.surg.2024.03.043

The Causes Behind the Survival Paradox in Non-Metastatic Colorectal Cancer Patients and its Impact on the Treatment Regimen: A Retrospective Cohort Study

Surgery. 2024 May 16:S0039-6060(24)00208-3. doi: 10.1016/j.surg.2024.03.043. Online ahead of print.

Authors

Pengwen Zheng¹, Guohui Xu², Zhaofu Qin², Shiqi Zhou², Hai Huang³, Dening Ma², Xinyi Gao⁴

Affiliations

¹ Zhejiang Chinese Medical University, Hangzhou, Zhejiang, China; Department of Colorectal Surgery, Zhejiang Cancer Hospital, Hangzhou, Zhejiang, China.
² Department of Colorectal Surgery, Zhejiang Cancer Hospital, Hangzhou, Zhejiang, China.
³ Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical University, Hangzhou, China.
⁴ Department of Radiology, Zhejiang Cancer Hospital, Hangzhou, Zhejiang, China. Electronic address: 10301010213@fudan.edu.cn.

PMID: 38760234
DOI: 10.1016/j.surg.2024.03.043

Abstract

Background: The survival paradox (ie, the prognosis of the population at earlier tumor stages is worse than that of the population at later stages) has been observed in colorectal cancer based on the American Joint Committee on Cancer Tumor-Nodes-Metastases staging system. We aimed to clarify the reason for the survival paradox and its impact on patient treatment.

Methods: We conducted a retrospective study analyzing eligible patients with colorectal cancer from the Surveillance, Epidemiology, and End Results database and Zhejiang Cancer Hospital between 2010 and 2019. Adjusting for confounders using propensity score matching allowed confirmation of the effect of staging on the survival paradox.

Results: Based on the Surveillance, Epidemiology, and End Results database, the subgroups with survival paradox might be IIB/C versus IIIA, IIA versus IIIA, and T4N0 (IIB/C) versus T3N1 (IIIB). After propensity score matching, stage IIB/C still had a worse prognosis than stage IIIA (5-year overall survival: 69.3% vs 78.5%, P < .001). Interestingly, the proportion of stage IIIA people receiving chemotherapy was higher than that of stage IIB/C (P < .001), and logistic regression models showed that staging was the reason for deciding whether a patient receives chemotherapy or not. These phenomena between stage IIB/C and IIIA were verified in the local database.

Conclusion: These results suggested that the survival paradox was mainly due to underestimation of stage T4 weights or overestimation of stage N1 weights, and the low proportion of chemotherapy in patients with T4N0M0 colorectal cancer (proven to be more malignant than stage IIIA) might be related to the assignment to earlier stages, resulting in a lack of attention and poor compliance to chemotherapy in these patients.