Survival outcomes for right-versus left-sided colon cancer and rectal cancer in England: A propensity-score matched population-based cohort study

Abstract

Introduction: The aim of this study was to evaluate 5-year overall survival (OS) in patients operated on for potentially curable right versus left-sided colon cancer and rectal cancer in England.

Materials and methods: A retrospective propensity-score matched population-based cohort study was performed using data from English Hospital Episode Statistics, Office for National Statistics and National Bowel Cancer Audit dataset. Patients ≥18 who underwent elective resection for right-colon, left-colon, or rectal cancer between 2000 and 2015 were included. Patients were matched using propensity scores with the dependant variable being site of primary tumour (right-colon, left-colon, or rectum) and independent variables age, Charlson comorbidity index, operation year and Duke's stage. The primary outcome was 5-year overall survival (OS).

Results: A total of 167,606 patients were included. Following propensity-score matching 26,662 patients remained in each group (right-colon, left-colon, and rectum). 5-year OS was significantly worse for patients with Duke's A-C right-sided primaries compared to left-sided and rectal cancers in the unmatched (58.8% vs 66.7% vs 70.0% p=<0.001) and matched cohorts (62.6%, 66.8%, 65.8% p=<0.001). Superior OS for patients with left-sided colon cancer was demonstrated across all stages (Duke's A HR 0.845, p = 0.003; Duke's B HR 0.947, p = 0.045; Duke's C HR 0.783, p < 0.0001). Secondary analysis demonstrated equivalent OS between study groups when ≥22 lymph nodes were harvested.

Conclusion: There is a significant difference in survival between comparable patients with right and left sided colon cancers in England. The quality and/or extent of mesenteric resection may be of particular importance in right-sided colon cancer.

Keywords: Colorectal cancer; Lymph node harvest; Survival; Tumour sidedness.