A national evaluation of the use and survival impact of adjuvant chemotherapy in Stage II colon cancer from the national cancer database

Aim Stage II colon cancers are a heterogeneous category, with controversy over use of adjuvant chemotherapy (AC). Patients with high-risk features may benefit from AC to improve overall survival (OS). Current guidelines do not routinely recommend AC in low-risk cases, but the actual use and benefit on OS in this cohort have not been fully examined on a national scale. We aimed to evaluate the use and impact of AC on OS in low-risk Stage II colon cancer. Methods The national cancer database was reviewed for Stage II colon cancers undergoing curative resection (2010-2015). Cases with preoperative radio-chemotherapy or high-risk features were excluded. Cases were stratified into 'AC' and 'no AC' cohorts, and then propensity score matched. Kaplan-Meier and Cox regression analysed OS. The main outcome measures were the incidence and impact of AC on OS in low-risk Stage II colon cancer. Results Of 39 926 patients evaluated, 8.2% (n = 3275) received AC. Matching resulted in 3275 cases per cohort. AC significantly improved 1-, 3- and 5-year OS versus no AC (P = 0.0017). The 5-year absolute risk reduction was 2.6%, relative risk reduction 12%, with a number needed to treat of 38. In the Cox model, AC remained significantly associated with increased OS (hazard ratio 0.816; 95% CI 0.713-0.934; P < 0.003). Conclusions From this dataset, AC was associated with improved OS in low-risk Stage II disease. These findings from a large-scale sample question current guidelines and the need for better risk stratification. Further study with more robust variables is warranted to determine AC best practices.

Automated summary

In Stage II colon cancer, receiving AC significantly improves 1, 3, and 5-year OS. The study findings suggest that AC is associated with improved OS in low-risk Stage II disease, questioning current guidelines and the need for better risk stratification. Further research with more robust variables is needed to determine best practices for AC.