K. M. Ramonell1, W. Liang2, Y. Liu2, T. Gillespie1, V. O. Shaffer1 2Emory University Rollins School Of Public Health,Department Of Biostatistics,Atlanta, GA, USA 1Emory University School Of Medicine,Department Of Surgery,Atlanta, GA, USA
Introduction: Colon cancer ranks fourth in incidence, and second in cancer mortality in the United States. After curative resection of the primary tumor, adjuvant chemotherapy (AC) may be considered for stage III or for high-risk stage II patients. However, the role of AC in stage II colon cancer remains controversial. This study sought to determine which demographic, socioeconomic, and histopathologic factors affect receipt of AC and how these factors affect overall survival (OS) in stage II colon cancer.
Methods: Data from the National Cancer Data Base (NCDB) Participant Use Files were used. All patients with stage II colon cancer diagnosed in the US between 2004 and 2012 who underwent curative resection were included. Univariate and multivariate regression analyses were performed using the chi-square test for categorical covariates and ANOVA for numerical covariates. Propensity score matching was also implemented to reduce treatment selection bias.
Results: Of the total stage II colon cancer patients included, 5,443 (16.9%) received adjuvant chemotherapy and 26,833 (83.1%) did not. An improvement in OS was found in stage II colon cancer patients who received AC (HR 0.78; p<0.001). The following factors were associated with receipt of AC: age <65, white race, positive surgical margins, Charleson-Deyo comorbidity score equal to 0 or 1, residing in a low-income neighborhood, and being insured by Medicare. Of those who received AC, the following factors were associated with worsened OS: male gender (HR 1.25; p<0.001), age at diagnosis >65 (HR 1.06; p<0.001), black race (1.36; p<0.001), having Medicaid as primary insurance provider (HR 1.37; p=0.016). Patients living in lower income areas (HR 1.22; p=0.001) and in areas with lower education rates (HR 1.13; p=0.012) had worse survival. Among histopathologic features analyzed, presence of perineural invasion and lymphovascular invasion were independently associated with a significantly worse OS (HR 1.27; p<0.001) and (HR 1.28; p<0.001), respectively.
Conclusion: The results of this study suggest that the OS benefit observed with the use of AC in stage II colon cancer patients is primarily associated with non-Black patients younger than 65; who are privately insured or have Medicare; and those with no evidence of perineural invasion or lymphovascular invasion. While clinical and histopathologic features are known to affect cancer prognosis; this analysis shows that both socioeconomic and demographic factors play key roles in important outcomes including OS. Thus it is imperative that future research explore underlying reasons for these findings and interventions be tested to improve outcomes among those at risk.