EP207 - Rectal Cancer Surgery and Survival in the Veterans Health Administration: Insights into Racial and Ethnic Equity in an Equal-Access System

Purpose/Background: Overall survival among patients with rectal cancer differs across population groups, with non-Hispanic Black (NHB) patients typically experiencing worse survival than non-Hispanic White (NHW) patients. Five-year relative survival for persons receiving care in non-Veterans Health Affairs (non-VA) settings is 64.8% for NHB patients versus 66.7% for NHW patients, with more pronounced disparities in advanced disease. Additionally, NHB patients have been shown to be less likely to undergo sphincter-preserving surgery (SPS) compared to NHW patients. The VA, an equal-access system, provides a unique setting to determine if equal-access systems mitigate surgical treatment and mortality differences.

Methods/Interventions: We conducted a retrospective cohort study of veterans diagnosed with rectal and rectosigmoid junction cancer from January 1, 2014 to July 15, 2025 using the VA Central Cancer Registry. The primary outcome was 5-year overall survival (OS) from diagnosis. Kaplan-Meier curves were estimated for groups defined by demographic and clinical factors and compared using log-rank tests. Multivariable Cox proportional hazards models were constructed in a step-wise fashion to evaluate the effects of age, sex, anatomic site (rectum vs. rectosigmoid), stage, and Gagne comorbidity index on survival, with results reported as hazard ratios (HR). A subgroup analysis using multivariable logistic regression was performed to evaluate a secondary outcome of SPS.

Results/Outcomes: Among 11,083 veterans, the mean age at diagnosis was 66.3. Race and ethnicity was 18.9% NHB, 65.6% NHW, and 15.5% all other groups. Stage distribution was: 0 (5.3%), I (21.0%), II (14.0%), III (19.7%), IV (14.2%), and unknown (25.8%). Five-year OS was 52.7% (95% CI 51.5-53.9%), ranging from 74.8% (95% CI 72.6-76.9%) for stage I to 9.6% (7.9-11.6%) for stage IV (p < 0.001), and 57.7% (95% CI 55.0-60.3%) and 51.2% (95% CI 49.7-52.6%) for NHB and NHW patients, respectively (p < 0.001). Figure 1 shows OS stratified by race and ethnicity. In the unadjusted analysis, NHB patients had decreased mortality risk by 18% (HR 0.82, 95% CI 0.76-0.89) compared to NHW patients (p < 0.001). Race and ethnicity was no longer reliably associated with survival after adjustment for age. Further adjustment for sex, anatomic site, stage, and comorbidity had no additional effect (Table 1). Among veterans who had surgery (N = 5,662), 92.6% of NHB patients underwent SPS compared to 89.1% of NHW patients (p < 0.01). This difference was largely driven by a high proportion of NHB patients undergoing local resection (31.3%) compared to NHW patients (17.4%). Multivariable logistic regression showed that NHB patients had 33% higher adjusted odds of SPS compared to NHW patients (odds ratio 1.33, 95% CI 1.03-1.70, p < 0.05). Excluding patients who underwent local resection eliminated the observed association between race and ethnicity and SPS.

Conclusion/Discussion: In this cohort of veterans with rectal cancer, survival was not reliably different between NHB and NHW patients after accounting for demographics, site, stage, and comorbidity, suggesting that racial and ethnic disparities in rectal cancer mortality in the general population may be minimized in equal-access systems. Similarly, NHB veterans had higher odds of undergoing SPS compared to NHW veterans - contrary to patterns reported in non-VA settings - further underscoring the potential of equitable access to mitigate racial and ethnic disparities in rectal cancer surgery and survival.

Figure 1. Overall survival by race and ethnicity groups.
Table 1. Cox proportional hazards models comparing risk of mortality from any cause.
** p < 0.01; AJCC, American Joint Committee on Cancer; CI, confidence interval; HR, hazard ratio; ref, reference