Factors Associated With Racial and Ethnic Disparities in Locally Advanced Rectal Cancer Outcomes

Abstract

IMPORTANCE Hispanic and non-Hispanic Black patients receiving neoadjuvant therapy and surgery for locally advanced rectal cancer (LARC) achieve less favorable clinical outcomes than non-Hispanic White patients, but the source of this disparity is incompletely understood. OBJECTIVE To assess whether racial and ethnic disparities in treatment outcomes among patients with LARC could be accounted for by social determinants of health and demographic, clinical, and pathologic factors known to be associated with treatment response. DESIGN, SETTING, AND PARTICIPANTS The National Cancer Database was interrogated to identify patients with T3 to T4 or N1 to N2 LARC treated with neoadjuvant therapy and surgery. Patients were diagnosed between January 1, 2004, and December 31, 2017. Data were culled from the National Cancer Database from July 1, 2022, through December 31, 2023. EXPOSURE Neoadjuvant therapy for rectal cancer followed by surgical resection. MAIN OUTCOMES AND MEASURES The primary outcome was the rate of pathologic complete response (pCR) following neoadjuvant therapy. Secondary outcomes were rate of tumor downstaging and achievement of pN0 status. RESULTS A total of 34 500 patient records were reviewed; 21 679 of the patients (62.8%) were men and 12 821 (37.2%) were women. The mean (SD) age at diagnosis was 59.7 (12.0) years. In terms of race and ethnicity, 2217 patients (6.4%) were Hispanic, 2843 (8.2%) were non-Hispanic Black, and 29 440 (85.3%) were non-Hispanic White. Hispanic patients achieved tumor downstaging (48.9% vs 51.8%; P = .01) and pN0 status (66.8% vs 68.8%; P = .02) less often than non-Hispanic White patients. Non-Hispanic Black race, but not Hispanic ethnicity, was associated with less tumor downstaging (odds ratio [OR], 0.86 [95% CI, 0.78-0.94]), less frequent pN0 status (OR, 0.91 [95% CI, 0.83-0.99]), and less frequent pCR (OR, 0.81 [95% CI, 0.72-0.92]). Other factors associated with reduced rate of pCR included rural location (OR, 0.80 [95% CI, 0.69-0.93]), lack of or inadequate insurance (OR for Medicaid, 0.86 [95% CI, 0.76-0.98]; OR for no insurance, 0.65 [95% CI, 0.54-0.78]), and treatment in a low-volume center (OR for first quartile, 0.73 [95% CI, 0.62-0.87]; OR for second quartile, 0.79 [95% CI, 0.70-0.90]; OR for third quartile, 0.86 [95% CI, 0.78-0.94]). Clinical and pathologic variables associated with a decreased pCR included higher tumor grade (OR, 0.58 [95% CI, 0.49-0.70]), advanced tumor stage (OR for T3, 0.56 [95% CI, 0.42-0.76]; OR for T4, 0.30 [95% CI, 0.22-0.42]), and lymph node–positive disease (OR for N1, 0.83 [95% CI, 0.77-0.89]; OR for N2, 0.73 [95% CI, 0.65-0.82]). CONCLUSIONS AND RELEVANCE The findings of this cohort study suggest that disparate treatment outcomes for Hispanic and non-Hispanic Black patients are likely multifactorial in origin. Future investigation into additional social determinants of health and biological variables is warranted.