Racial and socioeconomic disparities in time to chemotherapy and survival in patients with pancreatic cancer.

Abstract

4100 Background: Pancreatic cancer is traditionally known to be an aggressive malignancy, requiring timely diagnosis and treatment, including chemotherapy and/or surgery. Prior studies have evaluated variances in overall survival (OS) based on race or income; however, research regarding time to treatment based on racial and socioeconomic disparities is underwhelmingly studied. Our project sought to evaluate the impact of race and income on time to chemotherapy and OS in pancreatic cancer patients receiving neoadjuvant or adjuvant chemotherapy. Methods: The National Cancer Database (NCDB) from 2004-2020 was reviewed for patients with a diagnosis of pancreatic cancer, including all types and stages. Exclusion criteria included: no documented income level, time to first treatment surgery, or time to chemotherapy. Our patient population was then separated into two cohorts: neoadjuvant and adjuvant. Income level was stratified into the following: < $30,000, $30,000-34,999, $35,000-$45,999, and > $46,000. Race was also noted. We utilized a logistic regression to identify factors associated with longer time to chemotherapy (using median time to chemotherapy as a benchmark for early or later initiation of chemotherapy). A Cox proportional hazards model was used to identify factors associated with worse survival. A p-value < 0.05 was deemed significant. Results: Our final sample size included 64,640 patients. 6,887 (10.6%) of patients had an income less than $30,000. 55,340 (85.6%) of patients were Caucasian, 6,379 (9.9%) patients were African American, and 172 (0.002%) were American Indian. 18,679 (28.9%) received neoadjuvant chemotherapy, and 45,961 (71.1%) received adjuvant chemotherapy. Median time to chemotherapy was 27 [0-2048] days and 71 [0-1895] days in the neoadjuvant and adjuvant cohorts, respectively. Within both cohorts, income > $46,000 was noted to have significantly earlier time to chemotherapy compared to lower income levels. Those privately insured had significantly earlier time to chemotherapy compared to uninsured. The African American population was associated with a later time to first chemotherapy in comparison to Caucasians in both subgroups. A significantly less OS in the African American population was appreciated in the neoadjuvant cohort (HR 0.92, p-value 0.02), but not the adjuvant cohort. Conclusions: Our retrospective analysis emphasizes that minorities and/or those with a poor socioeconomic status have significant delay in time to chemotherapy in the neoadjuvant and adjuvant setting. Within the African American neoadjuvant cohort, a significantly worse OS was highlighted. This study emphasizes racial and socioeconomic disparities in our healthcare system. Limitations of our study include its retrospective nature and no information regarding the type of surgery the patients received, which could further account for differences in time to chemotherapy.