Social determinants of health and treatment outcomes among a retrospective cohort of patients receiving immunotherapy for cancer.

Abstract

e13744 Background: Disparities due to socioeconomic status persist in the treatment of cancer. Newer therapies, such as immune checkpoint inhibitors (ICIs), offer easier administration, lower risk of toxicity, and may improve accessibility compared to cytotoxic chemotherapy. Prior studies on ICI outcomes have shown mixed results at the county level, and there is a need to investigate social determinants of health at a more granular level. Our study aims to compare long-term ICI outcomes between socioeconomic groups at the census tract level among patients with cancer. Methods: The investigators compiled data from 2/1/2011 to 4/7/2022 on patients who received at least one dose of an ICI at a comprehensive cancer center and its outreach clinics to create a retrospective patient registry. Investigators used a secure, cloud-based REDCap registry, validated it with data quality rules, and resolved discrepancies. Clinical research specialists at Vasta Global captured most of the data. Investigators correlated patients’ ZIP codes with 2010 US Department of Agriculture Rural-Urban Commuting Area (RUCA) classifications and census tracts with 2020 Center for Disease Control Social Vulnerability Index (SVI) data. The univariate analyses used the ANOVA or Kruskal-Wallis tests, chi-square tests, and Kaplan-Meier methods. The multivariate analyses used Cox and logistic regressions, adjusting for age, race, ethnicity, type of cancer, smoking status, age-adjusted Charlson Comorbidity Index, and other comorbidities. Results: Our cohort consisted of 1,807 patients who were an average of 66 years old, predominantly male (60.8%), white (84%), non-Hispanic (98%), and had lung cancer (45.1%) as the most common tumor type. Univariate analyses found no associations between either RUCA or SVI and overall survival (OS), progression-free survival (PFS), immune-related adverse events (irAE), or time to irAE. For RUCA, multivariate analysis found that rural location of residence was associated with shorter PFS (all covariates, p 0.0386). For SVI, multivariate analysis showed no significant association with OS, PFS, number of irAEs, or time until irAE (including high-grade irAEs). Conclusions: Rural residence by ZIP code was associated with shorter PFS, indicating that there may be decreased access to ICI for these patients. However, the overall findings suggest similar ICI treatment outcomes across patients with widely variable social determinants of health by census tract. Further research is needed to understand how newer therapies can be best positioned to overcome disparities in cancer care.