Exploring social determinants of health and immunotherapy utilization in patients with stage III non-small cell lung cancer following definitive chemoradiation.

1580 Background: Since the approval of immunotherapy (IO) for maintenance therapy in stage III non-small cell lung cancer (NSCLC) in 2018, its use has expanded rapidly. This study aimed to evaluate patterns of IO use among diverse patient populations who received definitive chemoradiation (CRT) followed by adjuvant IO for stage III NSCLC. Methods: A retrospective analysis was performed using the National Cancer Database for patients ≥ 18 years old with stage III NSCLC diagnosed between 2018 to 2021. Patients receiving ≥ 60 Gy of radiation in ≥ 30 fractions and IO delivered 50-150 days from starting CRT were included. Patients who underwent surgery were excluded. Social determinants of health included race, ethnicity, insurance, treatment site, Charlson-Deyo comorbidity index (CCI), high school graduation rate (HSGR; lowest, low, mid, high per US Census data), income (categorized similarly to HSGR), sex, and age. Odds ratios (OR) were calculated for IO use. Logistic regression analyses were performed for each variable followed by multivariable analysis with statistically significant factors ( P < 0.05). IRB exemption was granted. Results: 25,746 patients were included: 21,848 White, 3,236 Black, 558 Asian, and 104 Native American. 620 were Hispanic. On univariate analysis, the following were predictive of IO receipt: Black vs White race (OR 0.86, P < 0.001), Hispanic vs non-Hispanic ethnicity (0.72, P < 0.001), Medicare or other governmental insurance vs private (OR 0.94, P = 0.037 and 0.85, P = 0.034), CCI 1 or 2 vs 0 (1.11, P < 0.001 and 1.13, P = 0.005), community site vs academic/research center (0.88, P = 0.007), lowest and low vs high HSGR (0.78, P < 0.001 and 0.87, P < 0.001), lowest vs high income (0.90, P = 0.005), female vs male sex (1.06, P = 0.028), and age (-0.88%/year, P < 0.001). On multivariate analyses, Black patients were 3.3% less likely to receive IO than White patients (0.88, P < 0.001). Hispanic patients were 8.2% less likely to receive IO than non-Hispanic patients (0.72, P < 0.001). Medicare insurance (0.93, P = 0.012) and treatment at community sites (0.86, P = 0.002) were associated with reduced IO use. Patients from areas with lower HSGR were less likely to receive IO than those with the highest HSGR (lowest: 0.75, P < 0.001; low: 0.83, P < 0.001; mid 0.91, P = 0.039). Higher CCI (CCI 1: 1.12, P < 0.001; CCI 2: 1.13, P = 0.004; CCI 3: 1.10, P = 0.046), lower income (lowest: 1.11, P = 0.048; low: 1.10, P = 0.041; mid: 1.10, P = 0.025), and female sex (1.05, P = 0.048) were associated with increased IO use. Conclusions: This study highlights disparities in IO use following CRT for stage III NSCLC patients. Black race, Hispanic ethnicity, Medicare or governmental insurance, treatment at community sites, and lower education were associated with decreased IO use. These findings emphasize the need for strategies to ensure equitable access to advanced cancer therapies.