Treatment Patterns and Survival Outcomes of Immune Checkpoint Inhibitors in Advanced Non-Small Cell Lung Cancer Survival at a Safety-Net Hospital

Background: Immune checkpoint inhibitors (ICI) have transformed the treatment of advanced non-small cell lung cancer (aNSCLC) without driver mutations. KEYNOTE 189 trial showed median overall survival (OS) of 21.8 months (m) for chemotherapy with ICI (CT-ICI) compared to 12.1m for chemotherapy (CT). Since a large percentage of underserved cancer patients in the United States receive care from safety-net hospitals (SNH), this poses the question: Are these patients benefitting from ICI? Herein, we report the treatment and survival patterns of patients with aNSCLC at John Peter Smith Hospital (JPS), a SNH in North Texas. Methods: Patient data were obtained from the JPS tumor registry for this retrospective study. Eligible patients were diagnosed at JPS from 1/1/2017 to 12/31/2021 with stages IIIB/IIIC/IV NSCLC. Patients with driver mutations were excluded. Electronic records were reviewed for programmed death ligand 1 (PD-L1) testing and first-line treatments. OS was calculated from diagnosis to death (if applicable) or the last chart entry before 5/31/2023. Covariates were sex, race, age at diagnosis, stage, histology and PDL-1. Log-rank tests were used to compare survival distributions. OS probability within each treatment group was modeled using a Kaplan-Meier curve. The log-normal accelerated failure time model was used to estimate the effects of covariates on survival. Results: 195 patients were included: 48% Non-Hispanic White, 35% Black, 12% Hispanic, and 6% Asian. 59% were males. 81% of patients had stage IV disease. Treatments were as follows: 29 CT, 27 CT-ICI, 15 ICI, and 15 chemoradiation (CRT). 106 (54%) patients did not receive any treatment (NT). CRT was used in 75% of treated stages IIIB/IIIC. Median OS for CT, CT-ICI, ICI, CRT, and NT were 6.6m, 22.6m, 23.6m, 23.4m, and 2.7m, respectively. The log-rank test detected a statistically significant difference in OS between CT-ICI and CT (p < 0.001) and between ICI and CT (p=0.003) but no difference between CT-ICI and ICI (p=1). 66% of patients underwent PDL-1 testing: PDL-1 <1%, PDL-1 1-49%, and PDL-1 ≥ 50% had longer OS with p=0.02, p=0.01 and p<0.001 respectively compared to those not tested. Conclusions: Our study is the first in SNH population, with OS similar to published trials. aNSCLC patients who received first-line CT-ICI or ICI had better survival compared to CT alone regardless of PDL-1 results. We should study the reasons why most patients did not receive any treatment and extend the benefit of ICIs to as many patients with aNSCLC as possible.