Impact of social determinants of health on genomic testing among patients with advanced/metastatic non-small cell lung cancer.

Guideline-concordant care for non-small cell lung cancer (NSCLC) is highly dependent on genomic testing to inform targeted therapy selection. Studies have reported persistent gaps in testing associated with racial and ethnic factors. Understanding the impact of social determinants of health (SDOH) can help define how to approach closing these gaps. In this study, we evaluated factors associated with completion of genomic testing among patients with NSCLC across SDOH factors.

Adults (≥18 years at diagnosis) with advanced/metastatic NSCLC diagnosed between Jan 2021 and Jan 2026 were identified in ConcertAI’s Patient360, a U.S. de-identified database integrating health records, claims, genomics, and SDOH. Molecular testing performed for EGFR mutation, ALK fusion, and ROS1 fusion at any point during the treatment was captured and available for analysis. SDOH were available from Patient360’s pre-derived SDOH data elements. 

Of 81,950 patients with advanced/metastatic NSCLC included in this analysis, 10.2% were Black, and 4.3% were Hispanic. Biomarker results were available for 47.5%. Overall testing rates in this cohort for EFGR, ALK, and ROS1 were 44.6%, 42.3%, and 38.3%, respectively. There were no differences in testing rate for EGFR, ALK, ROS1, and combined based on race (White, Black, Other) or ethnicity (Hispanic and non-Hispanic) (p>0.05). The distribution of SDOH variables is presented in Table 1. Across evaluated SDOH domains, testing rates were not different within categories of education, marital status, household income, insurance, and urban/rural status. Biomarker testing (%) was different across occupations (White collar: 54.5; Blue collar: 55.8; Retired: 58.6; Other: 57; Unknown: 52.7; p=0.015).

This analysis of EHR data from US practices did not demonstrate differences in genomic testing based on SDOH factors except for the patient’s occupation. This could reflect systems-based practices in institutions represented in the dataset. Future work can focus on time to biomarker testing and use of appropriate targeted treatments.