Real-world clinical utility of liquid biopsy in NSCLC in a community oncology practice: Concordance, discordance, and impact on precision therapy.

e15068 Background: Precision oncology in NSCLC relies on tissue NGS to identify actionable mutations, but in community oncology, tissue limitations—including insufficient samples, technical failures, and delays—often restrict access to targeted therapies. Liquid biopsy (LB), analyzing circulating tumor DNA, provides a minimally invasive complementary approach. Although tissue-LB concordance is well described in clinical trials, real-world data from community oncology practices remain limited. Evaluating feasibility, concordance, and clinical impact of LB in this setting is critical to optimize precision therapy. Methods: We retrospectively analyzed 47 NSCLC patients (median age 65; 18F/29M; 83% smokers) diagnosed in 2024 at a community oncology practice. Tissue NGS was attempted in all patients; 38 returned results, with 5 QNS and 4 missing/uncertain. LB was performed in 14 patients, including 10 with paired tissue results. Actionable alterations assessed included KRAS, MET exon 14 (METex14), ROS1, ERBB2, BRAF, and co-mutations. Endpoints were tissue NGS feasibility, mutational yield, and tissue-LB concordance. Results: Tissue NGS returned evaluable results in 38/47 patients (81%), with 5 QNS (11%) and 4 missing/uncertain (9%), highlighting real-world attrition. Among evaluable tissue, 14 patients (37%) harbored actionable mutations: KRAS (n = 4), KRAS+ROS1 (n = 1), KRAS+METex14 (n = 1), METex14 (n = 3), ROS1 (n = 2), ERBB2 (n = 2), BRAF (n = 1). LB detected actionable mutations in 6/14 patients (43%), including 2 METex14 alterations in patients without tissue NGS, enabling targeted therapy. Among 10 paired tissue-LB cases, concordance was 80% (8/10). Discordances (20%) included one KRAS-positive LB with negative tissue, and one ALK-positive LB versus BRAF-positive tissue, highlighting intra-patient heterogeneity and potential clinical impact on therapy selection. Conclusions: In this community oncology cohort, tissue NGS demonstrated ~19% attrition. LB is a highly effective complementary tool, rescuing actionable mutations in patients without evaluable tissue. Discordances between tissue and LB (20%) underscore the importance of a combined testing approach and multidisciplinary review before initiating targeted therapy. These results highlight the real-world clinical utility of circulating tumor DNA analysis in expanding access to precision therapy in NSCLC outside academic centers. However, there is a need for further larger studies to validate its clinical benefit and optimize precision therapy in community practice.