Biomarker Discovery and Outcomes for Comprehensive Cell-Free Circulating Tumor DNA Versus Standard-of-Care Tissue Testing in Advanced Non-Small-Cell Lung Cancer.

Abstract

Treatment guidelines for advanced non-small-cell lung cancer (aNSCLC) recommend broad molecular profiling for targeted therapy selection. This study prospectively assessed comprehensive next-generation sequencing (NGS) of cell-free circulating tumor DNA (cfDNA) compared with standard-of-care (SOC) tissue-based testing to identify guideline-recommended alterations in aNSCLC. Patients with treatment-naïve aNSCLC were tested using a well-validated NGS cfDNA panel, and results were compared with SOC tissue testing. The primary objective was noninferiority of cfDNA vs. tissue analysis for the detection of two guideline-recommended biomarkers (<i>EGFR</i> and <i>ALK</i>) and an additional six actionable biomarkers. Secondary analyses included tissue versus cfDNA biomarker discovery, overall response rate (ORR), progression-free survival (PFS) to targeted therapy, and positive predictive value (PPV) of cfDNA. The primary objective was met with cfDNA identifying actionable mutations in 46 patients versus 48 by tissue (<i>P</i> < .05). In total, 0/186 patients were genotyped for all eight biomarkers with tissue, compared with 90.8% using cfDNA. Targetable alterations or <i>KRAS</i> were identified in 80.7% when cfDNA was used first versus 57.1% when tissue was used first. PPV for cfDNA-detected <i>EGFR</i> was 100.0% (25/25). ORR and PFS in patients receiving targeted therapy based on tissue or cfDNA were similar to those previously reported. This prospective study confirms a previous report that comprehensive cfDNA testing is noninferior to SOC tissue testing in detecting aNSCLC-recommended biomarkers. Furthermore, cfDNA-based first-line therapy produced outcomes similar to tissue-based testing, demonstrating the clinical utility of comprehensive cfDNA genotyping as the initial genotyping modality in patients with treatment-naïve aNSCLC when tissue is insufficient or when all actionable biomarkers cannot be rapidly assessed.