A retrospective observational analysis examining liquid biopsies with tissue biopsies in patients with advanced non-small cell lung cancer.

Abstract

e15049 Background: Non-small cell lung cancer (NSCLC) is genetically diverse, harboring mutations targetable with FDA-approved therapies. Identifying actionable mutations through tissue- or blood-based next-generation sequencing (BB-NGS) is critical. Tissue-based NGS (TB-NGS), the gold standard, involves invasive sampling and long turnaround times (2–4 weeks), delaying treatment. BB-NGS, a less invasive technique, analyzes circulating tumor DNA from blood samples and provides faster results, though it is newer and less widely used. Methods: This single-institution, retrospective, IRB-approved study analyzed the concordance of BB-NGS and TB-NGS in patients with metastatic NSCLC, as well as the time from BB-NGS to treatment initiation. Eligible patients with metastatic NSCLC were identified from Northwell Health Cancer Institute's records between January 1, 2016, and March 31, 2020. Patients included received both BB-NGS and TB-NGS within one month of each other. Data collection included demographics, biopsy results, and clinical timelines. Concordance analysis was performed to compare mutation detection rates and descriptive statistics were used to calculate time to treatment initiation. Results: Concordance analysis revealed that the two modalities identified the same actionable mutation in 57.1% of cases but in 42.9% of cases it was only identified by one modality. BB-NGS identified actionable mutations in 11 patients (26.2%) that TB-NGS did not detect, enabling treatment initiation using a non-standard method. Of note, five of these mutations found only on BB-NGS were in KRAS. Conversely, TB-NGS identified mutations missed by BB-NGS in seven cases (16.7%). Of the 42 patients, 13 were initiated on treatment based on BB-NGS due to its faster turnaround time or its ability to detect mutations missed by TB-NGS. The median time to treatment initiation from blood collection for these 13 patients was 14 days (range: 7–24 days). Conclusions: Our findings highlight the complementary value of combining BB-NGS and TB-NGS to optimize treatment strategies. Mutations not detected on TB-NGS were likely due to heterogeneity or suboptimal tissue sampling and processing. BB-NGS, though not as widely performed, may facilitate faster treatment initiation, particularly when TB-NGS results are delayed or insufficient . Future directions include expanding the patient cohort, analyzing mutation-specific detection rates, and evaluating clinical outcomes associated with incorporating BB-NGS into routine care. Concordance between BB- and TB- NGS in detecting actionable mutations. TB+ TB- BB + 18 11 BB - 7 6 (AM = actionable mutation; BB = blood-based; TB = tissue-based).