Purpose: Cerebrospinal fluid (CSF) has revealed the unique genetic characteristics of leptomeningeal metastasis (LM) from non-small cell lung cancer (NSCLC). However, the research in this area is still very limited.
Methods: Patients with LM from NSCLC (n = 80) were retrospectively analyzed. Circulating tumor DNA (ctDNA) in CSF was tested by next-generation sequencing (NGS), with paired extracranial tissue or plasma samples included for comparison. An independent non-LM cohort (n = 100) was also analyzed for comparative purposes. Clinical outcomes were compared with Kaplan-Meier log-rank test and Cox proportional hazards methodologies.
Results: An overwhelming 93.8% of patients carried druggable mutations in NSCLC LM, with EGFR (78.8%) being the most prevalent. Notably, 4 patients who tested negative for driver genes in extracranial samples surprisingly showed EGFR mutations in their CSF and subsequently benefited from targeted therapy. There was a clear difference in genetic profiles between CSF and extracranial samples, with CSF showing more driver gene detections, increased Copy Number Variations (CNVs), and varied resistance mechanisms among individuals. Abnormalities in cell-cycle regulatory molecules were highly enriched in LM (50.9% vs 31.0%, p = 0.017), and CDKN2A/2B deletions were identified as an independent poor prognostic factor for LM patients, with a significant reduction in median OS (p = 0.013), supported by multivariate analysis (HR 2.63, 95% CI 1.32-5.26, p = 0.006).
Conclusions: CSF-based ctDNA analysis is crucial for detecting and characterizing genetic alterations in NSCLC LM. The distinct genetic profiles in CSF and extracranial tissues emphasize the need for personalized treatment approaches.
Keywords: Cerebrospinal fluid; Circulating tumor DNA; Leptomeningeal metastasis; NGS; Non-small cell lung cancer.
© 2023. The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.