Baseline CSF-Detected Driver Status and Outcomes in Lung Adenocarcinoma With Leptomeningeal Metastasis Treated With Intraventricular Pemetrexed

Abstract Background Baseline cerebrospinal fluid (CSF) genomic findings may reflect both leptomeningeal tumor biology and access to genotype-matched therapy. Their prognostic meaning within cohorts receiving intraventricular pemetrexed remains uncertain. Methods This single-center retrospective cohort included 137 patients with lung adenocarcinoma and LM who underwent baseline CSF next-generation sequencing and received Ommaya-based intraventricular pemetrexed. Patients were classified as CSF EGFR-positive, other oncogenic driver-positive, or having no detectable actionable driver.Intracranial progression-free survival (iPFS) and overall survival (OS) were calculated from the date of LM diagnosis. Survival outcomes were estimated using Kaplan–Meier analysis and Cox proportional hazards models. Results The cohort comprised 102 CSF EGFR-positive patients, 15 with other oncogenic drivers, and 20 with no detectable actionable driver. Over a median follow-up of 10.4 months, 89 iPFS events and 59 deaths occurred. Median iPFS was 6.6 months and median OS was 12.9 months. CSF-detected molecular subgroup was associated with both endpoints after baseline adjustment (global P < 0.001 for each). Versus the EGFR-positive group, the other-driver group had shorter iPFS (adjusted HR, 2.283; 95% CI, 1.166–4.467), but its OS difference was not significant. The no-detectable-driver group had shorter iPFS (adjusted HR, 2.704; 95% CI, 1.525–4.795) and OS (adjusted HR, 4.257; 95% CI, 2.192–8.266). Estimates were similar with an alternative time origin. Conclusions In this selected cohort, CSF-detected driver status was associated with iPFS and OS under treatment as observed. The findings may reflect tumor biology and matched-treatment access and should not be interpreted as differential sensitivity to intraventricular pemetrexed.