Serial monitoring of cerebrospinal fluid in patients with leptomeningeal metastases in lung cancer via the Ommaya reservoir as a predictive indicator of therapeutic efficacy and clinical prognosis.

Leptomeningeal metastasis (LM) is a severe neurological complication in patients with lung adenocarcinoma (LUAD), typically indicating terminal stage and poor survival. Optimal treatments for patients with LM-LUAD rely on accurately predicting treatment efficacy and dynamically adjusting regimens. Here, we collected 301 cerebrospinal fluid (CSF) samples from 125 patients with LM-LUAD via the Ommaya reservoir at baseline and during the treatment course to systematically explore the clinical significance of CSF circulating tumor DNA (ctDNA). At baseline, the positive detection rate of CSF ctDNA (99.2%) was significantly higher than that of plasma ctDNA (57.6%). By evaluating the association between varying magnitudes of ctDNA dynamic changes and clinical responses at each time interval during the treatment, we discovered that when the change in ctDNA reached 20%, the consistency with clinical assessment could exceed 90%. Moreover, we further analyzed the significance of different values of CSF ctDNA mutation frequencies at baseline and various time points, as well as the trends of ctDNA changes relative to the baseline levels, in predicting patient prognosis. Accordingly, we constructed a CSF ctDNA-based prognostic map and developed a multi-feature model that exhibited outstanding performance in predicting and assessing the clinical responses of patients. To the best of our knowledge, this work represents the most systematic large-scale study on CSF ctDNA-based dynamic monitoring of patients with LM-LUAD, which is of great significance for guiding clinical treatments to extend patient lives.