Efficacy and Safety of Intraventricule Pemetrexed Disodium Administered Via Ommaya Reservoir

This is a open-label, multi-center prospective observation study for the efficacy and safety of intraventricle pemetrexed disodium via ommaya reservoir in the treatment of leptomeningeal metastasis with lung cancer who have failed at least one targeted therapy. In detail: At least the treatment failure was after third-generation EGFR-TKIs in EGFR-mutated lung cancer; or at least the treatment failure was after second-generation ALK-TKIs in ALK-mutated lung cancer; or at least the treatment failure was after one-line of targeted-TKIs in ROS1-mutated non-squamous non-small lung cancer.

Participants were randomizedly assigned in cohort 1 who will be given pemetrexed 20mg every 24 hours for 72 hours every 2 weeks,or in cohort 2 will be given pemetrexed 30mg once a week. It was designed for two stages on treatment course. Induction therapy: Efficacy evaluation every 2 cycles, and confirm the efficacy at the next cycle. If at the next cycle, the efficacy result changed (such as SD or PR after initial PD; or PD after initial SD or PR; the second time of efficacy assess was required）. If the CSF cytology at the time of the efficacy evaluation was negative, one more cycle was required to confirm the CSF cytology. Consolidation: If the efficacy is remission(including complete remission, obvious remission, or partial remission) or stable disease for initial timepoint and the timepoint of the confirmed evaluation; then the patient will be advanced to the stage of consolidation treatment. Participants in cohort 1 who will be given pemetrexed 20mg every 24 hours for 72 hours every 3 weeks. Participants in cohort 2 will be given pemetrexed 30mg every 3 weeks. Until the toxicity is intolerable, or disease progression. Cross over： Compared the first 4 enrolled cases in each group; the comprehensive evaluation(including efficacy, cytology negative, toxicity) in better group(50% higher efficacy or 50% less toxicity) will be the following cohort, to which the other one will crossover.