Osmotic blood-brain barrier disruption with mannitol followed by methotrexate, rituximab, and carboplatin in treating patients with newly diagnosed primary central nervous system lymphoma

2077 Background: Osmotic blood-brain barrier disruption (BBBD) uses intra-arterial (IA) mannitol to open the neuro-vascular unit to enhance delivery of therapeutic agents to the brain. We report the results of a multi-institutional, single arm, phase 2 study evaluating the efficacy and safety of BBBD-enhanced chemotherapy in combination with immunotherapy in patients with newly diagnosed primary central nervous system lymphoma (PCNSL). PCNSL is a rare non-Hodgkin’s lymphoma that is limited to the brain, cerebrospinal fluid, spinal cord and/or the vitreoretinal compartment. Methods: Non-immunocompromised subjects with newly diagnosed PCNSL received rituximab IV over 5 hours on day 1. On days 2 and 3, under general anesthesia in the angiography suite, subjects received IA mannitol immediately followed by IA methotrexate over 10 minutes, and then IA carboplatin over 10 minutes. Subjects received sodium thiosulfate IV for hearing protection at 4 and 8 hours after each carboplatin dose. Treatment was repeated every month for up to 12 courses in the absence of disease progression or unacceptable toxicity. Per institutional recommendations, further accrual was terminated during the COVID-19 pandemic. Contrast MRI was used for response assessment based on IPCG criteria. Results: Twenty subjects (10 F, 10 M, Median KPS = 80; range 40-100) received a median of 8 cycles at 3 US institutions. Sixteen radiographic complete responses (80%) and 1 partial response were documented, 3 subjects were not evaluable (2 withdrew consent and 1 unrelated death). Median PFS was 55.2 months (95% CI :10.3 – not reached) and Median Overall survival was 82.8 months (95% CI :19.8-145.8). Fourteen patients (82% of evaluable patients) had grade 3 or 4 adverse events (AE). Grade 4 AEs (n = 14 in n = 7 patients) included neurologic toxicities (ischemia, cervical cord inflammation, neuropathy) and blood/bone marrow toxicities (lymphopenia, neutropenia, granulocytopenia). There were no treatment-related deaths. Conclusions: We demonstrate the feasibility of performing BBBD in a multi-institutional setting with excellent PFS and OS in patients without whole brain radiation or autologous transplant. Our CR rates, PFS and OS are comparable to other IV high-dose MTX regimens, with acceptable toxicity profile and no documented treatment-related deaths. Clinical trial information: NCT00293475 .