BACKGROUND: DIPG remains a uniformly fatal diagnosis with most children dying of disease between six months and two years from initial diagnosis. Radiation therapy remains the only intervention that alters the natural history of the disease, but is not curative. Given the frequent imaging feature of a non-contrast enhancing mass at diagnosis, the extent of systemic delivery of chemotherapeutics to the pons is uncertain. Attempts have been made to circumvent this issue by various methods including convention-enhanced delivery utilizing catheters inserted into the pons, blood-brain barrier disruption, and intra-arterial (IA) administration of chemotherapy into the vertebral circulation. This pilot study explores the safety, tolerability, and initial efficacy of highly selective intra-arterial administration of melphalan via the basilar artery at the time of clinical progression following initial radiotherapy for DIPG. METHODS: Subjects with a radiologic or biopsy proven diagnosis of DIPG are eligible at the time of symptom and/or radiographic progression in the months following initial irradiation. Consensus of the multidisciplinary Pediatric Neuro-Oncology conference faculty is required. Subjects with a documented hypercoagulable disorder or vasculopathy are not eligible, nor are subjects who have been re-irradiated. Two cycles of IA chemotherapy are planned 4 weeks apart. Under general anesthesia, the vertebral artery (right or left) is catheterized with a 4-French guide catheter. Following a baseline angiogram, a 1.5-French microcatheter is advanced into the mid basilar artery. After angiographic confirmation of positioning, IA delivery of 4 or 6 mg of melphalan is delivered over 30 minutes. Following completion of the procedure, subjects are observed overnight in the Pediatric Intensive Care prior to discharge the following day. Subjects are monitored for toxicity between, and following, IA administrations. RESULTS: To date, 2 children, of a planned cohort of 5 children, have been treated with IA melphalan. Both children tolerated the actual IA administrations without any procedural associated toxicity including the absence of stroke, hemorrhage, or other technical complications. The first child received one of two planned IA administrations but clinically deteriorated shortly before the second IA infusion was planned. The second child received both planned IA administrations and is in early follow-up. CONCLUSIONS: Highly selective IA administration of melphalan to the basilar artery is feasible and acutely tolerated. Updated experience and follow-up with enrolled patients will be presented. SECONDARY CATEGORY: Clinical Neuro-Oncology.