Blood-brain barrier disruption with implantable ultrasound to enhance paclitaxel delivery: A Phase 1-2 clinical trial in recurrent glioblastoma

Drug therapy in glioblastoma (GBM) frequently fails in patients despite showing good efficacy in preclinical models and in other cancers. Commonly, the reason for this failure is poor drug penetration across the blood- brain barrier (BBB). Pulsed-ultrasound (US) with concomitant injection of intravenous microbubbles transiently disrupts the BBB, enhancing delivery of drugs to the brain. Translation of this approach to humans requires ultrasound waves to bypass the thick human skull. To overcome this, an US that is implanted into skull window has been successfully tested in a Phase 1 clinical trial for recurrent GBM. This study showed BBB disruption, and a longer progression-free survival for patients in whom the BBB was successfully opened for carboplatin delivery. Yet, the effect of US-based BBB disruption on drug concentrations in peri-tumoral human brain remains unknown, which is key for targeting GBM infiltration beyond surgical margins. Paclitaxel (PTX), a microtubule- stabilizing drug, is exquisitely potent against GBM (120-fold more potent than carboplatin and 1400-fold more temozolomide). Studies exploring PTX’s role in human gliomas showed that in peri-tumoral brain, an essential compartment for effective treatment of infiltrative disease, this drug was undetectable. Moreover, cremophor, the solvent used in conventional PTX formulations is neurotoxic. Thus, whereas PTX remains one of the most potent drugs against GBM, it cannot be exploited due to poor BBB penetration and vehicle-related toxicity. We demonstrated that a novel FDA-approved formulation of albumin-bound PTX (ABX) that does contain cremophor, is well tolerated and exhibits better brain penetration than conventional PTX. Following systemic ABX administration into mice, US-based BBB disruption increased PTX brain tissue concentrations by 5-fold, achieving substantially higher levels than IC50 concentrations for most glioma cell lines. Our premise is that PTX will be effective against human GBM if sufficient tumor and brain concentrations are achieved. We hypothesize that US-based delivery of ABX will be tolerated, increase PTX concentrations in peri-tumoral brain, and provide a survival benefit for recurrent GBM patients. To investigate this, we will conduct a Phase 1-2 trial of US-based delivery of ABX for recurrent GBM patients to determine safety and early-signs of efficacy of this therapy (Aim 1). To maximize chances of efficacy, we will use a 2nd generation implantable US device that covers a 9-fold broader sonication volume than previous prototypes, and use ABX, a drug that is far more potent than the previously studied carboplatin. We will measure PTX concentrations in peri-tumoral human brain following US- based BBB disruption (Aim 2). We will also perform MRI-based determination of whether the field of BBB disruption is associated with local disease control and prevents progression (Aim 3). These studies will 1) determine the safety of US-based BBB disruption with concomitant ABX infusion, 2) provide clinical and radiographic assessment of efficacy, and 3) quantify the effect of BBB disruption on PTX brain concentrations.