Too late or too little? Potential for Alzheimer's disease passive immunotherapy via targeted intracerebroventricular delivery

Abstract

Listen

Translate article

Anti-amyloid-beta (Aβ) passive immunotherapy holds promise as a disease-modifying therapy for AD, despite failures to prevent cognitive and functional decline in recent phase III clinical trials. Treatment with two independent anti-Aβ antibodies showed impact on AD biomarker levels, and small cognitive and functional benefit in a subset of mild AD patients. Leading hypotheses for lack of efficacy confront issues with therapy being “too late”, where administration occurs at a disease state when the brain is already damaged beyond measurable repair, and “too little”, where insufficient amounts of antibody reach the required central nervous system exposure. The presence of the blood brain barrier combined with dose-limiting safety margins pose major challenges for systemic delivery of anti-Aβ antibodies to reach clinically significant efficacy. Device-mediated continuous intracerebroventricular (ICV) delivery of antibodies directly into the cerebrospinal fluid (CSF) helps circumvent this issue. We characterize the pharmacokinetics, pharmacodynamics, tolerability and brain uptake following continuous unilateral ICV delivery of humanized anti-Aβ antibody. ICV antibody delivery is achieved in non-human primates (NHPs) using the fully implantable SynchroMed®II infusion system, and in Tg2576 transgenic mice using Alzet® pumps. We have previously demonstrated prolonged ICV infusion of anti-Aβ antibodies dose-dependently reduce plaque burden, astrogliosis, and dystrophic neurites at doses 10- to 50-fold lower than used with systemic delivery of the same antibody in the Tg2576 transgenic mouse model. We have also shown significant bilateral reduction of Aβ42-positive plaques in the cortex of aged Stump-tailed macaques after 3 months of ICVanti-Aβ antibody infusion. Here, we show continuous ICV delivery of low antibody doses (0.3mg/day) in NHPsachieves anti-Aβ antibody levels of ∼3–5μg/mL in the CSF. We also characterize the brain-region specific anti-Aβ antibody distribution patterns following 3-month ICV delivery. Our pilot studies employing s table- i sotope- l abeled k inetics (SILK™) methodology are indicative of antibody engagement of the target (Aβ) in the CSF. Overall, our data continue to support device-mediated targeted infusion as a promising avenue to address the “too little” issue by delivering anti-Aβ antibodies directly into the CSF and maximizing therapy effects in the pre-clinical proof-of-concept phase and potentially in the clinical management of AD.