P4-291: Anti-Abeta immunotherapy: Quantitative evaluation of acute effects in young TG2576 mice

Abstract

Active immunization with Abeta and passive immunization with anti–Aβ mAbs attenuates Abeta amyloid deposition, behavioral deficits, and other–AD–like pathologies in Alzheimer disease (AD) mouse models. A plethora of non–exclusive hypotheses regarding how anti–Abeta antibodies alter amyloid deposition have been postulated; yet, there is still no consensus as to how either form of Abeta immunotherapy alters Abeta deposition. In this study we have examined the in vivo binding properties, pharmacokinetics, brain penetrance and effects on Abeta levels following a single peripheral dose of an anti–Abeta1–16 mAb (mAb9) to both wild type and young non–Abeta depositing APP mice. Our studies show that the rapid rise in plasma Abeta following mAb administration is attributable to essentially irreversible binding of the anti–Abeta antibody to Aβ and that the rise in plasma Abeta, measured by capture ELISA, peaks in ∼6 hours. Once the mAb:Abeta complex is formed it has a half–life of ∼5 days that is nearly identical to the plasma half–life of the free mAb; indicating that a classic multivalent immune complex is not formed. At most, only a small fraction (0.1% or less) of the mAb enters the brain, and despite the dramatic increase in plasma Abeta, we find no evidence that total brain Abeta levels are significantly altered. Surprisingly, CSF Abeta levels transiently rise, and when mAb9:Abeta complex is directly injected into the lateral ventricles of mice the complex is rapidly cleared from the brain into the plasma where it remains stable. These data provide a framework to evaluate the possible mechanisms of Abeta attenuation mediated by peripheral administration of an anti–Aβ mAb previously shown to effectively attenuate Abeta deposition in long–term studies. When evaluated in conjunction with estimates of the total daily turnover of Abeta such quantitative data make it unlikely that anti–Abeta antibodies are working by binding soluble monomeric Abeta, but instead indicate that the mAbs are either activating non–Fc dependent clearance of Abeta or targeting a low abundance aggregation intermediate.