148: PK/PD Modeling of Progranulin Elevation in Blood & CSF to Support AL101 Dose Selection & Phase 2 Study Design

Statement of Purpose: A model informed approach, within an integrated PK/PD modelling and simulation framework, is applied to AL101 Phase 1 results to accelerate its clinical development. PK/PD model and simulation tools can be used in the well-known “learn and confirm” paradigm to predict PD responses and design clinical trials to further inform more robust decision making.

Description of Methods & Materials: Progranulin (PGRN) regulates cell growth, survival, repair and inflammation with a major CNS role in regulation of lysosomal function and microglia responses. Heterozygous loss-of-function progranulin gene (GRN) mutations and GRN variants that result in PGRN deficiency are associated with a higher risk for a variety of neurodegenerative conditions such as Alzheimer's disease (AD), Frontotemporal dementia (FTD) and Parkinson’s disease (PD). AL101 is an anti-sortilin monoclonal antibody that elevates and restores plasma and CSF PGRN levels by blocking sortilin, a PGRN degradation receptor, potentially slowing progression in neurodegenerative diseases. In a Phase 1 study, single and repeated administration of AL101 to healthy participants caused saturable but dose-dependent increases of PGRN in plasma and CSF, with levels sustained for a longer duration at higher dose levels. An integrated population PK/PD model was developed using AL101 and PGRN concentrations in serum, plasma and CSF from a Phase 1 study in healthy volunteers. AL101 PK profiles, defined by a 3-compartment model including the CSF compartment and a slow perfusion peripheral compartment, were linked directly to PGRN turnover through an E_max model. Body weights and baseline PGRN levels were identified as statistically significant covariates.

Data & Results: The PK/PD model well described the target-mediated drug disposition effect due to the AL101-sortilin complex formation. The saturation of the effect in terms of progranulin increase from baseline is predicted to happen at AL101 concentrations of approximately 300 ng/mL and approximately 30 ng/mL in plasma and CSF respectively. The PK/PD model was qualified and simulations were run for predicting AL101 PK and PGRN concentrations following different dose levels, frequencies of administration and evaluating the difference between body weight adjusted and flat dosing. The simulated PK/PD profiles are suitable for chronic neurological conditions, such as AD and PD where PGRN deficiency is implicated. These simulations were used to help determine potential Phase 2 dose levels by exploring the dose/exposure-response relationship based on the estimated average PGRN elevation in patients.

Interpretation, Conclusion or Significance: The PK/PD model well described the target-mediated drug disposition effect due to the AL101-sortilin complex formation. The saturation of the effect in terms of progranulin increase from baseline is predicted to happen at AL101 concentrations of approximately 300 ng/mL and approximately 30 ng/mL in plasma and CSF respectively. The PK/PD model was qualified and simulations were run for predicting AL101 PK and PGRN concentrations following different dose levels, frequencies of administration and evaluating the difference between body weight adjusted and flat dosing. The simulated PK/PD profiles are suitable for chronic neurological conditions, such as AD and PD where PGRN deficiency is implicated. These simulations were used to help determine potential Phase 2 dose levels by exploring the dose/exposure-response relationship based on the estimated average PGRN elevation in patients.

Disclosures: MG, AG, RL, and DR are employees of GSK and may have an equity interest in GSK. AJ and CH are employees of PPD, part of Thermo Fisher Scientific, and may have an equity interest in Thermo Fisher Scientific. MW was an Alector employee at the time of study conception and may have an equity interest in Alector, Inc. LP and BB are employees of Alector LLC and may have an equity interest in Alector, Inc.