Himanshu Naik1, Lauren Stevenson1, Chase Shen1, Catherine Barbey2, Cristina Musselli1, Mirjam N. Trame3, Adam Meyers1, Ivan Nestorov1, Parul Gulati1, Dania Rabah1, and Nathalie Franchimont1
1Biogen, Cambridge, MA, 2Biogen, Zug, Switzerland,3Center for Pharmacometrics and Systems Pharmacology, University of Florida, Orlando, FL
Objectives: BIIB059 (anti-BDCA2) is a humanized, immunoglobulin G1 (IgG1) monoclonal antibody currently under development for treatment of SLE including CLE. BIIB059 is targeted against BDCA2, a receptor expressed on the surface of the human and non-human primate plasmacytoid dendritic cells ( pDCs). The objective of this work was to select dosing regimens for BIIB059 in the ongoing phase 2 study (NCT02847598).
Method: Dose selection was performed using a stepwise approach including in-vitro and in-vivo data as follow: 1) Determination of in-vitro EC90 and IC90 values for human BDCA2 internalization and for interferon alpha (IFNα) inhibition in human whole blood assays; 2) Establishment of the relationship between EC90 and IC90 values for BDCA2 internalization and IFNα inhibition based upon in-vitro data obtained from 10 healthy human donors; 3) Estimation of in-vivo EC90 value for BDCA2 internalization using the population PK-PD model developed based upon phase 1 PK and PD (BDCA2 internalization) data in healthy volunteers and SLE subjects (NCT02106897); 4) Prediction of in-vivo IC90 value for IFNα inhibition using the in-vivo estimated EC90 value of BDCA2 internalization in humans and the in-vitro relationship established between BDCA2 and IFNα; 5) The final PK-PD model was used to perform simulations and select doses for inclusion in the phase 2 study which are expected to keep the concentration of BIIB059 above EC90 of BDCA2 internalization and IC90 IFNα inhibition.
Results: The estimated in-vitro EC90 and IC90 values for BDCA2 internalization and IFNα inhibition based upon average of 10 human donors were 0.051 μg/mL and 0.67 μg/mL, respectively. The relationship between BIIB059 concentration and BDCA2 was well characterized by a two compartment PK model with linear and non-linear elimination and an indirect response pharmacodynamic model with stimulation of Kout (rate of elimination). The estimated in-vivo EC90 value for BDCA2 internalization from the population PK-PD model was 0.9 µg/mL. The estimated in-vivo IC90 IFNα inhibition based upon in-vitro relationship between the EC90 of BDCA2 and IC90 of IFNα and in-vivo EC90 values estimated based upon the PK-PD model in SLE patient was 11.7 µg/mL. The doses selected for the phase 2 study were: 50, 150 and 450 mg SC Q4W as based upon simulations using the final PK model.
- The low dose of 50 mg SC Q4W is expected to achieve plasma concentrations sufficient to maintain 90% BDCA2 internalization (EC90 = 0.9 µg/mL), for the duration of the dosing interval.
- The middle dose of 150 mg SC Q4W is expected to achieve plasma concentrations similar to or in excess of the calculated IC90 for IFNα (11.7 µg/mL) for the majority of the dosing interval.
- The top dose of 450 mg SC Q4W is expected to achieve Cmin levels similar to 3-fold of the calculated IC90 for IFNα inhibition (i.e. >35.1 µg/mL)
The safety margin compared to 450 mg SC Q4W dose based upon 6 month toxicological studies in cynomolgus monkeys is 25-fold and 48-fold for AUC and Cmax, respectively. In addition, the exposure of 450 mg SC Q4W is also expected to be lower than 20 mg/kg IV dose administered in phase 1 study and hence the selected doses for phase 2 study are considered safe and tolerable.
Conclusion: In-vivo and in-vitro data in combination with pharmacometric approaches allowed the selection of phase 2 doses for BIIB059.
Reference: PAGE 27 (2018) Abstr 8769 [www.page-meeting.org/?abstract=8769]
Poster: Drug/Disease Modelling - Endocrine