Inna Vainshtein

PK-PD modeling of protein drugs: implications in assay development

Pharmacokinetic-pharmacodynamic (PK-PD) modeling is an integral part of the preclinical and clinical development of protein drugs. Bioanalytical data from appropriately selected and well-characterized PK and PD biomarker assays can be incorporated into mechanistic PK-PD models and allow a quantitative relationship between protein drug exposure, target modulation, and biochemical, physiological and pathophysiological effects to be established. The selection of PD biomarkers that assess target engagement and modulation in the extracellular milieu and downstream cellular effects can provide proof-of-mechanism and define the magnitude and duration of target modulation following drug administration. The PK-PD data can provide an important link between magnitude of target modulation and clinical efficacy and safety outcomes, and guide the selection of doses and dosing schedules for clinical trials. In this article, approaches to the selection and development of fit-for-purpose, PK and PD assays for protein drugs are reviewed, and the applications of the assay results in PK-PD models are discussed.

Pharmacogenomics and Translational Simulations to Bridge Indications for an Anti-Interferon-α Receptor Antibody

A type I interferon (IFN) gene signature shared by systemic lupus erythematous (SLE) and systemic sclerosis (SSc) was used to evaluate an anti‐type I IFN‐α receptor (IFN‐αR) monoclonal antibody, MEDI‐546, in a phase I trial in SSc. MEDI‐546 suppressed IFN signature in blood and skin of SSc patients in a dose‐dependent manner. To bridge clinical indications to SLE, we developed a model incorporating (i) pharmacokinetics (PK) and pharmacodynamics (PD) in SSc patients, (ii) internalization kinetics of MEDI‐546/IFN‐αR complex, and (iii) the different IFN signatures between SSc and SLE. Simulations predicted that i.v. administration of MEDI‐546 at 300‐ or 1,000‐mg monthly doses could suppress IFN signature in blood to levels of healthy subjects in 53 and 68% of SLE patients, respectively. An innovative approach utilizing a novel biomarker characterized the PD of MEDI‐546 by modeling and simulation and allowed rapid progression of MEDI‐546 from a phase I study in SSc to a randomized, multiple‐dose phase II trial.

Mechanistic modeling of antigen sink effect for mavrilimumab following intravenous administration in patients with rheumatoid arthritis.

Mavrilimumab is a fully human monoclonal antibody that binds to granulocyte‐macrophage colony stimulating factor receptor α (GM‐CSFRα) with high affinity and specificity and has potential application in various inflammatory diseases. The objective of this investigation was to develop a mechanistic population model to characterize the pharmacokinetics of mavrilimumab, the GM‐CSFRα–mediated clearance, and receptor occupancy following single intravenous dosing to patients with rheumatoid arthritis. The internalization rate of mavrilimumab—GM‐CSFRα complex was fixed to a value determined from quantitative confocal fluorescent imaging. The estimated typical first‐order clearance and the central and peripheral distribution volumes were 3.79 mL/kg/d, 39.6 mL/kg, and 50.3 mL/kg, respectively. The systemic GM‐CSFRα expression level was estimated to be 0.0782 nM, and the equilibrium dissociation constant (0.103 nM) was in good agreement with the monovalent affinity determined by surface plasmon resonance. By fitting to the observed pharmacokinetic data, the mechanistic model predicted that systemically greater than 90% GM‐CSFRα blockade by mavrilimumab was achieved and maintained up to 4, 7, and 11 weeks following single 1‐, 3‐, and 10‐mg/kg administrations, respectively. Posterior visual predictive check and bootstrapping suggest that the mechanistic model is reasonably robust and can be used to predict mavrilimumab exposure under various scenarios for future clinical trial design.

Receptor occupancy assessment by flow cytometry as a pharmacodynamic biomarker in biopharmaceutical development

Receptor occupancy (RO) assays are designed to quantify the binding of therapeutics to their targets on the cell surface and are frequently used to generate pharmacodynamic (PD) biomarker data in nonclinical and clinical studies of biopharmaceuticals. When combined with the pharmacokinetic (PK) profile, RO data can establish PKPD relationships, which are crucial for informing dose decisions. RO is commonly measured by flow cytometry on fresh blood specimens and is subject to numerous technical and logistical challenges. To ensure that reliable and high quality results are generated from RO assays, careful assay design, key reagent characterization, data normalization/reporting, and thorough planning for implementation are of critical importance during development. In this article, the authors share their experiences and perspectives in these areas and discuss challenges and potential solutions when developing and implementing a flow cytometry‐based RO method in support of biopharmaceutical drug development.

Interference in immunoassays to support therapeutic antibody development in preclinical and clinical studies

During preclinical and clinical studies, immunoassays are used to measure the concentration of the therapeutic antibody, anti-drug antibodies and soluble protein biomarkers. The reliability of these assays is crucial since the results are routinely used for safety assessment and dose selection. Furthermore, soluble protein biomarkers can provide information about target engagement, proof of mechanism, proof of principle and prediction of response. Study samples mostly consist of complex matrices that can exhibit considerable interference, resulting in inaccurate measurements. This perspective discusses the source of interference and strategies to mitigate or eliminate interference in immunoassays used during preclinical and clinical drug development of drugs with a focus on the development of therapeutic antibodies.

Quantitative Measurement of the Target-Mediated Internalization Kinetics of Biopharmaceuticals

PurposeMeasurement of internalization of biopharmaceuticals targeting cell surface proteins can greatly facilitate drug development. The objective of this study was to develop a reliable method for determination of internalization rate constant (kint) and to demonstrate its utility.MethodsThis method utilized confocal imaging to record the internalization kinetics of fluorescence-tagged biopharmaceuticals in live-cells and a quantitative image-analysis algorithm for kint determination. Kint was incorporated into a pharmacokinetic-pharmacodynamic (PK-PD) model for simulation of the drug PK profiles, target occupancy and the displacement of endogenous ligand.ResultsThe method was highly sensitive, allowing kint determination in cells expressing as low as 5,000 receptors/cell, and was amenable to adherent and suspension cells. Its feasibility in a mixed cell population, such as whole blood, was also demonstrated. Accurate assessment of the kint was largely attributed to continuous monitoring of internalization in live cells, rapid confocal image acquisition and quantitative image-analysis algorithm. Translational PK-PD simulations demonstrated that kint is a major determinant of the drug PK profiles, target occupancy, and the displacement of endogenous ligand.ConclusionsThe developed method is robust for broad cell types. Reliable kint assessment can greatly expedite biopharmaceutical development by facilitating target evaluation, drug affinity goal setting, and clinical dose projection.

Pharmacokinetic and Pharmacodynamic Comparability Study of Moxetumomab Pasudotox, an Immunotoxin Targeting CD22, in Cynomolgus Monkeys

Moxetumomab pasudotox is an immunotoxin currently being investigated in patients for the treatment of CD22-expressing B-cell malignancies. A single-cycle pharmacokinetic (PK)-pharmacodynamic (PD) study was conducted in cynomolgus monkeys for PK comparability assessment and population PK-PD modeling after major manufacturing process and site changes. Primates were randomized by body weight and baseline CD22 lymphocyte counts to receive intravenous administrations of 1 mg/kg moxetumomab pasudotox (n = 12/group) on Days 1, 3, and 5. PK and B-lymphocyte count data were modeled using a population approach. The 90% confidence intervals of the geometric mean ratios of PK exposure were within the 80%-125% range. The B lymphocytes were depleted to a similar extent, and the immunogenicity incidences were similar across the two groups. The B-cell depletion was described by a novel lifespan model in which moxetumomab pasudotox induced random destruction of B cells in each aging compartment. The endogenous de novo influx from bone marrow was subject to a negative feedback mechanism. The estimated B cell apparent lifespan was 51 days. Covariate analysis confirmed that the manufacturing change had no impact on PK or PD of moxetumomab pasudotox. Results from this study supported continued clinical investigation of moxetumomab pasudotox using the new material.

Characterisation of anifrolumab, a fully human anti-interferon receptor antagonist antibody for the treatment of systemic lupus erythematosus

Objective We investigated the mechanistic and pharmacological properties of anifrolumab, a fully human, effector-null, anti-type I interferon (IFN) alpha receptor 1 (IFNAR1) monoclonal antibody in development for SLE. Methods IFNAR1 surface expression and internalisation on human monocytes before and after exposure to anifrolumab were assessed using confocal microscopy and flow cytometry. The effects of anifrolumab on type I IFN pathway activation were assessed using signal transducer and activator of transcription 1 (STAT1) phosphorylation, IFN-stimulated response element–luciferase reporter cell assays and type I IFN gene signature induction. The ability of anifrolumab to inhibit plasmacytoid dendritic cell (pDC) function and plasma cell differentiation was assessed by flow cytometry and ELISA. Effector-null properties of anifrolumab were assessed in antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) assays with B cells. Results Anifrolumab reduced cell surface IFNAR1 by eliciting IFNAR1 internalisation. Anifrolumab blocked type I IFN-dependent STAT1 phosphorylation and IFN-dependent signalling induced by recombinant and pDC-derived type I IFNs and serum of patients with SLE . Anifrolumab suppressed type I IFN production by blocking the type I IFN autoamplification loop and inhibited proinflammatory cytokine induction and the upregulation of costimulatory molecules on stimulated pDCs. Blockade of IFNAR1 suppressed plasma cell differentiation in pDC/B cell co-cultures. Anifrolumab did not exhibit CDC or ADCC activity. Conclusions Anifrolumab potently inhibits type I IFN-dependent signalling, including the type I IFN autoamplification loop, and is a promising therapeutic for patients with SLE and other diseases that exhibit chronic dysfunctional type I IFN signalling.

An immunoinhibition approach to overcome the impact of pre-existing antibodies on cut point establishment for immunogenicity assessment of moxetumomab pasudotox.

Immunogenicity can impact PK, PD, efficacy and safety of biopharmaceuticals, and is often evaluated as a secondary objective in clinical studies. Methods to detect anti-drug antibodies (ADA) and neutralizing ADA (NAb) are semi-quantitative and utilize cut points to determine positive or negative samples. Assay cut points are established by the statistical analysis of treatment-naïve subject specimens that are assumed ADA and NAb-negative. Pre-existing antibodies to various biopharmaceuticals have been observed in treatment-naïve subjects and may artificially elevate the cut point, resulting in compromised assay sensitivities, inaccuracy in immunogenicity reporting and ultimately misleading assessment of the impact of immunogenicity on clinical outcomes. Although several approaches such as removal of pre-existing antibody samples or increasing the sample dilution could be used for cut point establishment to mitigate impact of pre-existing antibodies, they each have limitations, especially when a high prevalence of pre-existing antibodies is observed. Here we describe an innovative approach used to establish cut points for ADA and NAb assays of moxetumomab pasudotox (moxetumomab), a recombinant anti-CD22 immunotoxin, to which a high prevalence of pre-existing antibodies was observed. In order to overcome the challenges associated with this high prevalence and prevent establishment of an artificially elevated cut point, we developed an immunoinhibition approach that allowed generation of pseudo ADA and NAb-negative populations for cut point determination. Immunoinhibition was performed by adding excess moxetumomab (for ADA) or a non-CD22 binding PE38-containing immunotoxin, CAT-5001 (for NAb), to treatment-naive samples prior to evaluating samples for cut point establishment. This approach successfully eliminated pre-existing antibody activity in treatment-naive samples, enabling establishment of more accurate ADA and NAb assay cut points. A comparative analysis of the clinical immunogenicity results using cut points derived with immunoinhibition and without immunoinhibition (conventional method) demonstrated that the immunoinhibition approach markedly improved detection sensitivity and accuracy of immunogenicity characterization in patient samples. This innovative approach provides an alternative, practical solution for immunogenicity assay cut point establishment when biopharmaceuticals have a high prevalence of pre-existing antibodies.

Multiplexing of receptor occupancy measurements for pharmacodynamic biomarker assessment of biopharmaceuticals

Receptor occupancy (RO) assays measure drug target engagement, and are used as pharmacodynamic (PD) biomarkers. RO assays are commonly performed by flow cytometry and often require multiplexing for assessment of multiple PD biomarkers when specimen volumes are limited. We present multiplexed RO assays for an IGF1R‐EGFR bispecific antibody (Bs‐Ab) and a CTLA4‐Ig recombinant fusion protein to demonstrate key considerations for accurate RO assessment.