Application of quantitative protein mass spectrometric data in the early predictive analysis of membrane-bound target engagement by monoclonal antibodies.

Model-informed drug discovery advocates the use of mathematical modeling and simulation for improved efficacy in drug discovery. In the case of monoclonal antibodies (mAbs) against cell membrane antigens, this requires quantitative insight into the target tissue concentration levels. Protein mass spectrometry data are often available but the values are expressed in relative, rather than in molar concentration units that are easier to incorporate into pharmacokinetic models. Here, we present an empirical correlation that converts the parts per million (ppm) concentrations in the PaxDb database to their molar equivalents that are more suitable for pharmacokinetic modeling. We evaluate the insight afforded to target tissue distribution by analyzing the likely tumor-targeting accuracy of mAbs recognizing either epidermal growth factor receptor or its homolog HER2. Surprisingly, the predicted tissue concentrations of both these targets exceed the Kd values of their respective therapeutic mAbs. Physiologically based pharmacokinetic (PBPK) modeling indicates that in these conditions only about 0.05% of the dosed mAb is likely to reach the solid tumor target cells. The rest of the dose is eliminated in healthy tissues via both nonspecific and target-mediated processes. The presented approach allows evaluation of the interplay between the target expression level in different tissues that determines the overall pharmacokinetic properties of the drug and the fraction that reaches the cells of interest. This methodology can help to evaluate the efficacy and safety properties of novel drugs, especially if the off-target cell degradation has cytotoxic outcomes, as in the case of antibody-drug conjugates.

Toward systems-informed models for biologics disposition: covariates of the abundance of the neonatal Fc Receptor (FcRn) in human tissues and implications for pharmacokinetic modelling.

Biologics are a fast-growing therapeutic class, with intertwined pharmacokinetics and pharmacodynamics, affected by the abundance and function of the FcRn receptor. While many investigators assume adequacy of classical models, such as allometry, for pharmacokinetic characterization of biologics, advocates of physiologically-based pharmacokinetics (PBPK) propose consideration of known systems parameters that affect the fate of biologics to enable a priori predictions, which go beyond allometry. The aim of this study was to deploy a systems-informed modelling approach to predict the disposition of Fc-containing biologics. We used global proteomics to quantify the FcRn receptor [p51 and ß2-microglobulin (B2M) subunits] in 167 samples of human tissue (liver, intestine, kidney and skin) and assessed covariates of its expression. FcRn p51 subunit was highest in liver relative to other tissues, and B2M was 1-2 orders of magnitude more abundant than FcRn p51 across all sets. There were no sex-related differences, while higher expression was confirmed in neonate liver compared with adult liver. Trends of expression in liver and kidney indicated a moderate effect of body mass index, which should be confirmed in a larger sample size. Expression of FcRn p51 subunit was approximately 2-fold lower in histologically normal liver tissue adjacent to cancer compared with healthy liver. FcRn mRNA in plasma-derived exosomes correlated moderately with protein abundance in matching liver tissue, opening the possibility of use as a potential clinical tool. Predicted effects of trends in FcRn abundance in healthy and disease (cancer and psoriasis) populations using trastuzumab and efalizumab PBPK models were in line with clinical observations, and global sensitivity analysis revealed endogenous IgG plasma concentration and tissue FcRn abundance as key systems parameters influencing exposure to Fc-conjugated biologics.

In vivo biodistribution and pharmacokinetics of sotrovimab, a SARS-CoV-2 monoclonal antibody, in healthy cynomolgus monkeys.

PURPOSE: Sotrovimab (VIR-7831), a human IgG1κ monoclonal antibody (mAb), binds to a conserved epitope on the SARS-CoV-2 spike protein receptor binding domain (RBD). The Fc region of VIR-7831 contains an LS modification to promote neonatal Fc receptor (FcRn)-mediated recycling and extend its serum half-life. Here, we aimed to evaluate the impact of the LS modification on tissue biodistribution, by comparing VIR-7831 to its non-LS-modified equivalent, VIR-7831-WT, in cynomolgus monkeys. METHODS: 89Zr-based PET/CT imaging of VIR-7831 and VIR-7831-WT was performed up to 14 days post injection. All major organs were analyzed for absolute concentration as well as tissue:blood ratios, with the focus on the respiratory tract, and a physiologically based pharmacokinetics (PBPK) model was used to evaluate the tissue biodistribution kinetics. Radiomics features were also extracted from the PET images and SUV values. RESULTS: SUVmean uptake in the pulmonary bronchi for 89Zr-VIR-7831 was statistically higher than for 89Zr-VIR-7831-WT at days 6 (3.43 ± 0.55 and 2.59 ± 0.38, respectively) and 10 (2.66 ± 0.32 and 2.15 ± 0.18, respectively), while the reverse was observed in the liver at days 6 (5.14 ± 0.80 and 8.63 ± 0.89, respectively), 10 (4.52 ± 0.59 and 7.73 ± 0.66, respectively), and 14 (4.95 ± 0.65 and 7.94 ± 0.54, respectively). Though the calculated terminal half-life was 21.3 ± 3.0 days for VIR-7831 and 16.5 ± 1.1 days for VIR-7831-WT, no consistent differences were observed in the tissue:blood ratios between the antibodies except in the liver. While the lung:blood SUVmean uptake ratio for both mAbs was 0.25 on day 3, the PBPK model predicted the total lung tissue and the interstitial space to serum ratio to be 0.31 and 0.55, respectively. Radiomics analysis showed VIR-7831 had mean-centralized PET SUV distribution in the lung and liver, indicating more uniform uptake than VIR-7831-WT. CONCLUSION: The half-life extended VIR-7831 remained in circulation longer than VIR-7831-WT, consistent with enhanced FcRn binding, while the tissue:blood concentration ratios in most tissues for both drugs remained statistically indistinguishable throughout the course of the experiment. In the bronchiolar region, a higher concentration of 89Zr-VIR-7831 was detected. The data also allow unparalleled insight into tissue distribution and elimination kinetics of mAbs that can guide future biologic drug discovery efforts, while the residualizing nature of the 89Zr label sheds light on the sites of antibody catabolism.

Application of physiologically based pharmacokinetic models for therapeutic proteins and other novel modalities.

The past two decades have seen diversification of drug development pipelines and approvals from traditional small molecule therapies to alternative modalities including monoclonal antibodies, engineered proteins, antibody drug conjugates (ADCs), oligonucleotides and gene therapies. At the same time, physiologically based pharmacokinetic (PBPK) models for small molecules have seen increased industry and regulatory acceptance.This review focusses on the current status of the application of PBPK models to these newer modalities and give a perspective on the successes, challenges and future directions of this field.There is greatest experience in the development of PBPK models for therapeutic proteins, and PBPK models for ADCs benefit from prior experience for both therapeutic proteins and small molecules. For other modalities, the application of PBPK models is in its infancy.Challenges are discussed and a common theme is lack of availability of physiological and experimental data to characterise systems and drug parameters to enable a priori prediction of pharmacokinetics. Furthermore, sufficient clinical data are required to build confidence in developed models.The PBPK modelling approach provides a quantitative framework for integrating knowledge and data from multiple sources and can be built on as more data becomes available.

Physiologically-based pharmacokinetic model for pulmonary disposition of protein therapeutics in humans.

Lung related disorders like COPD and Asthma, as well as various infectious diseases, form a major therapeutic area which would benefit from a predictive and adaptable mathematical model for describing pulmonary disposition of biological modalities. In this study we fill that gap by extending the cross-species two-pore physiologically-based pharmacokinetic (PBPK) platform with more detailed respiratory tract that includes the airways and alveolar space with epithelial lining fluid. We parameterize the paracellular and FcRn-facilitated exchange pathways between the epithelial lining fluid and lung interstitial space by building a mechanistic model for the exchange between the two. The optimized two-pore PBPK model described pulmonary exposure of several systemically dosed mAbs for which data is available and is also in agreement with the observed levels of endogenous IgG and albumin. The proposed framework can be used to assess pharmacokinetics of new lung-targeting biologic therapies and guide their dosing to achieve desired exposure at the pulmonary site-of-action.

Application of quantitative protein mass spectrometric data in the early predictive analysis of target engagement by monoclonal antibodies.

Model-informed drug discovery is endorsed by the US Food and Drug Administration (FDA) to improve the flow of medicines from bench to bedside. In the case of monoclonal antibodies, this necessitates taking into account not only the pharmacokinetic (PK) properties of the drug, but also the tissue distribution, concentration, and turnover of the target to guide dose and affinity selection, as well as serve as a link to downstream pharmacology. Relevant information (e.g., tissue proteomic data from quantitative mass spectrometry), is increasingly available from public domain data repositories, although not necessarily in the form that is directly usable for the purpose of quantitative, predictive, and mechanistic PK/pharmacodynamic (PD) modeling based on molarity or similar frameworks instead. Using secreted plasma protein concentrations measured both by immunochemical methods and mass spectrometry, we addressed this gap and derived an optimized nonlinear empirical function that establishes the correlation between the two data sets and validated the approach taken using a wider data set of all proteins found in plasma. In addition, we present a semimechanistic framework for the plasma half-life of soluble proteins where clearance is expressed as a nonlinear function of the molecular weight of the protein. Finally, we apply the approach to two established therapeutic antibody targets: complement factor C5 and PCSK9 to demonstrate how the described framework can be applied to predictive PK/PD modeling.

Novel Selection Approaches to Identify Antibodies Targeting Neoepitopes on the C5b6 Intermediate Complex to Inhibit Membrane Attack Complex Formation.

The terminal pathway of complement is implicated in the pathology of multiple diseases and its inhibition is, therefore, an attractive therapeutic proposition. The practicalities of inhibiting this pathway, however, are challenging, as highlighted by the very few molecules in the clinic. The proteins are highly abundant, and assembly is mediated by high-affinity protein-protein interactions. One strategy is to target neoepitopes that are present transiently and only exist on active or intermediate complexes but not on the abundant native proteins. Here, we describe an antibody discovery campaign that generated neoepitope-specific mAbs against the C5b6 complex, a stable intermediate complex in terminal complement complex assembly. We used a highly diverse yeast-based antibody library of fully human IgGs to screen against soluble C5b6 antigen and successfully identified C5b6 neoepitope-specific antibodies. These antibodies were diverse, showed good binding to C5b6, and inhibited membrane attack complex (MAC) formation in a solution-based assay. However, when tested in a more physiologically relevant membrane-based assay these antibodies failed to inhibit MAC formation. Our data highlight the feasibility of identifying neoepitope binding mAbs, but also the technical challenges associated with the identification of functionally relevant, neoepitope-specific inhibitors of the terminal pathway.

Cross-species/cross-modality physiologically based pharmacokinetics for biologics: 89Zr-labelled albumin-binding domain antibody GSK3128349 in humans.

Two-pore physiologically-based pharmacokinetics (PBPK) for biologics describes the tissue distribution and elimination kinetics of soluble proteins as a function of their hydrodynamic radius and the physiological properties of the organs. Whilst many studies have been performed in rodents to parameterize the PBPK framework in terms of organ-specific lymph flow rates, similar validation in humans has been limited. This is mainly due to the paucity of the tissue distribution time course data for biologics that is not distorted by target-related binding. Here, we demonstrate that a PBPK model based on rodent data provided good to satisfactory extrapolation to the tissue distribution time course of 89Zr-labeled albumin-binding domain antibody (AlbudAb™) GSK3128349 in healthy human volunteers, including correct prediction of albumin-like plasma half-life, volume of distribution, and extravasation half-life. The AlbudAb™ used only binds albumin, and hence it also provides information about the tissue distribution kinetics and turnover of that ubiquitous and multifunctional plasma protein.

Computer-assembled cross-species/cross-modalities two-pore physiologically based pharmacokinetic model for biologics in mice and rats.

Two-pore physiologically-based pharmacokinetic (PBPK) models can be expected to describe the tissue distribution and elimination kinetics of soluble proteins, endogenous or dosed, as function of their size. In this work, we amalgamated our previous two-pore PBPK model for an inert domain antibody (dAb) in mice with the cross-species platform PBPK model for monoclonal antibodies described in literature into a unified two-pore platform that describes protein modalities of different sizes and includes neonatal Fc receptor (FcRn) mediated recycling. This unified PBPK model was parametrized for organ-specific lymph flow rates and the endosomal recycling rate constant using an extended tissue distribution time-course dataset that included an inert dAb, albumin and IgG in rats and mice. The model was evaluated by comparing the ab initio predictions for the tissue distribution and elimination properties of albumin-binding dAbs (AlbudAbsTM) in mice and rats with the experimental observations. Due to the large number of molecular species and reactions involved in large-scale PBPK models, we have also developed and deployed a MatlabTM script for automating the assembly of SimBiologyTM-based two-pore biologics PBPK models which drastically cuts the time and effort required for model building.

The biodistribution and clearance of AlbudAb, a novel biopharmaceutical medicine platform, assessed via PET imaging in humans.

Conjugation or fusion to AlbudAbs™ (albumin-binding domain antibodies) is a novel approach to extend the half-life and alter the tissue distribution of biological and small molecule therapeutics. To understand extravasation kinetics and extravascular organ concentrations of AlbudAbs in humans, we studied tissue distribution and elimination of a non-conjugated 89Zr-labeled AlbudAb in healthy volunteers using positron emission tomography/computed tomography (PET/CT). METHODS: A non-conjugated AlbudAb (GSK3128349) was radiolabeled with 89Zr and a single 1 mg (~ 15 MBq) dose intravenously administered to eight healthy males. 89Zr-AlbudAb tissue distribution was followed for up to 7 days with four whole-body PET/CT scans. 89Zr-AlbudAb tissue concentrations were quantified in organs of therapeutic significance, measuring standardized uptake value and tissue/plasma ratios. Plasma pharmacokinetics were assessed by gamma counting and LC-MS/MS of blood samples. RESULTS: 89Zr-AlbudAb administration and PET/CT procedures were well tolerated, with no drug-related immunogenicity or adverse events. 89Zr-AlbudAb rapidly distributed throughout the vasculature, with tissue/plasma ratios in the liver, lungs, and heart relatively stable over 7 days post-dose, ranging between 0.1 and 0.5. The brain tissue/plasma ratio of 0.025 suggested minimal AlbudAb blood-brain barrier penetration. Slowly increasing ratios in muscle, testis, pancreas, and spleen reflected either slow AlbudAb penetration and/or 89Zr residualization in these organs. Across all tissues evaluated, the kidney tissue/plasma ratio was highest (0.5-1.5 range) with highest concentration in the renal cortex. The terminal half-life of the 89Zr-AlbudAb was 18 days. CONCLUSION: Evaluating the biodistribution of 89Zr-AlbudAb in healthy volunteers using a low radioactivity dose was successful (total subject exposure ~ 10 mSv). Results indicated rapid formation of reversible, but stable, complexes between AlbudAb and albumin upon dosing. 89Zr-AlbudAb demonstrated albumin-like pharmacokinetics, including limited renal elimination. This novel organ-specific distribution data for AlbudAbs in humans will facilitate a better selection of drug targets to prosecute using the AlbudAb platform and significantly contribute to modeling work optimizing dosing of therapeutic AlbudAbs in the clinic.

Modeling bispecific monoclonal antibody interaction with two cell membrane targets indicates the importance of surface diffusion.

We have developed a mathematical framework for describing a bispecific monoclonal antibody interaction with two independent membrane-bound targets that are expressed on the same cell surface. The bispecific antibody in solution binds either of the two targets first, and then cross-links with the second one while on the cell surface, subject to rate-limiting lateral diffusion step within the lifetime of the monovalently engaged antibody-antigen complex. At experimental densities, only a small fraction of the free targets is expected to lie within the reach of the antibody binding sites at any time. Using ordinary differential equation and Monte Carlo simulation-based models, we validated this approach against an independently published anti-CD4/CD70 DuetMab experimental data set. As a result of dimensional reduction, the cell surface reaction is expected to be so rapid that, in agreement with the experimental data, no monovalently bound bispecific antibody binary complexes accumulate until cross-linking is complete. The dissociation of the bispecific antibody from the ternary cross-linked complex is expected to be significantly slower than that from either of the monovalently bound variants. We estimate that the effective affinity of the bivalently bound bispecific antibody is enhanced for about 4 orders of magnitude over that of the monovalently bound species. This avidity enhancement allows for the highly specific binding of anti-CD4/CD70 DuetMab to the cells that are positive for both target antigens over those that express only one or the other We suggest that the lateral diffusion of target antigens in the cell membrane also plays a key role in the avidity effect of natural antibodies and other bivalent ligands in their interactions with their respective cell surface receptors.

Pharmacokinetic characteristics, pharmacodynamic effect and in vivo antiviral efficacy of liver-targeted interferon alpha.

Interferon alpha (IFNα) is used for the treatment of hepatitis B virus infection, and whilst efficacious, it is associated with multiple adverse events caused by systemic exposure to interferon. We therefore hypothesise that targeting IFN directly to the intended site of action in the liver would reduce exposure in blood and peripheral tissue and hence improve the safety and tolerability of IFNα therapy. Furthermore we investigated whether directing IFN to the reservoir of infection in the liver may improve antiviral efficacy by increasing local concentration in target organs and tissues. Our previous results show that the mIFNα2 fused to an ASGPR specific liver targeting antibody, DOM26h-196-61, results in a fusion protein which retains the activity of both fusion partners when measured in vitro. In vivo targeting of the liver by mIFNα2-DOM26h-196-61, hereafter referred to as targeted mIFNα2, was observed in microSPECT imaging studies in mice. In this study we show by pharmacokinetic analysis that antibody mediated liver-targeting results in increased uptake and exposure of targeted mIFNα2 in target tissues, and correspondingly reduced uptake and exposure in systemic circulation, clearance organs and non-target tissues. We also show that cytokine activity and antiviral activity of liver-targeted IFN is observed in vivo, but that, contrary to expectations, liver-targeting of mIFNα2 using ASGPR specific dAbs actually leads to a reduced pharmacodynamic effect in target organs and lower antiviral activity in vivo when compared to non-targeted mIFNα2-dAb fusions.

Development of a physiologically based pharmacokinetic model for a domain antibody in mice using the two-pore theory.

Domain antibodies (dAbs) are the smallest antigen-binding fragments of immunoglobulins. To date, there is limited insight into the pharmacokinetics of dAbs, especially their distribution into tissues and elimination. The objective of this work was to develop a physiologically-based pharmacokinetic model to investigate the biodisposition of a non-specific dAb construct in mice. Following a single IV administration of 10 mg/kg dummy dAb protein to twenty four female mice, frequent blood samples were collected and whole body lateral sections were analyzed by quantitative whole-body autoradiography. The model is based on the two-pore hypothesis of extravasation where organ-specific isogravimetric flow rates (Jorg,ISO) and permeability-surface area products (PSorg) are expressed as linear functions of the lymph flow rate (Jorg) and the kidney compartment is modified to account for glomerular filtration of dAb. As a result, only Jorg, glomerular filtration coefficient and the combined volume of Bowman's capsule, proximal and distal renal tubules and loop of Henle were optimized by fitting simultaneously all blood and organ data to the model. Our model captures the pharmacokinetic profiles of dAb in blood and all organs and shows that extravasation into interstitial space is a predominantly diffusion-driven process. The parameter values were estimated with good precision (%RMSE ≈ 30) and low cross-correlation (R(2) < 0.2). We developed a flexible model with a limited parameter number that may be applied to other biotherapeutics after adapting for size-related effects on extravasation and renal elimination processes.

Liver-targeting of interferon-alpha with tissue-specific domain antibodies.

Interferon alpha (IFNα) is used for the treatment of hepatitis C infection and whilst efficacious it is associated with multiple adverse events including reduced leukocyte, erythrocyte, and platelet counts, fatigue, and depression. These events are most likely caused by systemic exposure to interferon. We therefore hypothesise that targeting the therapeutic directly to the intended site of action in the liver would reduce exposure in blood and peripheral tissue and hence improve the safety and tolerability of IFNα therapy. We genetically fused IFN to a domain antibody (dAb) specific to a hepatocyte restricted antigen, asialoglycoprotein receptor (ASGPR). Our results show that the murine IFNα2 homolog (mIFNα2) fused to an ASGPR specific dAb, termed DOM26h-196-61, could be expressed in mammalian tissue culture systems and retains the desirable biophysical properties and activity of both fusion partners when measured in vitro. Furthermore a clear increase in in vivo targeting of the liver by mIFNα2-ASGPR dAb fusion protein, compared to that observed with either unfused mIFNα2 or mIFNα2 fused to an isotype control dAb VHD2 (which does not bind ASGPR) was demonstrated using microSPECT imaging. We suggest that these findings may be applicable in the development of a liver-targeted human IFN molecule with improved safety and patient compliance in comparison to the current standard of care, which could ultimately be used as a treatment for human hepatitis virus infections.

Cell-free selection of domain antibodies by in vitro compartmentalization.

Efficient identification of antibodies, or any fragments thereof, displaying desired specificity and affinity is critical for the development of novel immunotherapeutics. Here we describe the adaptation of in vitro compartmentalization for the cell-free selection of Vκ and VH domain antibodies (dAbs™) from large combinatorial libraries. The dAbs™ are in vitro expressed in fusion to the N-terminus of single-chain variant of phage P22 Arc repressor DNA-binding domain that links the compartmentally expressed protein molecules to their encoding PCR fragment-based genes via cognate operator sites present on the DNA. Libraries of up to 10(10) in size can be rapidly assembled and selected for improved affinity in equilibrium and off-rate conditions.

Next generation immunotherapeutics--honing the magic bullet.

Most therapeutic antibodies in the clinic today are based on fully humanised immunoglobulins. They have proven to be outstandingly effective, especially for the treatment of cancer, autoimmune and inflammatory diseases where the target is a single, well-defined and accessible molecule. Many diseases however are complex, involving multiple mediators or signalling pathways that could be targeted simultaneously to maximise clinical benefit. There is also a wealth of validated intracellular and CNS-based targets which are currently inaccessible to monoclonal antibody therapy. A spectrum of next generation immunotherapeutics is in development to address these issues and a number of them have also entered clinical trials.

Cell-free selection of zinc finger DNA-binding proteins using in vitro compartmentalization.

We have exploited emulsion-based in vitro compartmentalization (IVC) to devise a method for the selection of zinc finger proteins (ZFPs) on the basis of their DNA-binding specificity. A library of ZFPs fused to a C-terminal peptide tag is encoded by a set of DNA cassettes that are prepared wholly in vitro. In addition to the ZFP gene, each DNA cassette also carries a given DNA target binding site sequence for which one wishes to isolate ZFP binders. An aliquot of the library is added to bacterial S30 extract and emulsified in mineral oil so that most of the aqueous droplets contain, on average, no more than one gene. If an intra-compartmentally expressed ZFP binds specifically to its encoding DNA via the target binding site, the complex can be purified by affinity capture via the peptide tag after breaking the emulsion, thus rescuing the gene. We present proof-of-principle for this IVC selection method by selecting a specific high-affinity ZFP gene from a high background of a related gene. We also propose that high-affinity ZFPs can be used as genotype-phenotype linkages to enable selection of other proteins using IVC.

Microbead display by in vitro compartmentalisation: selection for binding using flow cytometry.

In vitro compartmentalisation in an emulsion was used to physically link proteins to the DNA that encodes them via microbeads. These microbeads can be selected for catalysis, or, as demonstrated here, for binding. Genes encoding a peptide containing an epitope (haemagglutinin) were enriched to near purity from a 10(6)-fold excess of genes encoding a different peptide by two rounds of selection using flow cytometry, indicating approximately 1000-fold enrichment per round. Single beads can be isolated using flow sorting and the single gene on the bead amplified by polymerase chain reaction. Hence, the entire process can be performed completely in vitro.