Toripalimab, a therapeutic monoclonal anti-PD-1 antibody with high binding affinity to PD-1 and enhanced potency to activate human T cells.

Over the past decade, US Food and Drug Administration (FDA)-approved immune checkpoint inhibitors that target programmed death-1 (PD-1) have demonstrated significant clinical benefit particularly in patients with PD-L1 expressing tumors. Toripalimab is a humanized anti-PD-1 antibody, approved by FDA for first-line treatment of nasopharyngeal carcinoma in combination with chemotherapy. In a post hoc analysis of phase 3 studies, toripalimab in combination with chemotherapy improved overall survival irrespective of PD-L1 status in nasopharyngeal carcinoma (JUPITER-02), advanced non-small cell lung cancer (CHOICE-01) and advanced esophageal squamous cell carcinoma (JUPITER-06). On further characterization, we determined that toripalimab is molecularly and functionally differentiated from pembrolizumab, an anti-PD-1 mAb approved previously for treating a wide spectrum of tumors. Toripalimab, which binds the FG loop of PD-1, has 12-fold higher binding affinity to PD-1 than pembrolizumab and promotes significantly more Th1- and myeloid-derived inflammatory cytokine responses in healthy human PBMCs in vitro. In an ex vivo system employing dissociated tumor cells from treatment naïve non-small cell lung cancer patients, toripalimab induced several unique genes in IFN-γ and immune cell pathways, showed different kinetics of activation and significantly enhanced IFN-γ signature. Additionally, binding of toripalimab to PD-1 induced lower levels of SHP1 and SHP2 recruitment, the negative regulators of T cell activation, in Jurkat T cells ectopically expressing PD-1. Taken together, these data demonstrate that toripalimab is a potent anti-PD-1 antibody with high affinity PD-1 binding, strong functional attributes and demonstrated clinical activity that encourage its continued clinical investigation in several types of cancer.

Biacore surface matrix effects on the binding kinetics and affinity of an antigen/antibody complex.

To characterize a proprietary therapeutic monoclonal antibody (mAb) candidate, a rigorous biophysical study consisting of 53 Biacore and kinetic exclusion assay (KinExA) experiments was undertaken on the therapeutic mAb complexing with its target antigen. Unexpectedly, the observed binding kinetics depended on the chip used, suggesting that the negatively charged carboxyl groups on CM5, CM4, and C1 chips were adversely affecting the Biacore kinetic results. To study this hypothesis, Biacore solution-phase and KinExA equilibrium titrations, as well as KinExA kinetic measurements, were performed to establish accurate values for the affinity and kinetic rate constants of the binding reaction between antigen and mAb. The results revealed that as the negative charge on the biosensor surface decreased, the binding kinetics and K(D) approached the accurate binding parameters more closely when measured in solution. Two potential causes for the artifactual Biacore surface-based measurements are (i) steric hindrance of antigen binding arising from an interaction of the negatively charged carboxymethyldextran matrix with the mAb, which is a highly basic protein with a pI of 9.4, and (ii) an electrostatic repulsion between the negatively charged antigen and the carboxymethyldextran matrix. Importantly, simple diagnostic tests can be performed early in the measurement process to identify these types of matrix-mediated artifacts.

Miscalculating Equilibrium Constants for Multivalent Protein Complexes: Site-Binding Concentration or Protein Molecular Concentration?

Because the authors continue to note instances in the scientific literature of failure to use the correct receptor binding site concentration for determining binding constants, herein we discuss the fundamental concepts that need to be considered to determine correct binding constants or conversely calculate accurate reactant concentrations with known equilibrium constants. We also show the derivation and analytical solutions of the cubic and quartic equations that give the exact free ligand concentration in bivalent and trivalent receptor systems at equilibrium as a function of the macroscopic equilibrium dissociation constants and the total concentrations of ligand and multivalent protein. These equations and solutions strongly reemphasize the critical dependency of deriving the correct concentrations of bound or free ligand and multivalent protein on the choice of the correct concentration basis for the multivalent protein, which is in turn dependent upon the type of equilibrium constant used. These results demonstrate the importance of choosing the proper multivalent protein concentration for the determination of either valid microscopic or valid macroscopic equilibrium dissociation constants from binding isotherms of ligand-multivalent protein complexes.

A rigorous multiple independent binding site model for determining cell-based equilibrium dissociation constants.

A new 4-parameter nonlinear equation based on the standard multiple independent binding site model (MIBS) is presented for fitting cell-based ligand titration data in order to calculate the ligand/cell receptor equilibrium dissociation constant and the number of receptors/cell. The most commonly used linear (Scatchard Plot) or nonlinear 2-parameter model (a single binding site model found in commercial programs like Prism(R)) used for analysis of ligand/receptor binding data assumes only the K(D) influences the shape of the titration curve. We demonstrate using simulated data sets that, depending upon the cell surface receptor expression level, the number of cells titrated, and the magnitude of the K(D) being measured, this assumption of always being under K(D)-controlled conditions can be erroneous and can lead to unreliable estimates for the binding parameters. We also compare and contrast the fitting of simulated data sets to the commonly used cell-based binding equation versus our more rigorous 4-parameter nonlinear MIBS model. It is shown through these simulations that the new 4-parameter MIBS model, when used for cell-based titrations under optimal conditions, yields highly accurate estimates of all binding parameters and hence should be the preferred model to fit cell-based experimental nonlinear titration data.

Detection of high- and low-affinity antibodies against a human monoclonal antibody using various technology platforms.

The effect of monoclonal antibody (mAb) affinity on the detection limit of enzyme-linked immunosorbent assay (ELISA), surface plasmon resonance (SPR), and electrochemiluminescence (ECL) methods was evaluated using a panel of murine mAbs with affinities ranging from 0.057 to 340 nM. M1 and M7 are anti-idiotypic mAbs against a human mAb, ABX10, with dissociation equilibrium constant (KD) values of 0.057 and 7.2 nM, respectively. HP6030 and HP6002 are anti human IgG mAbs with KD values of 30 and 340 nM, respectively. The limit of detection (LOD) for these mAbs was determined using ELISA, SPR, and ECL technologies and was generally correlated with the rank order of their affinities. The LODs for M1, M7, HP6030, and HP6002 by ELISA were 17 +/- 13, 26,000 +/- 9,020, 344,000 +/- 271,000, and 792,000 +/- 1,050,000 ng/ml, respectively. According to an industry-suggested detection limit of 500 ng/ml, the ELISA was not sensitive enough for detecting M7, HP6030, and HP6002, demonstrating its limitation for detection of low- affinity mAbs. The SPR method lowered the LOD for M7 to 3,900 ng/ml, which was above the industry requirement. The ECL method lowered the LOD for all antibodies tested. Importantly, the ECL method lowered the LOD for M7 to 570 +/- 370 ng/ml, which is close to the industry requirement. Since the ECL method had demonstrated a high serum tolerance, its detection capability may be improved by using a higher percentage of serum in the assay matrix. Although a hook effect was observed with ECL methods, the methods could still detect anti-drug antibody (ADA) concentrations greater than 1 mg/ml, which minimizes concerns that high-titer ADA responses could be missed. The results demonstrated the superiority of an ECL method in detecting high- and low- affinity antibodies when compared to the ELISA and SPR methods.

A strategic and systematic approach for the determination of biosensor regeneration conditions.

Determining the optimal conditions for surface regeneration is fundamental for performance of efficient and robust protein-protein interaction kinetic studies. We devised a systematic methodology comprised of an automated seven-cycle analyte and buffer injection Biacore scheme and data interpretation algorithm. The efficiency and utility is illustrated using an antigen/monoclonal antibody interaction that required ultimately six pulses of acid for regeneration. This technique has broad applicability to any biosensor assay that requires regeneration of a surface.

Translational strategies for development of monoclonal antibodies from discovery to the clinic.

Successful strategies for the development of monoclonal antibodies require integration of knowledge with respect to target antigen properties, antibody design criteria such as affinity, isotype selection, Fc domain engineering, PK/PD properties and antibody cross-reactivity across species from the early stages of antibody development. Biophysical measurements are one of the critical components necessary for the design of effective translational strategies for lead selection and evaluation of relevant animal species for preclinical safety and efficacy studies. Incorporation of effective translational strategies from the early stages of the antibody development process is a necessity; when considered it not only reduces development time and cost, but also fosters implementation of rational decision-making throughout all phases of antibody development.

Surrogate approaches in development of monoclonal antibodies.

When cross-reactivity of a lead antibody across species is limited, antibody development programs require the generation of surrogate molecules or surrogate animal models necessary for the conduct of preclinical pharmacology and safety studies. When surrogate approaches are employed, the complexities and challenges for translation of preclinical safety and efficacy results to the clinic are undoubtedly enhanced. Because there are no currently established criteria or regulatory guidance regarding the application of surrogate approaches, a science-based strategy for translation of preclinical information to the clinic is vital for effective development of the lead antibody.

Application of analytical detection concepts to immunogenicity testing.

The cut point and detection limit of any immunogenicity assay are two of the most important quantities that define the adequacy of an assay for detecting anti-drug antibodies against therapeutic proteins. To date in the immunogenicity testing literature, only the type I (alpha) error (i.e., the false positive) rate of the assay has been considered for establishing cut points. The "sensitivity" of an immunogenicity assay is usually reported as the concentration of a monoclonal or polyclonal anti-drug antibody standard corresponding to the signal at the cut point. We propose that a more traditional and rigorous analytical chemistry definition of the detection capability be utilized wherein both type I and type II (beta, false negative) error rates are considered. Specifically, the Hubaux-Vos technique of calculating cut points and limits of detection from predication intervals on calibration curves is recommended as a statistically rigorous approach. The utility of using receiver-operator characteristic curves for managing the type I and II error rates of an immunogenicity assay is also presented. In addition, we illustrate how a soluble receptor, sMUC18, for the therapeutic mAb ABX-MA1 can result in false positives by Biacore methodology. This result suggests that immunogenicity confirmatory experiments must be carefully designed, preferably with a smaller type I and II error rate than in the primary screening if an acceptable limit of detection can be maintained.

Screening antibody-antigen interactions in parallel using Biacore A100.

Label-free optical biosensor technology has become a standard tool for characterizing monoclonal antibodies for therapeutic and diagnostic applications. The availability of high-quality binding data at an early stage greatly improves the ability to select antibodies for further development. This article shows how Biacore A100, a protein interaction array system, is capable of providing high-quality data with increased throughput. In a 12-h automated run, we analyzed 386 crude hybridoma samples to identify those with the desired kinetic profiles. Selected antibodies were further characterized by higher resolution analysis, and binding interactions were studied under a range of buffer conditions. We demonstrate how this new parallel processing system significantly expands the throughput of protein interaction analysis while maintaining data quality.

Characterizing high-affinity antigen/antibody complexes by kinetic- and equilibrium-based methods.

Two biophysical methods, Biacore and KinExA, were used to kinetically and thermodynamically characterize high-affinity antigen/antibody complexes. Three to five independent experiments were performed on each platform with three different antigen/antibody complexes possessing nanomolar to picomolar equilibrium dissociation constants. By monitoring the dissociation phase on Biacore for 4 h, we were able to measure dissociation rate constants (kd) on the order of 1 x 10(-5)s(-1). To characterize high-affinity interactions by KinExA, samples needed to be equilibrated for up to 35 h to reach equilibrium. In the end, we show that similar kinetic rate constants and affinities were determined with both solution-phase and solid-phase methodologies. These results help further validate both interaction technologies and illustrate their suitability for characterizing extremely high-affinity interactions.