Uwe Wessels

Subset analysis of patients experiencing clinical events of a potentially immunogenic nature in the pivotal clinical trials of tocilizumab for rheumatoid arthritis: Evaluation of an antidrug antibody ELISA using clinical adverse event-driven immunogenicity testing.

BACKGROUND Tocilizumab (TCZ) is a recombinant humanized monoclonal antibody directed against the interleukin-6 receptor. In Europe, TCZ is approved for use in combination with methotrexate in the treatment of adult patients with moderate to severe rheumatoid arthritis (RA) who have failed to respond to or were unable to tolerate previous therapy with one or more disease-modifying antirheumatic drugs or tumor necrosis factor (TNF) antagonists; in the United States, it is approved for the treatment of adult patients with moderate to severe active RA who have failed to respond to one or more TNF antagonists. As part of the Phase III clinical development program, the immunogenicity of TCZ was evaluated using a bridging ELISA; however, this assay is considered limited in detecting low-affinity or immunoglobulin G4 subisotype antidrug antibodies (ADAs). OBJECTIVE This study assessed the validity of the ELISA for detecting anti-TCZ ADAs by using complementary bioanalytic assays to test samples from a subgroup of patients with clinical adverse events (AEs) of a potentially immunogenic nature, who were considered highly likely to have ADAs. The goal was to determine whether use of these additional assays led to detection of ADAs not found on the ELISA, thus minimizing the risk of false-negative results. METHODS The Phase III program for TCZ consisted of 5 core studies in which adult patients with RA received either TCZ 4 or 8 mg/kg IV or control every 4 weeks, with or without concomitant antirheumatic therapy. Blood samples obtained at baseline and at regular intervals thereafter were tested using the ELISA for ADA screening and confirmation. Regardless of the results on ADA screening, samples from patients who developed clinical AEs of a potentially immunogenic nature (ie, falling within predefined system organ classes, occurring during or within 24 hours of TCZ infusion, considered related to TCZ therapy, or leading to study withdrawal) were subjected to additional testing with a surface plasmon resonance (SPR) assay for isotyping and epitope localization and a standard ImmunoCAP immunoglobulin E (IgE) assay made specific for TCZ. RESULTS The 5 core studies and their open-label, longterm extension studies enrolled a total of 4199 adult patients with RA (82.1% female; 74.0% white; mean age, 52.0 years [range, 18-89 years]; mean weight, 73.4 kg [range, 35-150 kg]); 2928 patients received TCZ and 1271 received control. Of the 2816 samples from TCZ-treated patients tested, 64 (2.3%) had samples that tested positive at least once on the ELISA screening and confirmation assay, 48 (75.0%) of them at baseline. A clinical AE of a potentially immunogenic nature occurred during TCZ treatment in 21 patients, 8 of whom had an anaphylactic reaction. Eleven of the samples from these 21 patients had tested negative for AD As on the screening ELISA. Only 1 of these 11 patients tested positive for ADAs on both additional assays; all others tested negative. The results of the ELISA, SPR, and IgE assays were consistent for 16 of 18 tested patients (88.9%) who provided data on at least 2 of the 3 assays. CONCLUSIONS Based on the findings of this analysis in a relevant patient population, the bridging-type screening and confirmation ELISA was a valid method of detecting anti-TCZ ADAs. Immunogenicity testing of samples from patients with clinical AEs of a potentially immunogenic nature using assays complementary to the ELISA added valuable information about the incidence and character of ADAs.

Evaluation of an immunoassay for human-specific quantitation of therapeutic antibodies in serum samples from non-human primates

Pharmacokinetic characterization of therapeutic antibodies plays an important role during preclinical and clinical development. However, accurate pharmacokinetic evaluation of therapeutic antibodies in serum samples from non-human primates is often complicated by insufficient specificity of the assays to measure drug levels. The present paper describes the use of a murine monoclonal antibody in an immunoassay format to specifically and quantitatively measure human therapeutic antibodies in serum from non-human primates. This murine antibody is directed against a unique epitope on the constant region CH2 domain of all isotypes of human immunoglobulin G (IgG). The antibody, designated anti-human Fcgamma-pan: R10Z8E9, does not cross-react with serum from mouse, rat, and the non-human primates marmoset, rhesus macaque, cynomolgus monkey and baboon when using an enzyme-linked immunosorbent assay (ELISA) or surface plasmon resonance technology (Biacore) format for measurement of the therapeutic antibody. Use of the antibody anti-human Fcgamma-pan: R10Z8E9 as capturing and detection reagent allowed human-specific quantitation of total therapeutic antibody anti-IGF-1R in spiked cynomolgus monkey serum via a Sandwich ELISA format. In contrast, a commercially available polyclonal antibody (PAB) directed to the Fcgamma fragment of human IgG only specifically measured the therapeutic antibody in buffer samples, but not in serum from cynomolgus monkeys. This generic human IgG assay was already applied in several pharmacokinetic studies in cynomolgus monkeys to determine serum levels of different therapeutic antibodies, including the anti-IGF-1R. Validation of the assay for a humanized IgG1 therapeutic antibody against a membrane protein revealed a lower limit of quantitation of 8 ng/mL in undiluted serum. Intra-assay and inter-assay precision was characterized by a coefficient of variation of less than 10% and accuracy was within 15%. Dilutional linearity was evidenced by a recovery of 98.7-114% of expected concentrations. In conclusion, the monoclonal antibody anti-human Fcgamma-pan: R10Z8E9 provides a standard means for human-specific quantitation of therapeutic antibodies with high sensitivity in serum samples from non-human primates in a generic human IgG assay.

Evaluation of a generic immunoassay with drug tolerance to detect immune complexes in serum samples from cynomolgus monkeys after administration of human antibodies.

Current state of the art bridging ELISA technologies for detection of anti-drug antibodies (ADAs) against therapeutic antibodies bear the risk of false-negative results due to interference by circulating drug. Methods to remove the drug in the sample or sample pre-treatment techniques such as acid dissociation of the immune complexes are limited, laborious and may destroy ADAs resulting again in false-negative results. The immune complex ELISA described in this publication provides a simple solution. It is designed to analyze samples from cynomolgus monkeys dosed with human antibodies; it can be used for all human antibodies since it is independent of the specific antibody and its target. The generic applicability of the ADA assay is enabled by the use of (1) a murine anti-human Fc monoclonal antibody (MAb) as capture reagent; (2) a murine anti-cynomolgus monkey IgG MAb as detection reagent; and (3) an ADA positive control conjugate consisting of cynomolgus IgG complexed with human IgG. In its basic version, the generic ADA ELISA specifically detects only immune complexes formed in vivo. Validation of the ADA assay revealed a lower limit of quantitation of 15.6 ng/mL in serum samples. Intra-assay and inter-assay precision was characterized by a coefficient of variation of less than 10% and accuracy was within 8%. Matrix effects were low as evidenced by a mean recovery of 95%. In vitro pre-incubation of the serum samples with drug makes also the free ADA in the sample amenable to measurement by the immune complex ELISA as demonstrated by analysis of ADAs from two cynomolgus monkey studies with two different antibodies. The generic and versatile nature of this ADA assay favors its use in pilot pharmacokinetic and safety studies in cynomolgus monkeys during candidate selection of antibodies. The assay can help to explain unexpected drug clearance profiles, loss of efficacy or safety events caused by immune complexes and guide further development.

Impact of molecular processing in the hinge region of therapeutic IgG4 antibodies on disposition profiles in cynomolgus monkeys

The IgG4 isotype antibody is a potential candidate for immunotherapy when reduced effector functions are desirable. However, antigen binding fragment (Fab) arm exchange leads to functional monovalency with potentially reduced therapeutic efficacy. Mutagenesis studies suggested that the CH3 domain and not the core hinge is dominantly involved in in vivo molecular processing. This work investigated whether stabilization of the core hinge of a therapeutic IgG4 antibody by mutation of Ser228 to Pro (S228P) would be sufficient to prevent in vivo Fab arm exchange. In vitro experiments evaluated the influence of different levels of oxidation-reduction conditions in buffer and serum on Fab arm exchange (swapping) of wild-type (WT) IgG4 and IgG1 and of IgG4 S228P, which included a sterically neutral second mutation (Leu235 replaced by Glu). The objective of single-dose pharmacokinetic experiments in cynomolgus monkeys was to determine whether the mutation reduced IgG4 swapping in vivo. The results indicated that S228P mutation did not completely prevent Fab arm exchange in vitro in buffer under reducing conditions relative to IgG4 WT. The immunoassay findings were confirmed by mass spectrometry measurements. Results of the in vivo studies suggested that the therapeutic IgG4 WT antibody exchanged Fab arms with endogenous cynomolgus monkey IgG4, resulting in bispecific IgG4 antibodies with monovalency for the therapeutic target. In contrast, serum from cynomolgus monkeys dosed with the IgG4 mutant was virtually free of swapped IgG4. In conclusion, the results indicated that IgG4 swapping in vivo was markedly attenuated by S228P mutation.

Evaluation of a biosensor immunoassay for simultaneous characterization of isotype and binding region of human anti-tocilizumab antibodies with control by surrogate standards

This article describes the simultaneous Biacore analysis of human anti-human antibodies (HAHAs) with respect to the binding region and the isotype by a combination of 11 single measurements per sample. The multiplexing single assay setup made efficient use of the four parallel flow cells on one biosensor chip by immobilization of full-length antibody and its constant (Fc) and antigen binding (Fab) fragments for differential binding analysis of anti-drug antibodies (ADAs). Thereby, a complete time-specific immunogenicity profile (intensity, isotype, specificity, and kinetics) of a patient could be obtained by assessing the response patterns of serially collected samples analyzed in a single measurement run. The use of functionally active standard conjugates allowed control of the assay performance throughout the whole procedure. The positive control standard conjugates mimicking polyclonal human ADAs of different isotypes were obtained by conjugating polyclonal rabbit antibodies against the therapeutic antibody to human immunoglobulin (Ig) M, IgG, or IgE. In this article, the qualification of the assay is demonstrated and the application of the methodology to six representative rheumatoid arthritis patients treated with the therapeutic humanized IgG1 antibody tocilizumab (anti-IL-6R) is shown to illustrate the versatility of the assay. The presented method allows one to differentiate specific ADAs from drug-unspecific responses (e.g., rheumatoid factors). In addition, the method can be used to discriminate between isotype responses of the IgG, IgM, and IgE types and, thereby, allows one to describe the time course of specific ADA formation and its disappearance on the single patient level.

Characterization of murine anti-human Fab antibodies for use in an immunoassay for generic quantification of human Fab fragments in non-human serum samples including cynomolgus monkey samples.

Generic immunoassay formats in animal serum have been described for pharmacokinetic (PK) analysis of human full-length antibodies, but not of human antigen binding fragment (Fab) proteins. Here we characterize two murine monoclonal antibodies (mAb) raised against human immunoglobulin G (IgG) which bind to unique epitopes in the Fab region of human IgG. mAb M-1.7.10 is directed against the constant domain of the kappa light chain and mAb M-1.19.31 binds to the constant domain 1 (CH1) of the heavy chain. Surface plasmon resonance analysis showed that mAb M-1.7.10 does not cross-react with sera from mouse, rat, rabbit, dog, marmoset, rhesus macaque, baboon and cynomolgus monkey, but binds to human and chimpanzee serum (dissociation constant K(D) of 6.8 × 10(-12) and 3.1 × 10(-11)M, respectively). mAb M-1.19.31 shows a higher K(D) for human and chimpanzee IgG (2.0 × 10(-9)M and 5.8 × 10(-10)M, respectively), but also does not bind to serum of the other species. Therefore, mAb M-1.7.10 was used as capture and mAb M-1.19.31 as detection reagent in a generic enzyme linked immunosorbent assay (ELISA) to quantify the human anti-IGF-1R Fab in mouse serum. The generic human Fab assay showed a limit of detection of 31.5 ng/mL anti-IGF-1R Fab. Intra- and inter-assay precision was less than 12% and the accuracy range for all controls was within ±20% of the target concentration. The generic human Fab ELISA was applied to determine serum levels of human anti-IGF-1R Fab after intravenous (iv) administration of 10mg/kg to mice. The resulting concentration-time profile was nearly identical to that obtained by analysis with a validated specific ELISA for anti-IGF-1R Fab. The mean relative concentration of anti-IGF-1R Fab analyzed by the generic assay was 82-118% of that of the specific assay. This equivalence was confirmed in a cynomolgus monkey study with the full length human mAb anti-TROP-2 IgG. Both specific ELISAs used mAb M-1.7.10 as detection reagent and their targets for capturing. In conclusion, the two murine anti-human Fabs are versatile tools as capture and detection reagents for human antibodies in generic and specific PK ELISA formats for animal studies. Their use in specific ELISAs as detection reagents allows the usage of Fc-fusion proteins as capture reagents.

Novel drug and soluble target tolerant antidrug antibody assay for therapeutic antibodies bearing the P329G mutation

AIM Bridging immunoassays for detection of antidrug antibodies (ADAs) are typically susceptible to high concentrations of residual drug. Sensitive drug-tolerant assays are, therefore, needed. MATERIALS & METHODS An immune complex assay to detect ADAs against therapeutic antibodies bearing Pro329Gly mutation was established. The assay uses antibodies specific for the Pro329Gly mutation for capture and human soluble Fcγ receptor for detection. RESULTS When compared with a bridging assay, the new assay showed similar precision, high sensitivity to IgG1 ADA and dramatically improved drug tolerance. However, it was not able to detect early (IgM-based) immune responses. CONCLUSION Applied in combination with a bridging assay, the novel assay serves as orthogonal assay for immunogenicity assessment and allows further characterization of ADA responses.

An immunodepletion procedure advances free angiopoietin-2 determination in human plasma samples during anti-cancer therapy with bispecific anti-Ang2/VEGF CrossMab

Bispecific monoclonal IgG antibodies offer increased efficacy by antagonizing two different targets. Assessing drug mechanisms, target engagement and biomarker features, the quantification of free target levels is essential. The anti-Ang2/VEGF-CrossMab (anti-A2V) recognizes soluble vascular endothelial growth factor-A (VEGF-A) and soluble angiopoietin-2 (Ang2). However, an assay for reliable free Ang2 determination is missing. Here, we describe an immunodepletion procedure that allows for selective quantification of free Ang2 target levels by taking into advantage the bispecificity of the therapeutic antibody. The specificity for VEGF was utilized to efficiently eliminate drug-bound Ang2 from plasma samples prior to an established Ang2 measurement. The magnetic bead-based depletion procedure used an anti-idiotypic monoclonal antibody (mAb) specific for the VEGF binding site of anti-A2V (anti-Id-anti-VEGF mAb) to capture the drug along with drug-bound Ang2. High efficiencies of 99.9% were obtained for anti-A2V depletion (concentration range 300 ng/mL to 10(6)ng/mL) reflecting a 1000-fold reduction of drug-bound Ang2. A significant impact of the interaction of anti-Id-anti-VEGF mAb with anti-A2V on the Ang2 binding could be excluded. Moreover, reliable quantification of free Ang2 concentrations in plasma samples was assured by interference testing. Performing advanced free Ang2 determination including the immunodepletion step in parallel to established Ang2 measurement without immunodepletion, we compared free with total Ang2 concentrations in human plasma samples obtained from an anti-A2V Phase 1 clinical study. Samples from untreated patients displayed rather low and equal values for both free and total Ang2. In contrast, samples from drug-treated patients showed a significant reduction of free Ang2 accompanied by an accumulation in total Ang2. These results underline the value of the novel immunodepletion procedure for reliable discrimination of free vs. total target quantification with particular importance for pre-clinical and clinical development of anti-A2V. Moreover, this approach may serve as universal concept for the determination of free target levels of bispecific therapeutic antibodies.

Immunogenicity testing of therapeutic antibodies in ocular fluids after intravitreal injection.

AIM High drug concentrations in ocular fluids after intravitreal administration preclude the use of drug-sensitive immunoassays. A drug-tolerant immunoassay is therefore desirable for immunogenicity testing in ophthalmology. EXPERIMENTAL Immune complex (IC) antidrug antibody (ADA) assays were established for two species. The assays were compared with the bridging assay in ocular and plasma samples from two preclinical studies. RESULTS The IC assays showed high drug tolerance, which enabled a reliable ADA detection in ocular fluids after intravitreal administration. The IC assays were superior to the bridging assay in the analysis of ocular fluids with high drug concentrations. CONCLUSION The IC assay allows a reliable ADA detection in matrices with high drug concentrations, such as ocular fluids.