Deborah Finco-Kent

Recommendations for the validation of immunoassays used for detection of host antibodies against biotechnology products

Most biological drug products elicit some level of anti-drug antibody (ADA) response. This antibody response can, in some cases, lead to potentially serious side effects and/or loss of efficacy. In humans, ADA often causes no detectable clinical effects, but in the instances of some therapeutic proteins these antibodies have been shown to cause a variety of clinical consequences ranging from relatively mild to serious adverse events. In nonclinical (preclinical) studies, ADA can affect drug exposure, complicating the interpretation of the toxicity, pharmacokinetic (PK) and pharmacodynamic (PD) data. Therefore, the immunogenicity of therapeutic proteins is a concern for clinicians, manufacturers and regulatory agencies. In order to assess the immunogenic potential of biological drug molecules, and be able to correlate laboratory results with clinical events, it is important to develop reliable laboratory test methods that provide valid assessments of antibody responses in both nonclinical and clinical studies. For this, method validation is considered important, and is a necessary bioanalytical component of drug marketing authorization applications. Existing regulatory guidance documents dealing with the validation of methods address immunoassays in a limited manner, and in particular lack information on the validation of immunogenicity methods. Hence this article provides scientific recommendations for the validation of ADA immunoassays. Unique validation performance characteristics are addressed in addition to those provided in existing regulatory documents pertaining to bioanalyses. The authors recommend experimental and statistical approaches for the validation of immunoassay performance characteristics; these recommendations should be considered as examples of best practice and are intended to foster a more unified approach to antibody testing across the biopharmaceutical industry.

Recommendations on risk-based strategies for detection and characterization of antibodies against biotechnology products

The appropriate evaluation of the immunogenicity of biopharmaceuticals is of major importance for their successful development and licensure. Antibodies elicited by these products in many cases cause no detectable clinical effects in humans. However, antibodies to some therapeutic proteins have been shown to cause a variety of clinical consequences ranging from relatively mild to serious adverse events. In addition, antibodies can affect drug efficacy. In non-clinical studies, anti-drug antibodies (ADA) can complicate interpretation of the toxicity, pharmacokinetic (PK) and pharmacodynamic (PD) data. Therefore, it is important to develop testing strategies that provide valid assessments of antibody responses in both non-clinical and clinical studies. This document provides recommendations for antibody testing strategies stemming from the experience of contributing authors. The recommendations are intended to foster a more unified approach to antibody testing across the biopharmaceutical industry. The strategies proposed are also expected to contribute to better understanding of antibody responses and to further advance immunogenicity evaluation.

Synthesis and Characterization of Rhenium and Technetium-99m Labeled Insulin

A (99m)Tc-labeled insulin analogue was synthesized through a direct labeling method in which the [(99m)Tc(CO)(3)](+) core was combined with a protected insulin derivative (9) bearing a M(I) chelate linked to the first amino acid of the B-chain (B1). Regioselective labeling was achieved by careful control over the pH and the reaction time. Following a TFA-anisole mediated deprotection step (decay-corrected yield of 30 +/- 11%, n = 4), the identity of the final (99m)Tc-labeled product was confirmed by HPLC. Displacement of (125)I-insulin from the insulin receptor (IR) by the Re analogue 6 was similar to that of native insulin (17.8 nM vs 11.7 nM, respectively). The extent of autophosphorylation and Akt activation, as indicated by production of phospho-Akt (pAkt), showed no statistical difference between 6 and native insulin in both assays. These results support the use of the reported (99m)Tc-insulin derivative as a tracer for studying insulin biochemistry in vivo.

Development and Validation of a Canine T-Cell-Dependent Antibody Response Model for Immunotoxicity Evaluation

A T-cell dependent antibody response (TDAR) model to evaluate compounds for potential immunotoxicity in dogs has not been reported. The objective of these studies was to develop and validate a dog TDAR model using the T-cell dependent antigen, keyhole limpet hemocyanin (KLH). Studies were conducted to determine the appropriate dose of KLH, immunization route and kinetics of the antibody response to KLH in the dog. To validate the sensitivity of this method, we investigated the TDAR to KLH in the dog with a known immunosuppressive drug, cyclosporine (Neoral). The results of this study demonstrate that a robust primary IgM and IgG response to KLH can be generated in dogs and the IgG response was sensitive to cyclosporine treatment.

Development and Validation of a Quasi-Quantitative Bioassay for Neutralizing Antibodies Against CP-870,893

The human monoclonal antibody CP-870,893 is a CD40 receptor agonist currently being developed for the treatment of cancer. A bioassay to measure neutralizing antibodies (Nab) to CP-870,893 in 5% human serum matrix was developed and validated utilizing the Daudi cell line and flow cytometric detection. Additionally, samples from CP-870,893 treated cynomolgus monkeys were analyzed in the bioassay and compared to results obtained using a competitive receptor-binding (CRB) Nab immunoassay to determine if the CRB assay may be used in place of the bioassay. Treatment of Daudi cells for 2 d with CP-870,893 leads to a concentration-dependent increase in CD54 cell surface expression. The presence of antidrug Nab attenuates CP-870,893 binding to CD40 and the induction of CD54. An anti-idiotype monoclonal antibody (Mab) and a monkey sera pool were identified as positive controls for neutralization of CP-870,893. During development, it was observed that the assay robustness was altered by culture media and FBS substitutions. For validation the following parameters were established: cutpoint factors in the presence (0.779) and absence (1.282) of 50 ng/ml CP-870,893, linear region of the concentration-response (1–100 ng/ml CP-870,893), intra- and inter-assay precision (CV ≤ 25%), specificity and recovery (±25%), sensitivity (∼500 ng anti-idiotype Mab per ml serum), technician to technician ruggedness (CV ≤ 25%), and stability (positive control, CD54 labeling, and cell line). A concentration dependent increase in CP-870,893 neutralization was observed in a 3-mo toxicity study in monkeys using both the Bioassay and CRB assay (R2 = 0.94) suggesting the CRB Nab assay may be a suitable alternative to a bioassay. Based on the precision, specificity, sensitivity, and robustness, the validated bioassay is suitable for quasi-quantitative analysis of neutralizing anti-CP-870,893 antibodies in human serum.

Optimization and validation of a flow cytometric method for immunophenotyping peripheral blood lymphocytes from cynomolgus monkeys (Macaca fascicularis)

BACKGROUND Guidelines published by the Food and Drug Administration and Center for Human Medicinal Products describe the need to assess immunotoxic effects in nonclinical studies that evaluate drug toxicity, including the use of immunophenotyping to measure immunotoxicity. We are not aware of previous studies, however, that have validated methods for immunophenotyping peripheral blood lymphocyte subsets in whole blood samples from cynomolgus monkeys. OBJECTIVE The purpose of this study was to optimize and validate a flow cytometric assay for immunophenotyping lymphocytes in the peripheral blood of cynomolgus monkeys. METHODS A series of prevalidation experiments were done to determine optimal reagents, volumes, timing, and other procedural details of the flow cytometric assay. Using the optimized method, we then determined precision, interindividual variation, laboratory-to-laboratory variability, and sample stability. Stabilized human blood was used as a positive control for staining, processing, and analysis. The percentage and number of pan-T cells (CD3+), T-helper cells (CD3+4+), T cytotoxic/suppressor cells (CD3+8+), natural killer cells (CD3-16+), and B-cells (CD3-20+) were determined in 146 male and 140 female, clinically healthy monkeys and reference intervals were calculated. RESULTS By doing 4-color staining with a lyse-wash method, intra- and interassay precision were <5% for all lymphocyte subsets. Variability between technicians and laboratories was minimal (CVs<3%). Samples were stable for up to 24 hours after staining and fixing. CONCLUSIONS The validated method is extremely robust and can be performed under good laboratory practice conditions to support nonclinical studies. Reference intervals for lymphocyte subsets were similar to those previously reported.

Development and Validation of a Radioligand Binding Assay to Measure Insulin Specific IgG Subclass Antibodies in Human Serum

Abstract: The objective was to develop and validate a radioligand binding assay for insulin antibodies (IABs) of the IgG1, IgG2, IgG3, and IgG4 subclasses in human serum. The validation studies focused on determining specificity, capacity, linearity, sensitivity, and precision of each assay. It was seen that our assay for IAB IgG subclasses is specific and has sufficient capacity to measure each of the subclasses in human serum. Moreover, the linear region and limits of detection and quantitation for each assay are clearly determined.