Evolution of Antidrug Antibody Assays During the Development of Anti-Tissue Factor Pathway Inhibitor Monoclonal Antibody Marstacimab.

Tissue factor pathway inhibitor (TFPI) is an endogenous inhibitor of the extrinsic coagulation pathway. In patients with hemophilia A or B, inhibition of TFPI is an alternative therapeutic approach that augments the extrinsic coagulation pathway. Marstacimab is an investigational fully human monoclonal antibody that binds and neutralizes TFPI and is being evaluated as a prophylactic treatment to prevent or reduce the frequency of bleeding episodes in patients with severe hemophilia A or B, with or without inhibitors (antibodies against coagulation factors). However, the efficacy, safety, and pharmacokinetics of marstacimab may be affected by the induction of antidrug antibody (ADA) responses. Here, we describe the evolution and validation of three quasi-quantitative electrochemiluminescence-based methods to detect marstacimab ADAs, starting from their use in a first-in-human phase 1 study to their use in phase 2 and 3 clinical studies of patients with severe hemophilia. For all three methods, validation criteria evaluated the performance of the assays in screening and confirmatory cut points, precision, selectivity, drug tolerance, target interference, and stability. Additional criteria for validation were dilution linearity (Methods 1 and 2) and low positive control concentration, prozone effect, plate homogeneity, and robustness (Method 3). The three methods met validation criteria and are a potentially valuable tool in detecting the induction of marstacimab ADAs during treatment in patients with hemophilia.

Anti-drug Antibody Assay Validation: Improved Reporting of the Assay Selectivity via Simpler Positive Control Recovery Data Analysis.

Anti-drug antibody (ADA) assay selectivity is evaluated during assay validation to assess the potential for individual matrices to interfere with detection of ADA. While current EMA and FDA guideline documents suggest comparative analysis with and without matrix, they do not provide specific recommendations on the acceptance criteria such as an acceptable percent positive control (PC) recovery range or positive rate. Industry has adopted an approach where recovery of PC spiked sample is expected to fall within ± 20% (80 to 120%) vs. that for the PC material spiked in negative control matrix or assay buffer. Here, it is proposed that ADA assay selectivity evaluated using a qualitative assessment of PC recovery vs. a PK-like quantitative method may be more appropriate. The PC recovery test should focus on the reliability of the method to detect the low PC level in individual samples and avoid false-negative ADA reporting. Therefore, it is proposed that assessment of high PC level as well as the assessment of quantitative percent recovery (within ± 20%) should not be included in the test. The recovery test may be viewed as acceptable should a pre-selected number of individual samples (for example at least 8 or 9 out of 10) prepared at the low PC concentration of the assay score as ADA positive.

Neutralizing Antibody Assay Development with High Drug and Target Tolerance to Support Clinical Development of an Anti-TFPI Therapeutic Monoclonal Antibody.

Immunogenicity is a major challenge for protein therapeutics which can potentially reduce drug efficacy and safety and is often being monitored by anti-drug antibody (ADA) and neutralizing antibody (NAb) assays. Circulating targets and residual drugs in matrices can have significant impacts on accuracy of results from ADA and NAb assays, and sufficient drug and target tolerance for these assays are necessary. Here, we report the development of a competitive ligand binding (CLB) NAb assay for an anti-TFPI (tissue factor pathway inhibitor) monoclonal antibody (PF-06741086) with high drug and target tolerance to support ongoing clinical studies. A double acid affinity capture elution approach was used to mitigate drug interference, and a robust target removal strategy was employed to enhance target tolerance. The validated NAb assay has sensitivity of 313 ng/mL, drug tolerance of 50 µg/mL, and target tolerance of 1200 ng/mL. A step-by-step tutorial of assay development is described in this manuscript along with the rationale for using a high drug/target tolerant NAb assay. The NAb assay cut point factor obtained was 0.78. Other assay performance characteristics, e.g., precision and selectivity, are also discussed. This validated method demonstrated a superior drug and target tolerance to warrant specific and precise characterization of the NAb responses in support of ongoing clinical studies.

Approaches to Resolve False Reporting in Neutralizing Antibody Assays Caused by Reagent Leaching from Affinity Capture Elution Solid Phase.

Insufficient drug tolerance presents a major challenge in the development of neutralizing antibody (NAb) assays for biotherapeutics. Sample pre-treatment using solid-phase extraction with acid dissociation (SPEAD) is widely reported to improve drug tolerance. In this paper, a case study is presented in which SPEAD was used in conjunction with a competitive ligand binding NAb assay format. A significant degree of biotin-drug conjugate leaching was observed resulting in the reporting of both false positive and false negative results in NAb assay. Mitigation steps have been evaluated to address drug/biotin-drug conjugate leaching. These steps included assessment of the streptavidin-coated plate in conjunction with biotin-drug conjugates at various biotin molar challenge ratios (MCR). In addition, an alternative method based on covalent capture of the drug on an aldehyde-activated plate was assessed. Both approaches were compared for the degree of drug/biotin-drug conjugate leaching during the second elution step of the SPEAD procedure. Moreover, the impact of various conditions on the assay performance was assessed, including elution pH, sample incubation time, and biotin MCR. For the covalent drug capture method, capture conditions were evaluated. Optimized conditions in both streptavidin capture and covalent capture methods enabled a significant reduction of drug/biotin-drug conjugate leaching. A streptavidin high binding capacity approach using biotin-drug conjugate with a MCR of 50:1 was chosen as the optimal method yielding a NAb assay with a fit for purpose sensitivity (153 ng/mL) and a drug tolerance of up to 50 µg/mL with 500 ng/mL PC.

A Simple Approach to Determine a Curve Fitting Model with a Correct Weighting Function for Calibration Curves in Quantitative Ligand Binding Assays.

In ligand binding assays (LBA), the concentration to response data is a nonlinear relationship driven by the law of mass action. Four parameter logistic (4PL) and five parameter logistic (5PL) curve fitting models are two widely accepted and validated models for LBA calibration curve data. Selection of the appropriate regression model and weighting function are key components of LBA development. Assessment of selected model and weighting function should be performed during assay development and confirmed later during validation. There has been limited published work on practical approaches to determining an appropriate weighting function and selection of a regression model for ligand binding assays. Herein, a structured scheme is presented to determine both. By applying commonly available software, assay performance data were analyzed to determine weighting functions and associated choice of a curve fitting model in three presented case studies. As a result, assay ranges of quantification were improved by reducing lower limit of quantification (from 1.00 to 0.317 ng/mL in one case study and from 2.06 to 1.37 ng/mL in another) or extending both low and upper limits of quantification(e.g., 1.04 to 48.3 ng/mL improved to 0.602 to 145 ng/mL). In addition, assay calibration curve performance demonstrated improved assay accuracy (%RE) and precision (%CV). Recommendations on decision flow when determining appropriate weighting function and curve fit model are presented.

Paradigm of combination biologics: analytical challenges related to pharmacokinetic assays and interpretation of pharmacokinetic and immunogenicity results.

BACKGROUND: Combination biologic therapy is an emerging area of clinical development and the physiological and analytical impact of one treatment on the other requires careful assessment. Significant analytical challenges are associated with developing the corresponding pharmacokinetic assays and further challenges arise in interpreting the subsequent in vivo data, which may be confounded by immunogenicity to one or both of the biologics. RESULTS: A case study of two monoclonal antibody therapeutics, given in combination, is presented where the immunogenicity rates differed significantly when the drug(s) were administered as monotherapy or in combination. CONCLUSION: The interpretation of the in vivo data is inextricably linked to an in-depth understanding of the formats and performance attributes of the associated pharmacokinetic and immunogenicity assays.

A novel gamma-fitting statistical method for anti-drug antibody assays to establish assay cut points for data with non-normal distribution.

In recent years there has been growing recognition of the impact of anti-drug or anti-therapeutic antibodies (ADAs, ATAs) on the pharmacokinetic and pharmacodynamic behavior of the drug, which ultimately affects drug exposure and activity. These anti-drug antibodies can also impact safety of the therapeutic by inducing a range of reactions from hypersensitivity to neutralization of the activity of an endogenous protein. Assessments of immunogenicity, therefore, are critically dependent on the bioanalytical method used to test samples, in which a positive versus negative reactivity is determined by a statistically derived cut point based on the distribution of drug naïve samples. For non-normally distributed data, a novel gamma-fitting method for obtaining assay cut points is presented. Non-normal immunogenicity data distributions, which tend to be unimodal and positively skewed, can often be modeled by 3-parameter gamma fits. Under a gamma regime, gamma based cut points were found to be more accurate (closer to their targeted false positive rates) compared to normal or log-normal methods and more precise (smaller standard errors of cut point estimators) compared with the nonparametric percentile method. Under a gamma regime, normal theory based methods for estimating cut points targeting a 5% false positive rate were found in computer simulation experiments to have, on average, false positive rates ranging from 6.2 to 8.3% (or positive biases between +1.2 and +3.3%) with bias decreasing with the magnitude of the gamma shape parameter. The log-normal fits tended, on average, to underestimate false positive rates with negative biases as large a -2.3% with absolute bias decreasing with the shape parameter. These results were consistent with the well known fact that gamma distributions become less skewed and closer to a normal distribution as their shape parameters increase. Inflated false positive rates, especially in a screening assay, shifts the emphasis to confirm test results in a subsequent test (confirmatory assay). On the other hand, deflated false positive rates in the case of screening immunogenicity assays will not meet the minimum 5% false positive target as proposed in the immunogenicity assay guidance white papers.