(M1530-05-29) Sample Analysis of a Multidomain Fusion Protein Drug with Linker Peptides: Tiered Domain Characterization of Neutralizing and Anti-Drug Antibodies in Correlation to Pharmacokinetic Profiles

Purpose: This study evaluated anti-drug antibody (ADA) and neutralizing antibody (NAb) responses in patient serum samples after treatment with a multidomain fusion protein drug containing linker peptides. The fusion protein includes a receptor interacting domain (Domain A), a modulator domain (LBDA), and stabilizing peptides 1 and 2, which enable selective receptor activation to stimulate antitumor immune responses (Figure 1). The objective was to identify not only the presence of ADA/NAb but also to characterize the specific domains of the drug targeted by these antibodies and to understand how they affect Pharmacokinetic (PK) profiles and drug efficacy in patients.

Methods: Patient serum samples were evaluated for anti-drug antibodies (ADA) and neutralizing antibodies (NAb) using a tiered ECL ligand-binding assay. Characterization tiers for Domain A and LBDA were fully validated, while fit-for-purpose validation was performed for the Pep1 and Pep2 tiers due to limited reactivity with the drug-specific PCs.  For validated tiers, low positive control (LPC) levels were refined through targeted LPC determination with and without domain-specific confirmatory reagents.  For each domain tier, a specific confirmatory cut point (CCP) was set to ensure specificity (Figure 2). ADA evaluation followed the standard three-tier testing process. NAb testing was conducted on confirmed positive samples using a functional cell-based NAb assay. NAb-positive samples with ratios above the cutoff advanced to domain characterization, which used a four-tiered assay to identify ADA binding to specific epitopes on the multidomain fusion protein. A single set of positive controls (PCs), made of rabbit IgG affinity-purified against the drug, was used across all ADA tiers.

Results: The multi-tiered assay strategy provided detailed immunogenicity profiles for each patient, capturing both the presence and specificity of ADA and NAb responses at different timepoints throughout the study.  Among NAb-positive samples with ratios above the cutoff criteria, domain characterization revealed distinct binding patterns across the four targeted epitopes of the multidomain fusion protein.  This enabled classification of patient responses into domain-specific immunogenicity profiles, highlighting variability in immune recognition across the study samples.  The integration of ADA/NAb data with PK profiles supports a more informed interpretation of immunogenicity impact on drug disposition and therapeutic efficacy.  Notably, Domain A was more frequently targeted, indicating different immunogenic potential across the drug’s structural regions. Interestingly, the pharmacokinetic impact of the ADA and its neutralizing action appears to be further linked to the LBDA domain, suggesting a critical MOA for LBDA in receptor interaction.  As expected, ADA/NAb responses tend to be more common the longer a patient is on treatment. Towards the end of the study, most NAb-positive samples had ADAs targeting two or more of the four domains’ tiers.  In reviewing a patient’s PK Profile and NAb Ratio over the course of treatment, along with a heat map of Inhibition/CCP ratio for domain characterization tiers 1-4 (Figure 3), shows that NAb activity may not be equally correlated across all domains. For the patient in Figure 3, the ADA for Domain A remained high but did not significantly affect drug concentration as much as the ADA for LBDA, which led to a reduction in PK drug levels.

Conclusion: This optimized assay design and domain-specific characterization provides a more nuanced understanding of ADA/NAb responses.  This approach enhances the interpretation of immunogenicity data by identifying the specific regions of the drug targeted by patient antibodies.  The use of one ADA method with multiple tiers allows for efficient analysis of a high sample count and efficiently utilizes method reagents.  The ability to distinguish which specific domains of the drug are targeted by patient-derived antibodies provides mechanistic insight into the nature of the immune response.  This is particularly valuable for complex biologics, where immunogenicity may be driven by discrete structural regions in this case on the multidomain fusion protein drug containing linker peptides.  The determination of which domain ADA targets have PK implications is critical in monitoring patient safety and drug efficacy during treatment.

References: International Council for Harmonisation (ICH). M10 Bioanalytical Method Validation and Study Sample Analysis. Step 4. 2022

Acknowledgements: Affiliations: Labcorp Early Development Laboratories Inc. (Chantilly, VA, USA)

Figure 1. Mechanism of Action of the Multidomain Fusion Protein Drug Containing Linker Peptides

Figure 2. Validation Parameters Established for ADA Method with 4 Characterization Tiers (2 Fully Validated, 2 Qualified)

Figure 3. PK Profile and NAb Ratio Over the Course of Treatment and Heat Map of Inhibition/CCP Ratio for Domain Characterization Tiers 1-4.