Biopharmaceuticals, including monoclonal antibody therapeutics, have become an important tool in the treatment of various diseases [1], [2], [3]. However, due to their size and complex structure, biotherapeutics are prone to chemical and physical degradation [4], [5]. To ensure structural integrity throughout production, storage, transport and patient administration, identifying the ideal formulation composition is a crucial part of the biologic development process, which requires a comprehensive understanding of the candidate molecule. This task becomes more and more challenging since the continuous growth in this pharmaceutical area consequently increases the complexity of formulation development.

In the last years, AbbVie has developed a robotized, fully automated high throughput liquid formulation screening (HTS) line. This automation line enables a quality by design (QbD)-based approach by parallel testing of multiple stress conditions in combination with formulation excipient parameters [6]. As in early development stages the drug substance availability is often limited, the HT-screenings are performed in a miniaturized assay format in multi-well plates, thus only small amounts of the candidate protein are required. To evaluate the protein stability, various stress assays are performed, simulating the “real-life” degradation risk of a therapeutic protein, such as freeze/thaw, interfacial and temperature stress. They mimic freeze/thaw procedures during storage and transport, mechanical stress during manufacturing processes as well as incubation at storage (5 °C), accelerated and stress conditions (25 and 40 °C) as proposed by the ICH guidelines. To identify and quantify the impact of different stress factors, the standardized workflow includes a set of analytical methods, compatible with HT applications and multi-well formats. Each method is characterized by fast execution, universal applicability for all candidate molecules and low sample consumption. Suitable methods are chromatography-based or plate reader analytics [7]. Compared to a standard manual formulation development approach, HTS formulation development facilitates the generation of large data sets that allow comprehensive characterization of the candidate molecule, the identification of critical formulation parameters, and hence the definition of the corresponding design space. Further, due to standardization of the screening process, data comparison across multiple development projects is possible, creating opportunities for data science and prediction models in the future.

While the existing HTS workflow already considers the most common stress conditions, it currently does not include routine screenings for oxidation, leaving potential molecule liabilities undiscovered. Oxidation, one of the most common chemical degradation pathways of therapeutic proteins, can be induced by a large variety of mechanisms. Examples are oxidative contaminations during the manufacturing process or oxidants derived from excipients, such as hydrogen peroxide [8], [9]. Further causes of oxidation are the presence of transition metals [10], [11], leading to metal-catalyzed oxidation (MCO) or exposure to visible and UV light (photooxidation) [12], [13], [14]. Each of the examples described above, promotes the generation of reactive oxygen species (ROS) and/or free radicals. Depending on the extent and location, oxidation of amino acid residues can lead to modifications of the protein structure and thus changes in the protein’s physicochemical properties. As a consequence, the oxidized product might have altered pharmacokinetics, a reduced biological activity and an increased immunogenicity compared to the native form [15]. The different oxidation mechanisms and their consequences have been already extensively reviewed in literature [16], [17]. To ensure quality and safety of the product, oxidation susceptibility of candidate molecules is routinely assessed in forced degradation studies, using chemical oxidants such as H2O2 or tBHP. Other strategies are treatment with the free radical generating compound 2,2-azo-bis(2-amidinopropane)dihydrochloride (AAPH) or photostability studies according the ICH Q1B guideline [18], [19]. These studies, as part of the early-stage development, aim to identify potential degradation pathways of the molecule, hence harsh stress conditions are used. However, these conditions do not necessarily reflect practical oxidation risks. Additionally, oxidation experiments are often conducted using a limited set of oxidants or non-representatively low protein concentrations. Consequently, the oxidation susceptibility assessed in early stage forced oxidation studies cannot be applied for predicting molecule liabilities during formulation development.

The aim of the present study was the development of an HT oxidation screening strategy which allows comprehensive characterization of the oxidation susceptibility under relevant conditions within a short period of time and with minimal sample consumption. To allow integration in the already existing automated HTS workflow, it was necessary to define effective assay conditions considering representative oxidation risks of therapeutic proteins using the established equipment in the HTS line. To identify universally applicable HTS oxidation assays, in this present study we exposed five different monoclonal antibodies at an HTS standard protein concentration of 100 mg/ml to H2O2, tBHP, AAPH, MCO (Fenton reaction) visible and UV-A light. Further, the suitability of HTS analytics to detect the impact of oxidation on the proteins under representative stress conditions was evaluated. Therefore, the outcome of the oxidation experiments was characterized using Size Exclusion Chromatography, Cation Exchange Chromatography, Protein A Chromatography, Intrinsic Tryptophan Fluorescence spectroscopy and Reversed-Phase Chromatography of antibody subunits. Based on the data set obtained in this study, we established a universally applicable oxidation HT screening strategy for monoclonal antibodies.