Monoclonal antibodies and antibody-derived molecules (e.g., bispecifics, multispecifics, fragments) represent a rapidly expanding therapeutic modality across multiple therapy areas. The term ‘developability’ encompasses the feasibility of molecules to successfully progress to development via evaluation of their physicochemical properties, tendency for self- interaction, aggregation, thermal stability, colloidal stability and optimization of their properties through sequence engineering. Selection of multi-parameter optimized antibody molecules, taking into consideration biological function, safety, and developability, allows for a streamlined and successful development. We developed an efficient and practical high-throughput developability workflow (100’s-1,000’s of molecules) implemented during the early phase of antibody generation and screening is critical to guide the selection process of the best lead candidates. This workflow allows elimination of antibodies with suboptimal properties early in the discovery process and rank ordering of molecules for further evaluation in the candidate selection process. Mining of our large high-quality datasets identified novel patterns and correlations between biophysical assays. These patterns and correlations represent the basis for training deep neural networks and establishing machine learning algorithms for in silico interrogation and prediction of developability profiles from a massive diversity of antibody sequences available in the discovery space.

A preclinical candidate was selected from a diverse pool of engineered protease constructs expressed in mammalian and bacterial hosts. Selection from this diverse pool required generic analytical methods and candidate ranking based on developability/QbD principles with an emphasis on production levels and intrinsic stability. Analytical methods were further developed as the program progressed with a fit-for-purpose approach based on performance monitoring and streamlined method verification.

• What kind of data integrity operational elements are appropriate for R&D vs. GxP labs?
• How can it be confirmed that everyone in the lab understands the rationale and justification of quality practices for R&D CMC labs?
• How can a regulatory affairs reviewer assure all of the R&D data included in a product dossier are in fact authentic, complete, accurate, etc.?
• What should be done if errors or omissions are discovered in key R&D CMC study reports?

Excipients in drug products fulfill a range of functions. In parenteral protein biotherapeutics, excipients provide physical and chemical stability, while contributing to osmolality. Excipients can however display complex behavior and under certain circumstances, may even destabilize the active protein molecule. Additionally, excipients may not be pharmacologically inert. Excipients must therefore be selected with care, and their control considered during development of the control strategy for the product.

Physical degradation and aggregation of proteins at interfaces (e.g., solid-liquid, liquid-liquid, and air-liquid) can negatively impact the manufacturability, shelf-life stability, and administration of protein therapeutics.  Despite the critical impact of surfaces on protein stability, the mechanisms of interfacial degradations remain poorly understood and highly speculative in the pharmaceutical literature.  In this work, we have collaborated with the National Institute of Standards and Technology (NIST) and University of Manchester to implement and develop state of the art metrology and modeling tools to investigate protein interfacial degradation at pharmaceutically relevant surfaces (e.g., stainless steel, glass, and silicone oil).  Our improved understanding of protein interactions and degradation at interfaces could lead to establishing novel and prospective mitigation strategies, spanning from protein engineering, formulation development and manufacturing process controls. 

The generation of chimeric antigen receptor (CAR)-expressing T cells from patient lymphocytes commonly relies on viral vectors for transgene delivery. Viral vector production occurs in a producer cell line and the manufacturing process requires partitioning of intact viral particles away from host cell impurities. Although a nuclease treatment step aids in size reduction and improves the clearance of residual DNA species, there is still a potential for nucleic acid impurities to exist in the vector product, to be transferred into the CAR T drug product manufacturing process, and subsequently to be present in the CAR T cell infusion. Residual DNA impurities are a critical quality attribute. As such, a risk-based analytical strategy for the characterization of residual DNA impurities will be presented.

Distinguishing aggregated API from other particle types is important for understanding the root cause of instability. Existing methods are unreliable, too cumbersome and difficult to use in many workflows. With Aura, you can now finally count, size, and characterize aggregates and identify them as proteins, non-proteins, or other molecules.

Nuclear magnetic resonance (NMR) spectroscopy provides site-specific atomic-level information about protein drugs, and hence can detect any changes in local protein structure and dynamics that are often not detectable by global probes, such as circular dichroism and fluorescence. We will present recent case studies in which we used 2D NMR techniques to probe antibody-polysorbate interactions and the effect of chemical modifications on therapeutic proteins.

Development of high-resolution techniques for defining the higher order structure (HOS) of biotherapeutics has emerged as a priority in the pharmaceutical industry. This talk will describe applications of nuclear magnetic resonance (NMR) for HOS assessment, with a focus on mAb drugs. It will cover the extent to which NMR can detect and assign HOS differences and examples of chemometric methods for automated spectral analysis. 

Multi-Attribute Method (MAM) is a peptide–mapping based method that provides targeted monitoring of product quality attributes and non-targeted new peak detection. MAM has been implemented in the pharmaceutical industry for process development and is advancing into the Quality Control (QC) environment in alignment with Quality-by-Design principles. This talk describes the qualification of MAM as a potential platform method for QC at Genentech.

Protein aggregation and self-association are key attributes that may impact safety, efficacy and long term storage of biotherapeutics. This presentation will provide case studies using sub-visible particle analysis, composition gradient multi-angle light scattering and isothermal hold dynamic light scattering to demonstrate the power of high resolution techniques to detect protein self-association propensity and aggregation that may be otherwise missed by conventional techniques such as chromatography, electrophoretic methods and mass spectrometry.

The development of mass spectrometric (MS) methods for characterization of adeno-associated virus (AAV) capsid proteins allows for the complete structural elucidation of constituent viral capsids in gene therapy development. Conventional peptide map and intact protein analysis, as well as a recently developed ZipChip capillary electrophoresis (CE)-MS method will be discussed in the presentation for various AAV capsid analyses, including serotype identification, confirmation of mutation and characterization of post-translational modifications.

Methods for characterization and stability assessment of therapeutic monoclonal antibodies (mAbs) are well established. Same applies to protein subunit vaccine candidates. However, when working with conjugated products, this does not hold true. Here we present the challenges, findings, and at the end bordering of our expectation, in evaluation and identification of assays suitable for characterization, formulation and stability assessment of a conjugated vaccine candidate. We will present our experience with the conjugated HIV vaccine candidate.

Stream-lining ADC process development activities is key for enabling fast-to-clinic timelines.  Increasingly, we find the need to implement mass spectrometry to characterize our ADCs during pre-clinical development activities.  The challenge is how to adapt the MS peptide map used for characterization to a high-throughput environment where it can be used to drive process development decisions.  We focused on alleviating two methodological pain-points: inefficient, time-consuming data analysis and sample preparation workflows. We leveraged a vendor-neutral mass spectrometry data analysis software suite to perform unsupervised data analysis in conjunction with automated peptide map sample preparation on a liquid handling robotic system.  In this talk, we will focus on how we’ve implemented these workflows and discuss the strategies we have utilized to ensure that we can increase capacity.

Analytical Ultracentrifugation is a versatile analytical method that provides data for the determination of the mass, hydrodynamic and thermodynamic properties of macromolecules like e.g. proteins and other Biotherapeutics. This presentation gives an update on the example of the characterization of genome loaded viral capsids.