A novel SPR method testing interaction of Fc gamma receptors with Antibody-Antigen immunocomplex was developed. Although assay orientation had some effect, antigen had even more impact on the binding affinity of IgGs to most activating FcgRs, especially on IgG1-FcgRI interaction, and IgG2-FcgRIIIa (158F) interaction. These data suggest it may be useful to evaluate the IgG-FcgR interaction in the presence of antigen to help design safer and more effective biotherapeutics.

Identification of mAbs with low levels of self-association has traditionally required relatively concentrated protein solutions, confounding identification of such molecules during early discovery campaigns. We have developed charge-stabilized self-interaction nanoparticle spectroscopy (CS-SINS), a colormetric assay that measures colloidal self-interaction of antibodies in ultra-dilute solutions (0.01 mg/mL). CS-SINS results are predictive of high concentration properties such as viscosity and opalescence, which only emerge at orders of magnitude higher concentrations (100+ mg/mL).

Advances in mass spectrometry (MS) make this technology uniquely suited for characterizing protein therapeutics and cell therapies as well as fulfilling current unmet needs for such product knowledge. The San Jose Mass Spectrometry Center of BGI Americas is an innovation lab where it offers one of the industry’s most comprehensive multi-omics service portfolios. This presentation focuses on the Center’s MS techniques and applications for these emerging therapeutics.

Biolayer interferometry (BLI) and surface plasmon resonance (SPR) are leading technologies in real time label-free protein analysis, each with distinct features and benefits. One major difference, however, are the variety of injections offered by different SPR platforms. This presentation will discuss features of both technologies, the range of novel injections featured on the new Octet® SF3 SPR, and their potential applications across a number of different research areas.

Opalescence of biopharmaceutical solutions can indicate suboptimal colloidal stability which is generally an undesirable attribute that requires investigation and remediation. While there are numerous instrumentation options available for measuring opalescence, cross-instrument comparisons and detailed knowledge of analytical biases are limited. We highlight key findings from a multi-instrument investigation and demonstrate the advanced application of a 90° angle light scattering instrument as a suitable approach for making low volume, temperature-controlled, nephelometric measurements of opalescence. Moreover, we demonstrate how this approach enables the simultaneous evaluation of key physical properties that are pertinent to investigations of opalescent biopharmaceuticals.

Successful biologics must exhibit a series of suitable properties which depend on both protein sequence and buffer composition. In the context of this high-dimensional optimisation problem, we will show how machine learning algorithms can accelerate the design of biopharmaceutical formulations. These methods provide high speed of converging to optimal combinations of excipients, the ability to transfer prior knowledge, and the identification of trade-offs in optimising multiple biophysical properties.

We describe techniques that combine flow imaging microscopy with machine learning to analyze populations of particles, protein aggregates, and cells. Convolutional neural networks can be used to create “particle facial recognition” algorithms that allow detection and classification of particles that may appear during bioprocessing, even in samples that contain a background of pre-existing particles. Applications will be discussed, including root-cause analysis of aggregates in monoclonal antibody formulations, detection of particles formed during fill-finish operations for adjuvanted vaccines, and detection of microbial infections in blood.   

Like viral vectors, lipid nanoparticle formulations can be characterized by multi-wavelength analytical ultracentrifugation in high resolution to differentiate heterogeneous mixtures based on nucleic acid loading state, morphology, and particle size. In this talk, I will discuss how a D2O density matching approach can be employed to characterize particles based on their partial specific volume, or loading density. This adds a third characterization dimension to the hydrodynamic and spectral separation.

Therapeutic monoclonal antibodies (mAbs) are increasingly used in hospitals. However, mAbs treatments represent high immunological risk due to denaturation and aggregation issues that may occur during manipulations at a hospital. Our purpose is to provide information on mAbs states before administration to patients. Native mass spectrometry (MS) is a promising approach to characterize mAbs using near-physiological conditions. We have investigated whether this technique coupled to capillary electrophoresis or to size exclusion chromatography could allow not only mAbs identification but most importantly evaluation of mAbs conformational states. Another approach based on electrophoretic methods coupled to multivariate analysis has been also studied.

Subvisible particles are critical quality attributes for biologics. Gene and cell therapies present unique stability challenges since they are more complex than protein therapeutics. Here we show how the Aura counts, images, sizes, and identifies subvisible biologic aggregates and extrinsic particles in a rapid and low-volume format. We present case studies overviewing protein therapeutic stability from cell line development through manufacturing, AAV stability, and cell therapy aggregation and particle purity. 

Proteins have personalities, reflected in stability challenges and formulation selections. Investigating >100 diverse proteins, my appreciation continues to increase for the depth of data that can be gleaned from a well-equipped formulation/biophysical assessment lab. This presentation will review some of the informative science behind several key biophysical tools and applications for challenging molecules. Together, we will take a deeper dive into information-rich biophysical data used in successful protein development.

Patients would rather have a subcutaneous injection than an IV infusion, so biologics are often formulated at high concentration to enable self-administration in a small volume.  Many antibody drugs have viscosity limitations and there is a need for new ideas to address this.  Caffeine can mask protein-protein interactions, leading to reduced viscosity.  This presentation will include examples and a discussion of the safety and regulatory aspects of caffeine as an excipient. 

Structural heterogeneity remains a formidable problem limiting the utility of intact-mass MS in characterization of protein therapeutics and non-biological complex drugs. In many cases, this problem can be effectively addressed using novel chromatographic methods coupled with online MS detection and by supplementing MS measurements with tools from the arsenal of gas phase ion chemistry. These approaches will be illustrated using extensively glycosylated proteins, synthetic vaccines, and large immune complexes.

We use mass spectrometry (MS) to collaborate and support assay development of new immunoassays. Our efforts support development of proteins for small molecule (vitamin D) and large molecule (HIV, SARS-CoV-2) assays. This talk will explore recent efforts and approaches used to facilitate protein development in R&D phases of immunoassay development with an array of mass spectrometers and techniques as well as recent progress on quality assurance MS testing.

Introducing Microfluidic Diffusional Sizing (MDS) Technology which brings a new tool to the analytical characterization toolbox: a different approach that enables the analysis of protein interactions close to in vivo conditions.

I will present an end-to-end automated MS analytics workflow developed at Roche Innovation Center Munich. It covers all aspects of the protein MS analytics, starting with sample registration, wet lab processing, and MS measurement, as well as data analysis, management, and reporting applied on heterogeneous antibody-based drug candidate samples. Data and metadata are ultimately consolidated and stored in a way that enables mining and efficient and informed decision-making.

The presentation will summarize current progress, compare the arc of developing the deep learning approaches with the conventional methods, and describe concepts behind current strategies that may lead to potential future opportunities. For the presentation I'll dive into the details of how methods including RoseTTAFold, AlphaFold and recent large language models work.

Cryo-EM has become a powerful technique to speed up pharmaceutical discovery in the past few years. As a direct visualization technology that can reach atomic resolution, cryo-EM has broad applications in antibody research. Here we provide examples of utilizing EM to characterize antibodies and complexes, to provide key information for epitope mapping, cross-linking, and oligomerization. Therefore, cryo-EM demonstrates strong potential in antibody engineering by generating information that was previously intractable.