Evolving analytics for developability assessment of highly engineered antibodies. With the advent of highly engineered antibodies in the therapeutic landscape, analytical methods for developability assessment have had to keep pace. The unique properties of multispecific antibodies and the challenges associated with their assessment will be discussed.

MEDI5752 is a monovalent bispecific IgG1 antibody (DuetMab), targeting the two clinically validated receptors; PD-1 and CTLA-4. The bispecific antibody introduces novel MOAs that may provide an improved therapeutic index when compared to the two monotherapies and mAb combinations. MEDI5752 is currently being clinically evaluated for safety and efficacy.

The pipeline for antibodies, the largest class of biotherapeutics today, is growing in diversity with variants such as bispecifics, conjugates, or fragments. Platform approaches for purification work well for many monoclonal antibodies (mAbs) on the market but selecting purification schemes can be challenging for antibody variants due to molecular diversity. We’ll discuss how to select affinity resins to achieve selectivity and purification outcomes using established and new ligand innovations.

We have developed machine learning models for identifying antibodies with optimal combinations of high affinity and drug-like biophysical properties. Our approach combines novel descriptors of antibody molecular features with neural networks to identify Pareto optimal antibody variants with different levels of affinity improvement while minimizing natural trade-offs involving other key properties (e.g., specificity). These methods greatly reduce the required experimentation needed during antibody engineering efforts for co-optimizing multiple antibody properties. 

We developed a practical high-throughput developability workflow (100’s-1,000’s of molecules) implemented during the early phase of antibody generation and screening to guide the selection process of the best lead candidates. Mining of our 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.

Effective delivery of protein therapeutics to the central nervous system (CNS) has been greatly restricted by the blood-brain barrier (BBB). We describe the engineering and characterization of a BBB transport vehicle (TV) comprising an engineered Fc fragment that exploits receptor-mediated transcytosis for CNS delivery of biotherapeutics. The TV platform readily accommodates numerous configurations, including bispecific antibodies and protein fusions, yielding a highly modular CNS delivery platform.

The blood-brain barrier presents a major obstacle to brain drug delivery.  We have developed several different enabling platforms for the identification of antibodies against blood-brain barrier resident receptors that could ultimately be used to ferry drug cargo into the brain.  Here we will describe our recent efforts using new screening paradigms to identify blood-brain barrier targeting antibodies capable of delivering therapeutic payloads to the central nervous system.

Multiple pathways (e.g., VEGF-A, Ang-2) drive retinal disease by promoting vascular instability; however, current treatments only target VEGF-A. In preclinical studies, faricimab, a bispecific antibody targeting VEGF-A + Ang-2 developed using CrossMab technology, promoted vascular stability. In Phase 3 clinical trials, faricimab resulted in noninferior vision gains, improved anatomic outcomes, and demonstrated potential for extended dosing and comparable safety compared with aflibercept, a VEGF inhibitor, in patients with retinal diseases.

Multifunctional antibodies represent a major trend in therapeutic modalities. More than 50 different concepts have been established and many of them reached advanced clinical studies. This presentation compares the leading approaches, discusses pros and cons, gives recommendations on design and manufacturing supported by examples from case studies and reviews the current multispecific antibody pipeline in the context of biological function and molecular design.

Typical development and manufacturing of monoclonal antibodies (mAbs) use well established platform methods and processes which evolved over the decades of the biopharmaceutical industry experience. This approach enables a desired combination of high yielding cell lines and rapid development timelines. Due to close similarity to mAbs, in structure and physicochemical properties, F-star’s mAb2 bispecific format presents an ideal format to harness the well-established mAbs platform approaches with minimal optimization, to save time and money in drug development. This talk will explore the application of the platform technologies in manufacturing process development for mAb2 bispecific format.