Generative machine learning (ML) has been postulated to be a major driver in the computational design of antigen-specific monoclonal antibodies (mAb). We used a lattice-based antibody-antigen binding simulation framework, which incorporates a wide range of physiological antibody binding parameters to examine arbitrarily large numbers of antibody sequences for their most critical design parameters: paratope, epitope, affinity, and developability. We found that a deep generative model, trained exclusively on antibody sequence (1D) data can be used to design native-like conformational (3D) epitope-specific antibodies, matching or exceeding the training dataset in affinity and developability variety.

The ability to degrade proteins with heterobifunctional *small* molecules has the potential to dramatically alter therapy. However, the relatively *large* size of these chimeric molecules often results in undesirable physicochemical properties (e.g., low aqueous solubility) and poor pharmacokinetics which may diminish their application to in vivo settings. To address this challenge we have demonstrated the first example of a targeted degrader (aka PROTAC) conjugate. This talk will focus on our discovery of degrader-antibody conjugates, their efficacy and safety, and how this general approach can expand the utility of directed protein degradation as both a biological tool and a therapeutic possibility.

Heavy-light chain mispairing is one of the major concerns for producing bispecific antibodies. bYlok™ technology is a design engineering approach that solves this whist retaining molecule’s closer-to-nature structure and simplified downstream purification. The presentation will also include how GS System^® is set to support development of bispecific molecules by streamlining vector construction process and improved titers using GS piggyBac^® transposon system.

Next-generation therapies for cancer require more complex modalities to increase specificity and limit off-target toxicities. Efficient discovery, optimization and manufacturing of multispecific modalities require significant improvements to current technologies. Our group has established efficient workflows to discover and optimize multispecific antibodies that can be produced from single cell lines under short timelines. Based on example of bispecific antibody for immuno-oncology we share our experience on discovery and development of human bispecific antibodies with a common light chain.

Our novel, automated high-throughput engineering platform enables the fast generation of large panels of multi-specific variants (up to 10.000) giving rise to large data sets (more than 100.000 data points). By combining data science and structure-based design workflows we leverage the potential of our unique data sets to guide the computational engineering of our next-generation antibody therapeutics.

We have developed and established a unique class of multivalent and multispecific antibody formats using our proprietary Quad technology to produce the next-generation of antibody-based therapeutics. We use simple protein engineering, in a plug-and-play fashion, to generate a plethora of novel antibody formats with varying size, shape, flexibility and binding domain valences. The technology has wide ranging potential for generating novel therapeutics with enhanced binding, functionality and efficacy.

The continued spillover of coronaviruses from zoonotic reservoirs and the emergence of resistance SARS-CoV-2 variants in the human population highlights the need for broadly active counter measures. Here I will describe the identification, engineering, and clinical development of broadly neutralizing monoclonal antibodies (bnAbs) that potently neutralize SARS-CoV-2 variants of concern and pre-emergent bat SARS-related viruses. The results support the development of bnAbs for the treatment of COVID-19 and future emerging sarbecovirus threats.

Sanyou Biopharmaceuticals is a leading international high-tech biotechnology company focusing on R&D and integrated solutions of innovative antibody drugs. Sanyou is dedicated to provide new technologies, products, services and solutions of " the best quality, fastest speed and most cost-efficiency". The presentation is going to introduce Sanyou’s distinct technologies for antibody drug discovery, validation and development, as well as the capability of the integrated platforms and an example project.

Polysarcosine (poly-N-methylglycine) is a non-toxic, biodegradable, aminoacid-based polymer with low immunogenicity that could serve as a replacement for polyethyleneglycol (PEG). Our company Polypeptide Therapeutic Solutions (PTS) provides a custom-made synthesis and GMP-grade manufacturing of the drug substance.

In less than a year, researchers have uncovered structure-function details for many of the proteins encoded by SARS-CoV-2. We report structures that reveal how the SARS-CoV-2 spike glycoprotein binds to de novo designed angiotensin-converting enzyme 2 (ACE2) receptor decoys, the specificities of polyclonal antibody responses in COVID-19-convalescent individuals, and how monoclonal neutralizing antibodies bind spike to prevent infection. Collectively, these structures will allow pairs of monoclonal antibodies to be chosen for treatment cocktails, improve antibody potencies through structure-based engineering to be effective at lower doses and/or are resistant to viral mutations, and facilitate epitope mapping to guide structure-based immunogen design.

The in vitro affinity and/or functional maturation of naïve antibodies is common practice. In most cases, targeted introduction of sequence diversity into a limited number of complementarity determining region (CDR) loops coupled with selection for improved variants through phage or ribosome display is sufficient to deliver the required affinity or functional improvements. Occasionally, ‘hard to mature’ clones are seen that are inherently intractable to optimization, necessitating a more heuristic, unbiased approach to achieve the desired improvements. In this talk, I will describe the use of ribosome display to optimize these ‘hard to mature’ clones, using the affinity optimization of an inhibitory antibody to human Arginase 2 as a case study. This work exemplifies the application of novel Shuffle and Shuffle/StEP libraries as well as pool maturation and error prone libraries to deliver significant improvements in potency, affinity, and mode of binding, that would not be achievable through more conventional methods.

Cytokines are potent regulators of the immune system, and their therapeutic potential has been established in animal disease models and clinical trials. Clinical application of native cytokines is limited by their short half-life and narrow therapeutic window. Xencor has pioneered the creation of potency-reduced cytokine-Fc fusions to promote increased in vivo persistence and superior tolerability. Examples to be discussed include potency-reduced IL15/Ra-Fc (Phase 1), IL2-Fc (Phase 1), and IL12-Fc (preclinical).

Affibody molecules are small engineered alternative scaffold affinity proteins that can be site-specifically loaded with cytotoxic drugs creating homogenous conjugates with a desired drug-to-carrier ratio. The presentation will explore the impact of affibody-carrier architecture, drug load, and peptide-linker composition on biodistribution and in vitro and in vivo cytotoxic efficacy.  

The limited chemical diversity of proteins constrains molecular recognition. Noncanonical amino acids and efficient conjugation reactions offer powerful alternatives for expanding capabilities. Here, we describe the preparation and characterization of diverse conjugates on yeast. Starting from recently described fibronectin-maleimide-sulfonamide conjugates, we identified examples of drastic amino acid and small molecule substitutions retaining isoform-specific carbonic anhydrase binding. These findings highlight promising opportunities for performing protein-based drug discovery on yeast.

We designed and introduced free cysteine residues into antibody Fc regions for antibody conjugation with drug-to-antibody ratio greater than two using THIOMAB. The top single cysteine mutants were selected and combined as double cysteine mutants. Most double cysteine mutants display good expression and low aggregation. PEGylation screening identified many top double cysteine mutants with high conjugatability and selectivity. These mutants can potentially be applied to site-specific antibody conjugation.

The role of B-cells in the tumor microenvironment has largely been under-investigated. B cell receptor high-throughput sequencing enabled profiling antibody repertoire signature of tumor-infiltrating lymphocyte B-cells (TIL-Bs) in comparison to B cells from three compartments in a mouse model of triple-negative breast cancer. TIL-Bs exhibit a distinct antibody repertoire, suggesting an antigen-driven B-cell response. The distribution of TIL-B across compartments indicated that they migrate to and from the TME.