Therapeutic antibody development requires selection and engineering of molecules with high affinity and other drug-like biophysical properties. Co-optimization of multiple antibody properties remains a difficult and time-consuming process that impedes drug development. Here we evaluate the use of machine learning to simplify antibody co-optimization for a clinical-stage antibody (emibetuzumab) that displays high levels of both on-target (antigen) and off-target (non-specific) binding. We mutate sites in the antibody complementarity-determining regions, sort the antibody libraries for high and low levels of affinity and non-specific binding, and deep sequence the enriched libraries. Interestingly, machine learning models trained on datasets with binary labels enable predictions of continuous metrics that are strongly correlated with antibody affinity and non-specific binding. These models illustrate strong tradeoffs between these two properties, as increases in affinity along the co-optimal (Pareto) frontier require progressive reductions in specificity. Notably, models trained with deep learning features enable prediction of novel antibody mutations that co-optimize affinity and specificity beyond what is possible for the original antibody library. These findings demonstrate the power of machine learning models to greatly expand the exploration of novel antibody sequence space and accelerate the development of highly potent, drug-like antibodies.

The pharmacokinetic properties of antibodies are largely dictated by the pH-dependent binding of the IgG fragment crystallizable (Fc) domain to the human neonatal Fc receptor (hFcRn). Engineered Fc domains that confer a longer circulation half-life by virtue of more favorable pH-dependent binding to hFcRn are of great therapeutic interest. Here we developed a pH Toggle switch Fc variant containing the L309D/Q311H/N434S (DHS) substitutions, which exhibits markedly improved pharmacokinetics relative to both native IgG1 and widely used half-life extension variants, both in conventional hFcRn transgenic mice and in new knock-in mouse strains. engineered specifically to recapitulate all the key processes relevant to human antibody persistence in circulation, namely: (i) physiological expression of hFcRn, (ii) the impact of hFcγRs on antibody clearance and (iii) the role of competing endogenous IgG. DHS-IgG retains intact effector functions, which are important for the clearance of target pathogenic cells and also has favorable developability.

There is significant interest in formulating antibody therapeutics as concentrated liquid solutions, but early identification of developable antibodies with optimal manufacturability, stability, and delivery attributes remains challenging. Traditional methods of identifying developable mAbs with low self-association in common antibody formulations require relatively concentrated protein solutions (>1 mg/mL), and this single challenge has frustrated early-stage and large-scale identification of antibody candidates with drug-like colloidal properties. Here, we describe charge-stabilized self-interaction nanoparticle spectroscopy (CS-SINS), an affinity-capture nanoparticle assay that measures colloidal self-interactions at ultradilute antibody concentrations (0.01 mg/mL), and is predictive of antibody developability issues of high viscosity and opalescence that manifest at four orders of magnitude higher concentrations (>100 mg/mL). CS-SINS enables large-scale, high-throughput selection of developable antibodies during early discovery.

The rapidly evolving nature of antibody drug development has resulted in technologies that generate vast numbers (hundreds to thousands) of lead antibody candidates during early discovery. These candidates must be rapidly pared down to identify the most drug-like candidates for in-depth analysis of their safety and efficacy, which can only be performed on a limited number of antibodies due to time and resource requirements. One key biophysical property of successful antibody therapeutics is high specificity, defined as low levels of nonspecific binding or polyspecificity. Although there has been some progress in developing assays for detecting antibody polyspecificity, most of these assays are limited by poor sensitivity or assay formats that require proprietary antibody surface display methods, and some of these assays use complex and poorly defined polyspecificity reagents. Here we report the PolySpecificity Particle (PSP) assay, a sensitive flow cytometry assay for evaluating antibody nonspecific interactions that overcomes previous limitations and can be used for evaluating diverse types of IgGs, multispecific antibodies and Fc-fusion proteins. Our approach uses micron-sized magnetic beads coated with Protein A to capture antibodies at extremely dilute concentrations (<0.02 mg/mL). Flow cytometry analysis of polyspecificity reagent binding to these conjugates results in sensitive detection of differences in nonspecific interactions for clinical-stage antibodies. Our PSP assay strongly discriminates between antibodies with different levels of polyspecificity using previously reported polyspecificity reagents that are either well-defined proteins or highly complex protein mixtures. Moreover, we also find that a unique reagent, namely ovalbumin, results in the best assay sensitivity and specificity. Importantly, our assay is much more sensitive than standard assays such as ELISAs. We expect that our simple, sensitive, and highthroughput PSP assay will accelerate the development of safe and effective antibody therapeutics.

Highlights d Directed evolution of nanobodies that potently neutralize SARS-CoV-2 d CDR-swapping mutagenesis facilitates large affinity and activity improvements d Nanobody binding to RBD competes with ACE2 and two classes of neutralizing mAbs d Neutralizing nanobodies display drug-like biophysical properties

The COVID-19 pandemic continues to be a severe threat to human health, especially due to current and emerging SARS-CoV-2 variants with potential to escape humoral immunity developed after vaccination or infection. The development of broadly neutralizing antibodies that engage evolutionarily conserved epitopes on coronavirus spike proteins represents a promising strategy to improve therapy and prophylaxis against SARS-CoV-2 and variants thereof. Herein, a facile multivalent engineering approach is employed to achieve large synergistic improvements in the neutralizing activity of a SARS-CoV-2 cross-reactive nanobody (VHH-72) initially generated against SARS-CoV. This synergy is epitope specific and is not observed for a second high-affinity nanobody against a non-conserved epitope in the receptor-binding domain. Importantly, a hexavalent VHH-72 nanobody retains binding to spike proteins from multiple highly transmissible SARS-CoV-2 variants (B.1.1.7 and B.1.351) and potently neutralizes them. Multivalent VHH-72 nanobodies also display drug-like biophysical properties, including high stability, high solubility, and low levels of non-specific binding. The unique neutralizing and biophysical properties of VHH-72 multivalent nanobodies make them attractive as therapeutics against SARS-CoV-2 variants.

The aggregation of amyloidogenic polypeptides is strongly linked to several neurodegenerative disorders, including Alzheimer’s and Parkinson’s diseases. Conformational antibodies that selectively recognize protein aggregates are leading therapeutic agents for selectively neutralizing toxic aggregates, diagnostic and imaging agents for detecting disease, and biomedical reagents for elucidating disease mechanisms. Despite their importance, it is challenging to generate high-quality conformational antibodies in a systematic and site-specific manner due to the properties of protein aggregates (hydrophobic, multivalent, and heterogeneous) and limitations of immunization (uncontrolled antigen presentation and immunodominant epitopes). Toward addressing these challenges, we have developed a systematic directed evolution procedure for affinity maturing antibodies against Alzheimer’s Aβ fibrils and selecting variants with strict conformational and sequence specificity. We first designed a library based on a lead conformational antibody by sampling combinations of amino acids in the antigen-binding site predicted to mediate high antibody specificity. Next, we displayed this library on the surface of yeast, sorted it against Aβ42 aggregates, and identified promising clones using deep sequencing. The resulting antibodies displayed similar or higher affinities than clinical-stage Aβ antibodies (aducanumab and crenezumab). Moreover, the affinity-matured antibodies retained high conformational specificity for Aβ aggregates, as observed for aducanumab and unlike crenezumab. Notably, the affinity-maturated antibodies displayed extremely low levels of nonspecific interactions, as observed for crenezumab and unlike aducanumab. We expect that our systematic methods for generating antibodies with unique combinations of desirable properties will improve the generation of high-quality conformational antibodies specific for diverse types of aggregated conformers.

There is intense and widespread interest in developing monoclonal antibodies as therapeutic agents to treat diverse human disorders. During early-stage antibody discovery, hundreds to thousands of lead candidates are identified, and those that lack optimal physical and chemical properties must be deselected as early as possible to avoid problems later in drug development. It is particularly challenging to characterize such properties for large numbers of candidates with the low antibody quantities, concentrations, and purities that are available at the discovery stage, and to predict concentrated antibody properties (e.g., solubility, viscosity) required for efficient formulation, delivery, and efficacy. Here we review key recent advances in developing and implementing high-throughput methods for identifying antibodies with desirable in vitro and in vivo properties, including favorable antibody stability, specificity, solubility, pharmacokinetics, and immunogenicity profiles, that together encompass overall drug developability. In particular, we highlight impressive recent progress in developing computational methods for improving rational antibody design and prediction of drug-like behaviors that hold great promise for reducing the amount of required experimentation. We also discuss outstanding challenges that will need to be addressed in the future to fully realize the great potential of using such analysis for minimizing development times and improving the success rate of antibody candidates in the clinic.