Harnessing Fc/FcRn Affinity Data from Patents with Different Machine Learning Methods.

Monoclonal antibodies are biopharmaceuticals with a very long half-life due to the binding of their Fc portion to the neonatal receptor (FcRn), a pharmacokinetic property that can be further improved through engineering of the Fc portion, as demonstrated by the approval of several new drugs. Many Fc variants with increased binding to FcRn have been found using different methods, such as structure-guided design, random mutagenesis, or a combination of both, and are described in the literature as well as in patents. Our hypothesis is that this material could be subjected to a machine learning approach in order to generate new variants with similar properties. We therefore compiled 1323 Fc variants affecting the affinity for FcRn, which were disclosed in twenty patents. These data were used to train several algorithms, with two different models, in order to predict the affinity for FcRn of new randomly generated Fc variants. To determine which algorithm was the most robust, we first assessed the correlation between measured and predicted affinity in a 10-fold cross-validation test. We then generated variants by in silico random mutagenesis and compared the prediction made by the different algorithms. As a final validation, we produced variants, not described in any patents, and compared the predicted affinity with the experimental binding affinities measured by surface plasmon resonance (SPR). The best mean absolute error (MAE) between predicted and experimental values was obtained with a support vector regressor (SVR) using six features and trained on 1251 examples. With this setting, the error on the log(KD) was less than 0.17. The obtained results show that such an approach could be used to find new variants with better half-life properties that are different from those already extensively used in therapeutic antibody development.

Insights into the IgG heavy chain engineering patent landscape as applied to IgG4 antibody development.

Despite being the least abundant immunoglobulin G in human plasma, IgG4 are used therapeutically when weak effector functions are needed. The increase in engineered IgG4-based antibodies on the market led us to study the patent landscape of IgG4 Fc engineering, i.e., patents claiming modifications in the heavy chain. Thirty-seven relevant patent families were identified, comprising hundreds of IgG4 Fc variants focusing on removal of residual effector functions (since IgG4s bind to FcγRI and weakly to other FcγRs), half-life enhancement and IgG4 stability. Given the number of expired or soon to expire major patents in those 3 areas, companies developing blocking antibodies now have, or will in the near future, access to free tools to design silenced, half-life extended and stable IgG4 antibodies.

[Which immunoglobulin heavy chains for which immunostimulatory antibodies?] / Quelles chaînes lourdes d'immunoglobulines pour quels anticorps d'immunostimulation ?

Immunostimulatory antibodies in cancer rely on different mechanisms of action, simple blocking of immune checkpoints, depletion of tumour-infiltrating regulatory T cells, agonistic action on lymphocyte activatory receptors, etc. Choosing the relevant heavy chain isotype or the adequate engineering variant of this heavy chain is critical for the therapeutic efficacy, since the aforementioned properties sometimes bring into play the Fc region and the hinge region. This short review tries to learn the lessons from the clinical experience.

TITLE: Quelles chaînes lourdes d'immunoglobulines pour quels anticorps d'immunostimulation ? ABSTRACT: En cancérologie, les anticorps conduisant à une immunostimulation, ou anticorps d'immunostimulation, relèvent de différents mécanismes d'action: simple blocage de récepteurs agissant comme points de contrôle de l'immunité, élimination des lymphocytes T régulateurs infiltrant les tumeurs, action agoniste sur des récepteurs activateurs des lymphocytes, etc. Dans la mesure où ces propriétés font parfois intervenir la région Fc et la région charnière, le choix du bon isotype de chaîne lourde ou de variants de cette chaîne lourde obtenus par ingénierie peut s'avérer déterminant pour l'efficacité thérapeutique. Cette brève revue tente de tirer les premières leçons de l'expérience clinique.

New structural formats of therapeutic antibodies for rheumatology.

Pharmaceutical companies strive continuously to develop better medications in order to remain competitive. In the arena of monoclonal antibodies and related biologics (fusion proteins containing an IgG Fc fragment), the thrust is not only toward identifying new targets, but also toward developing new molecular formats. Here, new-generation antibodies used to treat rheumatic diseases are discussed, with emphasis on relations linking structure to pharmacological effects and on the improvements expected from the new formats. Isotypic and allotypic antibody diversity has pharmacological implications and is already exploited in commercially available antibodies. Efforts to engineer the Fc fragment of the various immunoglobulin G subclasses are reviewed with reference to abatacept, ixekizumab, other mutated IgG4 antibodies currently in development, sapelizumab, anifrolumab, and tanezumab. Bispecific antibodies are a focus of increasing interest (particularly those binding to both IL-17 and TNFα) and may earn a place in the therapeutic armamentarium as a means of avoiding the use of antibody combinations. However, the construction and production of bispecific antibodies continues to raise major technological challenges. Other molecular formats involve the fusion of antibodies to cytokines or the use of nanobodies and peptibodies. These new formats are at the very early stages of development, and their clinical relevance remains unclear.

Rethinking the INN system for therapeutic antibodies.

In the context of a possible revision of the International Nonproprietary Names (INN) system of recombinant monoclonal antibodies, which is saturated, we propose several avenues of reflection driven by the primary goal of the INN, information of health-care professionals. Clinical considerations argue for an abandon of the substems A (target category) and B (origin category), which lengthen the INN without real added-value. On the contrary, new substems or suffixes are required to alert on the absence/presence of an Fc portion and/or multispecificity, which are essential from a pharmacological point of view. Moreover, we think it necessary to explicitly mention Fc variations since they could influence the pharmacology of these biopharmaceuticals, and hence their efficacy and side-effects. Besides indicating the subclass/isotype in the documents easily accessible to health care professionals, we propose to systematically describe both the natural variations (allotypes) by using the Gm (G marker) system, and the artificial variations by using a Ge (G engineering) system that is discussed here and could apply to all IgG constant domains (tentatively called the Fy portion).