Clearly, mutations at all sites had a significant impact on self-association, and some seemingly disrupted self-interaction to baseline kD values, similar to weakly associating W104 Fabs (4

Clearly, mutations at all sites had a significant impact on self-association, and some seemingly disrupted self-interaction to baseline kD values, similar to weakly associating W104 Fabs (4.5 mL/g) and mAbs (15.3 mL/g), as summarized inFigure 2A. dramatically affected self-association, demonstrating the utility of these predictive approaches, while revealing a highly specific and tunable nature of self-binding modulated by single point mutations. Mutagenesis at these same key HC/LC CDR positions that affect self-interaction also typically abolished target binding with notable exceptions, clearly demonstrating the difficulties yet possibility of correcting self-association through engineering. Clear correlations were also observed between different methods used to assess self-interaction, such as Dynamic Light Scattering (DLS) and Affinity-Capture Self-Interaction Nanoparticle Spectroscopy (AC-SINS). Our findings advance our understanding of therapeutic protein and antibody self-association and offer insights into its prediction, evaluation and corrective mitigation to aid therapeutic development. Keywords:antibody, protein, self-association, self-interaction, developability, in silico prediction, computational modeling, viscosity, dynamic light ITIC scattering == 1. Introduction == Monoclonal antibodies and biologics in general have enjoyed increasing success and utility as therapeutic agents addressing a variety of biological targets of interest. As of late 2019, 79 commercial monoclonal antibody or antibody-based therapeutics have been approved [1], with several hundred currently being evaluated in clinical development [2]. Central to a therapeutic antibodys selection and success is usually its developability profile, which is a key driver in pre-clinical and clinical lead nomination [3,4]. Previously, developability flags in therapeutic antibodies have been correlated to overall clinical success, clearly indicating that developability attributes may impact clinical development beyond drug product purity, stability and manufacturability [5]. The developability properties of therapeutic antibodies range from expression and purification amenability to its physicochemical stability and behavior, both in the drug product form and in vivo [6,7,8]. Other major developability properties, such as self-association, can directly impact manufacturability and formulation success [3,9], and even strongly correlate to non-specific binding and animal Pharmacokinetics and clearance, therefore, affecting its overall efficacy [10,11,12,13]. Therapeutic antibody self-association has been well studied from a rheological standpoint and is known to directly impact solution viscosity, injectability, and manufacturability [14,15]. Therapeutic antibody formulations as low as 13 mg/mL have been reported to appreciably self-associate, significantly increasing solution viscosity and decreasing solubility, precluding further development even at common dose concentrations and formulation conditions TEF2 [16]. Moreover, strong antibody self-interaction tends to manifest in high viscosities at higher formulation concentrations, such as 100 to 200 mg/mL (or approximately 0.71.4 mM for a typical monoclonal antibody) and beyond [3]. Increasing solution viscosity is due to a concentration-dependent oligomerization effect of self-interacting molecules, particularly antibodies, whereby effectively large polymeric structures give rise to dramatic changes in solution rheology [17,18]. A resulting increase in viscosity may be exponential, making process filtration and pumping operations difficult and infeasible, and in the drug product form, handling, injectability, and potentially even stability may ITIC be negatively impacted [3,19,20]. This behavior is usually a major unfavorable developability attribute that is difficult to predict from sequence or structure and correct through molecular engineering and can halt further development and the selection of even the most promising large-molecule candidates [21]. High-concentration rheological behavior has significant importance in the selection of lead therapeutic candidates [22]. In general, molecular properties such as pI, net charge, and hydrophobicity can affect the rheology of antibody solutions [23]. Particularly at higher drug concentrations, it was shown that hydrophobic and charged surface patches result in increased self-interaction and solution viscosity above 100 mg/mL and approaching 200 mg/mL [24,25]. Such self-interactions can even lead to additional undesirable outcomes, such as opalescence, phase separation and gelling [26,27]. Antibody self-interactions have been reported to occur between antibody variable domains [21,28] as well as variableconstant interactions [16]. To ITIC possibly predict or correlate molecular properties to rheological outcomes, in silico computational approaches have been employed to ascertain a molecules propensity to self-interact [29]. Surface behavior characteristics, such as zeta potential and net charge derived from modeling, have also been correlated to viscosity, [18,30,31] other self-interaction parameters, including AC-SINS and kD [32], and lead molecule selection and success [33,34]. In addition to the inherent or predicted properties of a molecule, temperature [35,36] and formulation conditions [9,28] can have a dramatic effect on solution rheology. Because of this, formulation approaches have been successful in mitigating self-association, such as modulating pH and ionic strength [37,38,39], or the addition of excipients, such as Arginine [40,41]. However, success is.