Separating clinical antibodies from repertoire antibodies, a path to in silico developability assessment

Abstract

Approaches for antibody discovery have seen substantial improvement and success in recent years. Yet, advancing antibodies into the clinic remains difficult because therapeutic developability concerns are challenging to predict. We developed a computational model to simplify antibody developability assessment and enable accelerated early-stage screening. To this end, we quantified the ability of hundreds of sequence- and structure-based descriptors to differentiate clinical antibodies that have undergone rigorous screening and characterization for drug-like properties from antibodies in the human repertoire that are not natively paired. This analysis identified 144 descriptors capable of distinguishing clinical from repertoire antibodies. Five descriptors were selected and combined based on performance and orthogonality into a single model referred to as the Therapeutic Antibody Developability Analysis (TA-DA). On a hold-out test set, this tool separated clinical antibodies from repertoire antibodies with an AUC = 0.8, demonstrating the ability to identify developability attributes unique to clinical antibodies. Based on our results, the TA-DA score may serve as an approach for selecting lead antibodies for further development. Abbreviations: Affinity-Capture Self-Interaction Nanoparticle Spectroscopy (AC-SINS), Area Under the Curve (AUC), Complementary-Determining Region (CDR), Clinical-Stage Therapeutics (CST), Framework (FR), Monoclonal Antibodies (mAbs), Observed Antibody Space (OAS), Receiver Operating Characteristic (ROC), Size-Exclusion Chromatography (SEC), Structural Aggregation Propensity (SAP), Therapeutic Antibody Developability Analysis (TA-DA), Therapeutic Antibody Profiler (TAP), Therapeutic Structural Antibody Database (Thera-SAbDab), Variable Heavy (VH), Variable Light (VL).

Identifying developability risks for clinical progression of antibodies using high-throughput in vitro and in silico approaches

“…Further experiments will be needed to confirm whether these predictions agree with in vitro assessments. Negron et al 24 assessed the ability of 910 descriptors to discriminate between 4929 repertoire and 339 clinical antibodies. The final Therapeutic Antibody Developability Analysis (TA-DA) score contained contributions from framework aggregation scores that are driven by hydrophobic clusters of atoms, light chain CDR positive patch energy, overall atomic contact energy, and amino-acid penalties based on their relative enrichment in ordered vs. intrinsically disordered proteins.…”

SUMO –In SilicoSequence Assessment Using Multiple Optimization Parameters

“…Once the sequence and structures of the candidate binders are predicted, their developability needs to be evaluated on several factors like binding affinity, human-likeness, chemical liabilities, PTMs, MHC-II binding, aggregation propensity, thermostability, PK, etc. using suitable in silico developability assessment [29][30][31][32][33][34][36][37][38][39] . This focused set of binders can then progress to an antigen binding screen more confidently and later be developed into an antibody format of interest.…”

“…Another approach, implemented as solubility predictor, is CamSol [14], which was also used to design antibody variants towards an improved developability [15]. In recent years, further approaches that investigated the in silico property distribution of clinical or marketed antibodies to identify the most descriptive and informative properties have allowed to assess whether a sequence might be developable or not [16][17][18]. In those studies, three-dimensional (3D) models of the antibodies were generated and used as input for the calculation of different structure-or sequence-based descriptors.…”

“…An in-depth study of the intrinsic physicochemical profile of marketed antibodies from Boehringer Ingelheim [16] yielded the following nonredundant descriptors as potential developability criteria: surface area buried between the variable light (VL) and heavy (VH) domains, the isoelectric point (pI), ratio of dipole moment to hydrophobic moment, ratio of surface areas of charged to hydrophobic patches and a scale for hydrophobic imbalance. Recently, the Therapeutic Antibody Developability Analysis (TA-DA) [17] approach suggests the following scores to efficiently separate clinical antibodies from repertoire antibodies: AggScore [19] predictions of the light chain framework1 region, AggScore predictions of the framework region of the entire antibody, positive patches of the light chain CDRs, an atomic contact energy that reflects burial of hydrophobic residues and exposure of charged and polar atoms and TOP-IDP, a summed amino acid scale reflecting the propensity for intrinsic disorder.…”

“…This option requires the ability to discriminate candidates that have liabilities. Many experimental [22][23][24][25][26][27][28] and computational [29][30][31][32][33][34][35][36][37][38][39] methods of candidate assessment are available and have been previously discussed at length [40][41][42][43] . The second option is to generate drug candidates de-risked for developability by design.…”