Search

Antibodies are one of the predominant treatment modalities for various diseases. To improve the characteristics of a lead antibody, such as antigen-binding affinity and stability, we conducted comprehensive substitutions and exhaustively explored their sequence space. However, it is practically unfeasible to evaluate all possible combinations of mutations owing to combinatorial explosion when multiple amino acid residues are incorporated. It was recently reported that a machine-learning guided protein engineering approach such as Thompson sampling (TS) has been used to efficiently explore sequence space in the framework of Bayesian optimization. For TS, over-exploration occurs when the initial data are biasedly distributed in the vicinity of the lead antibody. We handle a large-scale virtual library that includes numerous mutations. When the number of experiments is limited, this over-exploration causes a serious issue. Thus, we conducted Monte Carlo Thompson sampling (MTS) to balance the exploration-exploitation trade-off by defining the posterior distribution via the Monte Carlo method and compared its performance with TS in antibody engineering. Our results demonstrated that MTS largely outperforms TS in discovering desirable candidates at an earlier round when over-exploration occurs on TS. Thus, the MTS method is a powerful technique for efficiently discovering antibodies with desired characteristics when the number of rounds is limited.

Modulating the complement system is a promising strategy in drug discovery for disorders with uncontrolled complement activation. Although some of these disorders can be effectively treated with an antibody that inhibits complement C5, the high plasma concentration of C5 requires a huge dosage and frequent intravenous administration. Moreover, a conventional anti-C5 antibody can cause C5 to accumulate in plasma by reducing C5 clearance when C5 forms an immune complex (IC) with the antibody, which can be salvaged from endosomal vesicles by neonatal Fc receptor (FcRn)-mediated recycling. In order to neutralize the increased C5, an even higher dosage of the antibody would be required. This antigen accumulation can be suppressed by giving the antibody a pH-dependent C5-binding property so that C5 is released from the antibody in the acidic endosome and then trafficked to the lysosome for degradation, while the C5-free antibody returns back to plasma. We recently demonstrated that a pH-dependent C5-binding antibody, SKY59, exhibited long-lasting neutralization of C5 in cynomolgus monkeys, showing potential for subcutaneous delivery or less frequent administration. Here we report the details of the antibody engineering involved in generating SKY59, from humanizing a rabbit antibody to improving the C5-binding property. Moreover, because the pH-dependent C5-binding antibodies that we first generated still accumulated C5, we hypothesized that the surface charges of the ICs partially contributed to a slow uptake rate of the C5-antibody ICs. This idea motivated us to engineer the surface charges of the antibody. Our surface-charge engineered antibody consequently exhibited a high capacity to sweep C5 and suppressed the C5 accumulation in vivo by accelerating the cycle of sweeping: uptake of ICs into cells, release of C5 from the antibody in endosomes, and salvage of the antigen-free antibody. Thus, our engineered anti-C5 antibody, SKY59, is expected to provide significant benefits for patients with complement-mediated disorders., Competing Interests: All the authors are employees and/or shareholders of Chugai, which is conducting the clinical study of SKY59. The authors report the filing of the following patents relevant to the investigational drug SKY59. ZS and YR are the inventors of the patents, “Anti-C5 antibodies and methods of use” (WO/2016/098356) and “Anti-C5 antibodies and methods of use” (WO/2017/217524). ZS is the inventor of the patent, “Anti-C5 antibodies and methods of use” (WO/2017/104779). ZS, K. Haraya, and TF are listed as inventors of the patent, “A pharmaceutical composition for use in the treatment or prevention of a C5-related disease and a method for treating or preventing a C5-related disease” (WO/2018/143266). These do not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials.

In hemophilia A, routine prophylaxis with exogenous factor VIII (FVIII) requires frequent intravenous injections and can lead to the development of anti-FVIII alloantibodies (FVIII inhibitors). To overcome these drawbacks, we screened asymmetric bispecific IgG antibodies to factor IXa (FIXa) and factor X (FX), mimicking the FVIII cofactor function. Since the therapeutic potential of the lead bispecific antibody was marginal, FVIII-mimetic activity was improved by modifying its binding properties to FIXa and FX, and the pharmacokinetics was improved by engineering the charge properties of the variable region. Difficulties in manufacturing the bispecific antibody were overcome by identifying a common light chain for the anti-FIXa and anti-FX heavy chains through framework/complementarity determining region shuffling, and by pI engineering of the two heavy chains to facilitate ion exchange chromatographic purification of the bispecific antibody from the mixture of byproducts. Engineering to overcome low solubility and deamidation was also performed. The multidimensionally optimized bispecific antibody hBS910 exhibited potent FVIII-mimetic activity in human FVIII-deficient plasma, and had a half-life of 3 weeks and high subcutaneous bioavailability in cynomolgus monkeys. Importantly, the activity of hBS910 was not affected by FVIII inhibitors, while anti-hBS910 antibodies did not inhibit FVIII activity, allowing the use of hBS910 without considering the development or presence of FVIII inhibitors. Furthermore, hBS910 could be purified on a large manufacturing scale and formulated into a subcutaneously injectable liquid formulation for clinical use. These features of hBS910 enable routine prophylaxis by subcutaneous delivery at a long dosing interval without considering the development or presence of FVIII inhibitors. We expect that hBS910 (investigational drug name: ACE910) will provide significant benefit for severe hemophilia A patients. [ABSTRACT FROM AUTHOR]