Reduction of monoclonal antibody viscosity using interpretable machine learning.

Early identification of antibody candidates with drug-like properties is essential for simplifying the development of safe and effective antibody therapeutics. For subcutaneous administration, it is important to identify candidates with low self-association to enable their formulation at high concentration while maintaining low viscosity, opalescence, and aggregation. Here, we report an interpretable machine learning model for predicting antibody (IgG1) variants with low viscosity using only the sequences of their variable (Fv) regions. Our model was trained on antibody viscosity data (>100 mg/mL mAb concentration) obtained at a common formulation pH (pH 5.2), and it identifies three key Fv features of antibodies linked to viscosity, namely their isoelectric points, hydrophobic patch sizes, and numbers of negatively charged patches. Of the three features, most predicted antibodies at risk for high viscosity, including antibodies with diverse antibody germlines in our study (79 mAbs) as well as clinical-stage IgG1s (94 mAbs), are those with low Fv isoelectric points (Fv pIs < 6.3). Our model identifies viscous antibodies with relatively high accuracy not only in our training and test sets, but also for previously reported data. Importantly, we show that the interpretable nature of the model enables the design of mutations that significantly reduce antibody viscosity, which we confirmed experimentally. We expect that this approach can be readily integrated into the drug development process to reduce the need for experimental viscosity screening and improve the identification of antibody candidates with drug-like properties.

Effects of nonintubated thoracoscopic surgery on postoperative neurocognitive function: a randomized controlled trial.

OBJECTIVES: Postoperative neurocognitive disorder following thoracoscopic surgery with general anaesthesia may be linked to reduced intraoperative cerebral oxygenation and perioperative inflammation, which can potentially be exacerbated by mechanical ventilation. However, nonintubated thoracoscopic surgery, which utilizes regional anaesthesia and maintains spontaneous breathing, provides a unique model for studying the potential benefits of avoiding mechanical ventilation. This approach allows investigation into the impact on perioperative neurocognitive profiles, inflammatory responses and intraoperative cerebral oxygen levels. METHODS: In total, 110 patients undergoing thoracoscopic surgery were randomly equally assigned to the intubated group and the nonintubated group. Regional cerebral oxygenation was monitored during surgery. Serum neuroinflammatory biomarkers, including interleukin-6 and glial fibrillary acidic protein, were measured at baseline (before surgery) and 24 h after surgery. Postoperative complication severity was compared using the Comprehensive Complication Index. The primary outcome was perioperative changes in neurocognitive test score, which was assessed at baseline, 24 h and 6 months after surgery. RESULTS: Patients in the nonintubated group had higher neurocognitive test scores at 24 h (69.9 ± 10.5 vs 65.3 ± 11.8; P = 0.03) and 6 months (70.6 ± 6.7 vs 65.4 ± 8.1; P < 0.01) after surgery and significantly higher regional cerebral oxygenation over time during one-lung ventilation (P = 0.03). Patients in the intubated group revealed a significantly higher postoperative serum interleukin-6 level (group by time interaction, P = 0.04) and a trend towards a significantly higher serum glial fibrillary acidic protein level (group by time interaction, P = 0.11). Furthermore, patients in the nonintubated group had a significantly lower Comprehensive Complication Index (9.0 ± 8.2 vs 6.1 ± 7.1; P < 0.05). CONCLUSIONS: Nonintubated thoracoscopic surgery was associated with improved postoperative neurocognitive recovery, more stable intraoperative cerebral oxygenation, ameliorated perioperative inflammation and attenuated postoperative complication severity.

Simplifying complex antibody engineering using machine learning.

Machine learning is transforming antibody engineering by enabling the generation of drug-like monoclonal antibodies with unprecedented efficiency. Unsupervised algorithms trained on massive and diverse protein sequence datasets facilitate the prediction of panels of antibody variants with native-like intrinsic properties (e.g., high stability), greatly reducing the amount of subsequent experimentation needed to identify specific candidates that also possess desired extrinsic properties (e.g., high affinity). Additionally, supervised algorithms, which are trained on deep sequencing datasets obtained after enrichment of in vitro antibody libraries for one or more specific extrinsic properties, enable the prediction of antibody variants with desired combinations of extrinsic properties without the need for additional screening. Here we review recent advances using both machine learning approaches and how they are impacting the field of antibody engineering as well as key outstanding challenges and opportunities for these paradigm-changing methods.