Fluetsch Andrin, Trunzer Markus, Gerebtzoff Grégori, Rodríguez-Pérez Raquel
Novartis Biomedical Research, Novartis Campus, CH-4002 Basel, Switzerland.
Chem Res Toxicol. 2024 Apr 15;37(4):549-560. doi: 10.1021/acs.chemrestox.3c00305. Epub 2024 Mar 19.
Most drugs are mainly metabolized by cytochrome P450 (CYP450), which can lead to drug-drug interactions (DDI). Specifically, time-dependent inhibition (TDI) of CYP3A4 isoenzyme has been associated with clinically relevant DDI. To overcome potential DDI issues, high-throughput assays were established to assess the TDI of CYP3A4 during the discovery and lead optimization phases. However, machine learning models would enable an earlier and larger-scale assessment of TDI potential liabilities. For CYP inhibition, most modeling efforts have focused on highly imbalanced and small data sets. Moreover, assay variability is rarely considered, which is key to understand the model's quality and suitability for decision-making. In this work, machine learning models were built for the prediction of TDI of CYP3A4, evaluated prospectively, and compared to the variability of the experimental assay. Different modeling strategies were investigated to assess their influence on the model's performance. Through multitask learning, additional data sets were leveraged for model building, coming from public databases, in-house CYP-related assays, or other pharmaceutical companies (federated learning). Apart from the numerical prediction of inactivation rates of CYP3A4 TDI, three-class predictions were carried out, giving a negative (inactivation rate < 0.01 min), weak positive (0.01 ≤ ≤ 0.025 min), or positive ( > 0.025 min) output. The final multitask graph neural network model achieved misclassification rates of 8 and 7% for positive and negative TDI, respectively. Importantly, the presented deep learning-based predictions had a similar precision to the reproducibility of experiments and thus offered great opportunities for drug design, early derisk of DDI potential, and selection of experiments. To facilitate CYP inhibition modeling efforts in the public domain, the developed model was used to annotate ∼16 000 publicly available structures, and a surrogate data set is shared as Supporting Information.
大多数药物主要通过细胞色素P450（CYP450）进行代谢，这可能导致药物相互作用（DDI）。具体而言，CYP3A4同工酶的时间依赖性抑制（TDI）与临床相关的DDI有关。为了克服潜在的DDI问题，在药物发现和先导化合物优化阶段建立了高通量检测方法来评估CYP3A4的TDI。然而，机器学习模型能够对TDI潜在风险进行更早且更大规模的评估。对于CYP抑制作用，大多数建模工作都集中在高度不平衡的小数据集上。此外，很少考虑检测变异性，而这对于理解模型质量和决策适用性至关重要。在这项工作中，构建了用于预测CYP3A4 TDI的机器学习模型，进行了前瞻性评估，并与实验检测的变异性进行了比较。研究了不同的建模策略以评估它们对模型性能的影响。通过多任务学习，利用来自公共数据库、内部CYP相关检测或其他制药公司（联邦学习）的额外数据集进行模型构建。除了对CYP3A4 TDI失活率进行数值预测外，还进行了三类预测，输出为阴性（失活率<0.01分钟）、弱阳性（0.01≤失活率≤0.025分钟）或阳性（失活率>0.025分钟）。最终的多任务图神经网络模型对阳性和阴性TDI的误分类率分别为8%和7%。重要的是，所呈现的基于深度学习的预测与实验的可重复性具有相似精度，因此为药物设计、DDI潜在风险的早期降低以及实验选择提供了巨大机会。为了促进公共领域的CYP抑制建模工作，使用开发的模型对约16000个公开可用结构进行注释，并作为支持信息共享了一个替代数据集。
