Human CYP2C8 is a key member of the CYP2C family and metabolizes more than 60 clinical drugs. A number of active site residues in CYP2C8 have been identified based on homology modeling and site-directed mutagenesis studies. In the structure of CYP2C8, the large active site cavity exhibits a trifurcated topology that approximates a T or Y shape, which is consistent with the finding that CYP2C8 can efficiently oxidize relatively large substrates such as paclitaxel and cerivastatin. The active site cavity of CYP2C8 contains at least 48 amino acid residues and many of them are important for substrate binding. The structures of CYP2C8 in complex with distinct ligands have revealed that the enzyme can bind divergent substrates and inhibitors without extensive conformational changes. CYP2C8 is a major catalyst in the metabolism of paclitaxel, amodiaquine, troglitazone, amiodarone, verapamil and ibuprofen, with a secondary role in the biotransformation of cerivastatin and fluvastatin. CYP2C8 also metabolises endogenous compounds such as retinoids and arachidonic acid. Many drugs are inhibitors of CYP2C8 and inhibition of this enzyme may result in clinical drug interactions. The pregnane X receptor, constitutive androstane receptor, and glucocorticoid receptor are likely to involve the regulation of CYP2C8. A number of genetic mutations in the CYP2C8 gene have been identified in humans and some of them have functional impact on the clearance of drugs. Further studies are needed to delineate the role of CYP2C8 in drug development and clinical practice.

Many risk factors associated with adverse pregnancy outcomes can be identified and modified preconceptionally. Despite a broad consensus that preconception care should be provided to all couples of reproductive age, it has not been integrated in routine healthcare. There are several barriers to its implementation, and even in the most resourceful countries, it is only provided to some select high-risk groups, rather than being an organized healthcare service provision to all. Recently, China seems to be leading the way by implementing preconception care nationwide in all rural areas. Its National Free Preconception Health Examination Project is a unique model of comprehensive preconception care. Advantages of this ambitious project are now becoming evident and benefiting the most vulnerable sections of Chinese society. This commentary provides an overview of National Free Preconception Health Examination Project and highlights the concepts that could be further developed and adapted into a model of preconception care.

Background. Gestational diabetes mellitus (GDM) contributes to adverse pregnancy and birth outcomes. In recent decades, extensive research has been devoted to the early prediction of GDM by various methods. Machine learning methods are flexible prediction algorithms with potential advantages over conventional regression. Objective. The purpose of this study was to use machine learning methods to predict GDM and compare their performance with that of logistic regressions. Methods. We performed a retrospective, observational study including women who attended their routine first hospital visits during early pregnancy and had Down’s syndrome screening at 16-20 gestational weeks in a tertiary maternity hospital in China from 2013.1.1 to 2017.12.31. A total of 22,242 singleton pregnancies were included, and 3182 (14.31%) women developed GDM. Candidate predictors included maternal demographic characteristics and medical history (maternal factors) and laboratory values at early pregnancy. The models were derived from the first 70% of the data and then validated with the next 30%. Variables were trained in different machine learning models and traditional logistic regression models. Eight common machine learning methods (GDBT, AdaBoost, LGB, Logistic, Vote, XGB, Decision Tree, and Random Forest) and two common regressions (stepwise logistic regression and logistic regression with RCS) were implemented to predict the occurrence of GDM. Models were compared on discrimination and calibration metrics. Results. In the validation dataset, the machine learning and logistic regression models performed moderately (AUC 0.59-0.74). Overall, the GBDT model performed best (AUC 0.74, 95% CI 0.71-0.76) among the machine learning methods, with negligible differences between them. Fasting blood glucose, HbA1c, triglycerides, and BMI strongly contributed to GDM. A cutoff point for the predictive value at 0.3 in the GBDT model had a negative predictive value of 74.1% (95% CI 69.5%-78.2%) and a sensitivity of 90% (95% CI 88.0%-91.7%), and the cutoff point at 0.7 had a positive predictive value of 93.2% (95% CI 88.2%-96.1%) and a specificity of 99% (95% CI 98.2%-99.4%). Conclusion. In this study, we found that several machine learning methods did not outperform logistic regression in predicting GDM. We developed a model with cutoff points for risk stratification of GDM.