First-trimester nuclear magnetic resonance based metabolomic profiling increases prediction of gestational diabetes: Study
Gestational diabetes mellitus (GDM) affects between 5% and 18% of pregnancies globally and is associated with adverse perinatal outcomes. In the long-term, it has been linked to increased risk of type 2 diabetes, obesity, and metabolic disease in both mothers and offspring. Currently, GDM is not typically diagnosed until 24 to 28 weeks’ gestation, when an oral glucose tolerance test (OGTT) is conducted; by this time, the fetus has already been exposed to some degree of maternal hyperglycemia and subtler metabolic alterations that precede it. The early identification of women at high risk of developing GDM thus represents an opportunity to apply preventative and therapeutic strategies, potentially reducing the incidence and impact of disease on women and their offspring.
The simplest and most widely applied methods of GDM prediction are based on maternal risk factors, used either individually or integrated in predictive models. Numerous clinically available biomarkers have also been explored in the context of early GDM identification, with variable performance. More recently, the increasing accessibility of high-throughput metabolomic technologies has led to the exploration of maternal serum metabolites in GDM.
The objective of this first-trimester screening study was to investigate whether the maternal metabolomic profile, obtained through a high-throughput nuclear magnetic resonance (NMR) metabolomics platform, can be used for early prediction of GDM.
This was a prospective study of 20,000 women attending routine pregnancy care visits at 11 to 13 weeks’ gestation. Metabolic profiles were assessed using a high-throughput nuclear magnetic resonance metabolomics platform. To inform translational applications, authors focused on a panel of 34 clinically validated biomarkers for detailed analysis and risk modeling. All biomarkers were used to generate a multivariable logistic regression model to predict gestational diabetes mellitus.
The concentrations of several metabolomic biomarkers, including cholesterol, triglycerides, fatty acids, and amino acids, differed between women who developed gestational diabetes mellitus and those who did not.
Addition of biomarker profile improved the prediction of gestational diabetes mellitus provided by maternal demographic characteristics and elements of medical history alone (before addition: area under the receiver operating characteristic curve, 0.790; detection rate, 50% [95% confidence interval, 44.3%e55.7%] at 10% false positive rate; and detection rate, 63% [95% confidence interval, 57.4%e68.3%] at 20% false positive rate; after addition: 0.840; 56% [50.3%e61.6%]; and 73% [67.7%e77.8%]; respectively).
The performance of combined testing was better for gestational diabetes mellitus treated by insulin (area under the receiver operating characteristic curve, 0.905; detection rate, 76% [95% confidence interval, 67.5%e83.2%] at 10% false positive rate; and detection rate, 85% [95% confidence interval, 77.4%e90.9%] at 20% false positive rate) than gestational diabetes mellitus treated by diet alone (area under the receiver operating characteristic curve, 0.762; detection rate, 47% [95% confidence interval, 37.7%e56.5%] at 10% false positive rate; and detection rate, 64% [95% confidence interval, 54.5%e72.7%] at 20% false positive rate).
The calibration plot showed good agreement between the observed incidence of gestational diabetes mellitus and the incidence predicted by the combined risk model. In the sensitivity analysis excluding the women diagnosed with gestational diabetes mellitus before 20 weeks’ gestation, there was a negligible difference in the area under the receiver operating characteristic curve compared with the results from the entire cohort combined.
In this large first-trimester screening study involving 974 women with GDM and over 18,000 women unaffected by GDM, authors demonstrated that a set of 34 metabolomic biomarkers can be used to improve early prediction of GDM achieved through screening by maternal risk factors alone.
The combination of this metabolite panel with maternal risk factors showed good predictive ability, with ROC AUC of 0.84 and DRs of 56% and 73% at respective FPRs of 10% and 20%. This was a moderate improvement relative to history-based prediction, with an approximate 10% increase in DR.
The combined model performed particularly well when predicting GDM requiring insulin therapy, with ROC AUC of 0.90 and DRs of 76% and 85% at respective FPRs of 10% and 20%.
In accordance with national and local guidance, study included women diagnosed with dysglycemia at any point in pregnancy as GDM cases, except for those who met the diagnostic criteria for pregestational diabetes before 20 weeks’ gestation. Given the lack of universally agreed-upon criteria for diagnosis outside of usual screening windows, especially in cases of early-onset GDM, authors conducted a sensitivity analysis excluding these women. This resulted in the predictive ability of model being virtually unaffected, with changes in ROC AUC well within the CIs of the original results. This further confirms the robustness of model.
Study presented a GDM risk prediction model for the first trimester, based on a panel of clinically validated serum biomarkers and maternal clinical and demographic characteristics, which showed an approximately 10% improvement in DR of GDM cases. This constitutes a promising advancement toward standardization of metabolomic studies in GDM. Further studies using the same methodology are required for external validation.
Source: Borges Manna L, Syngelaki A, Wu¨rtz P, et al. First-trimester nuclear magnetic resonanceebased metabolomic profiling increases the prediction of gestational diabetes mellitus. Am J Obstet Gynecol 2025;233:71.e1-14
