BACKGROUND: Pre-term pre-eclampsia is a major cause of maternal and perinatal morbidity and mortality worldwide. A multi-centre randomized-controlled trial has shown that first-trimester screening followed by treatment of high-risk women with aspirin reduces the risk of pre-term pre-eclampsia. However, the biomarkers currently employed in risk prediction are only weakly associated with the outcome. METHODS: We conducted a case-cohort study within the Pregnancy Outcome Prediction study to analyse untargeted maternal serum metabolomics in samples from 12, 20, 28 and 36 weeks of gestational age (wkGA) in women with pre-eclampsia delivering at term (n = 165) and pre-term (n = 29), plus a random sample of the cohort (n = 325). We used longitudinal linear mixed models to identify candidate metabolites at 20/28 wkGA that differed by term pre-eclampsia status. Candidates were validated using measurements at 36 wkGA in the same women. We then tested the association between the 12-, 20- and 28-wkGA measurements and pre-term pre-eclampsia. We externally validated the association using 24- to 28-wkGA samples from the Born in Bradford study (25 cases and 953 controls). RESULTS: We identified 100 metabolites that differed most at 20/28 wkGA in term pre-eclampsia. Thirty-three of these were validated (P < 0.0005) at 36 wkGA. 4-Hydroxyglutamate and C-glycosyltryptophan were independently predictive at 36 wkGA of term pre-eclampsia. 4-Hydroxyglutamate was also predictive (area under the receiver operating characteristic curve, 95% confidence interval) of pre-term pre-eclampsia at 12 (0.673, 0.558-0.787), 20 (0.731, 0.657-0.806) and 28 wkGA (0.733, 0.627-0.839). The predictive ability of 4-hydroxyglutamate at 12 wkGA was stronger than two existing protein biomarkers, namely PAPP-A (0.567, 0.439-0.695) and placenta growth factor (0.589, 0.463-0.714). Finally, 4-hydroxyglutamate at 24-28 wkGA was positively associated with pre-eclampsia (term or pre-term) among women from the Born in Bradford study. CONCLUSIONS: 4-hydroxyglutamate is a novel biochemical predictor of pre-eclampsia that provides better first-trimester prediction of pre-term disease than currently employed protein biomarkers.