Metabolomic identification of novel diagnostic biomarkers in ectopic pregnancy

• Published in: Metabolomics Vol. 15; no. 11; pp. 143 - 11
• Main Authors: Turkoglu, Onur, Citil, Ayse, Katar, Ceren, Mert, Ismail, Kumar, Praveen, Yilmaz, Ali, Uygur, Dilek S., Erkaya, Salim, Graham, Stewart F., Bahado-Singh, Ray O.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01-11-2019; Springer Nature B.V
• Subjects: Acetic acid; Adult; Biochemistry; Biomarkers; Biomarkers - analysis; Biomarkers - metabolism; Biomedical and Life Sciences; Biomedicine; Case-Control Studies; Cell Biology; Developmental Biology; Diagnosis; Ectopic pregnancy; Female; Glucose; Humans; Lactic acid; Life Sciences; Magnetic resonance spectroscopy; Metabolites; Metabolomics; Molecular Medicine; NMR; Nuclear magnetic resonance; Original Article; Pregnancy; Pregnancy complications; Pregnancy, Ectopic - diagnosis; Pregnancy, Ectopic - metabolism; Prospective Studies; Proton Magnetic Resonance Spectroscopy; Pyruvic acid; Regression analysis; Turkey; Turkey; Metabolomics; Nuclear magnetic resonance spectroscopy; Biomarker; Ectopic pregnancy; Metabolite
• ISSN: 1573-3882; 1573-3890
• DOI: 10.1007/s11306-019-1607-1

Introduction
Ectopic pregnancy (EP) is a potentially life-threatening condition and early diagnosis still remains a challenge, causing a delay in management leading to tubal rupture.
Objectives
To identify putative plasma biomarkers for the detection of tubal EP and elucidate altered biochemical pathways in EP compared to intrauterine pregnancies.
Methods
This case–control study included prospective recruitment of 39 tubal EP cases and 89 early intrauterine pregnancy controls. Plasma samples were biochemically profiled using proton nuclear magnetic resonance spectroscopy (
1
H NMR). To avoid over-fitting, datasets were randomly divided into a discovery group (26 cases vs 60 controls) and a test group (13 cases and 29 controls). Logistic regression models were developed in the discovery group and validated in the independent test group. Area under the receiver operating characteristics curve (AUC), 95% confidence interval (CI), sensitivity, and specificity values were calculated.
Results
In total 13 of 43 (30.3%) metabolite concentrations were significantly altered in EP plasma (p < 0.05). Metabolomic profiling yielded significant separation between EP and controls (p < 0.05). Independent validation of a two-metabolite model consisting of lactate and acetate, achieved an AUC (95% CI) = 0.935 (0.843–1.000) with a sensitivity of 92.3% and specificity of 96.6%. The second metabolite model (
d
-glucose, pyruvate, acetoacetate) performed well with an AUC (95% CI) = 0.822 (0.657–0.988) and a sensitivity of 84.6% and specificity of 86.2%.
Conclusion
We report novel metabolomic biomarkers with a high accuracy for the detection of EP. Accurate biomarkers could potentially result in improved early diagnosis of tubal EP cases.