Objective:Biomarkers have become important in the prognosis and diagnosis of various diseases. High-throughput methods, such as RNA sequencing facilitate the detection of differentially expressed genes (DEGs), hence potential biomarker candidates. Individual studies suggest long lists of DEGs, hampering the identification of clinically relevant ones. Concerning preeclampsia – a major obstetric burden with high risk for adverse maternal and/or neonatal outcomes – limitations in diagnosis and prediction are still important issues. We, therefore, developed a workflow to facilitate the screening for biomarkers.Methods:On the basis of the tool DESeq2, a comprehensive workflow for identifying DEGs was established, analyzing data from several publicly available RNA-sequencing studies. We applied it to four RNA-sequencing datasets (one blood, three placenta) analyzing patients with preeclampsia and normotensive controls. We compared our results with other published approaches and evaluated their performance.Results:We identified 110 genes that are dysregulated in preeclampsia, observed in at least three of the studies analyzed, six even in all four studies. These included FLT-1, TREM-1, and FN1, which either represent established biomarkers at protein level, or promising candidates based on recent studies. For comparison, using a published meta-analysis approach, 5240 DEGs were obtained.Conclusion:This study presents a data analysis workflow for preeclampsia biomarker screening, capable of identifying promising biomarker candidates, while drastically reducing the numbers of candidates. Moreover, we were also able to confirm its performance for heart failure. This approach can be applied to additional diseases for biomarker identification, and the set of DEGs identified in preeclampsia represents a resource for further studies.

Objective: Measurement of the ratio between soluble fms-like tyrosine kinase-1 (sFlt-1) and placental growth factor (PlGF) supports the diagnosis of preeclampsia. sFlt-1/PlGF ratios of at least 85 and at least 110 have previously been suggested for diagnosis of early-onset and late-onset preeclampsia, respectively. However, angiogenic and antiangiogenic factors change throughout the process of aging, potentially influencing preeclampsia diagnosis. In this study, we therefore evaluated in detail the effect of maternal age on sFlt-1/PlGF ratios. Methods: A total of 2775 pregnant female patients were included in this retrospective cohort study, spread across three maternal age groups: 18–25 years, 26–35 years, and more than 35 years at delivery. Receiver-operating characteristic (ROC) curve analysis was employed to evaluate sFlt-1/PlGF ratio cutoffs for use in preeclampsia diagnosis. Results: Controls (2462 pregnant women) showed a significant difference in sFlt-1/PlGF ratios between the youngest and oldest age groups, which resulted in differences in the best-performing sFlt-1/PlGF ratio cutoffs: optimized cutoffs were 143.4 (52.9%, 98.2%), 8.6 (84.4%, 75.3%), and 22.9 (78.6%, 82.3%) in early-onset preeclampsia, and 46.4 (67.5%, 81.5%), 40.8 (77.3%, 73%), and 44.1 (65.1%, 74.5%) in late-onset preeclampsia in age groups, 1, 2, and 3, respectively. Conclusion: sFlt-1/PlGF ratios change with maternal age, which has important clinical implications for their use in the diagnosis of preeclampsia: Better differentiated sFlt-1/PlGF cutoffs should be used that take maternal and gestational age into account.