Is maternal plasma transcriptome analysis a new tool for monitoring high-risk pregnancies?

The analysis of circulating nucleic acids has been accepted for more than a decade as a novel noninvasive tool for prenatal diagnosis and monitoring of pregnancy-associated disorders, as well as for monitoring other diseases such as cancer and diabetes (1–3 ). In this issue of Clinical Chemistry, Tsui et al. present a study in which they successfully investigated the complete maternal plasma transcriptome during pregnancy by the use of RNA sequencing (RNA-seq) (4 ). The authors analyzed circulating RNA for first-, second-, and third-trimester pregnancies and found an increasing fetal contribution to maternal plasma RNAs during pregnancy of up to more than 11% during late gestation. Fetaland maternal-specific alleles were detected on the basis of single nucleotide polymorphisms in RNA-seq reads specific for the paternal or the maternal genome to estimate the fetal and maternal contribution to the circulating transcripts in maternal plasma, respectively. Genes containing informative single nucleotide polymorphisms were identified by the use of exome-enriched genomic DNA library sequencing of samples from the fetal part of the placenta and from maternal white blood cells. Furthermore, the ratio of fetaland maternal-specific alleles to the common allele was used to find genes with a high fetal or maternal contribution, which correspond to genes specifically expressed from fetal or maternal cells, respectively. Additional evidence for pregnancy-associated transcripts in maternal circulation was provided by the comparison of preand postdelivery maternal plasma samples, which revealed 131 genes that were upregulated during pregnancy. This comparison also showed a relatively fast clearance, particularly of the transcripts with a high fetal contribution, further supporting the specificity of these circulating RNAs for pregnancy. Because the study by Tsui et al. (4 ) was a proof-ofconcept study, only a very limited number of pregnancy cases were analyzed. However, the clear and convincing results suggest the need for further studies that investigate more cases to show reproducibility of the results and to validate the identified pregnancyassociated transcripts found in maternal plasma. Furthermore, samples from pathological cases, such as preeclampsia, intrauterine growth retardation, and preterm birth could be compared to normal control pregnancies to identify expression patterns and transcript biomarkers associated with disturbed pregnancies, i.e., placental pathologies. The analysis of circulating RNA of fetal origin as a biomarker for preeclampsia has already been suggested (5 ). However, the study by Tsui et al. opens a new field of research with significant clinical implications. The approach Tsui et al. used further confirmed the potential of circulating plasma RNAs as diagnostic markers for normal or diseased pregnancy that can be used for prenatal testing and research. In a recent study by the same group, the authors analyzed fetal DNA in maternal plasma by genome-wide sequencing to identify mutations in the fetal genome (6 ). The main advantage of these noninvasive approaches is that they carry no risk for inducing abortions, compared to conventional techniques that require amniotic fluid samples or biopsies from chorionic villi. The analysis of fetal DNA is most useful for the search for mutations associated with genetic diseases. The analysis of RNA also can be applied to the search for genetic variants and mutations, but in addition, it is applicable for the analysis of the expression levels of pregnancyassociated genes to monitor the development of the fetus and the placenta, particularly in case of high-risk pregnancies. Tsui et al. found that the expression of pregnancy-associated transcripts in the placenta and in maternal plasma were positively correlated, suggesting that expression in the placenta can be inferred from transcript levels in the maternal plasma. However, there is still a long way to go in bringing this technique to clinical diagnostic applications. First, reproducibility of the results must be demonstrated by the analysis of many more cases. Second, it will be critically important to standardize the collection and processing of plasma samples to isolate RNA with comparable quality (7 ). Moreover, low amounts and poor integrity of circulating RNA have been noted for several placental RNAs circulating in maternal plasma (8 ). 1 ETH Zurich, Animal Physiology, Institute of Agricultural Sciences, Department of Environmental Systems Science, Zurich, Switzerland. * Address correspondence to this author at: ETH Zurich, Animal Physiology, Institute of Agricultural Sciences, Universitaetstrasse 2 / LFW B 58.1, 8092 Zurich, Switzerland. E-mail [email protected]. Received April 28, 2014; accepted April 30, 2014. Previously published online at DOI: 10.1373/clinchem.2014.225136 © 2014 American Association for Clinical Chemistry 2 Nonstandard abbreviations: RNA-seq, RNA sequencing. Clinical Chemistry 60:7 914–915 (2014) Editorials