It is well known that prenatal testing requires invasive testing of the fetus if it wants to know whether the fetus has a genetic disease, but recent scientists have found that through the blood test of pregnant women, it is possible to have a comprehensive understanding of the growth and development of the fetus.
Biological research into human pregnancy is challenging. Invasive diagnostic procedures can be used to take DNA from the placenta or fetus for genetic analysis, or to take RNA samples to get a snapshot of what was developing at the time. However, for ethical and practical reasons, it is not possible to repeat samples of the growing fetus and placenta to monitor development and pregnancy health throughout pregnancy. Rasmussen et al. wrote in the journal Nature to collect and analyze free-circulating RNA molecules in the blood called cell-free RNA (cfRNA) from more than 1,800 pregnant women of different ages, body mass indexes, and races, from different continents, at different stages of pregnancy. These analyses provide a basis for understanding changes in gene expression in normal fetuses, placenta, and mothers during pregnancy. The researchers then used knowledge of normal cfRNA features during pregnancy to predict the development of a potentially dangerous pregnancy complication called preeclampsia.
This invasive test can be eliminated later
The recognition that the release of cell-free DNA and RNA from maternal, fetal, and placental tissue and detectable in maternal plasma, a component of blood, is a milestone in pregnancy diagnosis2. Next-generation sequencing techniques for detecting cell-free fetal DNA (cffDNA) in maternal plasma are now widely used to non-invasively screen for common fetal chromosomal abnormalities in pregnancy and are highly sensitive and specific. While analyzing cffDNA helps diagnose specific fetal genetic conditions, cfRNA is a more dynamic nucleic acid that provides a snapshot of development in real time by gaining insight into genes currently being expressed.
Preeclampsia is characterized by the appearance of hypertension during pregnancy. It affects about 8% of pregnancies, can lead to damage to multi-organ systems, and is a leading cause of severe maternal and neonatal morbidity and death. Although the signs and symptoms of preeclampsia usually manifest in the last few weeks of pregnancy, the disease begins early in placental formation. Although preeclampsia has been known for hundreds of years, tests that predict the condition in the first trimester of pregnancy remain out of reach because interventions can alter the course of the disease.
Many studies of molecular "markers" of pregnancy-related disorders have focused on proteins in available maternal samples, such as blood or urine. However, Rasmussen and colleagues focused on analyzing cfRNAs from maternal, placental, and fetal tissue and released into the maternal bloodstream. The researchers assessed complement for cfRNA molecules found in maternal blood (transcriptome) from 2539 plasma samples from 1840 pregnancies in 8 separate groups, the largest and most diverse transcriptome pregnancy study ever conducted.
The researchers have shown from previous studies that cfRNA transcripts in maternal plasma can predict gestational age (the age of fetal development). They used a machine learning method to teach an algorithm that contains 1908 cfRNA atlases for predicting gestational age. On a separate test set of 474 samples, the algorithm can predict gestational age within 14.7 days. This accuracy is similar to an ultrasound in the second trimester. The ability of cfRNA to determine pregnancy dates is critical to establishing normal gene expression signatures at different gestational ages, and the researchers suggest that this method can be used as an adjunct to pregnancy date determination in cases where prenatal ultrasound is not available.
The researchers exhibited previous work demonstrating the potential of detecting fetal tissue-specific cfRNA in the maternal circulation. Rasmussen et al. defined specific subpopulations of cfRNA transcripts in maternal plasma that reflect maternal tissue, fetal organs, and placenta, and gained insight into the normal changes in these subpopulations during pregnancy. Maternal transcriptomes show increased expression of genes encoding proteins, such as the collagen that make up the extracellular matrix, which may reflect the remodeling of the cervix in preparation for and delivery. It is speculated that changes in RNA transcript abundance from fetal organs are consistent with known developmental trajectories —for example, transcript levels in the contracted portion of the fetal kidney also decline over time as pregnancy progresses. These findings show the exciting potential to use transcripts in maternal blood to non-invasively track pregnancy progression and fetal organ development.
Next, Rasmussen and colleagues assessed whether cfRNA signatures in the blood of pregnant women in the second trimester were able to predict future incidence of preeclampsia before clinical signs or symptoms appeared. The researchers compared cfRNA signatures in plasma samples of 72 pregnant women with preeclampsia (cases) and cfRNA features in 452 pregnant women (control group) who did not have preeclampsia. The samples were collected on average at 14.5 weeks before delivery. The researchers identified seven genes with consistent expression levels that distinguished cases from control groups. Four of these genes were previously associated with preeclampsia or placental development.
The researchers created a mathematical model that used cfRNA signatures to estimate the probability of preeclampsia. The model has a sensitivity of 75%, which means it can identify three-quarters of cases of eventual preeclampsia. In a group of pregnant individuals, almost 14 percent of the participants developed preeclampsia, and the model's positive predictive value (PPV) was 32 percent, meaning that about a third of the individuals predicted by the model that developed preeclampsia were later diagnosed with the disease. PPV increased sevenfold compared to the next most predictive preeclampsia detection described in literature 13. Notably, the authors found that the predictive power of the model was not affected by maternal race as a variable. This further supports diagnostic and therapeutic approaches that abandon racial bias, which have been found to have little or no utility and create racial disparities in health care.
One limitation of the investigators' work is that the incidence of preeclampsia in this group (almost 14%) is higher than the reported global incidence, which is about 2-8%; therefore, in the lower incidence group, the trial had a lower PPV. Any future clinical application will need to consider the benefits derived from early prediction of preeclampsia to prevent the potential harm of possible false-positive outcomes, such as unnecessary monitoring or intervention, and increased maternal anxiety.
A key direction for future work will be to determine whether cfRNA collected in the first trimester of pregnancy can be used to detect an increased risk of preeclampsia, as the only current preeclampsia prophylaxis treatment is low-dose aspirin, which may be more effective if started before 16 weeks of pregnancy. The widespread efficacy of low-dose aspirin in the prevention of preeclampsia (estimated 2–30% effectiveness) highlights the problems with the use of one treatment for a highly variable disease with a variety of manifestations. A molecular test could provide insight into the development of preeclampsia, a key step towards a more personalized approach to pregnancy treatment with transcriptomics or genomics. In addition, transcriptome data can be fed into computational resources such as Connectivity Map17 (an online database that records transcriptome signatures of various cell types after treatment with different treatments) to inform the repositioning of existing treatments to treat pregnancy-specific diseases.
This massive resource of transcriptomics from pregnant populations of different ethnicities and geographies is exciting for several reasons. Not only have researchers developed predictive tests for preeclampsia, but the study's findings have the potential to provide insights into typical pregnancy and fetal development, and advance the design of rational, precise therapies that can improve pregnancy care.