Although much is known about patterns in fetal development at the cellular and molecular level, we do not yet have molecular tests that can reliably determine the gestational age of an individual fetus.
Instead, gestational age is typically estimated based on the date of a woman’s last menstrual period and/or ultrasound imaging. However, women do not always remember exactly when their last period occurred, so this may be an imprecise measurement. Ultrasounds are expensive and unavailable in many parts of the developing world.
The lack of cheap, reliable options for assessing fetal development is problematic for several reasons. If gestational age is actually less than what doctors believe, doctors may unnecessarily induce pregnancy or perform a C-section before it is needed. Alternatively, if gestational age is later than doctors realize, they may fail to recognize a post-term pregnancy and provide the necessary interventions.
Moreover, current methods of estimating delivery date usually do not consider the likelihood of premature birth, which is a leading cause of neonatal death and birth injury. There are certain indications that a woman is at high risk for premature birth, such as cervical insufficiency, but many women go into preterm labor without having exhibited clinical warning signs. Thus far, research on genetic risk factors and causes has had limited success.
Research on Blood Tests to Predict Delivery Date
In a new study published by Science, Thuy T.M. Ngo and colleagues piloted the use of a blood test in predicting gestational age and premature birth. The authors set out to determine whether measurements of cell-free RNA (cfRNA) originating from fetal tissues, and obtained in samples of maternal blood, could predict gestational age and the risk of preterm delivery.
Methods and Results
Predicting Gestational Age in Full-Term Pregnancies
Thuy et al. recruited 31 pregnant women in Denmark. Each participant donated a blood sample once a week during her pregnancy. All of the subjects delivered full-term (≥37 weeks), with no major complications. Blood samples were analyzed using polymerase chain reaction (PCR). Thuy et al. found that cfRNA corresponding to placental genes may be able to provide an estimate of fetal age.
The authors then built a randomized model to predict the length of time between sample collection and birth. Thuy et al. compared their model to other tools for prediction of gestational age: date of last menstrual period and ultrasound technology. Their model was just as effective as ultrasound use, and did not require knowledge of a woman’s last period.
Attempt to Apply the Same Model to Predict Preterm Birth
In addition to verifying whether their random model could predict time until delivery for full-term pregnancies, Thuy et al. wanted to see whether it could be used to predict premature birth. They studied two cohorts of women who were at high risk for preterm delivery. The first, recruited by the University of Pennsylvania (n=15), had experienced premature contractions. The second, recruited by the University of Alabama (n=23), had previously given birth to a premature baby. Thuy et al. found that for these pregnancies, their model failed to accurately predict date of delivery. They explain that, “This suggests that the model’s content may not account for the various other physiological events that may lead to preterm birth.”
Creating a Blood Test to Predict Preterm Birth
In order to identify cfRNA transcripts that could be used to predict preterm birth, the authors compared RNA sequencing data of women who delivered full-term with those who delivered preterm (from the PA cohort). They found 38 genes that differed significantly between the two groups. Using this information, Thuy et al. developed a test to predict the likelihood of preterm delivery. Although it was not 100% accurate, their test had higher predictive value than previous metabolic tests, such as one that compared the ratio of two proteins in order to predict premature birth.
In order to verify the effectiveness of this test, a blinded clinical trial with larger, more diverse samples is needed. This should include an unselected population; two of the cohorts in this study were recruited because they were at higher risk for preterm birth.
Although more work is necessary before their test becomes part of clinical practice, Thuy et al.’s research is very promising. This type of blood test is much more affordable than ultrasound technology, and could be used in developing countries. It could also be used as a complement to ultrasound. These blood tests could enable physicians to identify women at high risk of early delivery and intervene as appropriate. Moreover, similar tests may be useful in recognizing congenital defects that could be treated while a baby is still in the womb.
Science – Noninvasive blood tests for fetal development predict gestational age and preterm delivery