The history of predicting the future: A review of prenatal screening, past and present

By Eva Martin, MD of Elm Tree Medical Inc.

Pregnant women these days may feel overwhelmed by the variety of options for prenatal genetic screening. From combinations of blood tests and ultrasounds to invasive testing with long needles, to the new cell free fetal DNA testing, expecting families have a lot of decisions to make surrounding whether they want genetic testing and which kind. However, it’s only in the last thirty years or so that predicting the risk of genetic abnormalities has been remotely possible. Where did all this screening start? And where have these new technologies taken us?

In 1984, Dr. Merkatz and colleagues published the first article on an association between a compound in mothers’ blood and chromosomal abnormalities in the growing fetus. They found that low alpha-fetoprotein in mother’s blood was linked to an increased risk of fetal chromosomal abnormalities. A flurry of research followed, identifying other markers in mothers’ blood. Ultrasounds of a little area at the back of the fetus’s neck called the nuchal translucency also increased predictive ability. In 2003, the FASTER trial (First and Second Trimester Evaluation of Risk Study) provided evidence on the best screening strategy. The results of this study supported a combined approach with screening at two time points. In the first trimester, trained ultrasound technicians measure the nuchal translucency, and a blood test measures a protein in mom’s blood called pregnancy-associated plasma protein A or PAPP-A. In the second trimester, a second blood test measures four more analytes: serum alpha-fetoprotein, human chorionic gonadotropin, unconjugated E3, and dimeric inhibin-A. Doing all of these screening tests will detect 92% of Trisomy 21 cases (Down Syndrome) if the thresholds are set to give a positive score for 5% of women who complete the screening. The idea of setting the “screen-positive” rate is important. The tests are not diagnostic or yes-no. The results exist along a spectrum and geneticists can only provide a risk score. Therefore, statisticians set the thresholds for a “screen-positive” result based on risk profiles. When the thresholds are set so that the 5% of women at the end of the spectrum get a screen positive result, only 2-6% of women who screen positive actually have a fetus with Down Syndrome. 

In 2011, the prenatal screening world was taken by storm when a new technique was adapted to isolate fragments of DNA from cells from the placenta. Initially dubbed “non-invasive prenatal testing,” the testing is now more appropriately called “cell-free DNA screening.” (Remember- the original set of blood tests and ultrasound are also noninvasive.) A 2011 study demonstrated that cell-free DNA screening could identify Down Syndrome. The original studies on this new technique used only women at high risk of carrying a fetus with a chromosomal abnormality. Therefore, the endorsement from the Society for Maternal-Fetal Medicine was limited to high risk groups. Now, new studies have shed light on cell-free DNA screening for low risk groups. These women are more likely to get a false positive result than women in high risk groups.

We’ve learned a few essential pieces of information about cell-free DNA screening in the few years it’s been in use. First, we’ve learned that many women don’t get a result at all. And, the women who get no-result have a higher risk of a chromosomal abnormality. Unfortunately, many studies have excluded anyone who got no-result from the analysis. We’ve also learned that abnormal results are not always due to actual abnormalities. For instance, a phenomenon called “mosaicism” can result in a positive screen. Mosaicism occurs when not all cells contain the same genetic make-up. For instance, some of the cells in the placenta may have the chromosomal abnormality while others do not. A vanishing twin with abnormal chromosomal make-up could also result in a false positive result.

A new study published in the June 2016 issue of the Gray Journal sheds light on the pros and cons of cell-free DNA screening for the general population of pregnant women. Dr. Norton and colleagues from the University of California, San Francisco compared the results of prenatal screening from sequential first and second trimester screening with estimated results of cell-free DNA screening. Importantly, they did not exclude cell-free DNA screenings that ended up with no-result. From 2009 to 2012, 452,901 women went through sequential screening through the California Prenatal Screening Program. The sequential screening approach detected 81.6% of all chromosomal abnormalities with a false-positive rate of 4.5%. If you count the “no-result” results from cell-free DNA as screen-positive results, cell-free DNA screening detected 77.1% of all chromosomal abnormalities with a false-positive rate of 3.7%. Cell-free DNA screening did detect more cases of Trisomy 21, but sequential screening methods were able to detect 54% of rare genetic abnormalities that cell-free DNA screening currently cannot detect. The authors conclude that physicians should not adopt cell-free DNA as the first-line screening method for the general population at this time. Can we use cell-free DNA screening as a follow up for an abnormal sequential screening result? Maybe not. For a pregnancy with a positive screening result from sequential screening, a fetus still has a 2 to 2.6% risk of carrying a chromosomal abnormality after a negative cell-free DNA screening result.

What do doctors do with cell-free DNA results? Two online calculators have been developed to help doctors and patients interpret their results. The first calculator is from the National Society of Genetic Counselors and the Perinatal Quality Foundation. The second calculator is from the University of North Carolina. The calculators will estimate the “post-test risk” or PTR. PTR is the likelihood that DNA from affected cells is present in the maternal circulation. The PTR does not calculate the risk the fetus has a chromosomal abnormality.  The PTR is also highly dependent on the accuracy of the default settings for the test parameters like sensitivity, specificity, prior risk, the false-positive rate, and the prevalence of the abnormality. For some abnormalities, these numbers are little more than a guess. It’s one thing to calculate the odds that a positive screen is actually positive for a whole population, but it’s problematic to calculate a PTR for a single test result.

Where does this leave us? In quite the quandary! For now, sequential screening is still the recommended option for the general population, but cell-free DNA screening could be the right choice for high risk women. Fortunately, genetic counselors are specially trained to help sort out the pros and cons of each screening approach, or none at all.