A paper published on Feb. 27, 2014, in the New England Journal of Medicine, titled “DNA Sequencing versus Standard Prenatal Aneuploidy Screening,” garnered a great deal of media attention focused on the use of maternal serum cell-free DNA screening, or noninvasive prenatal testing (NIPT), for aneuploidy in average-risk patients. The authors, supported by a commercial laboratory, compared NIPT with traditional screening for aneuploidy using serum analytes on a relatively small number of average-risk patients. The study was too small to compare detection rates, but the authors report that the false positive rate of NIPT is lower and therefore the test “merits serious consideration as a primary screening method for fetal autosomal aneuploidy.”
While this measured conclusion appears reasonable, the serious consideration that the authors propose requires further data, and the NEJM study must be viewed in the context of its many limitations. Most important, the study is underpowered to compare the detection rates, and it is generally not valid to compare false-positive rates in isolation.
NIPT uses cell-free DNA from maternal serum to screen for common fetal aneuploidies with high sensitivity and specificity. It also uses next generation sequencing to directly measure fetal DNA in the maternal circulation, and clinical tests are now available using this technique. While there are differences in the methodologies employed by the commercial laboratories, overall the reported performance is similar, with detection rates for Down syndrome above 99 percent and false positive rates that are less than one percent. This makes this screening test an attractive alternative to traditional serum screening for aneuploidy for patients.
Currently, the Society for Maternal-Fetal Medicine (SMFM) recommends that NIPT is most appropriate for high-risk patients. The five high-risk criteria currently include maternal age 35 years or older at delivery; sonographic findings indicating an increased risk of aneuploidy; history of a prior pregnancy with a trisomy; positive screening results for aneuploidy, including first trimester, sequential, integrated, or quadruple screen; or parental balanced Robertsonian translocation with increased risk for trisomy 13 or 21.
This recommendation has been based primarily on the more limited evidence regarding the utility of NIPT in low- or average-risk pregnant women, and validation studies that have generally been limited to high-risk populations. Understandably, there is a keen interest in performing the same type of validation studies in a low-risk population.
The study compared only false positive rates for trisomies 18 and 21, although all commercially available products also include testing for trisomy 13 and the sex chromosomes, which have higher false positive rates. A high percentage of the samples for NIPT were collected in the third trimester—at gestational ages when clinical aneuploidy screening is not performed and is not clinically relevant, yet when fetal DNA amounts are far higher, allowing better test performance. The authors compared NIPT with standard prenatal screening with a variety of first and second trimester tests that have a broad range of performance characteristics. Fewer than three percent of patients had integrated screening, which is the prenatal screening method with the best performance, including the lowest false positive rate. In addition, traditional screening can detect risk for a broad array of structural, chromosomal, and perinatal abnormalities.
The importance of these in a low-risk population may be far greater than the impact of trisomy 18 and 21, which are relatively rare in a younger maternal cohort. It is important to note that this study included only five Down syndrome cases. Five cases of trisomy 21 in the population of 1,909 patients represents a rate of 1 out of 382, which is substantially higher than the population risk of 1 out of 700 that would be expected in a truly “average” or “low-risk” cohort. Finally and importantly, when considering population screening, some patients who choose NIPT will fail to obtain a result. While that number was just under one percent in this study, rates as high as 12 percent have been published, especially in overweight and obese women.
SMFM has reviewed the evidence, including this recent paper, and feels that while NIPT is a promising new technology, and this new report is important and excellent news, it does not constitute sufficient reason to change current American Congress of Obstetricians abd Gynecologists (ACOG) and SMFM recommendations. Given that just eight aneuploidies were present in the entire cohort of patients, the true test performance is difficult to determine.
Further evidence comparing costs, false positive rates for all included analyses, ability to obtain a result, and overall test performance for all detectable abnormalities in larger numbers of truly average-risk patients is required to justify changing recommendations regarding population-based prenatal screening from just high-risk pregnancies, to all pregnancies. More studies are needed. One study, presented in March 2014 and scheduled for publication in the near future, is the NEXT study, by Drs. Norton, Wapner, and colleagues. This research screened more than 11,000 “low-risk” women younger than 35 years. We eagerly await the results of the ongoing research studies which will address these issues.
