by Lisa Demers and Stephanie Snow
Stephanie Snow, MS, CGC has 11 years of prenatal genetic counseling experience. She worked as a clinical site coordinator and genetic counselor for the FASTER study and as a clinical research coordinator for the NEXT study. Lisa Demers, MS, LGC has 12 years of prenatal genetic counseling experience and currently works with Ariosa Diagnostics as a Medical Science Liaison.
The landscape of prenatal screening is changing. The use of non-invasive prenatal testing (NIPT) in clinical practice is already common and is being adopted quickly by generalist obstetricians and maternal fetal medicine specialists. While the cell-free DNA technology is innovative and the impact on patient care is significant, there is a rising chatter about NIPT failures – the 1-8% (depending on the company) of reports that return without a test result. This is a dual issue – there’s the underlying “annoyance” that NIPT occasionally fails to produce a result, and then there are publications suggesting an association between fetal aneuploidy and test failure. The latter is a conversation for another day.
Although these “no call” results frustrate patients and their doctors, the negativity surrounding these failures is surprising. The concept of a test failing is not new in medicine, and certainly not within prenatal medicine. Increasing rates of maternal obesity are just one reason for limited prenatal surveillance, with one study demonstrating that 41% or less of fetal survey ultrasounds on patients with a BMI of 30 or higher were able to be completed on the first try. When it comes to first trimester measurement of nuchal translucency (NT), the FASTER trial noted an overall 7.5% failure rate, either because of an inability to measure or due to inaccurate measurement. In a review of patients within one clinic, where nearly 50% of patients had a BMI over 25 and 25% had a BMI of 30 or more, 4% of patients had an NT failure on the first attempt and of those who opted for a second attempt, 18% failed. Overall in this population, 2.7% of patients did not achieve a NT measurement.
This is not to say that test failures are necessarily bad. When an NIPT test fails, it is often because quality metrics are in place to ensure proper test performance – just as there are standards for NT measurement which are established by the Fetal Medicine Foundation (FMF) and the Nuchal Translucency Quality Review (NTQR) program. An NT may “fail” because a patient presents for screening outside of the appropriate gestational age requirements or because of suboptimal fetal positioning. The nuchal translucency measurement is critical in obtaining aneuploidy risk assessment when combined with serum biochemistry, and even the slightest over or under estimation dramatically impacts clinical care. Such is the case with NIPT quality metrics. These metrics are in place to ensure appropriate risk assessment for the pregnancy, with the most important of these being fetal fraction. Fetal fraction is greatly affected by maternal weight, with obese women less likely having the required minimum concentration of fetal DNA in circulation. Here again, maternal obesity reduces our ability to accurately assess the well-being of a fetus.
In reality, any test failure rate can be a nuisance to a busy clinic. Having patients return for an additional visit is inconvenient to patient and provider alike. However, there are biological and technical reasons for at least some NIPT tests to fail. The thoughtful provider will consider the various metrics involved with the NIPT options and select one that balances high quality metrics (including fetal fraction) and low rate of technical failures.
The DNA Exchange 
While I agree that the issue of “failed tests” is not unique to NIPT, I don’t get the sense that the chatter is so much about this being a clinical nuisance. It seems to me the real issue is that the failed/no-call results are not included in the outcomes data of the prenatal cfDNA studies which are then referenced to calculate test performance. Ideally these studies would show outcomes based on *intent to treat* – all enrolled study participants. With cfDNA studies, the participants that do not get results are excluded from the performance calculations. Multiple studies do more than “suggest” that there is a higher rate of aneuploidy in “no-call” samples, although the manuscripts don’t often highlight this. One example is the recent publication of the NEXT data, with the rate of aneuploidy in participants that were excluded due to no-call results being 2.7%. The fact that aneuploid pregnancies were not detected with the cfDNA assay (due to no-calls) should either be counted against the test sensitivity OR these no-call results should be considered as “screen-positives” and thus count against the specificity (thereby decreasing the PPV). After all, we consider maternal serum screens with a higher than ~0.5% chance of Down syndrome to be screen positive. Given that healthcare providers are looking to these studies for information about test performance, we need studies that accurately reflect the real life experience and include all study participants in the outcomes calculations.
It is unavoidable to have test failures. Even that will bring inconvenience and rework to any people, but for a high quality and good service, maybe it is not bad thing completely!
Ene Sena
http://www.creative-animodel.com/Animal-Model-Development/Knockin-Service.html
Hello. I have had two failed NIPT tests. The reason code that was checked off was “the sample did not meet the threshold for quality control”… any idea what this means? Thank you!