5 questions with our webinar speakers

Last week, in a 30 minute live webinar, authors of a recent study in Prenatal Diagnosis discussed the key differences between the two most common genome sequencing methods used for noninvasive prenatal screening (NIPS): whole-genome sequencing (WGS) and the single-nucleotide polymorphism (SNP) method.  
Co-authors of the paper and webinar presenters, Dale Muzzey, PhD and Carrie Haverty, MS, LCGC, discuss the answers to questions received during the live session.

1. Is NIPS recommended for average risk patients by societal guidelines?

Carrie Haverty, MS, LCGC: The 2016 ACOG Practice Bulletin No. 163 on “Screening for Fetal Aneuploidy”, written in conjunction with SMFM, states that “all women should be offered the option of aneuploidy screening or diagnostic testing for fetal genetic disorders, regardless of maternal age.” NIPS (or cell-free DNA) is listed as one of the screening tests now available for aneuploidy for pregnant women in all trimesters of pregnancy: “triple, quad, and penta screens; cell-free DNA; and ultrasonographic screening as single screening tests.”
Additionally, the 2016 ACMG statement, “Noninvasive prenatal screening for fetal aneuploidy”, recommends “informing all pregnant women that NIPS is the most sensitive screening option for traditionally screened aneuploidies.”
As NIPS moves into the average risk population, it will impact more people, which makes it all the more important to be well-informed about the key differences between different NIPS sequencing methodologies.  

2. Would you recommend a redraw for WGS at fetal fractions less than 1%, where sensitivity even for the WGS method drops to 80% or below for Trisomy 21?

Carrie Haverty, MS, LCGC: The guidelines referenced in the previous question, ACOG Practice Bulletin No. 163 on “Screening for Fetal Aneuploidy” recommends against redraws even for no-calls because it “may delay diagnosis of fetal aneuploidy, which may affect reproductive options for an abnormal result.”
It is important to note that a sensitivity of 80% is far better than that provided by maternal age or nuchal translucency screening or quad screen. Additionally, for trisomy 21, it has been shown that affected pregnancies actually have a higher average fetal fraction overall, so weighing the odds of a pregnancy having trisomy 21 and a fetal fraction less than 2% and being called a false negative against the clinical consequences of many no calls (additional invasive procedures, anxiety, delays in results, etc) is an important consideration and one of the motivations for doing this modeling and evaluation.
We are also aware that this is a screening test and the purpose is not to identify every affected pregnancy at any cost, but rather to most effectively identify the vast majority of affected pregnancies with the least number of unnecessary invasive procedures.

3. Doesn’t “no-calling” mitigate the risk of reporting false negatives?

Carrie Haverty, MS, LCGC: Guidelines state that patients with a “no-call” result should get counseling and strongly consider a diagnostic procedure. However, it is important to consider what happens in these “no call” situations. What do patients and providers do with this information (or lack thereof)?
Based on published clinical experience studies, we know the reality is many patients will neither get a second blood draw done nor will they get an invasive diagnostic procedure done. Ultimately, this leads to some of them having a baby born with a chromosome abnormality that was undiagnosed, an effective false negative.

4. How was fetal fraction calculated for both the WGS and SNP methods?

Dale Muzzey, PhD: For the modeling in our paper, fetal fraction was an input parameter used to generate the simulated data, not a value that needed to be inferred from patient data. We assumed for both methods that fetal fraction was known without error.
In a clinical setting, inference of fetal fraction depends on testing methodology. For the SNP method, fetal fraction can be calculated from the deflection of allele balance at SNPs where the mother is homozygous for one allele and the fetus inherits a different allele from the father. For the WGS method, fetal fraction can be calculated using the SeqFF method published by Kim et al in 2015, a well-established and widely used technique.

5. In the simulated population, what % of “patients” had a fetal fraction of 4% and below?  

Dale Muzzey, PhD: For this analysis, all simulated patients had low fetal fraction, i.e., ≤4% fetal fraction. As this was a simulation-based paper, we did not examine or model what percent of the US population have a low fetal fraction. This percent would range based on the laboratory and methodology used to measure fetal fraction. For example, Ryan et al (2016) noted that 3.8% of women would have a fetal fraction less than 2.8%, and fitting a model to the Nicolaides et al (2012) WGS-method paper suggests that only 0.33% of samples have <2.8% fetal fraction.
 

 
Not all cell-free DNA methodologies are created equal. The area where the two main sequencing methodologies differ in performance is in low fetal fraction samples. This analysis found that no-calling low fetal fraction samples has far greater implications than taking a slight dip in sensitivity.
Please note that all sources for this paper were third party data, patents or peer-reviewed publications, that are available in published scientific literature. Full transparency of our analysis has been provided by making all information publicly available (https://github.com/counsylresearch/artieri_et_al_nips_at_low_ff).
Access the recording of our live webinar here.
Only have 5 minutes? View our Counsyl Insights video, highlighting the main conclusions.
Please submit any questions to [email protected].
ABOUT US
At Counsyl, we strive to put patients first, put clinicians in control, and put costs in their place. Since 2007, we’ve worked with tens of thousands of providers to screen over half a million patients.

Our DNA tests include:
Foresight™ Carrier Screen (formerly Family Prep Screen): Unmatched detection of at-risk couples for serious conditions.
Prelude™  Prenatal Screen (formerly Informed Pregnancy Screen): Personalized insights into more patients’ pregnancies with our NIPS.
Reliant™ Cancer Screen (formerly Inherited Cancer Screen): Help patients get ahead of cancer with our hereditary cancer screen.

References: Artieri, C. G., Haverty, C., Evans, E. A., Goldberg, J. D., Haque, I. S., Yaron, Y., and Muzzey, D. (2017) Noninvasive Prenatal Screening at Low Fetal Fraction: Comparing Whole-Genome Sequencing and Single-Nucleotide Polymorphism Methods. Prenat Diagn, doi: 10.1002/pd.503