Prenatal genetic screening goes beyond trisomies

In the May 2015 issue of MLO [2015;47(5):14] we reviewed the evolution of prenatal genetic screening. The closing line of that article stated that “…rapid technological progress, particularly using SNPs, holds the promise of even greater improvements in test performance and safety.”¹ Now, we revisit the status of prenatal screening for chromosome abnormalities and the continued advancements in this area.

Non-invasive prenatal testing or screening (NIPT/NIPS) using cell-free DNA (cfDNA) has been commercially available for over five years. A number of peer-reviewed journal articles have consistently demonstrated its superior positive predictive value for common trisomies (21, 13, and 18) compared to traditional maternal serum screening. Norton et al demonstrated a 79 percent sensitivity and 3.4 percent positive predictive value (PPV) for trisomy 21 using standard first-trimester screening in all patients. In comparison, cfDNA screening showed a 100 percent sensitivity and 81 percent PPV for trisomy 21 for all patients.² Most important, a total of 11,994 women in that study were < 35 years of age, without additional risk factors for trisomy 21, and cfDNA showed 100 percent sensitivity and 76 percent PPV in that “low risk” group.

Both the American College of Obstetricians and Gynecologists (ACOG) and the American College of Medical Genetics (ACMG) have updated their screening statements to recommend that NIPT be made available to all pregnant woman, regardless of their prior risk for aneuploidy.^3,4 However, approximately half of women with private insurance and essentially no women with Medicaid are financially covered to choose NIPT as a first-line screen.

The fact that NIPT maintains its high performance standards in average-risk women in turn prompts continued discussion to “rethink screening.” There are many conditions that are unrelated to maternal age, and usually without a family history. For what additional conditions beyond trisomies is it reasonable/responsible to offer screening to all pregnant women, if non-invasive screening is possible with acceptable sensitivity, specificity, and positive predictive values? How is SNP technology for NIPT uniquely positioned to provide sensitive and specific expansion of prenatal genetic screening?

A logical next step in prenatal genetic screening is to consider smaller genetic changes in a chromosome, called copy number variants (CNVs). CNVs are typically microdeletions or microduplications less than 10 Mb that are associated with clinically significant outcomes and are unrelated to maternal age. In 2012, Wapner et al reported that clinically relevant deletions and duplications were found in six percent of pregnancies with ultrasound anomalies and 1.7 percent of pregnancies without risk factors.⁵ However, unlike the common trisomies, deletions and duplications are not part of routine serum screening or NIPT. There are different approaches to CNV screening: targeted and whole genome.

Targeted screening

Targeted CNV screening entails choosing specific deletion or duplication sites with known clinical outcome, and providing risk assessment particular to that associated condition. For example, the most common microdeletion condition, 22q11.2 deletion syndrome, has a published incidence at birth of 1/2000, though this may be an underestimate. It is a well-described clinical entity with known chromosomal breakpoints and clear diagnostic testing parameters. Therefore, screening for 22q11.2 deletion syndrome is a natural expansion of NIPT.

ACOG and ACMG guidelines are clear with respect to the necessary laboratory conditions that need to be met in order to offer CNV screening. ACMG states that “Laboratory requisitions and pretest counseling information should specify the DR, SPEC, PPV, and NPV of each CNV screened.”⁴ This requires analytical and clinical validation for each deletion or duplication, as well as accurate estimates of population incidence of each condition. Currently, different NIPT laboratories offer a range of microdeletion syndromes, but their reports often do not follow these guidelines.

A recent publication on the clinical experience of SNP-based microdeletion testing addresses the issues of expanded screening and the need for transparent follow-up. This publication extends the initial reporting of SNP-based NIPT screening for 22q, and highlights outcome data for the remainder of the microdeletion panel currently offered (1p36 deletion, Angelman, Prader-Willi, and cri-du-chat). Performance improvements to this SNP-based testing resulted in a decrease in false positive test rate (0.07 percent for 22q) and an increase in PPV (44.2 percent for 22q; 31.7 percent combined for others). While some publications have questioned the expansion of NIPT into microdeletions due to concerns about positive screen rate and low detection rate, the targeted nature of SNP-based NIPT screening is shown to have a higher sensitivity than other NIPT methodologies.⁶

Whole genome

The concept of targeted CNV screening can be further expanded to other less common and less well-described microdeletions and microduplications, as well as rare autosomal trisomies (RATs), by looking at genomic information across all chromosomes. However, ACMG specifically recommends not screening for genome-wide CNVs, stating (among other reasons) that “If this level of information is desired, then diagnostic testing followed by CMA is recommended.”³ The clinical performance of such test options is not clinically well validated, and reporting does not follow test metric guidelines, as described earlier.