NEW YORK (GenomeWeb) – Close on the heels of research that highlighted how widely the reported specificity and the positive predictive value of sequencing-based non-invasive prenatal tests can diverge, Natera, one of several companies offering clinical sequencing-based NIPT, has published a study of its Panorama test showing high PPV not only for higher-risk but also lower-risk women.
The study, which appeared online last week ahead of print in the American Journal of Obstetrics and Gynecology, evaluated NIPT results compared to karyotype or clinical evaluation in a subset of several hundred positive results among almost 30,000 samples tested by the company over a six-month period in 2013.
Solomon Moshkevich, Natera's VP of marketing and business development, told Clinical Sequencing News that the study researchers, led by Peer Dar at New York's Albert Einstein College of Medicine, are treating their finding that PPV for low-risk women was just as high as for high-risk women as "interesting," but not necessarily characteristic.
Statistically, since the prevalence of aneuploidies in lower-risk women should be lower than in high-risk women, the PPV would be expected to go down in that population. But in the Natera study, PPVs for both high- and low-risk women were about the same.
Within the cohort of tested women, the researchers reported that 507 received a high-risk result for trisomy 21, trisomy 18, trisomy 13, or monosomy X. The group selected 356 of these positives for follow up and comparison to karyotype or other outcome results.
Among those NIPT positive samples that were confirmed by cytogenetics, 84 were true positives, while 38 were false positives. Another 19 cases had no karyotype results, but ultrasound findings were suggestive of aneuploidy. There were 36 spontaneous abortions without karyotyping, 22 pregnancy terminations without confirmation, and 57 cases were lost to follow-up.
Based only on cases that had a cytogenetic diagnosis, the group calculated PPV across all aneuploidies at about 83 percent. For trisomy 21, PPV was 90 percent. The researchers also projected how including all the unconfirmed positive cases – assuming that every one was a false positive – would skew the PPV. In the case of trisomy 21 they reported the PPV could drop as low as 60 percent. For all the aneuploidies tested, PPV could be as low as 52 percent if all unconfirmed cases were false positives.
Even at those levels, the authors wrote, NIPT still far outperformed the standard biochemical screening methods currently used, which have PPVs in the single digits.
According to Moshkevich, Natera sees screening in the low-risk population as a significant growth opportunity. Test indications for NIPT are currently shifting from high-risk women – either older women, or those with other indications of elevated risk of having a fetal aneuploidy – to frontline screening, replacing the biochemical tests currently used in this setting.
Though professional societies have not endorsed NIPT as a frontline screening tool, Moshkevich said Natera has seen increasing numbers of physicians ordering the test in this context, as well as interest from women themselves, driven by increasing knowledge of available tests and technologies.
Interestingly, in the Natera study, when the researchers split their cohort into women over 35 and under 35, though the prevalence of positives was different, the predictive value of Panorama remained similar, actually rising slightly in the younger, lower-risk women.
Moshkevich said that though the Natera researchers don't know why they got such a high value in lower-risk women, or whether this result can be replicated, they have some hypotheses about what might explain it.
Often, NIPT false positives are due to maternal mosaicism. Because mosaicism is known to increase in women as they age, the researchers posited that a lower chance of placental mosaicism in younger women might contribute to a higher PPV in this lower-risk group.
Overall, the PPV reported in the study was higher than that of previous studies of NIPT using other sequencing-based approaches. For instance, a study by Illumina and Tufts researchers published earlier this year in the New England Journal of Medicine found positive predictive values for their NIPT were 45.5 percent for trisomy 21 and 40 percent for trisomy 18.
Though all are based on sequencing circulating DNA in a mother's blood, NIPT from the various companies on the market vary in their exact strategy. Unlike other tests that determine the presence of aneuploidies in a fetus by counting the number of chromosomes present in a sample, Natera uses a computational strategy whereby it creates thousands of virtual projections – possible recombinations of the genomes of a particular mother and father.
The company then performs targeted sequencing of SNPs in a maternal sample and compares these results to its virtual genomes to calculate a risk of a fetus having a particular aneuploidy.
According to Moshkevich, reporting results to patients in a way that communicates what a positive NIPT result means in terms of the actual risk of having a disordered fetus, and making clear that NIPT is a screening test and not a diagnostic is central to Natera's service.
However, about six percent of the subjects with positive Panorama results in the company's study – 22 women – decided to terminate their pregnancies without confirmatory cytogenetic testing, highlighting that there may remain room for improvement in the communication of the nature of NIPT and its predictive power.
Recently, Natera added testing for microdeletions to its Panorama service as an optional add-on. Moshkevich said that since the addition, over 80 percent of Panorama orders have opted in to the microdeletion testing, too. Communicating PPV and actual risk for microdeletions has also been crucial in this area.
"With microdeletions, the prevalence is even lower [than for aneuploidies]," Moshkevich said. "So it's a huge question for the community, how to report these results so that patients know what a positive result actually means."
PPV for Natera's microdeletion testing varies by the deletion, he added. For the 22q11.2 deletion, also called DiGeorge syndrome, he said, risk is about one in 19 for women who receive a positive result.
Circulating tumor DNA
As it looks toward increased use of NIPT in the frontline screening of pregnant women, Natera is also moving to apply its technology beyond prenatal testing, specifically to detection of circulating tumor DNA in the blood of cancer patients.
A recently announced partnership between Natera and the Feinstein Institute for Medical Research is one way this process has been accelerated.
Moshkevich said Natera believes the computation strategy that it has developed for NIPT can also provide extremely sensitive detection of ctDNA, especially in measuring copy number variation. He noted that the firm has found in early unpublished experiments that it can detect CNVs down to 0.5 percent and SNVs down to 0.05 percent concentration.
The company is also investigating using its technology in the context of transplant monitoring – to detect changes in circulating DNA from transplanted tissue.