This article has been updated with additional information about coverage of NIPT.
NEW YORK – Natera has found that the performance of its Panorama noninvasive prenatal screening (NIPS) test has stayed consistently high over several years, even as testing has shifted to a younger age group with lower risk for fetal aneuploidies. In addition, the firm is nearing completion of a prospective clinical study that it says will further bolster these results, not only for common fetal trisomies but also for some microdeletions.
In a quality assurance study involving more than a million tests conducted in Natera's laboratory over a four-year period, published last month in the Journal of Clinical Medicine, company researchers showed that based on positive test results for which they were able to obtain follow-up information, the Panorama test had an overall PPV of 93.1 percent for detecting trisomies 21, 18 and 13, as well as monosomy X. PPV differed little between women under the age of 35, considered at lower risk, and those over the age of 35 who have a higher risk. The company presented the results of the study earlier this year at the Society for Maternal-Fetal Medicine annual meeting.
Commercial test providers are hoping that studies like these will eventually persuade large commercial insurers, along with additional Medicaid state programs, to cover NIPS in all pregnancy risk groups rather than just in high-risk women.
Natera implemented the quality assurance program described in its study in late 2013 to ensure that Panorama test results met professional and internal standards. One goal of the study was to show that PPV can be used as a measure of quality, according to Chief Medical Officer Paul Billings.
For the study, Natera researchers looked at 1,035,844 test results reported from early 2014 to the end of 2017. During that time, test volumes increased from 33,700 per quarter to 86,800 per quarter.
They have risen further since — last month, Natera said it was processing 2,000 Panorama test per working day, or about 120,000 per quarter, and had processed more than 2 million tests in total. "It's fair to say that we're the leading provider of NIPS in the United States," Billings said.
Each quarter, the researchers randomly selected 200 to 400 test-positive cases for which they requested pregnancy outcome information from referring physicians or clinics. Using this information, they calculated PPVs for each condition tested, as well as an overall PPV.
Follow-up was not perfect: for the 4,071 test-positive cases for which they contacted physicians, confirmatory information by genetic testing, ultrasound, or pregnancy loss was only received for 2,044 cases, or about 15 percent of the overall 13,231 test-positive cases in the cohort.
PPV was overall high but differed between the conditions tested, from 79.6 percent for trisomy 13 to 95.7 percent for trisomy 21.
Over the course of the study, the percentage of women under the age of 35 who received testing gradually increased, from 49 percent at the beginning to 63 percent later on. Along with this shift, as expected, the test-positive rate decreased over time.
A lower disease incidence in a population is usually associated with a lower PPV for a screening test, but that's not what the researchers found. "The number of women below age 35 actually grew quite dramatically, yet the predictive values were pretty much the same" to those over age 35, Billings said. Specifically, overall PPV was 90.1 percent in women under 35 and 94.5 percent in those over 35.
"This is crucial because the reimbursement payment for these tests is not fully available for women below the age of 35," he said. "We wanted to be very clear that it's not the test performance which is limiting the availability of this test for younger women but rather other kinds of considerations and policies."
It is not entirely clear why PPV did not decrease, as expected, in women of younger age who presumably have a lower incidence of aneuploidy. Billings said one hypothesis is that in women above age 35, the rate of false positive results is higher than in younger women due to increased placental mosaicism and other factors. "It tends to kind of wash out, so it keeps the positive predictive value about the same" in the two age groups, he said. "That's our current explanation for it."
Natera continues to collect pregnancy outcomes data on an ongoing basis, collecting data for hundreds of cases each quarter, Billings said. "Obviously, false-negative results are even harder to get good data on because we don't know who to look for," he added.
Not everyone is convinced by the study's results, though. "We agree that quality monitoring is an important laboratory practice but urge caution in the use of a commercial laboratory experience report such [Natera's study] to assess test performance," Maximilian Schmid, head of medical affairs at Roche Sequencing Solutions, said in an email.
Because follow-up was incomplete in the study, the report "is prone to ascertainment bias," he said. "We advocate for the use of large prospective clinical studies with complete follow-up in well-defined populations, such as the NEXT study … to address test performance."
That study, which was funded by Roche's Ariosa Diagnostics and the Perinatal Quality Foundation, was led by Mary Norton at the University of California, San Francisco and Ronald Wapner at Columbia University. It was published in the New England Journal of Medicine in 2015 after being presented at a conference the previous year. The NEXT trial enrolled almost 19,000 pregnant women at 35 centers and analyzed about 16,000 results from Ariosa's Harmony test. It found that the test worked well in women of all ages, with a positive predictive value of 80.9 percent for trisomy 21.
Meanwhile, Natera has been working on its own prospective study for the Panorama test. Last November, the company said it had completed enrollment for the SMART (SNP-based Microdeletion and Aneuploidy Registry Trial) study, which will study the performance of the Panorama test in 20,000 pregnant women recruited at 21 medical centers. Notably, the study will look not only at common trisomies but also at a number of microdeletions. "We believe it is the largest registry-type prospective trial where we have complete follow-up," Billings said.
He said that it was important to include microdeletion syndromes because their incidence, in total, can be as high as that of trisomy 21 in a younger population. Not much performance data has been published for microdeletion tests so far, he said, "so we need to be cautious about what the positive predictive value of a microdeletion finding is for any particular laboratory until we have more published data."
Results from the SMART study will be available in the first half of 2020. "The preliminary analysis strongly supports the important role of an NIPT in the provision of accurate information," he said, consistent with the high performance seen for the Panorama test in the published quality assurance study.
The company is currently gathering the follow-up data for the study, he said. In addition, it is re-analyzing some of the data generated in the earlier phases of the study, which has been ongoing for three years, taking advantage of recent improvements in the analysis methods. Due to these new methods, the no-call rate — cases where the Panorama test could not yield a result — has dropped to 1 percent or less, he said.
Whether or not these studies will sway US insurers to start covering NIPS in the general population is unclear, though. About half the births in the US are in women covered by Medicaid, and at the moment, Medicaid programs in about 10 US states regularly pay for NIPS in women below the age of 35, Billings said, either with or without an official coverage policy. According to Natera, there are 14 additional states with managed Medicaid plans that offer average-risk coverage.
However, the large commercial insurers — in particular, Aetna and United Healthcare, who he said cover "a very significant fraction" of women of childbearing age — do not currently offer NIPS coverage in women under age 35.
"We believe absolutely, for the intent that most women are seeking information, that NIPS is the best screening test and performs beautifully, both in high-risk and in average-risk populations," he said.
Even if the American College of Obstetricians and Gynecologists were to come out with an endorsement of cell-free DNA analysis as a screening test for all age groups, major payors may "not necessarily" change their policies, Billings said.
While ACOG has historically been influential, he said, he would have expected the big insurers to cover NIPT widely by now, based on their own technology assessments and the available performance data. "So their reticence, or the delay in coverage, may have other explanations that we don't understand," he said.