NEW YORK (GenomeWeb) – The Netherlands has made low-cost noninvasive prenatal screening available on a nationwide scale as part of a new implementation study, one of the first countries to do so. Since April 1, under the TRIDENT-2 (Trial by Dutch Laboratories for Evaluation of Noninvasive Prenatal Testing) study, all pregnant women can choose between NIPT and the first-trimester combined test (FCT) for their prenatal screening.
The study, which received about 6,000 samples during the first month, is subsidized with €26 million from the Dutch Ministry of Health, Welfare, and Sport, so women have an out-of-pocket cost of just €175 ($190), similar to the €165 they currently have to pay for FCT, which combines an ultrasound with a biochemical blood test.
TRIDENT-2, which is organized by the Dutch NIPT Consortium, a group that includes many stakeholders involved in prenatal care in the country, aims to integrate NIPT seamlessly into the country's existing national prenatal screening program. "It is not only about doing the test – it's a whole chain" that involves midwives, gynecologists, clinical geneticists, and laboratories, said Erik Sistermans, head of genome diagnostics at the VU University Medical Center, who coordinates the study. "Depending on what we find, there is always proper follow-up and counseling of the pregnant woman. It's not just sending blood somewhere and getting an email back."
In the meantime, genetic testing firm Gendia of Belgium has been challenging the Dutch health ministry in court over the subsidies for TRIDENT-2 because it is not eligible to receive them, which it says violates European Union market regulations. The company argues that it has been unfairly shut out of a market that it has been serving for years. "It's not only, in our view, illegal state support, but, also, it discriminates against foreign companies," said Patrick Willems, Gendia's director.
Gendia, which does not perform NIPT in house but sends samples out to Ariosa's laboratory in the US, currently offers Ariosa's Harmony test for €450. The Antwerp-based firm has been providing NIPT to more than 20,000 Dutch women over the past several years, when the test was not widely available in the Netherlands. It also claims that its test provides more reliable results than the Dutch genetic centers, which the TRIDENT organizers deny.
Since the start of TRIDENT-2, Gendia has seen its NIPT business drop sharply: Instead of an average of 100 to 120 samples per week, it now receives fewer than 20 samples per week from the Netherlands. Until recently, Dutch samples accounted for about 50 percent of its overall NIPT business, Willems said.
Last week, Gendia suffered a legal defeat when a court in The Hague decided that the government subsidies are permitted because they help provide a "service of general economic interest," allowing pregnant women to decide freely between NIPT and the FCT, provided to them at about equal cost. The company is considering appealing the decision or taking its case to the European Commission, Willems said, though he noted that the latter might be a lengthy process.
The dispute highlights the opposing interests of countries wanting to implement new clinical tests in their national healthcare systems in an organized manner and commercial providers wanting access to the same patient population on a level playing field. It also showcases how European countries differ in their approach to regulating genetic testing – some relying on commercial providers more than others – despite the fact that they are all part of the European Union and its single market.
In Belgium, for example, genetic testing is generally performed by eight university genetics labs but is also permitted by private companies, and Willems said hat reimbursement for NIPT is now available even if the test was performed by a commercial provider.
In Germany, NIPT is provided by a variety of commercial testing labs – increasingly by local providers that do not send samples abroad – but is currently not generally reimbursed by the country's public health insurance system, though this might change in 2019, at least for high-risk pregnancies.
In the UK, on the other hand, NIPT is performed both by commercial labs and by some National Health Service laboratories. Late last year, the UK government approved a plan to introduce NIPT to the country's existing prenatal screening program over the next three years.
In the Netherlands, only eight genetic laboratories, all associated with university medical centers, are licensed by the government to provide genetic testing, meaning no commercial providers can offer NIPT in the country. "We're hoping to change this," Willems said. "There are only a few countries where genetic testing is limited to a university setting – in most countries, genetic testing can be done by private companies."
However, this does not mean that Dutch women have not had access to genetic testing from commercial laboratories. Since NIPT was only available on a very limited basis through the Dutch genetics centers until last month, Dutch women for the past several years have turned to both companies and academic centers in neighboring countries, in particular Belgium, paying for the test out of pocket. While some Dutch women literally traveled across the border for testing, many had their blood drawn by their healthcare provider at home, who sent their sample abroad for testing.
The Netherlands has been evaluating the implementation of NIPT into its national prenatal screening program since 2014, initially only for women with high-risk pregnancies. That year, the NIPT Consortium launched the nationwide TRIDENT study after obtaining a two-year license from the Dutch Ministry of Health.
The study, funded by the Netherlands Organization for Research and Development (ZonMw), made NIPT available to Dutch women with an increased risk of carrying a fetus with trisomy 21, 18, or 13, based on results of their FCT or their medical history.
For the study, NIPT was performed by six of the eight university genetic laboratories using standard whole-genome sequencing of maternal plasma DNA and an in-house-developed algorithm called Wisecondor. Initially, the labs used sequencing platforms from Illumina and Life Technologies but later moved to Illumina machines alone.
Last December, the consortium published results from the first five months of the TRIDENT study in Prenatal Diagnosis. According to the article, 86 percent of 1,413 women with a high-risk FCT result opted for NIPT, and another 179 women chose NIPT because of their medical history. NIPT was successful in 99.7 percent of cases, with a mean turnaround time of 14 days. It correctly predicted all cases of trisomy 21 except one, and all cases of trisomy 18 and 13. It also provided one or two false positive results for each trisomy, resulting in a positive predictive value of 94 percent for trisomy 21, 80 percent for trisomy 18, and 67 percent for trisomy 13. Overall, the authors estimated that invasive testing decreased 62 percent as a result of NIPT. So far, the researchers have not yet published any updated results for the TRIDENT study.
For TRIDENT-2, the number of testing laboratories has been consolidated to three, who have all ramped up their capacity: VU University Medical Center in Amsterdam, Erasmus Medical Center in Rotterdam, and Maastricht University Medical Center. According to Sistermans, labs built out their capacity assuming that 50 percent of the 170,000 or so women who give birth in the Netherlands each year might request NIPT. Historically, only about a quarter to a third of Dutch women have opted for FCT, though, so actual numbers for NIPT might be lower.
For the time being, the centers are using their existing testing platform, but the study is in the midst of a European tender for providers of whole-genome sequencing technology, and depending on the offers it receives, it may switch to a different platform. According to Sistermans, the study is only considering whole-genome sequencing solutions because besides results for trisomies 21, 18, and 13, it also reports other, rarer aneuploidies and large deletions and duplications, as the original TRIDENT study did.
So far, TRIDENT-2 has received about 6,000 samples, and for 80 percent of those analyzed so far, the turnaround time has been five or six business days from receipt of sample by the laboratories. Sistermans said the study had expected a large number of incoming samples during the first week because some women had waited for the program to start, so it had warned women that testing might take on the order of 10 business days during the first month. Samples indeed peaked during the first week, but the majority of reports were still issued within five to six days. "Everything went rather smoothly – we had no major problems," he said.
Because testing is now available to all women, the positive predictive value – which indicates how likely a positive test result is true and depends on how many positive cases there are in a population -- is expected to be lower than for the original TRIDENT study, which only included high-risk pregnancies. In a statement posted on behalf of the NIPT Consortium on the VU University Medical Center Amsterdam website, the organizers said they expect the PPV to be around 75 percent for trisomy 21. That number, Sistermans said, is an estimate, based on several published studies that used the whole-genome sequencing approach. By comparison, the FCT only has a PPV of 5 percent.
In addition, a brochure for genetic counselors put out by the Dutch National Institute for Public Health and the Environment (RIVM) in March, which was developed by the NIPT Consortium and others, estimates the PPV for trisomy 18 to be 24 percent and for trisomy 13 to be 23 percent.
Gendia maintains that the Ariosa Harmony test it provides, which runs on a microarray platform and only reports on trisomies 21, 18, and 13, yields better PPVs, citing numbers from a study it published last year in the Journal of Molecular Biomarkers & Diagnosis that analyzed results from 10,000 consecutive samples it received between 2013 and 2015. Of those, 39 percent were from women with advanced maternal age and 12 percent had an elevated risk based on the combined screening test. More than 6,500 of the samples came from the Netherlands. Overall, 99.3 percent of the tests were successful, with a turnaround time of about 10 days, which Willems said has since improved to an average of six business days. In total, 147 samples had a positive result, and for 121 of those, follow-up data was available. Based on those, the PPV for trisomy 21 was 99 percent, for trisomy 18 it was 80 percent, and for trisomy 13 it was 90 percent.
However, the NIPT Consortium argues that the TRIDENT-2 population and Gendia's study population are not comparable, since the Gendia study had a large number of high-risk pregnancies, while the consortium bases its PPV estimates on an all-risk population. According to the consortium's statement, Ariosa itself cites different PPVs than Gendia for its test, based on the NEXT study, published in the New England Journal of Medicine in 2015, which analyzed results from almost 16,000 women from a general prenatal screening population. That study reported a PPV of 81 percent for trisomy 21, of 90 percent for trisomy 18, and of 50 percent for trisomy 13. At least for trisomy 21, the most prevalent chromosomal aneuploidy, that PPV is similar to that of the Dutch test, according to the consortium.