Has noninvasive prenatal testing impacted termination of pregnancy and live birth rates of infants with Down syndrome?

Abstract Background Implementation of noninvasive prenatal testing (NIPT) as a highly accurate aneuploidy screening test has raised questions around whether the high uptake may result in more terminations of pregnancies and fewer births of children with Down syndrome (DS). Aim The aim of the study was to investigate the impact of NIPT on termination and live birth rates for DS. Methods Literature reporting pregnancy outcomes following NIPT was reviewed. Termination rates were calculated for women with a high‐risk NIPT result for DS. Two audits of pregnancy outcomes where NIPT indicated DS were conducted in the United Kingdom and Singapore. Results Fourteen studies from the United States, Asia, Europe, and the United Kingdom were included in the review. Live births of children with DS were reported in 8 studies. Termination rates following NIPT were unchanged or decreased when compared to termination rates prior to the introduction of NIPT. Audits found 15 of 43 women in the United Kingdom and 2 of 6 in Singapore continued pregnancies following a high‐risk NIPT result. Conclusions Termination rates following the detection of DS by NIPT are unchanged or decreased compared to historical termination rates. Impact on live birth rates may be minimal in settings where termination rates fall. Population‐based studies are required to determine the true impact.

increase in the number of parents seeking prenatal testing and termination of pregnancy with a resultant significant decrease in the number of children born with DS. 11,12,[15][16][17] Research with pregnant women 11,12 and parents of children with DS 16,17 identified concerns that fewer children being born with DS could result in a reduction of social supports and resources for children with disabilities and a society that is less tolerant of people who have children with disabilities. Furthermore, these key stakeholder groups felt that a less tolerant society could increase feelings of pressure for pregnant women to have testing and subsequently terminate an affected pregnancy. 11,12,16 Interest in and uptake of NIPT is high, 7 making it likely that the number of parents opting for prenatal testing for DS will grow as many parents who would not have previously opted for prenatal testing because of the risk of miscarriage would be willing to have NIPT. 11,18 It is not yet clear, however, whether the increase in the numbers of women having prenatal testing will directly result in more terminations of pregnancy. There will be wide variation in the number of children born with DS between and even within countries depending on attitudes to prenatal testing, disability, and termination that are influenced by religious, social, and cultural settings, costs of prenatal testing and access to termination of pregnancy. Ultimately, we will see the impact  23 and 55% fewer in China (2011). 24 As the numbers of older women giving birth have grown, the numbers of pregnancies affected with DS have also increased. However, in England and Wales 21 and in Europe 25 the live birth rate remained relatively unchanged between 1990 and 2009 even though prenatal screening and diagnosis became more common over the same period. In contrast, in the United States, the live birth rate has increased since the early 1990s through to 2007. 19 One small regional study conducted in the Hampton Roads area of Virginia reported that NIPT has not affected the number of children born with DS in this area. 7 More time is needed to see the impact of NIPT on the live birth rate more widely and conclusively.
While we await definitive population-based studies, there are several lines of evidence available to us now that can help assess the impact the introduction of NIPT might ultimately have on the numbers of parents choosing to continue their pregnancy following a prenatal diagnosis of DS. Here we examine the literature to look at reports of pregnancy outcomes following NIPT. We also describe 2 new audits of NIPT services in clinical practice that were conducted in England and in Singapore.

| Literature review
We reviewed the published literature reporting the number of live births of children with DS following screening with NIPT. A search of English-language articles from the time NIPT entered clinical practice (January 1, 2011, to September 25, 2017) was conducted. We searched the PubMed electronic database using the following search terms: "cell free fetal DNA," "NIPT" or "Non-invasive prenatal test*" or "noninvasive prenatal test*" or "NIPD" or "non-invasive prenatal diagnosis" or "noninvasive prenatal diagnosis." A manual search of the reference lists of included studies and relevant original and review articles was also performed. Publications were included if they described data on numbers of women having NIPT for DS and provided information on pregnancy outcomes, such as live births, pregnancy termination, fetal demise, or stillbirths. Studies that described What is already known about this topic?
• Noninvasive prenatal testing (NIPT) has been shown to be a highly accurate prenatal screening test for DS and is being implemented widely throughout the world.
• Introduction of NIPT has increased the prenatal detection of DS with a significantly reduced invasive testing rate, but the impact on rates of termination of pregnancy and the number of children born with DS is not yet known.
What does this study add?
• Introduction of NIPT has a variable effect on termination rates for DS, but rates have remained unchanged or decreased when compared to termination rates reported prior to the introduction of NIPT, with many parents using NIPT for information and continuing pregnancies when results show a high risk of DS.
• Practical and emotional support structures are needed for these families.
• Where termination rates fall NIPT may have a minimal impact on live birth rates for DS.
• Monitoring at population levels is required for a more accurate assessment of live birth rates. calculated as a proportion of all pregnancies with a high-risk NIPT result (excluding false positives and negatives). This includes cases confirmed with invasive testing and those confirmed at birth without confirmatory invasive testing. Where additional data were available, termination rates were calculated for women going directly to invasive testing.
A separate search was conducted to identify published termination of pregnancy rates prior to introduction of NIPT. Termination rates were sought for each of the countries where the studies included in the review were conducted. The PubMed electronic database was searched using the following terms: "country name" AND "Down* syndrome" AND "termination" or "abortion" or "live-birth rates." 2.2 | Audit of pregnancy outcomes following NIPT as a clinical service in Singapore NIPT has been offered at the National University Hospital Singapore

| Audit of pregnancy outcomes following NIPT offered as a clinical service in the United Kingdom
Following the development and validation of a cfDNA sequencing protocol by our NHS service laboratory (North East Thames Regional Genetics Service), NIPT was initially offered as part of a research trial in 8 UK maternity units, which demonstrated that NIPT could be successfully offered as a contingent test without increasing costs in the NHS. 35 One weakness of this study was that there was potentially

| Literature review of pregnancy outcomes following NIPT
Fourteen studies were included in the review, 8 were prospective 8,30-32,34-37,39 and 6 were retrospective audits. [26][27][28][29]33,38,40 Most studies were conducted in the United States (n = 5), 8,28,29,31,33 however, 2 studies from the United Kingdom [35][36][37] and China, 32,40 and individual studies from Hong Kong, 26,27 Taiwan, 30 the Netherlands, 34 Spain, 38 and France 39 were also identified. The 14 studies were diverse in their objectives and study design. Taking a broad overview of objectives, 4 studies reviewed the experience of offering NIPT at a single centre, looking at factors such as patient characteristics and uptake of NIPT and invasive testing, 28,29,31,38 one study was questionnaire based across 4 centres and explored women's views about NIPT and factors influencing decision making, including views on termination of pregnancy. 8 One study looked at the utility of using NIPT for twin pregnancies. 39 Four studies aimed to examine NIPT performance, 26,27,30,32,40 and 3 explored the impact of implementing NIPT as part of state supported health care services. [34][35][36][37] Only 1 study had the explicit aim of looking at clinical outcomes and patient choices following NIPT, including continuing an affected pregnancy or opting for termination of pregnancy. 33 Pregnancy outcome data from each of the studies are summarised in Table 1  Lau et al 26,27 Hong Kong   3 0 0 100% a Termination rates were calculated as a proportion of all pregnancies that had a high-risk result for DS from NIPT (false positives and negatives excluded). This may include cases confirmed with IPD and those confirmed at birth who did not undergo confirmatory IPD. Where additional data were available, termination rates were calculated for "NIPT"-those undergoing NIPT (including cases confirmed with IPD and those confirmed at birth without confirmatory IPD), "IPD"-those going directly to IPD without NIPT, and "Overall"-the termination rate for all cases of DS diagnosed prenatally. Comparison between termination rates reported prior to the introduction of NIPT with termination rates seen in the reviewed studies and audits suggests a general trend towards a decrease in termination rates compared with pre-NIPT rates ( cluded, as we did, that the overall termination rate was not higher than historical controls and stated that their findings argue against the concern that cfDNA screening would increase rates of pregnancy termination.
As one of the common reasons women have declined screening in the past is the miscarriage risk associated with invasive testing, 47,48 it is not surprising that NIPT is being adopted widely around the world. 1 However, our findings suggest that this increase may not impact greatly on the number of babies born with DS as many parents will use NIPT for information and not for decisions about termination of pregnancy. The decrease in termination rates compared with pre-NIPT FIGURE 1 Flowchart showing numbers of women and outcomes for the Singapore audit rates observed in the reviewed studies, presumably reflects, at least in part, the uptake of NIPT by women seeking information who would not have had prenatal testing in the past as they would not put their pregnancies at risk with invasive testing. We also found evidence that the termination rate for women opting for NIPT was lower when compared to women who chose to go directly to invasive testing following a high-risk screening result. In the Singapore audit the number of women opting for termination following NIPT (4/6-66.7%) was lower than those who chose invasive testing ( [35][36][37] This difference is most likely due to the variances in motivation for women choosing NIPT versus invasive testing, with the latter group perhaps being more likely to want diagnostic information to make decisions about termination. 49 We know from the literature that the uptake of NIPT is high, 34,35 making it likely that detection of DS will increase, but the lower termination rates following NIPT in some countries suggest that live birth rates may remain largely unchanged compared to termination rates prior to the introduction of NIPT. Termination rates did not, however, fall in all studies reviewed here and in settings where NIPT uptake is high and termination rates remain unchanged there will be an overall increase in numbers of terminations of pregnancy and a corresponding decrease in the live birth rate.
Ultimately, however, the research included here describes relatively small numbers of women and can only give insights into pregnancy outcomes following prenatal testing and we do not know how many children with DS were born to parents who chose not to have prenatal testing. Population-based studies of live birth rates are therefore essential to allow us to see the overall impact of NIPT. The importance of the population-based studies is highlighted by reports from countries such as the Netherlands where the termination rate following a prenatal diagnosis of DS is high, but the overall live birth rate has increased over time as many women opt not to have DS screening. 20 Uptake of DS screening is low in the Netherlands compared to other European countries. This may reflect the fact that parents must make a financial contribution to screening, also that the offer of DS screening is not presented as a routine test and is discussed in a way that emphasises the right not to know. 50 Moreover, several studies looking at hypothetical choices have shown that even with NIPT as an option many women will still choose not to have any prenatal testing. In a survey of 2666 women from 9 countries, there was a sizable proportion of women who said they would not have any prenatal testing for DS, including more than one third of women in the Netherlands and Israel. 51 Similarly, studies from the United States and the Netherlands found that around one third of people surveyed reported not wanting any tests for DS. 52,53 Research looking at women's hypothetical choices regarding how they would respond to an NIPT result that suggests DS is highly likely support our findings that many women will choose to continue their pregnancies. 18,49,52,54 While the data on hypothetical choices needs to be interpreted with caution, as parents may make different choices when faced with real-life situations, these studies indicate that a significant number of parents would use NIPT for information only, so that they could plan and prepare for the birth of an affected child rather than using the information to make decisions about termination of pregnancy. 18  opportunities to discuss the diverse range of feelings that accompany prenatal diagnosis. 59 As many parents will have prenatal testing with NIPT for information only, there will be more parents continuing pregnancies knowing that the baby has DS. As many of these women receive the diagnosis in early pregnancy it is important that their ongoing needs for emotional and clinical support are met. Furthermore, knowledge of fetal DS status may allow increased surveillance to prevent intrauterine death as highlighted by recent research from the United States, which found elevated rates of growth restriction, early delivery due to nonreassuring fetal status, and placental insufficiency in a cohort of 64 women continuing the pregnancy with a diagnosis of DS. 60 Provision of practical and psychosocial support for individuals with DS and their families will also continue to be needed.   21 Gil et al 36