Portal de Periódicos da CAPES

Non‐invasive prenatal diagnosis for Down syndrome: the paradigm will shift, but slowly

2012; Wiley; Volume: 39; Issue: 2 Linguagem: Inglês

10.1002/uog.11083

1469-0705

Peter Benn, Howard Cuckle, Eugene Pergament,

Gestational Trophoblastic Disease Studies

After decades of research with a wide range of putative methodologies, at last a commercially viable technique has emerged for the non-invasive prenatal diagnosis (NIPD) of Down syndrome1-4. Public health planners and clinicians providing healthcare to individual patients need to evaluate the efficacy of this technique and assess the consequences of this new option. Has the time come to replace invasive prenatal diagnosis by NIPD following a positive screening test? Can the aneuploidies other than Down syndrome that are currently detected in chorionic villus samples (CVS) and amniotic fluid cells also be detected? What about microdeletion/duplication syndromes that can now be detected by fluorescent in-situ hybridization (FISH) and microarray analysis of villi and amniocytes as part of invasive prenatal diagnosis? Can NIPD replace conventional screening? And, if so, will other potential benefits of routine first-trimester biochemical and ultrasound investigation (e.g. early detection of fetal structural abnormalities and pregnancy complications) be lost? The technique in question is 'massively parallel sequencing' (MPS). To evaluate whether MPS is a diagnostic test for Down syndrome, the meaning of 'diagnostic' must be defined. Diagnostic could describe a test which has 100% sensitivity and 100% specificity or it could simply signify a test whose performance characteristics are sufficiently high as to allow a definitive diagnosis without serious concern for an erroneous clinical conclusion. The results with MPS so far fail to indicate that it fulfils either meaning. Table 1 summarizes data from three published trials, all of which were carried out in high-risk pregnancies; the combined results yield a detection rate of 99.1% for a false-positive rate of 0.3%, although the data are still too few to allow exclusion of rates of 97% and 0.7%, respectively. Based on the tenuous assumption that the detection rate and false-positive rate are constant for all high-risk women, Table 2 presents the final risk when MPS is applied to women who are screen-positive according to a currently used screening protocol5. A negative MPS result substantially reduces the chance of an affected pregnancy, but there remains a residual risk. Especially for women with a relatively low prior risk, the possibility of a false-positive result is a significant and serious concern. As noted by Palomaki et al.4, invasive testing is recommended to confirm a positive result, i.e. while MPS is a highly sensitive screening test, it is not a diagnostic test. Improvements in the application of MPS are certainly likely (see below), but it needs to be recognized that a test based on cell-free fetal (cff) DNA derived from trophoblasts can only be expected to approach the diagnostic efficacy of direct chromosomal preparations of CVS, which is less than that possible by amniocentesis6. Confined placental mosaicism is well known and a relatively common confounder in CVS diagnoses; this could be even more of a problem for the MPS test because the cffDNA is derived from a non-viable cell population. The ability to detect fetal Down syndrome by MPS when there is a twin pregnancy and its application for mosaic cases also needs to be established. Finally, it should be noted that the MPS test will not distinguish between trisomy and unbalanced translocation; therefore, in the absence of a confirmatory invasive test, chromosomal analysis of both parents would usually be indicated to rule out segregation of a familial translocation. Down syndrome is the single most common clinically significant disorder associated with chromosomal abnormality present at birth, and has therefore been the main focus of prenatal screening and diagnosis. However, the established screening protocols currently in use also identify a broad range of other clinically significant cytogenetic abnormalities. This is achieved either through a specific protocol (e.g. reporting the risk of trisomies 18 and 13), or because some cytogenetic abnormalities are associated with marker patterns that are similar to those of trisomy 21, 13 or 18 (e.g. triploidy, XO and trisomy 16) and because many aneuploidies are associated with maternal age, which is a component of all reported risks. Among women who undergo invasive testing following the established prenatal screening protocols, Down syndrome constitutes only about half of the chromosomal abnormalities identified7. This estimate does not include many of the microdeletions and microduplications now detectable with FISH and microarray analyses. The use of MPS for Down syndrome only, in high-risk pregnancies, needs to be considered in the context of this broader range of fetal cytogenetic abnormalities. There are encouraging results suggesting that MPS can be developed for detection of trisomies 18 and 13, but these have not yet been the subject of rigorous clinical trials8. In theory, the approach should also be applicable for the identification of sex chromosomal aneuploidy and other major imbalances. However, in the short term, it appears unlikely that MPS can be developed to provide prenatal diagnosis of most microdeletion/duplication syndromes that are now detectable by FISH or microarrays. Invasive testing would also be indicated currently for women who are at risk for single gene disorders, although the non-invasive detection of a paternally inherited mutation is being developed. Therefore, the present published configuration of MPS for Down syndrome only, constitutes a considerable limitation in identifying the broad range of chromosomal abnormalities and genetic disorders currently identified through invasive testing. If such a limited configuration becomes an established paradigm, new methods will be needed to identify women for invasive testing who are specifically at high risk of chromosomal abnormalities other than Down syndrome. The current screening markers would be a component of this, together with detailed ultrasound findings. In many locations it is now common practice to provide women with limited information prior to prenatal screening and more extensive genetic counseling to high-risk women prior to any invasive testing. Because prenatal genetic counseling services are often integral components of the maternal–fetal obstetric departments where CVS and amniocentesis are performed, there is a concern that the replacement of invasive testing by an easily accessible blood test, such as MPS, could be associated with modified patient referral practices that might reduce or eliminate genetic counseling. The importance of comprehensive genetic counseling should not be underestimated; the provision of laboratory results that indicate a very high probability of abnormality, without prior explanation about the significance of the test, would be a seriously deficient medical practice9. In fact, the introduction of MPS would increase, not decrease, the need for genetic counseling. Women will now need to weigh the obvious advantage of potentially avoiding invasive testing against the limitations discussed above. Since a positive MPS result needs to be confirmed by an invasive test, women will need to consider the additional delay in obtaining a final, complete diagnostic result. This delay includes the period of time following a positive MPS test, which is likely to be particularly distressing because the risk for an affected pregnancy is high. Recognizing the complexities of the options now presented to high-risk women, the International Society for Prenatal Diagnosis has issued an advisory statement indicating the need for genetic counseling for women considering the MPS test10. Women's acceptance of test results based on MPS versus invasive testing is currently unknown. Of serious concern is the extent to which women with positive MPS results would choose to skip a confirmatory invasive test and elect immediately to terminate a pregnancy which could be unaffected. The advent of NIPD will be considered by many to open up the possibility of routine universal screening without regard to prior risk. It could be argued that this approach would increase considerably the Down syndrome detection rate compared with current screening protocols, whilst reducing the invasive testing rate. However, the studies of MPS published or presented so far have been in high-risk pregnancies; without direct evidence to the contrary, it is prudent to assume that test performance will be poorer in those at low risk. There are associations between placental insufficiency and various conventional screening markers, and screen-positive groups probably contain a relatively high proportion of women with abnormal placentation. The amount of circulating cffDNA is higher in these women11, 12. In low-risk women, the MPS test failure rate may therefore be substantially higher than the 1.4% seen in the high-risk trials (Table 1). Even if the performance of MPS is acceptable in low-risk women, cost may be a barrier to a universal policy. Palomaki et al.4 (in their supplementary material) provide an economic analysis that suggests that, even for high-risk women, the Down syndrome MPS test would add to prenatal healthcare costs if its price were similar to that of an invasive test. A full economic analysis for MPS would need to include the additional costs of genetic counseling and the expenses that are associated with undetected chromosomal abnormalities. Since the detection rate of current screening protocols, particularly those incorporating nasal bone and other specialist ultrasound markers, already exceeds 90–95%5, 13, 14, the marginal costs of any additional detection would be considerable. Healthcare costs vary considerably from country to country, but the pressure to constrain new expenses appears to be universal. In recent years, a number of first-trimester screening protocols for pre-eclampsia have been developed. Modeling predicts a very high detection rate for the more severe early-onset cases15 and meta-analysis indicates that low-dose aspirin halves the incidence of pre-eclampsia if initiated before 16 weeks' gestation16. Nicolaides17 has proposed a strategy of first-trimester triage for a wide range of adverse pregnancy outcomes, which he entitles 'a new pyramid of care'. His proposal is to use the results of tests and examinations carried out at the time of first-trimester aneuploidy screening to plan a sequence of further tests and visits tailored to the individual patient. MPS for Down syndrome could be incorporated in a contingent fashion, for those with intermediate risk following conventional screening, without compromising these new screening approaches. The role or timing of second-trimester ultrasound for the detection of additional fetal anatomic abnormalities and the use of the second-trimester maternal serum alpha-fetoprotein test will also need to be re-evaluated. Initially, access to MPS testing is likely to be limited. This test has been developed in the private sector, is protected by patents, and requires large investment to establish a testing facility. Concerns have been expressed about the proliferation of highly complex 'laboratory developed tests' in the United States and there have been calls for regulation18. Although, in theory, laboratories, including those in the private sector, should be capable of self-audit and adherence to defined protocols, it should be noted that some have stumbled badly in the past19. The possibility that MPS testing will be marketed directly to patients, bypassing healthcare providers, is another concern9. MPS technology is in its infancy and test improvements are a near certainty. Anticipated technical developments include more accurate sequencing, enrichment of cffDNA20, 21, selection of the most informative sequences for analysis, improved approaches for normalizing and defining euploidy and identifying other chromosomal abnormalities, and utilizing more sophisticated algorithms such as a Gaussian model (instead of the current fixed z cut-off) for assessing the risk for abnormality. These approaches should improve test efficacy, facilitate detection of smaller imbalances and reduce costs. Disorders with a Mendelian pattern of inheritance will also be diagnosable, for example, using a single nucleotide polymorphism (SNP) linkage approach22. Researchers continue to improve prenatal diagnosis using fetal cells (rather than cffDNA) isolated from the maternal circulation and this may also play a role in NIPD23. Meanwhile, advances in mutation discovery through techniques such as exome sequencing24 and improvements in carrier testing25 will also complement this new non-invasive approach to diagnosis of genetic disorders. As with all new developments in prenatal screening and diagnosis, MPS has the long-term potential both for improving public health and for expanding the reproductive choices available to women. However, there are considerable obstacles to its rapid implementation. Most importantly, test validation must be of the highest standards and the application of MPS must be part of protocols that are acceptable to healthcare providers and their patients. MPS can be judged a useful secondary screening test for Down syndrome in women found to be at high risk of this disorder by conventional screening. However, it is a poor secondary screening test when considering all forms of aneuploidy together and will lead to a considerable loss of detection of aneuploidies other than trisomy 21. Until trials have been published, it has no place in the prenatal diagnosis of low-risk women or even those of borderline risk based on conventional screening tests. P. Benn has no consulting agreements or other commercial interests in prenatal screening and diagnosis. H. Cuckle is a consultant to PerkinElmer Inc. and Aria Diagnostics Inc., a Director of Genome Ltd and Genome Screening Ltd, and holds International Patent number PCT/GB88/00 557 on 'The use of oestriol, progesterone, 16alpha-hydroxy-DHEAS and DHEAS in screening for Down's syndrome'. E. Pergament is a consultant to Perkin Elmer Inc., CellScape Inc. and DPR Equity Inc.