Scandinavian Journal of Clinical & Laboratory Investigation, 2014; 74(Suppl 244): 41–47
ORIGINAL ARTICLE
Screening for Down syndrome
KEVIN SPENCER
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
Prenatal Research Unit, Department of Clinical Biochemistry, Barking Havering & Redbridge University Hospitals NHS Trust, King George Hospital, Goodmayes, UK Abstract Screening for Down Syndrome was initially only related to maternal age and has successively developed by introducing biochemical markers and algorithms to estimate the risk for particularly trisomy 21 and 18. We now have a long experience of screening with four biochemical markers, alpha-fetoprotein, total hCG, unconjugated estriol and free β-hCG during the second trimester. Screening is now moving towards screening in the first trimester using a combination of ultrasound (Nuchal Translucency) and the maternal serum biochemical markers free β-hCG and Pregnancy Associated Plasma Protein-A (PAPP-A). This has become known as the combined test. Several maternal and pregnancy factors which can influence the concentrations of biochemical markers are discussed. The possibilities of screening for other aneuploidies in the first trimester and an outline of recent methods to improve overall screening performance are highlighted and the review will suggest some possible options for the future in which Cell Free DNA techniques may become part of an improved overall screening strategy. In conclusion it is emphasized that the time has come to invert the Pyramid of Antenatal Care to focus on the 11–13 week assessment. Key Words: Aneuploidy, first trimester, β-hCG, maternal serum, nuchal translucency, PAPP-A
Introduction
Occurrence
Screening is a service which detects the predisposition for a particular disease/condition or its early ‘treatable’ stages in people who are generally considered to be disease/condition free. A screening test is usually in itself not diagnostic, it defines a sub group of those tested who are at a higher risk of having the disease/ condition than the original population screened. This sub group needs further investigation with a diagnostic test, which is often more time consuming, expensive, and may be invasive and more risk prone than the screening test itself. The basic principles underpinning the requirements for a screening program are well established from the work of Wilson and Jungner [1]. The UK National Screening Committee has also extensively outlined criteria for appraisal of potential screening programs within the UK Health Service (http://www.screening.nhs.uk/criteria). In the context of screening for Down syndrome and other major aneuploidies, the primary aim is to provide specific information that allows couples to make informed reproductive decisions and is not focused on eradicating disability.
The natural frequency of chromosomal abnormalities at birth approximates to 6 per 1000 births amongst women without any form of antenatal screening. The autosomal aneuploidies are most frequent with Down syndrome (Trisomy 21) the most common of the group having a historical birth prevalence of 1 in 800, and Edward’s syndrome (Trisomy 18) at 1 in 6,500 and Patau’s syndrome (Trisomy 13) 1 in 12,500. The birth rates of the major autosomal trisomies increase with advancing maternal age. Thus the general prevalence of these conditions has increased over the past 25 years as a direct consequence of women postponing childbirth until later life, with the consequence that the expected life birth rate without screening is approaching 1 in 500. The actual prevalence of each trisomy at any one time during pregnancy varies due to the differing intrauterine lethality rates of the various conditions resulting in a significantly greater number of affected fetuses in early pregnancy compared with mid gestation or term. In addition to maternal age influencing the background preva-
Correspondence: Kevin Spencer, Prenatal Research Unit, Department of Clinical Biochemistry, Barking Havering & Redbridge University Hospitals NHS Trust, King George Hospital, Barley Lane, Goodmayes, IG3 8YB, UK. E-mail: [email protected] ISSN 0036-5513 print/ISSN 1502-7686 online © 2014 Informa Healthcare DOI: 10.3109/00365513.2014.936680
42
K. Spencer
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
lence there is also an increased risk of recurrence of around 0.75 % above the background maternal age risk in women who have delivered a fetus with an autosomal trisomy. Until around 1990 maternal age was the main screening modality for detecting pregnancies carrying a fetus with Down syndrome. If a mother was over the age of 35 she would have been offered amniocentesis and karyotyping (counting and matching chromosome pairs) of the fetal cells isolated and cultured from the amniotic fluid. In the 1980s this would have identified about 5 % of pregnant women who would need to consider amniocentesis and potentially would have identified about 30 % of cases with Down syndrome.
Developments in second trimester maternal serum screening In the mid to late 1980s a series of observations led to the proposed use of second trimester maternal serum biochemical markers in a screening program for Down syndrome which invariably built upon existing programs screening for Neural Tube Defects. In 1984 Merkatz et al. [2] initially observed low levels of α-fetoprotein (alpha-fetoprotein, AFP) in a group of pregnancies with Trisomy 21 or Trisomy 18. This was confirmed in a larger series by Cuckle et al. [3] who proposed a screening program based on specific AFP cut off ’s. In an attempt to improve screening performance Cuckle et al. [4] proposed a method of screening which combined a priori age risk with the likelihood ratio of the AFP result having come from a population of pregnancies in which the fetus had Down syndrome rather than from a population with a normal fetus. A risk cut off was then chosen to identify those women with sufficiently high enough risk to warrant invasive diagnostic testing. In 1987 Bogart et al. [5] observed increased maternal serum levels of total β-hCG in pregnancies with trisomy 21 and low levels in those with trisomy 18. One year later Canick et al. [6] observed reduced levels of unconjugated estriol in pregnancies with trisomy 21. The next logical step in screening was proposed by Wald et al. [7] in using the methodology outlined by Cuckle et al. [4] the three biochemical markers were combined together using their Gaussian distributions from affected and unaffected pregnancies to achieve a combined likelihood ratio which was then used to modify the a priori age risk; a procedure based on the well-known statistical procedure call Bayes Theorem.This multimarker assessment scheme became known as the Triple Test and heralded the introduction of biochemical screening for Down syndrome in the second trimester. In 1990 and 1991 two groups [8,9] independently reported that free β-hCG levels were increased in the second trimester of pregnancies with Down syndrome and that as a marker it was more discriminating than
Intact or Total hCG. This led to considerable discussion in the literature as to the merits of free versus intact hCG but eventually a consensus was reached agreeing that this marker was a better discriminator. Issues around patents limited its use in North America. Between 1992 and 1996 various publications initially using nonspecific methods for Inhibin [10] showed this marker and specifically the form Inhibin-A [11] was also increased in pregnancies with trisomy 21and with the development of more robust assays for this marker [12], it was gradually added to screening programs resulting in what now is known as the quadruple test. The information gained from each marker is not totally independent for example there is a significant correlation between hCG and Inhibin A of the order of r ⫽ 0.5; this may be because the two markers are primarily of placental origin. Such inter correlation can be taken into account in the algorithms used in screening. Other correlation coefficients are of the order of 0.15 or less. AFP is primarily produced by the fetal liver and arrives in the maternal circulation after fetal urination into the amniotic fluid and transplacental transfer. Unconjugated oestriol is derived from steroidogenesis in the fetal adrenal followed by modification in the fetal liver by conversion of dehydroepiandrosterone to its 16 hydroxlated form and transplacental conversion and transfer into the maternal circulation.
MoM One special feature of the measurement of most fetoplacental and related products is the noticeable change in levels in relation to the stage of pregnancy; some decreasing and others increasing. For this reason it is usually not possible to quote a single reference interval for a given analyte; instead, a separate reference interval has to be used for each week and sometimes day of pregnancy. This has two important implications. First, large numbers of samples are required to establish a reference interval for uncomplicated pregnancies. Second, it is almost impossible for the clinician or analyst to recall the interval for each week without some sort of prompt. Furthermore many of the quantities that are measured during pregnancy have no defined primary calibrator by which to standardize the assay method and furthermore the very nature of immunoassay (the usual method of choice for the measurement of most placental and pregnancy measurands) can introduce methodological biases. To remove gestational age variation effects and to allow some degree of standardization from center to center and between different methods, biochemical parameters and some biophysical (Ultrasound and Blood Pressure) parameters have been expressed as Multiples of the normal Median (MoM). MoM is the ratio between
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
Screening for Down syndrome the measured value and the median of normal pregnancies of the same gestational age. Thus if a MoM of 1.00 is normal, a value of 2.00 is elevated and a value of 0.50 is reduced. Thus, when the action limit for alpha-fetoprotein (AFP) in screening for Neural Tube Defects (NTDs) is stated as being ‘2.5 times the median’ or ‘2.5 MoM’ this will be the same regardless of the week of pregnancy and it will be universal from center to center and assay to assay. The median will, of course, change but the action limit expressed as a MoM will not. The relationship between analyte concentrations and gestational age places a heavy demand on the accuracy of gestational dating. A given marker value could be normal for 16 weeks but above the reference interval 1 week earlier. In clinical practice, dating errors of up to 4 weeks or more are relatively common. Dating is usually based on the first day of the last menstrual period (LMP). This can be confirmed by ultrasound measurements of the fetus either by crown-rump length (CRL) before 14 weeks of pregnancy; biparietal diameter (BPD) or head circumference (HC) thereafter, although it is recognized that the latter is more accurate. If the LMP is in doubt, or there is a gross discrepancy with ultrasound then the ultrasound date is accepted as the ‘gold standard’. When biochemical markers are being used in the context of aneuploidy screening it is usually common to base median calculations on the dating by gestational age or indeed on the CRL measurement itself. The latter is usually most preferred because there are a number of different ultrasound charts in use which would result in different gestational days being calculated for the same fixed CRL. The vast majority of markers are log-Gaussian distributed. Second trimester screening performance In the second trimester, screening for aneuploidies is primarily for trisomy 21 and to a lesser extent for trisomy 18 since the pattern of biochemical markers for trisomy 13 and other aneuploidies are similar to the normal pattern with perhaps the exception of triploidy (3 copies of each chromosome). The expected median concentrations of the major markers in pregnancies with trisomy 21 and 18 are shown in Table I.
Table I. Marker levels in second trimester cases of trisomy 18 and 21. Serum marker AFP Total hCG Unconjugate estriol Free β-hCG Inhibin A
T18 median MoM
T21 median MoM
0.65 0.32 0.42 0.33 0.87
0.75 2.06 0.72 2.20 1.92
43
Screening performance using a combination of the different markers shows that for a two marker protocol using AFP and Free β-hCG a detection rate of around 64 % can be achieved for a 5 % false positive rate. If a three marker protocol were used – including U-Estriol (U-E3) then this would increase to around 67 % and if a four marker protocol were used – including Inhibin-A then this would increase to around 72 %. In more than 20 published prospective intervention studies this modelled screening performance has been confirmed in routine practice [13]. The four marker protocol commonly referred to as the ‘Quad Test’ is generally the standard in use when screening is performed in the second trimester. However when comparing different reported detection rates and false positive rates one needs to bear in mind that the screened population age distributions may not all be the same and there is a need to use age standardized detection rates and false positive rates when comparing screening performance [14]. Detection rates for Trisomy 18 tend on the whole to be considerably lower at around 60 %. At the 18–22 week ultrasound anomaly scan the presence of a wide variety of ‘soft markers’ has also been proposed as a screening method for trisomy 21. The vast majority of these soft makers have been poorly evaluated [15] and it is only recently that appropriate likelihood ratios have been associated with the most useful of the markers such as ventriculomegaly, nuchal fold thickness, hypoplastic nasal bone and aberrant subclavian artery [16]. First trimester screening The past decade and a half has seen a considerable focus on moving screening earlier into the first trimester. Earlier screening is seen as providing women with an earlier reassurance and if termination of pregnancy is desired, this can be completed by a safer procedure and before fetal movements are usually evident. The fact that some aneuploid pregnancies detected in the first trimester will be spontaneously lost before term is not a valid argument against early screening. For these women it is important information to know, which can shape future reproductive decision making. Screening markers in the first trimester consist of the ultrasound marker Nuchal Translucency (NT) and the maternal serum biochemical markers free β-hCG and Pregnancy Associated Plasma Protein-A (PAPP-A) and form what has become known as the combined test. This test has now become the National Screening Standard in many countries. Although many biochemical markers have been investigated including those used in the second trimester only free β-hCG is of value across both trimesters. Both PAPP-A and free β-hCG are placental-derived markers. NT is the term used for an ultrasound measurement of the thickness of an echogenic area of fluid
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
44
K. Spencer
that exists in all fetuses between the fetal skin and the soft tissue overlying the cervical spine seen in a mid-sagittal view of the face of the fetus. NT is the single most discriminatory marker of trisomy 21 in the first trimester. The measurement of NT, unlike the measurement of serum markers, depends on operator skill especially when one considers the small size of the measurement of 1–3 mm being normal. The NT measurement may be made at the same time as the early pregnancy dating scan, between 11 weeks 0 days and 13 weeks 6 days. Most experienced sonographers can obtain the measurement within a few minutes. The crown-rump length (CRL) is measured during the same examination. NT increases with advancing pregnancy, and the CRL (a precise measure of gestational age) is used to standardize NT measurements for gestational age, either converting NT to MoM values (observed NT measurements divided by the median NT measurement for the corresponding CRL) or to delta values (observed NT measurements minus the median NT measurement for the corresponding CRL). The Fetal Medicine Foundation (www. fetalmedicine.com/fmf/) and Nuchal Translucency Quality Review Program (www.ntqr.org) are two organizations that provide ultrasonographer training, certification, and ongoing audit. Continuing audit of individual sonographers is required by many screening programs and in the UK this is carried out along with the monitoring of biochemical markers by the Downs Syndrome Screening Quality Assurance Support Service (DQASS) at a national level (http://fetalanomaly.screening.nhs.uk/dqass). Bias and inaccurate estimation of both NT and CRL have been shown to adversely affect detection rates and false positive rates [17]. Not only is increased NT a good marker for trisomy 21 it is also a very good marker for trisomy 13 and trisomy 18 and can be supra elevated in hydropic Turner’s syndrome. Increased NT is also associated with a number of adverse pregnancy outcomes and an increased risk for fetal cardiac defects. Combining first trimester maternal serum biochemistry and NT measurement is an effective screening procedure because the two modalities do not appear to be correlated [18]. Just as in the second trimester the likelihood ratio for each marker MoM can be derived from the overlapping Gaussian distributions from the affected and unaffected populations and this can be used to modify the a priori risk based on maternal age or history. In some algorithms used by the Fetal Medicine Foundation a mixture model approach is used for NT which is thought to give more accurate risks [19]. Retrospective and now many prospective studies have shown that 90 % of cases of T21 can be detected for a false positive rate of 5 % or less [13] and studies have also shown that approximately 90 % of all other major chromosomal anomalies can be detected for an
Table II. Changes in marker patterns in the first trimester of aneuploid pregnancies. Aneuploidy
NT
CRL
Fetal heart rate
Free β-hCG
PAPP-A
Trisomy 21 Trisomy 18 Trisomy 13 Turner’s Triploidy I Triploidy II
↑2.5 ↑3.5 ↑2.5 ↑7.0 ↑2.5 ↔
↔ ↓ ↔ ↔ ↔ ↓
↑ ↓ ↑ ↑ ↓ ↓
↑2.2 ↓0.3 ↓0.5 ↔ ↑8.0 ↓0.2
↓0.5 ↓0.5 ↓0.3 ↓0.5 ↓0.8 ↓0.2
additional 1 % false positive rate [20]. Table II summarizes the basic pattern of changes in the various markers associated with aneuploidy. Attempts to improve the accuracy of individualized risks in both the second and first trimester have been carried out by correcting many of the biochemical markers for pregnancy variables that influence the levels of these markers. Some of the variables are outlined in Table III. If these are not corrected for then considerable errors in individualized risk can be introduced [21].
National screening policy In many countries with state funded health care systems national policies on screening for Down syndrome have been developed. In the UK the National Institute of Clinical Excellence (NICE) have issued guidelines (http://www.nice.org.uk/nice media/live/11947/40115/40115.pdf http://www.nice. org.uk/nicemedia/live/11947/40145/40145.pdf) based on the work of the UKNSC and its Fetal Anomaly Screening program. The primary screening program across the whole of England is that of the combined test and the Model of Best Practice 2011– 2014 http://fetalanomaly.screening.nhs.uk/getdata. php?id ⫽ 11393 outlines the aim of the UK program in focusing on reducing the false positive rate whilst keeping the detection rate at a high level. The aspiration of this program is to achieve a Detection Rate of 90 % for a 2 % False/Screen positive rate by 2014. The success of this program in reducing the screen positive rate and hence the invasive testing rate has been quite remarkable [22] and similar results have also been achieved in the Danish program. However such success really requires an extensive program of training, auditing and quality monitoring of all aspects of the program.
Screening in twin pregnancies Screening in twin pregnancies is still considered by some to be problematical. Whilst the NT measure can allow for a fetus specific risk, the maternal serum biochemistry represents an average of the two fetuses and is therefore pregnancy specific. With maternal
Screening for Down syndrome
45
Table III. Maternal and pregnancy factors influencing biochemical marker concentrations. Factor Gestational age Maternal weight Multiple pregnancy IDDM Fetal sex Assisted conception Ethnicity
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
Smoking Gravidity/parity Previous pregnancy
First trimester
Second trimester
PAPP-A increased, free β-hCG decreased after 9 weeks All decreased with increasing weight Twice as high in twins – chorionicity effect PAPP-A decreased Free β-hCG and PAPP-A increased with a female fetus Free β-hCG increased PAPP-A decreased Afro-Caribbean PAPP-A 50 % higher Free β-hCG higher by 10 % PAPP-A increased Both increased as pregnancy number increases 2 to 3 times more likely to be high risk if previous was high risk
serum biochemistry it is argued that in the presence of one abnormal fetus and one normal fetus the biochemistry result from the abnormal fetus is clouded by the normal contribution from the normal fetus. Provided appropriate correction factors are used for the biochemical markers – which are gestation specific and chorionicity specific then detection rates approaching those in singleton pregnancies can be achieved in the first trimester of twins discordant for Down syndrome [23]. A more detailed discussion of screening in twin pregnancies can be found in the webcast http://web8.perkinelmer.com/LP ⫽ 7012?url ⫽ 33113641 or in Spencer 2005 [24]. One other issue arises with pregnancies that start as twins is the situation of a vanishing twin (i.e. when one of the twins has died in utero) where some of the maternal serum biochemistry may still be being contributed by the demised twin. Certainly for markers that have a long clearance half life – like PAPP-A, this can result in elevated concentrations even 4 weeks after demise [25]. In this situation it may be advisable to use NT alone, although there are instances when it may be appropriate to include biochemistry data [26].
Alternative screening strategies Alternative screening strategies have been described which are more complex than the simple Quad test or the Combined Test. One such test is the Integrate Test which attempts to combine the tests used in the first trimester with those used in the second trimester. Thus PAPP-A and NT are measured at 11–13 weeks, the patient is not given a risk assessment but then returns at 16 weeks for further biochemical tests which include AFP, Free β-hCG, U-E3 and Inhibin-A. Only when all the data are together is a risk produced. The theoretical benefits of this approach are a perceived increase in detection rate or lowering of the false positive rate based on data from two trials SURUSS [27] and FASTER [28]
AFP, uE3 increased, hCG decreased, inhibin-A little change All decreased with increasing weight, U-E3 least affected Twice as high in twins AFP decreased, U-E3 and hCG increased HCG increased, AFP decreased with female fetus HCG increased, U-E3 decreased AFP, hCG higher in Asian an Afro-Caribbean, inhibin-A lower in Afro-Caribbean HCG, U-E3 decreased, AFP increased and Inhibin-A increased by 60 % HCG decreased 3 to 5 times more likely to be high risk if previous was high risk
however the practicalities of this approach make screening more complex and potentially more costly (http://fetalanomaly.screening.nhs.uk/getdata. php?id ⫽ 11393). Also many have voiced serious ethical and moral issues with respect to withholding information at the end of the first trimester part of the test. In the UK this is not a nationally recommended screening test but is popular in the USA.
Timing Another proposal to improve performance overall has been as a result of the observation that in T21 the specificity of the individual markers change across the first and second trimester gestational windows [29,30]. Thus PAPP-A has better clinical discrimination between 9 and 11 weeks than between 11 and 13 whereas for free β-hCG this is reversed. Since NT is measured optimally at 11–13 weeks and the observation of other ultrasound features is best performed around 12 weeks it has been suggested that further improvements could be made by taking a blood sample at 9–10 weeks and using this to measure PAPP-A and then at 12 weeks collect a second sample at the time of NT for measuring free β-hCG [31]. Such protocols will allow detection rates of 90 % to be achieved with a false positive rate of less than 2 % – which is the aspirational detection rate for the UK National program for 2011. It is also becoming apparent that the various correction factors used in screening are very gestational age-dependent and require gestational-specific corrections as for twins [25], smoking [32] and ethnicity [33].
Contingent testing Other more complex screening strategies that have been proposed are based around setting risk cut offs so that those with very high risks (e.g. 1:50) are offered
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
46
K. Spencer
Figure 1. Inverting the Pyramid of Antenatal Care to focus on the 11–13 week assessment. (modified from Nicolaides [37]).
CVS and those with a very low risk (e.g. 1:1000) are reassured and those in the middle are offered further testing. This further contingent testing [34] can be to refer those women to a specialist fetal medicine/ Ultrasound unit for the measurement of nasal bone, facial angle, tricuspid regurgitation and Ductus venosus in the first trimester – this information is then combined with the earlier information from combined testing and a revised single risk calculated [35]. All of these latter markers require skills in measurement and in the case of the latter two high energy colour doppler. They are not advised as routine screening markers and are best left to skilled tertiary centers. A second alternative would be to offer the intermediate group second trimester biochemical screening with the Quad test and to combine all of the test elements together in a single risk. The modifications add little to the overall detection rate but can mean the false positive rate falls well below 2 %. More recently it has been suggested that cell free fetal DNA measured in the maternal circulation may also be a contender for use in a Contingent approach [36], harnessing the improved screening potential of this marker to dramatically reduce the screen positive rate yet only performing this expensive test on around 12 % of the pregnant population. In its current form cell-free fetal DNA testing is not diagnostic and is unlikely to replace CVS/Amniocentesis and QFPCR/karyotyping in the near future.
Inverting the pyramid of antenatal care Nicolaides [37] has proposed recently that ‘It is likely that the new challenge for improvement of pregnancy
outcome will be met by inverting the pyramid of prenatal care to introduce on a large scale and in a systematic fashion a new model of prenatal care which will be based on the results of a comprehensive assessment at 11 to 13 weeks’. This is because research over the past decade has shown that by using ultrasound, biophysical and biochemical markers and previous history it is possible to screen pregnancies at this stage for the major pregnancy complications which include aneuploidies, pre-eclampsia, babies small for gestational age, macrosomia, gestational diabetes, pre-term delivery, miscarriage and stillbirth and fetal structural anomalies (Figure 1). By the addition of placental growth factor (PlGF) and possibly AFP not only is it possible to improve the detection of aneuploidy by 3 % as a result of the lowered levels of PlGF seen with this condition [38] its use in conjunction with PAPP-A, mean arterial pressure and uterine artery Doppler velocimetry is able to identify more than 90 % of cases with early pre-eclampsia with a false positive rate of 5 % [39] – thus paving the way for potential therapeutic intervention with low dose aspirin [40]. It is probably this change to the way that antenatal care is delivered in the UK that will dictate how new technologies like cell-free fetal DNA are utilized in early pregnancy. Declaration of interest: The author reports no conflicts of interest. The author alone is responsible for the content and writing of the paper. References [1] Wilson JMG, Jungner G. Principles and practice of screening for disease. Public Health Paper Number 34. Geneva: WHO, 1968. [2] Merkatz IR, Nitowsky HM, Macri JN, et al. An association between low maternal serum alpha-fetoprotein and fetal chromosomal abnormalities. Am J Obstet Gynecol 1984; 148:886–94. [3] Cuckle HS, Wald NJ, Lindenbaum RH. Maternal serum alpha-fetoprotein measurement: a screening test for Down syndrome. Lancet 1984;1(8383):926–9. [4] Cuckle HS, Wald NJ, Thomson SG. Estimating a women’s risk of having a pregnancy associated with Down’s syndrome using her age and serum alphafetoprotein level. Br J Obstet Gynaecol 1987;94:387–402. [5] Bogart MH, Pandian MR, Jones OW. Abnormal maternal serum chorionic gonadotropin levels in pregnancies with fetal chromosome abnormalities. Prenat Diagn 1987;7: 623–30. [6] Canick JA, Knight GJ, Palomaki GE, et al. Low second trimester maternal serum unconjugated oestriol in pregnancies with Down’s syndrome. Br J Obstet Gynaecol 1988;95: 330–3. [7] Wald NJ, Cuckle HS, Densem JW, et al. Maternal serum screening for Down’s syndrome in early pregnancy. BMJ 1988;297:883–7. [8] Macri JN, Kasturi RV, Krantz DA, et al. Maternal serum Down syndrome screening: free beta protein is a more effective marker than human chorionic gonadotropin. Am J Obstet Gynecol 1990;163:1248–53.
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
Screening for Down syndrome [9] Spencer K. Evaluation of an assay of the free beta-subunit of choriogonadotropin and its potential value in screening for Down’s syndrome. Clin Chem 1991:37:809–14. [10] Spencer K, Wood PJ, Anthony FW. Elevated levels of maternal serum inhibin immunoreactivity in second trimester pregnancies affected by Down’s syndrome. Ann Clin Biochem1993;30:219–20. [11] Aitken DA, Wallace EM, Crossley JA, et al. Dimeric inhibin A as a marker for Down’s syndrome in early pregnancy. N Engl J Med 1996;334:1231–6. [12] Wallace EM, Crossley JA, Ritoe SC, et al. Evolution of an inhibin A ELISA method: implications for Down’s syndrome screening. Ann Clin Biochem 1998;35:656–64. [13] Spencer K. Aneuploidy screening in the first trimester. Am J Med Genet Part C Semin Med Genet 2007;145C18–32. [14] Cuckle H, Aitken D, Goodburn S, et al. Age-standardisation when target setting and auditing performance of Down syndrome screening programmes. Prenat Diagn 2004;24:851–6. [15] Lau TK, Evans MI. Second-trimester sonographic soft markers: what can we learn from experience of first trimester nuchal translucency screening? Ultrasound Obstet Gynecol 2008;32:123–5. [16] Agathokleous M, Chaveeva P, Poon LCY, et al. Metaanalysis of second trimester markers for trisomy 21. Ultrasound Obstet Gynecol 2013;41:247–61. [17] Kagan KO, Wright D, Etchegaray A, et al. Effect of deviation of nuchal translucency measurements on the performance of screening for trisomy 21. Ultrasound Obstet Gynecol 2009;33:657–64. [18] Spencer K, Souter V, Tul N, et al. A screening program for trisomy 21 at 10–14 weeks using fetal nuchal translucency, maternal serum free b-human chorionic gonadotropin and pregnancy-associated plasma protein-A. Ultrasound Obstet Gynecol 1999;13:231–7. [19] Wright D, Kagan KO, Molina FS, et al. A mixture model of nuchal translucency thickness in screening for chromosomal defects. Ultrasound Obstet Gynecol 2008;31:376–83. [20] Kagan KO, Wright D, Valencia C, et al. Screening for trisomies 21, 18 and 13 by maternal age, fetal nuchal translucency, fetal heart rate, free β-hCG and pregnancy-associated plasma protein-A. Hum Reprod 2008;23:1968–75. [21] Kagan KO, Wright D, Spencer K, et al. First-trimester screening for trisomy 21 by free beta-human chorionic gonadotropin and pregnancy-associated plasma protein-A: impact of maternal and pregnancy characteristics. Ultrasound Obstet Gynecol 2008;31:493–502. [22] Morgan S, Delbarre A, Ward P. Impact of introducing a national policy for prenatal Down syndrome screening on the diagnostic invasive procedure rate in England. Ultrasound Obstet Gynecol 2013;41:526–9. [23] Madsen HN, Ball S, Wright D, et al. A re-assessment of biochemical marker distributions in T21 affected and unaffected twin pregnancies in the first trimester. Ultrasound Obstet Gynecol 2011;37:38–47. [24] Spencer K. Non-invasive screening tests. In Blickstein I, Keith L (eds.) Multiple pregnancy: epidemiology, gestation and perinatal outcome. Oxford: Taylor & Francis. 2005. pp. 368–4.
47
[25] Spencer K, Staboulidou I, Nicolaides KH. First trimester aneuploidy screening in the presence of a vanishing twin: implications for maternal serum markers. Prenat Diagn 2010;30:235–40. [26] Sankaran S, Rozette C, Dean J, et al. Screening in the presence of a vanished twin: nuchal translucency or combined screening test? Prenat Diagn 2011;31;600–01. [27] Wald NJ, Rodeck C, Hackshaw AW, et al. First and second trimester antenatal screening for Down’s syndrome: the results of the Serum, Urine and Ultrasound Screening Study (SURUSS). Health Technol Assess 2003;7:1–77. [28] Malone FD, Canick JA, Ball RH, et al. First-trimester or second-trimester screening, or both, for Down’s syndrome. N Engl J Med 2005;353:2001–11. [29] Spencer K, Crossley JA, Aitken DA, et al. Temporal changes in maternal serum biochemical markers of Trisomy 21 across the first and second trimester of pregnancy. Ann Clin Biochem 2004;39:567–76. [30] Spencer K, Crossley JA, Aitken DA, et al. The effect of temporal variation in biochemical markers of trisomy 21 across the first and second trimester of pregnancy on the estimation of individual patient-specific risks and detection rates for Down’s syndrome. Ann Clin Biochem 2003;40:219–31. [31] Wright D, Spencer K, Kagan K, et al. First-trimester combined screening for trisomy 21 at 8–13 weeks. Ultrasound Obstet Gynecol 2010;30:404–11. [32] Spencer K, Cowans NJ. Correction of first trimester biochemical aneuploidy screening markers for smoking status: influence of gestational age, maternal ethnicity and cigarette dosage. Prenat Diagn 2013;33:116–23. [33] Cowans NJ, Spencer K. Effect of gestational age on first trimester maternal serum prenatal screening correction factors for ethnicity and IVF conception. Prenat Diagn 2013;33:56–60. [34] Wright D, Bradbury I, Benn P, et al. Contingent screening for Down syndrome is an efficient alternative to non-disclosure sequential screening. Prenat Diagn 2004;24:762–6. [35] Nicolaides KH, Spencer K, Avgidou K, et al. Multicentre study of first trimester screening for trisomy 21 in 75,821 pregnancies: results and estimation of the potential impact of individual risk orientated two stage first trimester screening. Ultrasound Obstet Gynecol 2005;25:221–6. [36] Chitty LS, Hill M, White H, et al. Noninvasive prenatal testing for aneuploidy – ready for prime time. Am J Obstet Gynecol 2012;206:269–75. [37] Nicolaides KH. Turning the pyramid of prenatal care. Fetal Diagn Ther 2011;29:183–96. [38] Pandya P, Wright D, Syngelaki A, et al. Maternal serum placental growth factor in prospective screening for aneuploidies at 8–13 weeks’ gestation. Fetal Diagn Ther 2012; 31:87–93. [39] Akolekar R, Syngelaki A, Poon L, et al. Competing risks model in early screening for preeclampsia by biophysical and biochemical markers. Fetal Diagn Ther 2013;33:8–15. [40] Roberge S, Villa P, Nicolaides K, et al. Early administration of low-dose aspirin for the prevention of preterm and term preeclampsia: a systematic review and meta-analysis. Fetal Diagn Ther 2012;31:141–6.
ORIGINAL ARTICLE
Screening for Down syndrome
KEVIN SPENCER
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
Prenatal Research Unit, Department of Clinical Biochemistry, Barking Havering & Redbridge University Hospitals NHS Trust, King George Hospital, Goodmayes, UK Abstract Screening for Down Syndrome was initially only related to maternal age and has successively developed by introducing biochemical markers and algorithms to estimate the risk for particularly trisomy 21 and 18. We now have a long experience of screening with four biochemical markers, alpha-fetoprotein, total hCG, unconjugated estriol and free β-hCG during the second trimester. Screening is now moving towards screening in the first trimester using a combination of ultrasound (Nuchal Translucency) and the maternal serum biochemical markers free β-hCG and Pregnancy Associated Plasma Protein-A (PAPP-A). This has become known as the combined test. Several maternal and pregnancy factors which can influence the concentrations of biochemical markers are discussed. The possibilities of screening for other aneuploidies in the first trimester and an outline of recent methods to improve overall screening performance are highlighted and the review will suggest some possible options for the future in which Cell Free DNA techniques may become part of an improved overall screening strategy. In conclusion it is emphasized that the time has come to invert the Pyramid of Antenatal Care to focus on the 11–13 week assessment. Key Words: Aneuploidy, first trimester, β-hCG, maternal serum, nuchal translucency, PAPP-A
Introduction
Occurrence
Screening is a service which detects the predisposition for a particular disease/condition or its early ‘treatable’ stages in people who are generally considered to be disease/condition free. A screening test is usually in itself not diagnostic, it defines a sub group of those tested who are at a higher risk of having the disease/ condition than the original population screened. This sub group needs further investigation with a diagnostic test, which is often more time consuming, expensive, and may be invasive and more risk prone than the screening test itself. The basic principles underpinning the requirements for a screening program are well established from the work of Wilson and Jungner [1]. The UK National Screening Committee has also extensively outlined criteria for appraisal of potential screening programs within the UK Health Service (http://www.screening.nhs.uk/criteria). In the context of screening for Down syndrome and other major aneuploidies, the primary aim is to provide specific information that allows couples to make informed reproductive decisions and is not focused on eradicating disability.
The natural frequency of chromosomal abnormalities at birth approximates to 6 per 1000 births amongst women without any form of antenatal screening. The autosomal aneuploidies are most frequent with Down syndrome (Trisomy 21) the most common of the group having a historical birth prevalence of 1 in 800, and Edward’s syndrome (Trisomy 18) at 1 in 6,500 and Patau’s syndrome (Trisomy 13) 1 in 12,500. The birth rates of the major autosomal trisomies increase with advancing maternal age. Thus the general prevalence of these conditions has increased over the past 25 years as a direct consequence of women postponing childbirth until later life, with the consequence that the expected life birth rate without screening is approaching 1 in 500. The actual prevalence of each trisomy at any one time during pregnancy varies due to the differing intrauterine lethality rates of the various conditions resulting in a significantly greater number of affected fetuses in early pregnancy compared with mid gestation or term. In addition to maternal age influencing the background preva-
Correspondence: Kevin Spencer, Prenatal Research Unit, Department of Clinical Biochemistry, Barking Havering & Redbridge University Hospitals NHS Trust, King George Hospital, Barley Lane, Goodmayes, IG3 8YB, UK. E-mail: [email protected] ISSN 0036-5513 print/ISSN 1502-7686 online © 2014 Informa Healthcare DOI: 10.3109/00365513.2014.936680
42
K. Spencer
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
lence there is also an increased risk of recurrence of around 0.75 % above the background maternal age risk in women who have delivered a fetus with an autosomal trisomy. Until around 1990 maternal age was the main screening modality for detecting pregnancies carrying a fetus with Down syndrome. If a mother was over the age of 35 she would have been offered amniocentesis and karyotyping (counting and matching chromosome pairs) of the fetal cells isolated and cultured from the amniotic fluid. In the 1980s this would have identified about 5 % of pregnant women who would need to consider amniocentesis and potentially would have identified about 30 % of cases with Down syndrome.
Developments in second trimester maternal serum screening In the mid to late 1980s a series of observations led to the proposed use of second trimester maternal serum biochemical markers in a screening program for Down syndrome which invariably built upon existing programs screening for Neural Tube Defects. In 1984 Merkatz et al. [2] initially observed low levels of α-fetoprotein (alpha-fetoprotein, AFP) in a group of pregnancies with Trisomy 21 or Trisomy 18. This was confirmed in a larger series by Cuckle et al. [3] who proposed a screening program based on specific AFP cut off ’s. In an attempt to improve screening performance Cuckle et al. [4] proposed a method of screening which combined a priori age risk with the likelihood ratio of the AFP result having come from a population of pregnancies in which the fetus had Down syndrome rather than from a population with a normal fetus. A risk cut off was then chosen to identify those women with sufficiently high enough risk to warrant invasive diagnostic testing. In 1987 Bogart et al. [5] observed increased maternal serum levels of total β-hCG in pregnancies with trisomy 21 and low levels in those with trisomy 18. One year later Canick et al. [6] observed reduced levels of unconjugated estriol in pregnancies with trisomy 21. The next logical step in screening was proposed by Wald et al. [7] in using the methodology outlined by Cuckle et al. [4] the three biochemical markers were combined together using their Gaussian distributions from affected and unaffected pregnancies to achieve a combined likelihood ratio which was then used to modify the a priori age risk; a procedure based on the well-known statistical procedure call Bayes Theorem.This multimarker assessment scheme became known as the Triple Test and heralded the introduction of biochemical screening for Down syndrome in the second trimester. In 1990 and 1991 two groups [8,9] independently reported that free β-hCG levels were increased in the second trimester of pregnancies with Down syndrome and that as a marker it was more discriminating than
Intact or Total hCG. This led to considerable discussion in the literature as to the merits of free versus intact hCG but eventually a consensus was reached agreeing that this marker was a better discriminator. Issues around patents limited its use in North America. Between 1992 and 1996 various publications initially using nonspecific methods for Inhibin [10] showed this marker and specifically the form Inhibin-A [11] was also increased in pregnancies with trisomy 21and with the development of more robust assays for this marker [12], it was gradually added to screening programs resulting in what now is known as the quadruple test. The information gained from each marker is not totally independent for example there is a significant correlation between hCG and Inhibin A of the order of r ⫽ 0.5; this may be because the two markers are primarily of placental origin. Such inter correlation can be taken into account in the algorithms used in screening. Other correlation coefficients are of the order of 0.15 or less. AFP is primarily produced by the fetal liver and arrives in the maternal circulation after fetal urination into the amniotic fluid and transplacental transfer. Unconjugated oestriol is derived from steroidogenesis in the fetal adrenal followed by modification in the fetal liver by conversion of dehydroepiandrosterone to its 16 hydroxlated form and transplacental conversion and transfer into the maternal circulation.
MoM One special feature of the measurement of most fetoplacental and related products is the noticeable change in levels in relation to the stage of pregnancy; some decreasing and others increasing. For this reason it is usually not possible to quote a single reference interval for a given analyte; instead, a separate reference interval has to be used for each week and sometimes day of pregnancy. This has two important implications. First, large numbers of samples are required to establish a reference interval for uncomplicated pregnancies. Second, it is almost impossible for the clinician or analyst to recall the interval for each week without some sort of prompt. Furthermore many of the quantities that are measured during pregnancy have no defined primary calibrator by which to standardize the assay method and furthermore the very nature of immunoassay (the usual method of choice for the measurement of most placental and pregnancy measurands) can introduce methodological biases. To remove gestational age variation effects and to allow some degree of standardization from center to center and between different methods, biochemical parameters and some biophysical (Ultrasound and Blood Pressure) parameters have been expressed as Multiples of the normal Median (MoM). MoM is the ratio between
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
Screening for Down syndrome the measured value and the median of normal pregnancies of the same gestational age. Thus if a MoM of 1.00 is normal, a value of 2.00 is elevated and a value of 0.50 is reduced. Thus, when the action limit for alpha-fetoprotein (AFP) in screening for Neural Tube Defects (NTDs) is stated as being ‘2.5 times the median’ or ‘2.5 MoM’ this will be the same regardless of the week of pregnancy and it will be universal from center to center and assay to assay. The median will, of course, change but the action limit expressed as a MoM will not. The relationship between analyte concentrations and gestational age places a heavy demand on the accuracy of gestational dating. A given marker value could be normal for 16 weeks but above the reference interval 1 week earlier. In clinical practice, dating errors of up to 4 weeks or more are relatively common. Dating is usually based on the first day of the last menstrual period (LMP). This can be confirmed by ultrasound measurements of the fetus either by crown-rump length (CRL) before 14 weeks of pregnancy; biparietal diameter (BPD) or head circumference (HC) thereafter, although it is recognized that the latter is more accurate. If the LMP is in doubt, or there is a gross discrepancy with ultrasound then the ultrasound date is accepted as the ‘gold standard’. When biochemical markers are being used in the context of aneuploidy screening it is usually common to base median calculations on the dating by gestational age or indeed on the CRL measurement itself. The latter is usually most preferred because there are a number of different ultrasound charts in use which would result in different gestational days being calculated for the same fixed CRL. The vast majority of markers are log-Gaussian distributed. Second trimester screening performance In the second trimester, screening for aneuploidies is primarily for trisomy 21 and to a lesser extent for trisomy 18 since the pattern of biochemical markers for trisomy 13 and other aneuploidies are similar to the normal pattern with perhaps the exception of triploidy (3 copies of each chromosome). The expected median concentrations of the major markers in pregnancies with trisomy 21 and 18 are shown in Table I.
Table I. Marker levels in second trimester cases of trisomy 18 and 21. Serum marker AFP Total hCG Unconjugate estriol Free β-hCG Inhibin A
T18 median MoM
T21 median MoM
0.65 0.32 0.42 0.33 0.87
0.75 2.06 0.72 2.20 1.92
43
Screening performance using a combination of the different markers shows that for a two marker protocol using AFP and Free β-hCG a detection rate of around 64 % can be achieved for a 5 % false positive rate. If a three marker protocol were used – including U-Estriol (U-E3) then this would increase to around 67 % and if a four marker protocol were used – including Inhibin-A then this would increase to around 72 %. In more than 20 published prospective intervention studies this modelled screening performance has been confirmed in routine practice [13]. The four marker protocol commonly referred to as the ‘Quad Test’ is generally the standard in use when screening is performed in the second trimester. However when comparing different reported detection rates and false positive rates one needs to bear in mind that the screened population age distributions may not all be the same and there is a need to use age standardized detection rates and false positive rates when comparing screening performance [14]. Detection rates for Trisomy 18 tend on the whole to be considerably lower at around 60 %. At the 18–22 week ultrasound anomaly scan the presence of a wide variety of ‘soft markers’ has also been proposed as a screening method for trisomy 21. The vast majority of these soft makers have been poorly evaluated [15] and it is only recently that appropriate likelihood ratios have been associated with the most useful of the markers such as ventriculomegaly, nuchal fold thickness, hypoplastic nasal bone and aberrant subclavian artery [16]. First trimester screening The past decade and a half has seen a considerable focus on moving screening earlier into the first trimester. Earlier screening is seen as providing women with an earlier reassurance and if termination of pregnancy is desired, this can be completed by a safer procedure and before fetal movements are usually evident. The fact that some aneuploid pregnancies detected in the first trimester will be spontaneously lost before term is not a valid argument against early screening. For these women it is important information to know, which can shape future reproductive decision making. Screening markers in the first trimester consist of the ultrasound marker Nuchal Translucency (NT) and the maternal serum biochemical markers free β-hCG and Pregnancy Associated Plasma Protein-A (PAPP-A) and form what has become known as the combined test. This test has now become the National Screening Standard in many countries. Although many biochemical markers have been investigated including those used in the second trimester only free β-hCG is of value across both trimesters. Both PAPP-A and free β-hCG are placental-derived markers. NT is the term used for an ultrasound measurement of the thickness of an echogenic area of fluid
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
44
K. Spencer
that exists in all fetuses between the fetal skin and the soft tissue overlying the cervical spine seen in a mid-sagittal view of the face of the fetus. NT is the single most discriminatory marker of trisomy 21 in the first trimester. The measurement of NT, unlike the measurement of serum markers, depends on operator skill especially when one considers the small size of the measurement of 1–3 mm being normal. The NT measurement may be made at the same time as the early pregnancy dating scan, between 11 weeks 0 days and 13 weeks 6 days. Most experienced sonographers can obtain the measurement within a few minutes. The crown-rump length (CRL) is measured during the same examination. NT increases with advancing pregnancy, and the CRL (a precise measure of gestational age) is used to standardize NT measurements for gestational age, either converting NT to MoM values (observed NT measurements divided by the median NT measurement for the corresponding CRL) or to delta values (observed NT measurements minus the median NT measurement for the corresponding CRL). The Fetal Medicine Foundation (www. fetalmedicine.com/fmf/) and Nuchal Translucency Quality Review Program (www.ntqr.org) are two organizations that provide ultrasonographer training, certification, and ongoing audit. Continuing audit of individual sonographers is required by many screening programs and in the UK this is carried out along with the monitoring of biochemical markers by the Downs Syndrome Screening Quality Assurance Support Service (DQASS) at a national level (http://fetalanomaly.screening.nhs.uk/dqass). Bias and inaccurate estimation of both NT and CRL have been shown to adversely affect detection rates and false positive rates [17]. Not only is increased NT a good marker for trisomy 21 it is also a very good marker for trisomy 13 and trisomy 18 and can be supra elevated in hydropic Turner’s syndrome. Increased NT is also associated with a number of adverse pregnancy outcomes and an increased risk for fetal cardiac defects. Combining first trimester maternal serum biochemistry and NT measurement is an effective screening procedure because the two modalities do not appear to be correlated [18]. Just as in the second trimester the likelihood ratio for each marker MoM can be derived from the overlapping Gaussian distributions from the affected and unaffected populations and this can be used to modify the a priori risk based on maternal age or history. In some algorithms used by the Fetal Medicine Foundation a mixture model approach is used for NT which is thought to give more accurate risks [19]. Retrospective and now many prospective studies have shown that 90 % of cases of T21 can be detected for a false positive rate of 5 % or less [13] and studies have also shown that approximately 90 % of all other major chromosomal anomalies can be detected for an
Table II. Changes in marker patterns in the first trimester of aneuploid pregnancies. Aneuploidy
NT
CRL
Fetal heart rate
Free β-hCG
PAPP-A
Trisomy 21 Trisomy 18 Trisomy 13 Turner’s Triploidy I Triploidy II
↑2.5 ↑3.5 ↑2.5 ↑7.0 ↑2.5 ↔
↔ ↓ ↔ ↔ ↔ ↓
↑ ↓ ↑ ↑ ↓ ↓
↑2.2 ↓0.3 ↓0.5 ↔ ↑8.0 ↓0.2
↓0.5 ↓0.5 ↓0.3 ↓0.5 ↓0.8 ↓0.2
additional 1 % false positive rate [20]. Table II summarizes the basic pattern of changes in the various markers associated with aneuploidy. Attempts to improve the accuracy of individualized risks in both the second and first trimester have been carried out by correcting many of the biochemical markers for pregnancy variables that influence the levels of these markers. Some of the variables are outlined in Table III. If these are not corrected for then considerable errors in individualized risk can be introduced [21].
National screening policy In many countries with state funded health care systems national policies on screening for Down syndrome have been developed. In the UK the National Institute of Clinical Excellence (NICE) have issued guidelines (http://www.nice.org.uk/nice media/live/11947/40115/40115.pdf http://www.nice. org.uk/nicemedia/live/11947/40145/40145.pdf) based on the work of the UKNSC and its Fetal Anomaly Screening program. The primary screening program across the whole of England is that of the combined test and the Model of Best Practice 2011– 2014 http://fetalanomaly.screening.nhs.uk/getdata. php?id ⫽ 11393 outlines the aim of the UK program in focusing on reducing the false positive rate whilst keeping the detection rate at a high level. The aspiration of this program is to achieve a Detection Rate of 90 % for a 2 % False/Screen positive rate by 2014. The success of this program in reducing the screen positive rate and hence the invasive testing rate has been quite remarkable [22] and similar results have also been achieved in the Danish program. However such success really requires an extensive program of training, auditing and quality monitoring of all aspects of the program.
Screening in twin pregnancies Screening in twin pregnancies is still considered by some to be problematical. Whilst the NT measure can allow for a fetus specific risk, the maternal serum biochemistry represents an average of the two fetuses and is therefore pregnancy specific. With maternal
Screening for Down syndrome
45
Table III. Maternal and pregnancy factors influencing biochemical marker concentrations. Factor Gestational age Maternal weight Multiple pregnancy IDDM Fetal sex Assisted conception Ethnicity
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
Smoking Gravidity/parity Previous pregnancy
First trimester
Second trimester
PAPP-A increased, free β-hCG decreased after 9 weeks All decreased with increasing weight Twice as high in twins – chorionicity effect PAPP-A decreased Free β-hCG and PAPP-A increased with a female fetus Free β-hCG increased PAPP-A decreased Afro-Caribbean PAPP-A 50 % higher Free β-hCG higher by 10 % PAPP-A increased Both increased as pregnancy number increases 2 to 3 times more likely to be high risk if previous was high risk
serum biochemistry it is argued that in the presence of one abnormal fetus and one normal fetus the biochemistry result from the abnormal fetus is clouded by the normal contribution from the normal fetus. Provided appropriate correction factors are used for the biochemical markers – which are gestation specific and chorionicity specific then detection rates approaching those in singleton pregnancies can be achieved in the first trimester of twins discordant for Down syndrome [23]. A more detailed discussion of screening in twin pregnancies can be found in the webcast http://web8.perkinelmer.com/LP ⫽ 7012?url ⫽ 33113641 or in Spencer 2005 [24]. One other issue arises with pregnancies that start as twins is the situation of a vanishing twin (i.e. when one of the twins has died in utero) where some of the maternal serum biochemistry may still be being contributed by the demised twin. Certainly for markers that have a long clearance half life – like PAPP-A, this can result in elevated concentrations even 4 weeks after demise [25]. In this situation it may be advisable to use NT alone, although there are instances when it may be appropriate to include biochemistry data [26].
Alternative screening strategies Alternative screening strategies have been described which are more complex than the simple Quad test or the Combined Test. One such test is the Integrate Test which attempts to combine the tests used in the first trimester with those used in the second trimester. Thus PAPP-A and NT are measured at 11–13 weeks, the patient is not given a risk assessment but then returns at 16 weeks for further biochemical tests which include AFP, Free β-hCG, U-E3 and Inhibin-A. Only when all the data are together is a risk produced. The theoretical benefits of this approach are a perceived increase in detection rate or lowering of the false positive rate based on data from two trials SURUSS [27] and FASTER [28]
AFP, uE3 increased, hCG decreased, inhibin-A little change All decreased with increasing weight, U-E3 least affected Twice as high in twins AFP decreased, U-E3 and hCG increased HCG increased, AFP decreased with female fetus HCG increased, U-E3 decreased AFP, hCG higher in Asian an Afro-Caribbean, inhibin-A lower in Afro-Caribbean HCG, U-E3 decreased, AFP increased and Inhibin-A increased by 60 % HCG decreased 3 to 5 times more likely to be high risk if previous was high risk
however the practicalities of this approach make screening more complex and potentially more costly (http://fetalanomaly.screening.nhs.uk/getdata. php?id ⫽ 11393). Also many have voiced serious ethical and moral issues with respect to withholding information at the end of the first trimester part of the test. In the UK this is not a nationally recommended screening test but is popular in the USA.
Timing Another proposal to improve performance overall has been as a result of the observation that in T21 the specificity of the individual markers change across the first and second trimester gestational windows [29,30]. Thus PAPP-A has better clinical discrimination between 9 and 11 weeks than between 11 and 13 whereas for free β-hCG this is reversed. Since NT is measured optimally at 11–13 weeks and the observation of other ultrasound features is best performed around 12 weeks it has been suggested that further improvements could be made by taking a blood sample at 9–10 weeks and using this to measure PAPP-A and then at 12 weeks collect a second sample at the time of NT for measuring free β-hCG [31]. Such protocols will allow detection rates of 90 % to be achieved with a false positive rate of less than 2 % – which is the aspirational detection rate for the UK National program for 2011. It is also becoming apparent that the various correction factors used in screening are very gestational age-dependent and require gestational-specific corrections as for twins [25], smoking [32] and ethnicity [33].
Contingent testing Other more complex screening strategies that have been proposed are based around setting risk cut offs so that those with very high risks (e.g. 1:50) are offered
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
46
K. Spencer
Figure 1. Inverting the Pyramid of Antenatal Care to focus on the 11–13 week assessment. (modified from Nicolaides [37]).
CVS and those with a very low risk (e.g. 1:1000) are reassured and those in the middle are offered further testing. This further contingent testing [34] can be to refer those women to a specialist fetal medicine/ Ultrasound unit for the measurement of nasal bone, facial angle, tricuspid regurgitation and Ductus venosus in the first trimester – this information is then combined with the earlier information from combined testing and a revised single risk calculated [35]. All of these latter markers require skills in measurement and in the case of the latter two high energy colour doppler. They are not advised as routine screening markers and are best left to skilled tertiary centers. A second alternative would be to offer the intermediate group second trimester biochemical screening with the Quad test and to combine all of the test elements together in a single risk. The modifications add little to the overall detection rate but can mean the false positive rate falls well below 2 %. More recently it has been suggested that cell free fetal DNA measured in the maternal circulation may also be a contender for use in a Contingent approach [36], harnessing the improved screening potential of this marker to dramatically reduce the screen positive rate yet only performing this expensive test on around 12 % of the pregnant population. In its current form cell-free fetal DNA testing is not diagnostic and is unlikely to replace CVS/Amniocentesis and QFPCR/karyotyping in the near future.
Inverting the pyramid of antenatal care Nicolaides [37] has proposed recently that ‘It is likely that the new challenge for improvement of pregnancy
outcome will be met by inverting the pyramid of prenatal care to introduce on a large scale and in a systematic fashion a new model of prenatal care which will be based on the results of a comprehensive assessment at 11 to 13 weeks’. This is because research over the past decade has shown that by using ultrasound, biophysical and biochemical markers and previous history it is possible to screen pregnancies at this stage for the major pregnancy complications which include aneuploidies, pre-eclampsia, babies small for gestational age, macrosomia, gestational diabetes, pre-term delivery, miscarriage and stillbirth and fetal structural anomalies (Figure 1). By the addition of placental growth factor (PlGF) and possibly AFP not only is it possible to improve the detection of aneuploidy by 3 % as a result of the lowered levels of PlGF seen with this condition [38] its use in conjunction with PAPP-A, mean arterial pressure and uterine artery Doppler velocimetry is able to identify more than 90 % of cases with early pre-eclampsia with a false positive rate of 5 % [39] – thus paving the way for potential therapeutic intervention with low dose aspirin [40]. It is probably this change to the way that antenatal care is delivered in the UK that will dictate how new technologies like cell-free fetal DNA are utilized in early pregnancy. Declaration of interest: The author reports no conflicts of interest. The author alone is responsible for the content and writing of the paper. References [1] Wilson JMG, Jungner G. Principles and practice of screening for disease. Public Health Paper Number 34. Geneva: WHO, 1968. [2] Merkatz IR, Nitowsky HM, Macri JN, et al. An association between low maternal serum alpha-fetoprotein and fetal chromosomal abnormalities. Am J Obstet Gynecol 1984; 148:886–94. [3] Cuckle HS, Wald NJ, Lindenbaum RH. Maternal serum alpha-fetoprotein measurement: a screening test for Down syndrome. Lancet 1984;1(8383):926–9. [4] Cuckle HS, Wald NJ, Thomson SG. Estimating a women’s risk of having a pregnancy associated with Down’s syndrome using her age and serum alphafetoprotein level. Br J Obstet Gynaecol 1987;94:387–402. [5] Bogart MH, Pandian MR, Jones OW. Abnormal maternal serum chorionic gonadotropin levels in pregnancies with fetal chromosome abnormalities. Prenat Diagn 1987;7: 623–30. [6] Canick JA, Knight GJ, Palomaki GE, et al. Low second trimester maternal serum unconjugated oestriol in pregnancies with Down’s syndrome. Br J Obstet Gynaecol 1988;95: 330–3. [7] Wald NJ, Cuckle HS, Densem JW, et al. Maternal serum screening for Down’s syndrome in early pregnancy. BMJ 1988;297:883–7. [8] Macri JN, Kasturi RV, Krantz DA, et al. Maternal serum Down syndrome screening: free beta protein is a more effective marker than human chorionic gonadotropin. Am J Obstet Gynecol 1990;163:1248–53.
Scand J Clin Lab Invest Downloaded from informahealthcare.com by University of Otago on 09/08/14 For personal use only.
Screening for Down syndrome [9] Spencer K. Evaluation of an assay of the free beta-subunit of choriogonadotropin and its potential value in screening for Down’s syndrome. Clin Chem 1991:37:809–14. [10] Spencer K, Wood PJ, Anthony FW. Elevated levels of maternal serum inhibin immunoreactivity in second trimester pregnancies affected by Down’s syndrome. Ann Clin Biochem1993;30:219–20. [11] Aitken DA, Wallace EM, Crossley JA, et al. Dimeric inhibin A as a marker for Down’s syndrome in early pregnancy. N Engl J Med 1996;334:1231–6. [12] Wallace EM, Crossley JA, Ritoe SC, et al. Evolution of an inhibin A ELISA method: implications for Down’s syndrome screening. Ann Clin Biochem 1998;35:656–64. [13] Spencer K. Aneuploidy screening in the first trimester. Am J Med Genet Part C Semin Med Genet 2007;145C18–32. [14] Cuckle H, Aitken D, Goodburn S, et al. Age-standardisation when target setting and auditing performance of Down syndrome screening programmes. Prenat Diagn 2004;24:851–6. [15] Lau TK, Evans MI. Second-trimester sonographic soft markers: what can we learn from experience of first trimester nuchal translucency screening? Ultrasound Obstet Gynecol 2008;32:123–5. [16] Agathokleous M, Chaveeva P, Poon LCY, et al. Metaanalysis of second trimester markers for trisomy 21. Ultrasound Obstet Gynecol 2013;41:247–61. [17] Kagan KO, Wright D, Etchegaray A, et al. Effect of deviation of nuchal translucency measurements on the performance of screening for trisomy 21. Ultrasound Obstet Gynecol 2009;33:657–64. [18] Spencer K, Souter V, Tul N, et al. A screening program for trisomy 21 at 10–14 weeks using fetal nuchal translucency, maternal serum free b-human chorionic gonadotropin and pregnancy-associated plasma protein-A. Ultrasound Obstet Gynecol 1999;13:231–7. [19] Wright D, Kagan KO, Molina FS, et al. A mixture model of nuchal translucency thickness in screening for chromosomal defects. Ultrasound Obstet Gynecol 2008;31:376–83. [20] Kagan KO, Wright D, Valencia C, et al. Screening for trisomies 21, 18 and 13 by maternal age, fetal nuchal translucency, fetal heart rate, free β-hCG and pregnancy-associated plasma protein-A. Hum Reprod 2008;23:1968–75. [21] Kagan KO, Wright D, Spencer K, et al. First-trimester screening for trisomy 21 by free beta-human chorionic gonadotropin and pregnancy-associated plasma protein-A: impact of maternal and pregnancy characteristics. Ultrasound Obstet Gynecol 2008;31:493–502. [22] Morgan S, Delbarre A, Ward P. Impact of introducing a national policy for prenatal Down syndrome screening on the diagnostic invasive procedure rate in England. Ultrasound Obstet Gynecol 2013;41:526–9. [23] Madsen HN, Ball S, Wright D, et al. A re-assessment of biochemical marker distributions in T21 affected and unaffected twin pregnancies in the first trimester. Ultrasound Obstet Gynecol 2011;37:38–47. [24] Spencer K. Non-invasive screening tests. In Blickstein I, Keith L (eds.) Multiple pregnancy: epidemiology, gestation and perinatal outcome. Oxford: Taylor & Francis. 2005. pp. 368–4.
47
[25] Spencer K, Staboulidou I, Nicolaides KH. First trimester aneuploidy screening in the presence of a vanishing twin: implications for maternal serum markers. Prenat Diagn 2010;30:235–40. [26] Sankaran S, Rozette C, Dean J, et al. Screening in the presence of a vanished twin: nuchal translucency or combined screening test? Prenat Diagn 2011;31;600–01. [27] Wald NJ, Rodeck C, Hackshaw AW, et al. First and second trimester antenatal screening for Down’s syndrome: the results of the Serum, Urine and Ultrasound Screening Study (SURUSS). Health Technol Assess 2003;7:1–77. [28] Malone FD, Canick JA, Ball RH, et al. First-trimester or second-trimester screening, or both, for Down’s syndrome. N Engl J Med 2005;353:2001–11. [29] Spencer K, Crossley JA, Aitken DA, et al. Temporal changes in maternal serum biochemical markers of Trisomy 21 across the first and second trimester of pregnancy. Ann Clin Biochem 2004;39:567–76. [30] Spencer K, Crossley JA, Aitken DA, et al. The effect of temporal variation in biochemical markers of trisomy 21 across the first and second trimester of pregnancy on the estimation of individual patient-specific risks and detection rates for Down’s syndrome. Ann Clin Biochem 2003;40:219–31. [31] Wright D, Spencer K, Kagan K, et al. First-trimester combined screening for trisomy 21 at 8–13 weeks. Ultrasound Obstet Gynecol 2010;30:404–11. [32] Spencer K, Cowans NJ. Correction of first trimester biochemical aneuploidy screening markers for smoking status: influence of gestational age, maternal ethnicity and cigarette dosage. Prenat Diagn 2013;33:116–23. [33] Cowans NJ, Spencer K. Effect of gestational age on first trimester maternal serum prenatal screening correction factors for ethnicity and IVF conception. Prenat Diagn 2013;33:56–60. [34] Wright D, Bradbury I, Benn P, et al. Contingent screening for Down syndrome is an efficient alternative to non-disclosure sequential screening. Prenat Diagn 2004;24:762–6. [35] Nicolaides KH, Spencer K, Avgidou K, et al. Multicentre study of first trimester screening for trisomy 21 in 75,821 pregnancies: results and estimation of the potential impact of individual risk orientated two stage first trimester screening. Ultrasound Obstet Gynecol 2005;25:221–6. [36] Chitty LS, Hill M, White H, et al. Noninvasive prenatal testing for aneuploidy – ready for prime time. Am J Obstet Gynecol 2012;206:269–75. [37] Nicolaides KH. Turning the pyramid of prenatal care. Fetal Diagn Ther 2011;29:183–96. [38] Pandya P, Wright D, Syngelaki A, et al. Maternal serum placental growth factor in prospective screening for aneuploidies at 8–13 weeks’ gestation. Fetal Diagn Ther 2012; 31:87–93. [39] Akolekar R, Syngelaki A, Poon L, et al. Competing risks model in early screening for preeclampsia by biophysical and biochemical markers. Fetal Diagn Ther 2013;33:8–15. [40] Roberge S, Villa P, Nicolaides K, et al. Early administration of low-dose aspirin for the prevention of preterm and term preeclampsia: a systematic review and meta-analysis. Fetal Diagn Ther 2012;31:141–6.
