CHAPTER 21 Lisa C. Zuckerwise1Karen Archabald2 and Joshua Copel1 Department of Obstetrics, Gynecology & Reproductive Sciences, Division of Maternal‐Fetal Medicine, Yale School of Medicine, New Haven, CT, USA Legacy Health, Portland, OR, USA Screening for aneuploidy has become an important part of routine obstetrical care. In 2007, the American College of Obstetricians and Gynecologists (ACOG) recommended that all women, regardless of maternal age, be offered aneuploidy screening before 20 weeks of gestation and be given the option of invasive testing [1, 2]. Options for screening and diagnosis depend on the gestational age at time of presentation for care as well as patient preference and availability of resources. This chapter will discuss the advantages and limitations of the various methodologies available. Cell‐free fetal DNA can be detected in maternal blood during pregnancy [3]. These small fragments of DNA actually derive from placental cells, so they more accurately reflect the DNA makeup of the placenta. The first report of cfDNA to diagnose Trisomy 21 occurred in 2008 and was by the massively parallel shotgun sequencing technique. Other techniques for analyzing cfDNA have been validated by various laboratories, including selective sequencing of target chromosomes and single nucleotide polymorphism (SNP)‐based methods. These all rely on next‐generation sequencing technology and advanced bioinformatics analysis and possess similarly high sensitivities and specificities for detection of Trisomy 21 and Trisomy 18 [46]. An early nested case control study of a cohort of 4664 high‐risk pregnancies for Trisomy 21 in 27 centers validated the use of cfDNA as a diagnostic tool. Women were classified as high‐risk based on maternal age, family history or a positive serum and/or sonographic screening test. cfDNA was compared to conventional karyotype analysis. Trisomy 21 was detected in 98.6% (209/212) of positive cases, the false‐positive rate was 0.20% (3/1471), and the testing failed in 13 pregnancies (0.8%). Subsequent larger studies and meta‐analyses have confirmed that cfDNA analysis is a very powerful screening tool for both Trisomy 21 and Trisomy 18 in the high‐risk population, with an overall sensitivity and specificity of >99% [7, 8]. In low‐risk women, the prevalence of aneuploidy is considerably lower. As a result, while the sensitivity and specificity of any given screening test are unchanged, the clinical significance of a positive test is altered. This is important when considering the use of circulating free DNA (cfDNA) in the low‐risk obstetric population. Based on this principle, both the Society for Maternal‐Fetal Medicine and the ACOGs currently recommend conventional screening methods as the most appropriate choice for first‐line screening in the routine obstetric population. [9]. Despite this recommendation, there is a growing body of evidence that cfDNA analysis remains a powerful tool for detecting Trisomy 21 and Trisomy 18 even in the general obstetric population. A primary series of almost 2000 women at multiple centers presenting for routine prenatal care compared performance of standard aneuploidy screening (serum biochemical assays with or without NT measurement) with cfDNA in order to primarily assess the rates of false positive results. In this low‐risk population with a mean age of 29.6 years, the false positive rates were significantly lower for cfDNA than with standard screening for Trisomy 21 (0.3% versus 3.6%) and for Trisomy 18 (0.2% versus 0.6%). Positive predictive values were also favorable for cfDNA compared to standard screening, with positive predictive value (PPV) of 45.5% versus 4.2% for Trisomy 21 and PPV of 40.0% versus 8.3% for trisomy 18, [4]. For counseling purposes as well as clinical applicability, it is important to recognize that based on these findings in the low‐risk (i.e. younger age) population, there is still less than 50% chance of a fetus actually having Trisomy 21 or trisomy 18 with a positive cfDNA result. A subsequent larger prospective, multicenter, blinded study compared standard screening (NT and biochemical analytes) with cfDNA testing in a routine obstetric population. The average age of the more than 15 000 women included was 30.7 years. In this study, the sensitivity of standard screening for detecting Trisomy 21 was 78.9%, compared to 100% (38 of 38) for cfDNA. False positive rates were 5.4% in the standard screening and 0.06% in the cfDNA groups, and positive predictive values were 3.4% for standard screening, compared to 80.9% for cfDNA. All of these results were statistically significant [10]. While the positive predictive value in this study was higher than previously reported, it is still a function of the overall prevalence in the given population. Individual patient positive predictive values can be calculated with a tool found at http://www.perinatalquality.org/Vendors/NSGC/NIPT. In order to obtain a meaningful result from cfDNA analysis, there must be an adequate fraction of fetal DNA (fetal fraction) in the maternal blood. In most patients, samples drawn after 10 weeks gestation will provide adequate fetal fractions, meaning >8% of fetal DNA; however, in some cases, there will be fetal fraction too low to report (0–4%) or an intermediate amount (4–8%), where a result can be given but with compromised sensitivity. Low fetal fraction is associated with maternal obesity, and multiple studies have demonstrated an inverse relationship between fetal fraction and maternal weight [11-13]. One large study of 22 384 pregnant patients who underwent cfDNA testing found that about 20% of women weighing over 130 kg and 30% weighing over 140 kg have a fetal fraction less than 4% [11]. While this is clinically important in itself, given the increasing prevalence of obesity in the general population, there is also evidence that a failed result for cfDNA is associated with aneuploidy. In a large prospective study on performance of cfDNA to detect trisomy, which included over 18 000 patients, those with failed results due to low fetal fraction (<4%) had an aneuploidy risk of 4.7%, which was significantly higher than the overall cohort rate of 0.4% [10]. This increased risk of aneuploidy in fetuses with failed cfDNA results has been reported in smaller studies as well [14], highlighting the importance of genetic counseling following failed results, including a recommendation for either repeat cfDNA analysis or invasive testing for aneuploidy. Serum screening for fetal aneuploidy was introduced in the 1980s when low serum AFP in the second trimester was noted to be associated with an elevated risk of Trisomy 21 [15]. The type of noninvasive risk assessment for aneuploidy recommended depends on the time a woman presents for prenatal care and the availability of laboratory and ultrasonographic services, as well as her aneuploidy risk. Testing options include a combination of first and/or second trimester maternal serum analytes with or without the addition of ultrasonographic assessment, ultimately leading to an assignment of risk of both Trisomy 21 and Trisomy 18. Based on the risk calculation and discussion with the patient, further screening or diagnostic testing can subsequently be offered. First trimester combined screening includes: (i) sonographic measurement of the fetal crown–rump length and NT; and (ii) serum levels of pregnancy‐associated plasma protein A (PAPP‐A) and either total or free‐β human chorionic gonadotropin (hCG). Prospective studies from the United States and Europe have revealed detection rates for Trisomy 21 with NT alone ranging from 54% to 79% (see Table 21.1). A meta‐analysis assessing the role of NT as a screening tool for aneuploidy including 30 studies and 16 311 patients found an overall detection rate of 77% with a 5.9% screen positive rate [15]. Compared to euploid pregnancies [16], Trisomy 21 pregnancies have decreased levels of PAPP‐A and increased levels of hCG [16, 17]; however, first trimester serum screening with these two serum analytes without ultrasound detects only 55–70% of cases (Table 21.1). When both serum markers and NT are combined, sensitivity and specificity of the screening for Trisomy 21 improves significantly (Table 21.1). In the United States, the First and Second Trimester Evaluation of Risk (FASTER) trial found that combined first trimester screening at 11 weeks increased the detection rate of Trisomy 21–87% with a screen positive rate of 5%. [18] Maintaining a 5% screen positive rate, the detection rate decreases marginally to 85% when testing is performed at 12 weeks gestation and to 82% at 13 weeks gestation [18]. Two additional large multicenter trials; the Biochemistry, Ultrasound, Nuchal Translucency (BUN) study from the United States and the Serum, Urine, and Ultrasound Screening Study (SURUSS) trial completed in the United Kingdom and Austria, found detection rates of Trisomy 21 of 86% and 79%, respectively with a set screen positive rate of 5%. For women who present early to prenatal care in centers where NT is available, first trimester combined screening is a powerful tool in early detection of Trisomy 21. Table 21.1 Detection rate of Trisomy 21 with first trimester screening given 5% screen positive rate. First trimester combined screening can also be used for the detection of Trisomy 18. In the first trimester, Trisomy 18 pregnancies have lower levels of maternal serum hCG and PAPP‐A and a higher NT than euploid counterparts [19]. The FASTER trial found that with combination of first trimester NT and a serum screen positive for either Trisomy 21 (risk of 1 : 300), Trisomy 18 (risk of 1 : 100), or a cystic hygroma there was an 82% detection rate of Trisomy 18 with a screen positive rate of 6% [20]. In the BUN trial, the combination of NT, serum screening, and maternal age detected 90.9% of Trisomy18 with a screen positive rate of 2% [21]. In addition to combined screening with NT and maternal serum markers in the first trimester, additional sonographic markers including absent nasal bone, abnormal flow in the ductus venosus and tricuspid regurgitation have been proposed as adjunct screening tools. In an initial observational study of 701 women who were high‐risk secondary to increased NT and maternal age, the nasal bone was noted to be absent in 73% of Trisomy 21fetuses and only in 0.5% of unaffected fetuses [22]. Presence or absence of the nasal bone has subsequently been well studied, with a wide range of sensitivities reported ranging from as low as 7.7% to up to 65% of aneuploidy [23, 24]. Overall, the available data suggests that in low‐risk women, nasal bone screening adds little to first trimester combined screening [25–30]. There are several different approaches to combined first and second trimester screening, which are detailed below. Table 21.2 provides a comparison of screening detection rates between these different options. Table 21.2 Detection rate of trisomy 21 with combined first and second trimester screening. The full integrated screen includes first trimester combined screening with ultrasound measurement of NT and maternal serum PAPP‐A between 10 and 13 weeks as well as second trimester assessment of α‐fetoprotein (AFP), unconjugated estriol (uE3), hCG and inhibin‐A [31, 32]. Between 15 and 19 weeks, pregnancies with Trisomy 21 are associated with lower maternal serum alpha‐fetoprotein (MSAFP) [15], higher hCG [33], reduced levels of uE3 [34] and higher inhibin A than their euploid counterparts [35]. In contrast, pregnancies with Trisomy 18 are associated with decreased levels of AFP and uE3 [36, 37]. For true “Fully Integrated Screening,” results are only given after both first and second trimester testing is completed. The FASTER trial found that the fully integrated test has a detection rate of Trisomy 21 of 95% with a 5% screen positive rate [18]. The SURUSS trial found similar detection rate of 94% with a 5% screen positive rate [38]. (Table 21.2) A more recent prospective study comparing methods of combined first and second trimester screening found a detection rate of 90% with a screen positive rate of 3.4% [39]. The major drawback to this method of screening is that results are not given until the second trimester when diagnosis by CVS is no longer available and when pregnancies are more visible to others. Women have different reasons for undergoing aneuploidy screening; some women prefer early detection in order to facilitate safer elective termination while others desire reassurance regarding the health of the fetus. Depending on the wishes of the woman, waiting until the second trimester for results may not be the most logical option [40]. The serum integrated test involves first trimester PAPP‐A combined with a second trimester quadruple screen (AFP, uE3, hCG, inhibin‐A). This test is most appropriate in places where NT is not readily available, and can also be used for women who have had blood drawn for first trimester screening at the appropriate time, but whose fetuses cannot have NT measured for technical reasons (usually either due to fetal position or because the crown rump length (CRL) of the fetus is above accepted cutoffs). The FASTER trial found an 86% detection rate of Trisomy 21 at a 5% screen positive rate [18] for serum integrated screening, while the SURUSS trial found an 87% detection rate at a 5% screen positive rate [38]. A meta‐analysis of serum integrated screening for Trisomy18 found that by combining first trimester PAPP‐A with second trimester AFP, uE3, and hCG at a risk cutoff of 1: 100 the detection rate was 90% with a screen positive rate of 0.1% [41]. Stepwise sequential screening involves the initial calculation of risk from the ultrasound measurement of NT, maternal PAPP‐A and hCG between 11 and 13 weeks. Women with pregnancies at high‐risk are offered immediate invasive prenatal diagnosis with CVS, while the remainder go on to have second trimester testing. After completion of second trimester testing, the risk is recalculated to include second trimester markers and a new risk is assigned. This form of testing has the advantage of providing an early diagnosis for a substantial proportion of affected pregnancies. Analysis of the FASTER data reports a detection rate of Trisomy 21 of 92% with a 5% screen positive rate. A computer model based on the SURUSS data found a detection rate of 90% with a 2.25% screen positive rate. However, the majority of women have testing in both trimesters. Contingent screening involves the initial calculation of risk from the ultrasound measurement of NT, PAPP‐A and hCG between 11 and 13 weeks. Based on these results, pregnancies are classified as high, medium, or low‐risk. Women at high‐risk can immediately be offered CVS. Women with a low‐risk (negative) screen have no further testing. Women with borderline initial risk, which should be a predetermined range, go on to have quadruple screening with AFP, hCG, uE3 and inhibin at 15–18 weeks. A final risk is calculated that combines the first and second trimester screening results. The final result is considered positive if the risk is greater than 1 in 270. Predetermination of risk cutoffs is essential to the success of this approach. The initial first trimester cutoff should identify the majority of cases of Trisomy 21 while maintaining a low false positive rate. Studies to date have defined a first trimester false positive rate of 0.5% or a risk of >1 : 30 as high‐risk [39, 42, 43]. The FASTER trial found a 60% detection rate for Trisomy 21 in the first trimester with a screen positive rate of 1.2%. Modeling from the SURUSS trial predicted a 66% detection rate with a false positive rate of 2.42%. Finally, Guanciali found a detection rate of 82% with a screen positive rate of 4% [39]. By deferring subsequent testing in the low‐risk group only, this should minimize the number of affected pregnancies missed, while also limiting the number of patients who go on to have second trimester testing. Cutoffs used in the above studies range from <1 : 1200 [39] to >1 : 1500 [42]. The intermediate cutoff is the range between these two cutoffs (e.g. between 1 : 30 and 1 : 1500) and should identify many of the remaining cases of aneuploidy while minimizing the number of invasive procedures. Contingent screening decreases the number of women who require second trimester testing. Cost effectiveness models from both the SURUSS and FASTER trials have found that contingent screening may be the most cost effective method, because it decreases the number of women who need second trimester blood work, while maintaining a high aneuploidy detection rate [44, 45]. At present, the most sensitive conventional risk assessment tool for low‐risk women who present for prenatal care after 14 weeks of gestation is the Quadruple Screen. The Quadruple Screen measures serum levels of MSAFP, hCG, uE3, and inhibin‐A. Using a risk cutoff of 1 : 300 as positive, the detection rate of Trisomy 21 using the quadruple screen in the FASTER trial was 85% at an 8.5% screen positive rate [18]. Data from the same trial evaluated detection of Trisomy 18. Using a risk cutoff of 1 : 100 as positive detection the Quadruple Screen detection rate was 100% with a screen positive rate of 0.3% [20]. Although these detection rates are not as high as the first trimester or integrated screens, in a patient who presents for prenatal care in the second trimester they provide valuable information that can assist when counseling a patient regarding the advisability of diagnostic procedures.
Prenatal diagnosis
Background
Clinical questions
First trimester combined screening
NT
1st trimester serum
1st trimester combined
Snijders 1998 (n = 96 127) (83)
77%
Wald 2003 SURUSS (n = 47 053) (20)
63%
86%
Wapner 2005(BUN Trial) (n = 8216) (84)
67%
69%
79%
Malone 2005 (FASTER) (n = 33 557) (7)
70%
70%
87%
Bindra 2002 (OSCAR) (n = 14 383) (85)
79%
60%
90%
Crossley 2002 (n = 17 229)(86)
54%
55%
82%
Ghaffari 2011 (n = 13 706) (87)
94% (4.8% SPR)
Guanciali‐Franchi 2011 (n = 7292) (28)
81% (4% SPR)
Combined first and second trimester screening
1st trimester combined
Full integrated
Serum integrated
Stepwise sequential
Contingent
Quad screen
Malone 2005 (FASTER) (n = 33 546) (7)
87%
96%
88%
95%
81%
Cuckle 2008 (FASTER) (n = 32 355) (31)
93%
93%
92%
Wald 2003 (SURUSS) (n = 47 053) (20)
86%
94%
87%
83%
Wald 2006 (SURUSS) Model (32)
90% (2.25% SPR)
90% (2.42% SPR)
Platt 2004 (BUN) (n = 4325) (88)
79%
98% (17% SPR)
Guanciali‐Franchi 2011 (n = 7292) (28)
81% (4% SPR)
90% (3.4% SPR)
90% (5.2%SPR)
90% (2.6% SPR)
Fully integrated screening
Serum integrated screening
Stepwise sequential screening
Contingent screening
Second trimester screening
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
