Tests for probability of chromosomal anomalies

50.1 Approaches to testing for high probability of chromosomal anomalies

A range of biochemical tests and ultrasound techniques has been developed that can significantly increase the identification of pregnancies with a high probability of chromosomal anomalies such as trisomy 21 (Down syndrome), trisomy 18 (Edwards syndrome) and trisomy 13 (Patau syndrome) (see Glossary). A high probability test result leads to the offer of a diagnostic test (chorionic villus sampling or amniocentesis) (see Chapter 51).

The suitability of any test depends on the gestational stage. Extensive pre- and post-test information and counselling are required, with consideration also being given to the woman’s preferences, availability of testing facilities, costs to the woman and, for ultrasound, operator expertise.

Current practice in Australia is that testing for chromosomal anomalies is done in the first trimester. The combined first trimester test comprises:

• ultrasound measurement of fetal nuchal translucency thickness between 11 weeks and 13 weeks 6 days gestation (when the fetus has a crown-rump length of 45–84 mm) combined with
• maternal plasma testing of pregnancy-associated placental protein-A (PAPP-A) and free beta-human chorionic gonadotrophin (β-hCG) between 9 weeks and 13 weeks, 6 days gestation.

An emerging practice is the use of cell-free deoxyribonucleic acid (cfDNA) testing (also referred to as non-invasive prenatal testing [NIPT]). cfDNA testing can be performed for detection of fetal anomaly from 10 weeks gestation. The test involves sequencing DNA fragments in maternal serum, mapping each DNA sequence to a reference genome to determine its chromosome of origin, and counting the number of fragments arising from each chromosome. If the fetus is affected by trisomy, a greater than expected number of the relevant chromosome fragments will be present in maternal serum.

Cell-free DNA testing has been used as a first-tier test, as a second-tier test (with women with increased probability on combined first trimester screening offered cfDNA or diagnostic testing) or in a contingent model (where women with an intermediate probability on combined first trimester screening are offered cfDNA testing and those with a very high probability are offered diagnostic testing). Evaluations of the implementation of contingent cfDNA testing in national screening programs have found improved performance of the program Chitty et al 2016, Gil et al 2016, Oepkes et al 2016. Use of cfDNA as a first-tier test may be appropriate for women with infections where an invasive procedure carries an increased risk of mother-to-child transmission.

Later in pregnancy (14 to 20 weeks), the triple test (maternal serum testing of a-fetoprotein [AFP], free β-hCG [or total hCG] and unconjugated estriol) or the quadruple test (which also includes inhibin A) is used to assess the risk of fetal chromosomal anomaly. While this is an important publicly funded option for women who present later in pregnancy or for whom specialist ultrasound is not available (eg in rural and remote areas) or who cannot afford cfDNA testing (until this becomes publicly funded), the evidence for these tests has not been reviewed as part of the development of these Guidelines. As cfDNA testing can be performed at any gestation from 10 weeks, it should be discussed along with second trimester serum screening for women who have missed the gestational age window for combined first trimester screening.

50.2 Effectiveness of tests for probability of chromosomal anomaly

Offering testing for probability of fetal chromosomal anomaly to all women in the first trimester, regardless of maternal age, is recommended in the United Kingdom (NICE 2008), the United States ACOG 2007 and Australia RANZCOG 2015.

50.2.1 Combined first trimester tests

The combined first trimester test identifies factors that are known to be associated with fetal chromosomal anomalies and that are independent of each other.

The probability of chromosomal and other anomalies and fetal and postnatal death increases with nuchal translucency thickness. Favourable outcomes have been observed in 92% of babies with nuchal translucency of 3.4 mm (95th centile) compared to 18% of those with nuchal translucency of ≥6.5 mm (Ayras et al 2013). In some situations, the ultrasound component of first-trimester testing may be difficult or impossible (eg due to high BMI, fetal positioning).

Combining nuchal translucency assessment with testing of maternal serum increases the predictive value (Alexioy et al 2009). Recent evidence on the sensitivity of the combined test had the following findings.

• A systematic review (65 studies) found detection rates of 91.9% for trisomy 18 (false positive rate 3.5%), 83.1% for trisomy 13 (false positive rate 4.4%) and 70.1% for monosomy X (false positive rate 5.4%) (Metcalfe et al 2014).
• Cohort studies found detection rates of:
  • 92.2% for trisomy 21 (false positive rate 8.0%) (n=675,332) (Kagan et al 2015b).
  • 91.3% for trisomy 21, 97.1% for trisomy 18, 92.3% for trisomy 13, 80% for sex chromosome aneuploidies and 87% for atypical aneuploidies (n=21,052) (Kagan et al 2015a)
  • 87% for trisomy 21, 91.8% for trisomies 13 and 18, 86.0% for monosomy X, 8.1% for other sex chromosome aneuploidies, 89.3% for triploidy and 13.0% for other high-risk outcome (n=14,684) (Syngelaki et al 2014)

The pooled rate of invasive procedures was 59 per 1,000 pregnancies tested Susman et al 2010, Syngelaki et al 2014, Kagan et al 2015a.

As fetal nuchal translucency thickness increases with crown-rump length Pandya et al 1995, Edwards et al 2003 and the detection rate in serum is influenced by maternal age (Grati et al 2010), these factors are included in assessment algorithms. The inclusion of age in the calculation, either alone or in combination with serum test results, increases identification of the probability of chromosomal anomalies Wapner et al 2003, Scott et al 2004, Centini et al 2005, Soergel et al 2006, Gebb & Dar 2009, Hagen et a 2010, Schmidt et al 2010. The maternal serum variables are also influenced by gestational age, maternal weight, ethnicity, smoking, in vitro fertilisation, parity and diabetes, the background risk for each being calculated and then included in the algorithm with nuchal transluceny and maternal age. A history of a previous trisomy 21 pregnancy increases the chance of an abnormal test result for trisomy 21.

50.2.2 Cell-free DNA testing

cfDNA testing has a higher detection rate for the more common trisomies than the combined first-trimester test: relative risk of detection 1.13 (1.08 to 1.18) for trisomy 21 and 1.22 (1.18 to 1.26) for trisomies 18 and 13 Petersen et al 2014, Syngelaki et al 2014, Gyselaers et al 2015, Kagan et al 2015a, Kagan et al 2015b, McLennan et al 2016. Fewer invasive procedures are required (10 per 1,000 women tested) Susman et al 2010, Syngelaki et al 2014, Kagan et al 2015a and rates of procedure-related miscarriage are lower Morris et al 2014, Gyselaers et al 2015, Mersy et al 2015.

However, cfDNA testing may not detect less common chromosomal anomalies identified through ultrasound assessment: relative risk of detection 0.23 (0.16 to 0.33) for sex chromosome aneuploidies Syngelaki et al 2014, Kagan et al 2015a, McLennan et al 2016 and 0.01 (0.00 to 0.04) for atypical aneuploidies Petersen et al 2014, Syngelaki et al 2014, Kagan et al 2015a if these are not included in the cfDNA test panel. As well, the economic costs of incorporating cfDNA testing for trisomy 21 as a first-tier test into Australian practice are currently higher than those for combined first trimester testing (costs associated with cfDNA testing for other chromosomal anomalies have not been investigated in Australia) O’Leary et al 2013, Ayres et al 2014. However, its use as a contingent screen for trisomy 21 at specific thresholds may be more cost-effective than combined first trimester testing (Maxwell et al 2017).

As cfDNA testing is available in Australia, health professionals need to have an understanding of the test, including that:

• the test may be conducted from 10 weeks onwards (Gil et al 2015a)
• the test is not diagnostic; a positive result requires confirmation by invasive procedures Gil et al 2015a, Meck et al 2015, Neufeld-Kaiser et al 2015, McLennan et al 2016
• diagnosis of fetal structural or genetic anomalies may be delayed or missed if the 11–13 week ultrasound is not performed in conjunction with cell-free DNA testing RANZCOG 2015
• although the false positive rate is lower than for combined first trimester testing, both false positives and false negatives occur Gil et al 2015b, Mackie et al 2016, Taylor-Phillips et al 2016
• low fetal fraction of DNA in the maternal circulation Benachi et al 2015, Gil et al 2015b, Neufeld-Kaiser et al 2015, McLennan et al 2016, which is common among women with a BMI >30 kg/m2 Benachi et al 2015, Gil et al 2015b, Neufeld-Kaiser et al 2015, McLennan et al 2016, may yield an unreportable result; depending on the timing of the test, this may mean that women with a test failure miss the window for the combined first trimester test
• in rare circumstances, the test may raise suspicions of maternal or fetal conditions other than the fetal anomalies for which the test is being performed (Sachs et al 2015)
• the test is not currently covered by Medicare or private health insurance; costs to women are $400–$500, depending on location.

50.3 Discussing tests with women

At the first antenatal visit or as early as possible in pregnancy, the availability of testing for probability of chromosomal anomalies should be discussed and women given relevant written information or other appropriate materials (eg video, DVD) (see Section 53.1). Providing information is particularly important, due to the complexity of the process and the level of decision-making that may be required. A systematic review found levels of knowledge adequate for decision-making were at times not being achieved despite information leaflets and video having some effect (Green et al 2004). Studies in which knowledge about genetic testing is increased have not observed any corresponding increase in anxiety (Green et al 2004).

In discussing the tests so that women can give informed consent, it is important to talk in terms of ‘probability’ or ‘chance’ rather than ‘risk’ and to explain:

• it is the woman’s/couple’s decision whether any testing takes place
• the chromosomal anomalies for which testing is available and the differences between these conditions
• the different pathways for testing (ie combined first trimester test alone, cfDNA testing as first-tier or second-tier test or in a contingent model; see Section 50.1) and the risks and benefits of each approach
• the testing pathway, the decisions that need to be made at each point and their consequences
• the need for accurate assessment of gestational age so that tests are conducted at the appropriate time
• that results of these tests alone indicate a probability of fetal chromosomal anomaly but do not give a definitive diagnosis of any anomalies
• the sensitivity, specificity and positive predictive value for the woman’s age of the test and a full explanation of the reporting format of the test (eg high probability/low probability, 1 in 10, 1 in 300, 1 in 1,000)
• the options for women who receive a high-probability result, including information about chorionic villus sampling and amniocentesis (see Section 51.2)
• a large nuchal translucency associated with normal chromosomes may indicate other anomalies which may be structural (eg diaphragmatic hernia, cardiac anomaly) or genetic (eg Smith-Lemli-Opitz syndrome, Noonan syndrome)
• factors that increase the probability of fetal chromosomal anomalies (advanced maternal age, family history of chromosomal anomalies)
• where and how tests can be accessed if the woman chooses to have them
• the availability of evaluated decision aids (eg the Ottawa Decision Framework) Arimori 2006, Nagle et al 2006, Nagle et al 2008 (see Section 53.1)
• the costs involved for the woman and the timeframe for receiving results.

Women may choose not to have a test or may choose to proceed directly to a diagnostic procedure instead (eg due to a preference to receive definitive information and/or concerns about the sensitivity of available tests). The choice a woman and her partner make about testing should not influence the subsequent care she receives.

50.4 Care for women with a high probability of fetal chromosomal anomaly

Following a result that suggests a higher probability of having a baby with a chromosomal anomaly, the offer of referral to a health professional (eg a genetic counsellor) is an important consideration.

Antenatal care for women with a high probability of having a baby with a chromosomal anomaly should be supportive and respectful of women’s choices about continuation of pregnancy.