Detection rates and residual risk for a postnatal diagnosis of an atypical chromosome aberration following combined first‐trimester screening

To determine the detection rates of all types of chromosome aberrations and the residual risk for postnatal diagnosis of an atypical chromosome aberration depending on the strategy for further investigation with either noninvasive prenatal testing (NIPT) or invasive testing in pregnancies with increased risk following combined first‐trimester screening (cFTS).


| INTRODUCTION
The introduction of noninvasive prenatal testing (NIPT) has challenged the policies and practices for standard of care concerning prenatal screening for chromosome aberrations in the first trimester. There are different ways of implementing NIPT into clinical practice, for example, used as a second-line test for women with an increased risk following combined first-trimester screening (cFTS), 1 or as a primary screening tool for all pregnant women. 2 However, the question still remains regarding how NIPT can best be applied in an efficient and cost-effective way. There is an increasing trend that NIPT analysing the common trisomies is replacing invasive prenatal testing with an analysis of all chromosomes using either karyotype or array-CGH.
One of the key questions is what we miss by using only targeted NIPT tests for trisomies 21, 18, 13 and sex chromosome aneuploidies?
Studies reviewing the expected performance of primary screening using NIPT in large population-based cohorts have shown that 17%-25% of fetal chromosome aberrations are clinically significant atypical chromosome aberrations considered nondetectable by NIPT. [3][4][5][6] However, nondetectable in early pregnancy is not equivalent to the likelihood of giving birth to a child with an atypical chromosome aberration. Other adverse pregnancy outcomes such as miscarriage/intrauterine fetal death (IUFD) and malformations detected at the second trimester scan may intervene and affect the residual risk for a postnatal diagnosis. We lack population-based studies providing information on the residual risk of giving birth to a child with an atypical chromosome aberration if prenatal screening only involves the analysis of common trisomies as the estimations published so far are based on extrapolations. [3][4][5][6] The aims of this study were to compare the residual risk of a postnatal diagnosis if either NIPT or invasive testing was used for further investigation in women with an increased risk following cFTS. Also, to estimate the detection rates for all types of chromosome aberrations depending on which strategy is used for further investigation. What's already known about this topic?
• Studies reviewing the expected performance of primary screening using NIPT in large population-based cohorts have shown that 17%-25% of fetal chromosome aberrations are clinically significant atypical chromosome aberrations considered not detectable by NIPT. However, they were limited by the absence of information on outcome of pregnancy and did not report pregnancy complications and could not study rates of "missed" atypical abnormalities subsequently diagnosed at birth.

What does this study add?
• This is the first population-based study on residual risk for a postnatal diagnosis of an atypical chromosome aberration when miscarriages and malformations detected on ultrasound have been accounted for.
• Continuing to offer cFTS and reinstating invasive testing would allow the majority of atypical aberrations and 95% of aneuploidies to be detected.
• A total of 40% of the atypical aberrations are detected later in pregnancy due to miscarriage/IUFD or a malformation detected on ultrasound. detection rates of trisomies 21 and 13/18 were 88% and 85%, respectively.

All genetic analyses were performed at the Department of Clinical
Genetics, Karolinska University Hospital. QF-PCR was performed according to routine clinical protocol using a panel of specific short tandem repeats from chromosomes 13, 18, 21, X, and Y as previously described. 9,10 CMA was performed using an array-based comparative genomic hybridization platform, a 180 K oligonucleotide array with evenly distributed whole-genome coverage (Oxford Gene Technology). Analysis of copy number variants (CNVs) was performed using the CytoSure Interpret Software (Oxford Gene Technology). Karyotype analysis was performed with conventional G-banding using standard cytogenetic procedures.

| RESULTS
The cohort consisted of 129 493 pregnancies examined with cFTS at 10 ultrasound units in the Stockholm county. There were 870 (0.7%) chromosome aberrations in the whole cohort, also including chromosome aberrations found later in pregnancy, following miscarriages/ IUFD or after birth. Eighteen atypical aberrations were considered not clinically significant and excluded from further analyses resulting in 852 clinically significant aberrations in the cohort. Common trisomies (13, 18, 21, and) or SCAs constituted 88% (754) of these, in our study presumed to be possible to detect by NIPT analysis, and 12% (98) were atypical chromosome aberrations not detectable by targeted NIPT (Table 1). The proportion of chromosome aberrations in each cFTS risk group is seen in Figure 1.    (Table 3). The number of atypical aberrations, miscarriage/IUFD, and malformations detected by ultrasound depending on cFTS risk group category

| DISCUSSION
In previous studies describing the residual risk for an atypical chromosome aberration, the pregnant women with an increased risk after cFTS have been offered a full karyotype or CMA analysis prenatally. [3][4][5][6] Hence, the majority of pregnancies where an atypical aberration was detected presumably had a termination and it will be The NIPT-analysis performance is improving continuously and when CNV detection based on cfDNA is comparable to CMA performance, the need for an invasive test will dissolve. Detection of segmental imbalances as well as submicroscopic CNVs has been reported and is feasible for some NIPT platforms. [13][14][15][16][17][18][19][20][21][22][23][24][25] Yet, the resolution of CNV detection based on cfDNA analysis is today not comparable to CMA and a large part of the atypical chromosome aberrations in our cohort would have been missed.
We also investigated the detection rates of all types of chromo- with an offer of low-dose aspirin to women at high-risk and thereby reducing the risk of preterm PE. 26 It seems apparent that first trimester biochemistry is here to stay for the near future and also that the procedure-related risk following invasive testing is considerably lower than previously believed. [27][28][29][30] Increased NT seems to have a stronger association with fetal abnormalities and genetic syndromes in general rather than to be associated with atypical chromosome aberrations. 31 It has also been proposed that the presence of pathogenic CNVs in association with high fetal NT are due to the presence of other malformations rather than to the high fetal NT per se. 32 In a large population-based study of 1.3 million pregnant women the investigators found that 17% of chromosome aberrations were not detectable by noninvasive testing and that 35% of atypical aberrations had an increased risk (>1:100) following cFTS which is consistent with our results with increased NT >3.5 mm in 18% of pathogenic CNVs, and 55% of atypical aberrations having an increased risk (>1:200). 4 Clearly, the combined first trimester test including biochemical markers, seems more sensitive in terms of detecting the more uncommon atypical chromosome aberrations compared to only using a NT above 3.5 mm as an indication for further investigation.
Limitations to our study is that malformations were not recorded at the time of the first trimester scan and that miscarriages before 22 weeks may have occurred without receiving a genetic diagnosis.
The fact that strategies to offer prenatal diagnosis varied during the study period and that CMA was used first after 2012 is not believed to have affected results as postnatal follow-up of children was virtually complete through the Swedish Pregnancy Registry.
In conclusion, our data in this study support that we should keep the cFTS examination and use it as a window of opportunity to not only detect common aneuploidies but also atypical chromosome aberrations. These are more uncommon but often cause more morbidity compared to Down syndrome.