Abstract
Objective
To determine the perinatal outcomes of fetuses diagnosed with a pathogenic copy number variant (CNV) or variant of uncertain significance (VUS); and to characterize these variants in terms of testing indication, genomic location, size, and inheritance.
Methods
Retrospective study of singleton pregnancies with a pathogenic CNV or VUS from a single laboratory during 2012–2018. Probabilistic record linkage between the prenatal diagnosis dataset and perinatal outcome data for births from 20 weeks gestation was performed. If no birth record was found, this implied a pregnancy loss <20 weeks.
Results
We included 6945 prenatal microarray results; a pathogenic CNV was detected in 230 (3.3%, 95% CI: 2.9%–3.8%) and a VUS in 483 (7.0%, 95% CI: 6.4%–7.6%). Of pregnancies with a pathogenic CNV, 20.0% (95% CI: 15.3%–25.6%) had a live birth, 3.0% (95% CI: 1.5%–6.2%) had a perinatal death (stillbirth or neonatal death), and 77% (95% CI: 71.1%–81.9%) had no birth record. Of those with a VUS, 64.4% (95% CI: 60.0%–68.5%) had a live birth, 1.8% (95% CI: 1.0%–3.5%) had a perinatal death, and no birth record was found for 33.7% (95% CI: 29.7%–38.1%). Most pathogenic CNVs (61.1%) were <7 Mb in size. The most common microdeletion syndromes were DiGeorge, Wolf‐Hirschhorn, and Cri‐du‐chat syndromes.
Conclusion
This study provides an overview of perinatal outcomes and frequency of recurrent CNVs observed in the prenatal microarray era.
Key points
What is already known about this topic?
Fetal copy number variants (CNVs) detected on chromosomal microarray have a wide range of clinical significance.
Large studies reporting perinatal outcome rates after a prenatal diagnosis of a CNV are rare.
What does this study add?
A live birth was recorded for 20.0% of fetuses with a pathogenic genomic CNV and for 64.4% of fetuses with a variant of uncertain significance.
The live birth rate was higher in pregnancies without an ultrasound abnormality and in those with an inherited CNV.
1. INTRODUCTION
The routine use of chromosomal microarray (CMA) in prenatal diagnosis (chorionic villus sampling [CVS] and amniocentesis) has become the standard of care in many high income settings because of its superior diagnostic yield over G‐banded karyotyping. 1 , 2 , 3 In a 2012 landmark trial, CMA identified a clinically relevant copy number variant (CNV) in 6.0% of fetuses with an ultrasound abnormality and a normal karyotype. 3 Consequently, professional guidelines now recommended CMA as a first‐tier diagnostic tool to investigate fetuses with a structural abnormality. 4 , 5 , 6
In addition to pathogenic copy number variants (pCNVs), CMA also detects variants of uncertain or unknown significance (VUS). These present specific genetic counseling challenges due to lack of a prenatal phenotype to guide variant interpretation, limited or conflicting evidence regarding pathogenicity, and need for parental testing to assist interpretation. Variable expressivity and incomplete penetrance can also make it difficult to predict the postnatal phenotype. Also, as new scientific and clinical evidence arises, VUS maybe re‐classified as pathogenic or benign. 7 , 8 , 9 Prenatal VUS therefore can create significant clinical workload, despite having a less severe prognosis than pCNVs.
There is a dearth of current literature on perinatal outcomes of fetal pCNVs and VUS, in terms of the number of live births and perinatal deaths. Published studies are limited by small sample sizes, selection bias due to limited indications for prenatal diagnosis, and incomplete reporting of perinatal outcomes. 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 Perinatal outcomes are likely influenced by indication for prenatal diagnosis, such as an ultrasound abnormality, as well as genomic location and CNV size. Information on perinatal outcomes is important for patient counseling, health service planning, and understanding reproductive choices in our population.
In this study, we aimed (i) to determine the perinatal outcomes of fetuses with pCNVs or VUS by performing individual record linkage between a large prenatal diagnostic dataset and state government birth data and (ii) to characterize these prenatal CNVs in terms of testing indication, genomic location, size, and mode of inheritance.
2. METHODS
2.1. Study population
This is a retrospective cohort study of all women resident in the Australian state of Victoria who had a prenatal diagnostic sample analyzed by the Victorian Clinical Genetics Services (VCGS) from January 2012 to December 2018 inclusive. The VCGS is a not‐for‐profit laboratory with a wide referral base of tertiary maternity hospitals, private ultrasound practices, and regional centers in Victoria.
All pregnant women in Australia are offered screening for fetal chromosome and structural anomalies. Victoria has approximately 73,000 births per year with a median maternal age of 31.6 years. 20 The uptake of prenatal screening is 83.6%, with at least 20.2% of women electing to self‐fund non‐invasive prenatal testing (NIPT) as a primary screening test. 21 First trimester combined screening, second trimester serum screening, and mid‐trimester morphology scan are government‐subsidized but involve variable out‐of‐pocket costs to the pregnant woman. Access to genetic counseling and prenatal diagnostic tests are provided through public and private fetal medicine units. State‐wide, between 2012 and 2018, 65.6% of all prenatal diagnostic tests were analyzed by CMA. 22 Termination of pregnancy (TOP) is lawful on maternal request up to 24 weeks and after 24 weeks if two medical practitioners deem it appropriate considering the women's physical and mental wellbeing.
2.2. Data sources
2.2.1. Prenatal diagnosis results from the Victorian Clinical Genetics Service
Prenatal CMA results over the 7‐year study period were reviewed and all women with a prenatal diagnosis of a pCNV or VUS in a singleton pregnancy were included. Prenatal CNVs were classified in accordance with established guidelines. 23 , 24 In this study, we use the term “pCNV” to refer to pathogenic and likely pathogenic CNVs and “VUS” to denote CNVs of uncertain or unknown clinical significance. Pregnancies with a CNV and concurrent whole chromosome aneuploidy, triploidy, or mosaic result were excluded (Figure 1). CMAs in use during the study period were high‐resolution whole genome single‐nucleotide polymorphism microarrays (Affymetrix CytoScan 750K array [Affymetrix], Illumina CytoSNP‐12 array, and Illumina Infinium GSA‐24 v1.0 [Illumina]).
FIGURE 1.
Study flowchart of fetuses with a copy number variant diagnosed from 2012 to 2018. The flowchart illustrates the selection of fetuses with a copy number variant of pathogenic or uncertain significance in this cohort. CMA, chromosomal microarray; CNV, copy number variant; FISH, fluorescent in situ hybridization; pCNV, pathogenic copy number variant; VUS, variant of uncertain or unknown significance.
Repeat diagnostic tests in the same pregnancy, for example, CVS followed by amniocentesis, were collapsed into a single record. Multiple pregnancies were excluded because of their higher baseline rates of miscarriage, and perinatal morbidity and mortality.
Ultrasound abnormalities included structural anomalies, “soft markers” (such as increased nuchal translucency and hypoplastic nasal bone), and abnormalities of fetal growth or amniotic fluid.
The genomic characteristics evaluated were size (in megabases), parent‐of‐origin status, and association with a haploinsufficient or dosage sensitive gene or a known microdeletion syndrome. CNVs were categorized as <7 or ≥7 Mb in size to correlate with the detection threshold for whole‐genome NIPT assays that screen for segmental aneuploidy. CNVs were labeled as “frequent” if they occurred at least five times, excluding clusters of familial CNVs.
2.2.2. Perinatal outcomes from the Victorian Perinatal Data Collection
The Victorian Perinatal Data Collection is a state government agency that is mandated to collect data on every birth in Victoria of at least 20 weeks' gestation or 400 g birth weight (if gestation is unknown). For this study, births up to 31 December 2018 were available for linkage. The prenatal diagnosis dataset was transferred to the Centre for Victorian Data Linkage in the Department of Health and Human Services to undergo individual probabilistic record linkage with the perinatal dataset to obtain perinatal outcomes.
The de‐identified prenatal diagnosis dataset contained the following variables for each record: year of prenatal diagnosis, prenatal sample type (CVS or amniocentesis), CNV classification, and testing indication. Linkage to the perinatal dataset was attempted for each prenatal diagnosis record to obtain perinatal outcome, birth weight, gestational age group at birth, and infant sex (Figure S1).
Perinatal outcomes were coded as live births (discharged from place of birth or transferred to another hospital and survived at least 28 days) or perinatal deaths (stillbirth [≥20 weeks'] or neonatal death [death <28 days of life]).
TOP after 20 weeks is recorded by the Victorian Perinatal Data Collection unit as a stillbirth. Low birth weight was defined as <2.5 kg in an infant of at least 37 completed weeks of gestation. Pre‐term birth was defined as birth before 37 completed weeks of gestation. If no birth record was identified via linkage, this was considered a presumptive pregnancy loss prior to 20 weeks (miscarriage or TOP) or maternal migration out of Victoria before birth. Interstate and overseas migration from Victoria was 2.0% per annum for residents aged 20–39 years in 2018. 25
2.3. Statistical analysis
The chi‐square test was used to assess differences in proportions in Stata version 17 (StataCorp, 2021) 26 . Wilson score method was used to calculate 95% confidence intervals using Epitools (Sergeant, 2018) 27 .
2.4. Ethics approval
This study received Human Research Ethics Committee approval from the Royal Children's Hospital HREC (Reference No. 2019.038) on April 9, 2019. A waiver for the requirement for patient consent as well as use and retainment of personal identifiers for the purpose of record linkage was granted in accordance with the National Health and Medical Research Council National Statement on Ethical Conduct in Human Research 2007 section 2.3.10.
3. RESULTS
During the 7‐year study period, 6945 prenatal diagnosis samples were analyzed by CMA; a fetal CNV was detected in 713 samples (10.3%, 95% CI: 9.6%–11.0%), including 230 pCNVs (3.3%, 95% CI: 2.9%–3.8%) and 483 VUS (7.0%, 95% CI: 6.4%–7.6%) (Figure 1). Median gestational age at prenatal diagnosis in the CNV group was 16 weeks (range 10–35 weeks), and median maternal age was 33 years (range 18–47 years). The most common indications for prenatal diagnosis were ultrasound abnormality (59.7%, 95% CI: 56.1%–63.3%) and positive (“high risk”) first trimester combined screening result (22.7%, 95% CI: 19.8%–25.9%) (Table 1).
TABLE 1.
Indications for prenatal diagnostic testing for 713 fetuses with a copy number variant
| Indications | pCNV (N = 230 fetuses, 279 indications a ) | VUS (N = 483 fetuses, 548 indications a ) | Total (N = 713 fetuses, 827 indications a ) |
|---|---|---|---|
| n (% fetuses) | n (% fetuses) | n (% fetuses) | |
| Ultrasound abnormality b | 156 (67.8) | 270 (55.9) | 426 (59.7) |
| First trimester combined screening | 31 (13.5) | 131 (27.1) | 162 (22.7) |
| Family history chromosomal condition | 33 (14.3) | 36 (7.5) | 69 (9.7) |
| Single gene condition | 16 (7.0) | 23 (4.8) | 39 (5.5) |
| Non‐invasive prenatal test | 20 (8.7) | 15 (3.1) | 35 (4.9) |
| Second trimester screening | 4 (1.7) | 20 (4.1) | 24 (3.4) |
| Other testing indications | 19 (8.3) | 53 (11.0) | 72 (10.1) |
Abbreviations: pCNV, pathogenic copy number variant; VUS, variant of uncertain or unknown clinical significance.
More than one indication could be recorded for each pregnancy. There were 827 indications recorded for 713 pregnancies, hence the totals exceed 100%.
Ultrasound abnormality included a soft marker on ultrasound such as increased nuchal translucency and hypoplastic nasal bone.
3.1. Perinatal outcomes
Record linkage identified a perinatal outcome for 373/713 (52.3%) fetuses with a CNV. There was no birth record for 177/230 fetuses with a pCNV (77.0%, 95% CI: 71.1%–81.9%) and 163/483 fetuses with a VUS (33.7%, 95% CI: 29.7%–38.1%) (Figure 2). Overall, there were more male live births than female, 202 (56.6%) versus 155 (43.4%).
FIGURE 2.
Birth outcomes of fetal copy number variants. The flowchart depicts the birth outcomes for fetuses with a copy number variant of pathogenic or uncertain significance including the number with a prenatal ultrasound abnormality.
Of the 230 fetuses with a pCNV, 46 (20.0%, 95% CI: 15.3%–25.6%) were live born and 7 (3.0%, 95% CI: 1.5%–6.2%) were perinatal deaths (Figure 2A). Of the 46 live born infants, 10 (21.7%, 95% CI: 12.3%–35.6%) were born preterm and 17 (37.0%, 95% CI: 24.5%–51.4%) were of low birthweight (Table 2).
TABLE 2.
Rate of low birthweight in live born infants by gestational age
| Gestational age at birth | Birthweight, n (%) | Total n (%) | |
|---|---|---|---|
| <2.5 kg | ≥2.5 kg | ||
| pCNV | |||
| Preterm (<37 weeks) | 6 (60.0) | <5 a (<50.0) | 10 (100.0) |
| Term (≥37 weeks) | 11 (30.6) | 25 (69.4) | 36 (100.0) |
| Total | 17 | 29 | 46 |
| VUS | |||
| Preterm (<37 weeks) | 35 (71.4) | 14 (28.6) | 49 (100.0) |
| Term (≥37 weeks) | 14 (5.7) | 247 (94.3) | 262 (100.0) |
| Total | 50 | 261 | 311 |
Abbreviations: pCNV, pathogenic copy number variant; VUS, variant of uncertain/unknown significance.
Categories with 1–4 affected pregnancies are reported as “<5” to protect patient privacy according to the conditions of the data release from the Victorian Department of Health.
Of the fetuses with a VUS, 311 (64.4%, 95% CI: 60.0%–68.5%) were live born and 9 (1.9%, 95% CI: 1.0%–3.5%) were perinatal deaths (Figure 2B). Of the 311 live born infants, 49 (15.8%, 95% CI: 12.1%–20.2%) were born preterm and 50 (16.1%, 95% CI: 12.4%–20.6%) were of low birthweight (Table 2).
Fetuses with a pCNV had a lower live birth rate compared with those with a VUS (20.0% vs. 64.4%, χ 2 = 122.8, p < 0.001). Live infants were more likely to be born preterm in the presence of a pCNV compared with VUS (21.7% vs. 15.8%, χ 2 = 1.04, p = 0.308). The proportion of term infants that had low birth weight was higher in those with a pCNV compared with a VUS (37.0% vs. 16.1%, p < 0.001).
3.1.1. Live birth rates for fetuses with an ultrasound abnormality and a CNV
Among fetuses with a pCNV, those without an ultrasound abnormality were more likely to be live born than those with an ultrasound abnormality (25.6% vs. 16.4%, respectively, p = 0.091) (Table 3).
TABLE 3.
Birth outcomes of pregnancies with a copy number variant by presence of ultrasound abnormality
| Indication for prenatal diagnosis | Birth outcome, n (%) | Total n, (%) | |
|---|---|---|---|
| Live birth | Stillbirth, neonatal death, or no birth record | ||
| pCNV | |||
| Ultrasound abnormality a | 23 (16.4) | 117 (83.6) | 140 (100.0) |
| All other indications | 23 (25.6) | 67 (74.4) | 90 (100.0) |
| Total | 46 | 184 | 230 |
| VUS | |||
| Ultrasound abnormality a | 145 (57.8) | 106 (42.2) | 251 (100.0) |
| All other indications | 166 (71.6) | 66 (28.4) | 232 (100.0) |
| Total | 311 | 172 | 483 |
| Total CNVs | |||
| Ultrasound abnormality a | 168 (43.0) | 223 (57.0) | 391 (100.0) |
| All other indications | 189 (58.7) | 133 (41.3) | 322 (100.0) |
| Total | 357 | 356 | 713 |
Abbreviations: CNV, copy number variant; pCNV, pathogenic copy number variant; VUS, variant of uncertain/unknown significance.
Prenatal ultrasound abnormality included structural anomalies and soft markers (increased nuchal translucency and absent nasal bone).
Similarly, among fetuses with a VUS, those without an ultrasound abnormality were more likely to be live born that those with an ultrasound abnormality (71.6% vs. 57.8%, respectively, p = 0.02). This difference in frequency of live birth in the presence of a VUS and an ultrasound abnormality was more pronounced than that seen with a pCNV.
3.1.2. Live birth rates for fetuses with an inherited CNV
Seventy‐two percent of pregnancies with a CNV had parent‐of‐origin testing (Table 4). Pregnancies with an inherited pCNV had a live birth outcome significantly more often than those with a de novo pCNV (33.3% vs. 17.2%, respectively, p = 0.038).
TABLE 4.
Birth outcomes by copy number variant inheritance status
| CNV origin | Outcome of pregnancy, n (%) | Total n (%) | |
|---|---|---|---|
| Live birth | Stillbirth, neonatal death or unknown | ||
| pCNV | |||
| Inherited | 20 (33.3) | 40 (66.7) | 60 (100.0) |
| De novo | 11 (17.2) | 53 (82.8) | 64 (100.0) |
| Undetermined | 15 (14.2) | 91 (85.8) | 106 (100.0) |
| Total pCNVs | 46 | 184 | 230 |
| VUS | |||
| Inherited | 240 (69.4) | 106 (30.6) | 346 (100.0) |
| De novo | 24 (52.2) | 22 (47.8) | 46 (100.0) |
| Undetermined | 47 (51.6) | 44 (48.4) | 91 (100.0) |
| Total VUS | 311 | 172 | 483 |
| Total CNVs | |||
| Inherited | 260 (64.0) | 146 (36.0) | 406 (100.0) |
| De novo | 35 (31.8) | 75 (68.2) | 110 (100.0) |
| Undetermined | 62 (31.5) | 135 (68.5) | 197 (100.0) |
| Total CNVs | 357 | 356 | 713 |
Abbreviations: CNV, copy number variant; pCNV, pathogenic copy number variant; VUS, variant of uncertain/unknown significance.
Similarly, among fetuses with a VUS, those with an inherited VUS had a live birth outcome significantly more often than those with a de novo VUS (69.4% vs. 52.2%, respectively, p = 0.02).
3.2. Range and genomic characteristics of CNVs
Among the 230 pregnancies with a pCNV, 68 had 2 or more CNVs, resulting in a total of 316 individual pCNVs. Of these pCNVs, 70.6% were deletions and 23.1% were inherited (Table 5). Most of these inherited pCNVs (60.3%, 41/68) involved both a deletion and a duplication, suggesting an unbalanced structural rearrangement that would not appear in the parent.
TABLE 5.
Genomic characteristics of individual copy number variants
| CNV group | Dosage, n (%) | Size, n (%) | Inheritance, n (%) | Total CNVs, n (%) | |||||
|---|---|---|---|---|---|---|---|---|---|
| Del | Dup | <7 Mb | ≥7 Mb | Mat | Pat | De novo | Unknown | ||
| pCNVs (n = 230 fetuses) | 223 (70.6) | 93 (29.4) | 193 (61.1) | 123 (38.9) | 49 (15.5) a | 24 (7.6) | 81 (25.6) | 162 (51.3) | 316 (100.0) |
| VUS (n = 483 fetuses) | 182 (34.6) | 344 (65.4) | 518 (98.5) | 8 (1.5) | 202 (38.4) a | 171 (32.5) | 54 (10.3) | 99 (18.8) b | 526 (100.0) |
| Total CNVs (n = 713 fetuses) | 405 (48.1) | 437 (51.9) | 711 (84.4) | 131 (15.6) | 251 (29.8) | 195 (23.2) | 135 (16.0) | 261 (31.0) | 842 (100.0) |
Abbreviations: CNV, copy number variant; del, deletion; dup, duplication; mat, maternal; pat, paternal; pCNV, pathogenic copy number variant; VUS, variant of uncertain/unknown significance.
Includes a CNV of either maternal or paternal inheritance.
Includes a CNV where inheritance status was withheld.
Among the 483 pregnancies with a VUS, 38 had 2 or more CNVs, resulting in a total of 526 individual VUS. Of these VUS, 65.4% were duplications and 70.9% were inherited.
3.2.1. Most frequent pCNVs
The majority of pCNVs (61.1%) were <7 Mb in size (Table 5). A positive NIPT result was the indication for testing in 5.2% (10/193) of pregnancies with a pCNV <7 MB and 14.6% (18/123) of those with pCNV ≥7 Mb (Table 6). The NIPT results in these 28 pregnancies were positive for suspected segmental aneuploidy (n = 17), suspected autosomal or sex chromosomal aneuploidy (n = 8), and “inconclusive for sex chromosomes” (n = 1). Two NIPT results were unspecified.
TABLE 6.
Indications for prenatal diagnostic testing for fetuses with a pathogenic copy number variant by genomic size
| Indications | pCNV <7 Mb (N = 193 pCNVs, 240 indications), n (% a ) | pCNV ≥7 Mb (N = 123 pCNVs, 157 indications), n (% a ) | Total (N = 316 PCNVs, 397 indications), n (% a ) |
|---|---|---|---|
| Ultrasound abnormality b | 137 (71.0) | 80 (65.0) | 227 (71.8) |
| History chromosomal condition | 25 (13.0) | 22 (17.9) | 47 (14.9) |
| First trimester combined screening | 25 (13.0) | 20 (16.3) | 45 (14.2) |
| Non‐invasive prenatal test | 10 (5.2) | 18 (14.6) | 28 (8.9) |
| Other testing indications | 43 (22.3) | 7 (5.7) | 50 (15.8) |
Abbreviation: pCNV, pathogenic copy number variant.
More than one indication could be recorded for each fetus with a pCNV. There were 397 indications recorded for 316 pCNVs, hence the totals exceed 100%.
Prenatal ultrasound abnormality included structural anomalies and soft markers (increased nuchal translucency and absent nasal bone).
The most frequent pCNV was 22q11.2 deletion (DiGeorge syndrome), accounting for 13.5% (31/230) of all pregnancies with a pCNV (Table 7).
TABLE 7.
Most frequent pathogenic copy number variants
| CNV region | Gene content (syndrome, OMIM #) | Coordinate range of included cases (GRCh37/hg19) | Size range (Mb) | Del/dup | n | Inheritance | Indication for prenatal diagnostic testing | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Mat or pat | De novo | Und | US abnormality | Positive screening result a | Other | ||||||
|
Multiple (DiGeorge syndrome, 188400) | 18,648,855–21,811,991 | 2.15–3.15 | Del | 31 | 1 | 19 | 11 | 25 | 5 | 1 |
| 4p16.3 | WHSCR (Wolf‐Hirschhorn syndrome, 194190) | 35,030–24,837,342 | 2.3–24.8 | Del | 9 | 0 | 4 | 5 | 5 | 2 | 2 |
| 5p15.33 | Multiple (Cri‐du‐Chat syndrome, 123450) | 38,139–33,799,699 | 8.9–33.8 | Del | 7 b | 2 | 1 | 4 | 5 | 1 | 1 |
| 17p12 | PMP22 (Hereditary liability to pressure palsies, 162500) | 14,063,251–15,457,054 | 1.31–1.38 | Del | 6 | 3 | 0 | 3 | 2 | 2 | 2 |
| Xp21.1 | DMD | 31,750,506–32,244,312 | 0.08–0.22 | Del | 5 | 2 | 0 | 3 | 1 | 1 | 3 |
Abbreviations: CNV, copy number variant; Del, deletion; dup, duplication; mat, maternal; pat, paternal; und, undetermined; US, ultrasound.
Positive (“high risk”) screening result included non‐invasive prenatal testing, combined first trimester screening, and second trimester serum screening.
Two cases of 5p13.33 were from the same paternally derived translocation t(1;5)(q31.1;p15.1).
The second most frequent pCNV was the 4p16.3 deletion (Wolf‐Hirschhorn syndrome), comprising 3.0% (9/230) of pregnancies with a pCNV. Five of these cases presented with ultrasound anomalies, including fetal growth restriction, small or absent kidneys, and fluid abnormalities (oligohydramnios, hydrops, cystic hygroma, and pleural effusion).
The third most frequent pCNV was 5p15.33 deletion (Cri‐du‐chat syndrome), accounting for 3.0% (7/230) of pregnancies with a pCNV. Ultrasound anomalies were recorded in five cases including ventral septal defects, cystic hygroma, ventriculomegaly, microcephaly, cerebellar hypoplasia, and omphalocele.
3.2.2. Most frequent VUS
The most frequent VUS was the 15q11.2 deletion, representing 6.4% (31/483) of all pregnancies with a VUS (Table 8). 15q11.2 is a known neuro‐susceptibility locus between low copy number repeat regions proximal to the Prader‐Willi and Angelman syndrome region. The majority of 15q11.2 deletions were inherited (23/31, 74.2%) and were incidentally detected following a positive first trimester combined screening, NIPT, or second trimester serum screening result.
TABLE 8.
Most frequent copy number variants of uncertain significance
| CNV region | Gene content | Coordinate range of included cases (GRCh37/hg19) | Size range (Mb) | Del/dup | n | Inheritance | Indication for prenatal diagnostic testing | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Mat or pat | De novo | Und | US abnormality | Positive screening result a | Other | ||||||
| 15q11.2 | NIPA1, NIPA2, CYFIP1, TUBGCP5 | 22,754,322–23,625,785 | 0.38–0.86 | Del | 31 | 23 | 0 | 8 | 11 | 16 | 4 |
| 22q11.2 | Proximal 22q11.2 recurrent region includes TBX1 | 18,631,364–21,947,644 | 2.54–3.3 | Dup | 10 | 5 | 5 | 0 | 4 | 6 | 0 |
| 18,294,211–18,626,637 | 0.33 | Dup | 1 | 1 | 0 | 0 | 0 | 1 | 0 | ||
| Distal 22q11.2 recurrent region | 20,225,536–21,462,353 | 0.43–1.02 | Dup | 5 | 3 | 0 | 2 | 3 | 1 | 1 | |
| 1q21.1 | Recurrent region includes RBM8A | 145,382,123–145,988,238 | 0.38–0.61 | Dup | 5 | 4 | 1 | 0 | 4 | 1 | 0 |
| 1q21.1q21.2 | Recurrent region includes GJA5 | 145,416,559–147,929,115 | 1.32–2.41 | Dup | 7 | 2 | 2 | 3 | 6 | 0 | 1 |
| 16p11.2 | Recurrent region includes SH2B1 | 28,371,467–29,339,889 | 0.21–0.97 | Dup | 7 | 5 | 1 | 0 | 4 | 2 | 1 |
| Recurrent region includes TBX6 | 29,351,826–30,302,348 | 0.57–0.95 | Dup | 3 | 1 | 1 | 1 | 0 | 1 | 2 | |
| 1q21.1 | Recurrent region includes GJA5 | 146,470,899–147,828,939 | 0.25–1.35 | Del | 7 | 5 | 0 | 2 | 2 | 3 | 2 |
| Recurrent region includes RBM8A | 145,382,123–145,958,362 | 0.50–0.58 | Del | 2 | 2 | 0 | 0 | 1 | 0 | 1 | |
| 16p13.11 | Recurrent region includes NDE1, MYH11 | 14,892,975–18,196,549 | 0.74–2.70 | Dup | 7 | 5 | 1 | 1 | 3 | 3 | 1 |
| 16p12 | EEF2K | 21,841,353–22,442,007 | 0.45–0.60 | Del | 6 | 4 | 0 | 2 | 3 | 2 | 1 |
| 8p23 | CSMD1 | 3,686,944–5,962,953 | 2.22–2.26 | Dup | 5 b | 5 | 0 | 0 | 1 | 2 | 2 |
Abbreviations: CNV, copy number variant; Del, deletion; dup, duplication; mat, maternal; pat, paternal; und, undetermined; US, ultrasound.
Positive (“high risk”) screening result included non‐invasive prenatal testing, combined first trimester screening, and second trimester serum screening.
Three cases were different pregnancies from the same person.
The second most common VUS was the 22q11.21 duplication, representing 3.3% of all pregnancies with a VUS (10/483). This microduplication is in the same region as 22q11.21 deletion syndrome and is associated with variable expressivity.
4. DISCUSSION
4.1. Principal findings
Our large prenatal CMA dataset with linkage to state‐wide perinatal outcomes demonstrated that one in five fetuses with a pCNV resulted in a live birth, suggesting a pregnancy loss rate, either through TOP or miscarriage, as high as 80%. Among live born infants, a prenatal diagnosis of a pCNV was also associated with rates of preterm birth (21.7%) and low birthweight (37.0%) that are several fold higher than the background population rates of 7.3% and 6.9% in Victoria, respectively. 28
4.2. Clinical implications in the context of what is known
4.2.1. Pathogenic CNVs
The prevalence of pCNVs in our cohort was 3.3%, which was similar to that reported in a published review of 23,865 pCNVs. 29 Our study of 230 pCNVs is the largest to include birth outcomes, with previous publications containing smaller cohorts (range 5–93) and/or only reporting perinatal outcomes for a small subset. 10 , 11 , 12 , 13 , 14 , 18 Our summary statistics on preterm birth and low birth weight provide general prognostic information that can be used for counseling couples about obstetric outcomes for pCNVs, particularly for rare pCNVs where data on obstetric outcomes are lacking and planning ongoing management in a high risk model of prenatal care. The association with term low birth weight corroborates findings from a case‐control study of placental genomic imbalances and small for gestational age fetuses. 30
The finding that a live birth outcome was more likely for an inherited pCNV suggests that parent‐of‐origin information influences couples' reproductive decisions. It is possible that an inherited pCNV may have a milder phenotype than a de novo pCNV, such as an X‐linked condition in a female fetus, which would influence the pregnancy outcome.
Our results also provide data to assess the theoretical clinical utility of expanded NIPT. The majority (83.2%) of pCNVs in our cohort were not one of the six common syndromes included in NIPT microdeletion panels. Furthermore, over half (61.1%) of pCNVs were <7 Mb which is below the resolution of many genome‐wide NIPT platforms. 31 , 32 These results suggest that most clinically relevant fetal pCNVs are currently undetectable by NIPT.
4.2.2. Variants of uncertain significance
Our cohort of 483 fetal VUS is the second largest reported to date. Our live birth rate for fetuses with a VUS (64%) was higher than for those with a pCNV (20%) and was within the range reported in other fetal VUS cohorts (35%–90%). 10 , 11 , 12 , 13 , 15 , 16 , 19 Similarly to Chen et al., who followed up 721 fetal VUS, we found that an inherited VUS was more likely to result in a live birth than a non‐inherited VUS. 17 However, unlike Chen et al., we were able to analyze our dataset to show that an ultrasound abnormality was significantly associated with lower live birth rates in the VUS cohort.
An ultrasound abnormality was found in 59.7% of our CNV cohort. Had CMA only been performed in pregnancies with an ultrasound abnormality, 27.9% (90/322, Table 3) of pCNVs in our cohort would have gone undetected before birth. 3
Our VUS prevalence of 7.0% was higher than some studies, 1 , 3 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 19 , 33 , 34 but comparable to 7.2% recorded in the Belgium national prenatal cohort. 33 The most common VUS in our study was the 15q11.2 deletion, which has a highly variable phenotype including neurodevelopmental delay, autism spectrum disorders, and obsessive‐compulsive disorders. 35 For this reason, we no longer report this penetrant CNV. The detection of susceptibility loci for neurodevelopmental conditions with reduced penetrance and variable expressivity could be viewed as one of the drawbacks of liberal application of CMA in prenatal diagnosis. Parents and health care professionals may not anticipate the psychological stress, anxiety, and emotional burden of having to make reproductive decisions based on uncertain results. 16 , 36 Providing couples a choice of whether to receive prenatal VUS results has been examined in various research settings but proven difficult to implement in routine clinical practice. 37
Recent studies suggest a higher diagnostic and clinical significance of low‐pass genome sequencing compared with CMA. 38 However, the improved resolution of sequencing increases the number of CNVs identified, potentially increasing the complexity of clinical interpretation and management. 39 In the future, it is likely that a sequencing‐based test will become the standard approach in clinical care. 40 , 41
4.3. Strengths and limitations
The prenatal cohort from the VCGS is drawn from a large geographical region from a variety of practice settings, including tertiary hospitals and private ultrasound practices. We were able to obtain complete ascertainment of registered births from 20 weeks' gestation through linkage to government datasets to allow us to confidently determine perinatal outcomes. This allowed us to create the first estimates of perinatal outcomes for fetal CNVs in a population where CMA is the standard test for prenatal diagnosis for a range of indications, and TOP is lawful without a fixed upper gestational age limit.
The major limitation of our study is that data on perinatal outcomes before 20 weeks were unavailable as this information is not routinely collected by government agencies. We could therefore only surmise that pregnancies with a missing birth record ended in either miscarriage or TOP before 20 weeks. Interstate and overseas migration out of Victoria is uncommon (2.0% p.a. 25 ) therefore we assume that maternal relocation would account for only a small amount of the missing birth records. Ideally, record linkage using a control group with normal CMA results would be done to help clarify such potential outcomes.
We were not able to analyze or report perinatal outcomes by specific CNVs due to ethical protections around potentially identifying information. We were also unable to calculate the rate of fetal growth restriction as exact weight and gestational age at birth were not provided by the Victorian Perinatal Data Collection.
4.4. Research implications
Perinatal outcomes for the common autosomal trisomies and selected structural anomalies were last reported for our population in the Victorian Congenital Anomalies Report, 2015–2016. 42 The live birth rate for trisomy 21 was 21.7% (107/492) in that 2‐year reporting period, which is similar to our findings here for the pCNV cohort. Reporting of perinatal outcomes for a wider range of genomic variants including sex chromosome conditions and pCNVs would provide a more complete picture of the impact of prenatal screening and diagnosis in the genomic era.
5. CONCLUSION
This linkage study provides an overview of perinatal outcomes in the era of widespread adoption of prenatal CMA. One in five fetuses with a pCNV resulted in a live birth; these infants had high rates of preterm birth and low birthweight. A live birth outcome was more likely if the CNV was inherited, and in the absence of an associated ultrasound abnormality. These data have important clinical implications for prenatal genetic counseling, maternity, and pediatric health services planning, and understanding the reproductive choices of our population.
CONFLICT OF INTEREST
The authors declare no potential conflict of interest.
Supporting information
Figure S1
ACKNOWLEDGMENTS
We thank the VCGS for contributing the prenatal diagnosis dataset. We are grateful to Consultative Council on Obstetric and Paediatric Mortality and Morbidity (CCOPMM) for providing access to the data used for this project and for the assistance of the staff at the Safer Care Victoria. The conclusions, findings, opinions, and views or recommendations expressed in this paper are strictly those of the authors. They do not necessarily reflect those of CCOPMM. Associate Professor Lisa Hui funded by National Health and Medical Research Council, Grant/Award Numbers: 1021252, 1105603; University of Melbourne Faculty of Medicine, Dentistry and Health Sciences Fellowship. The funding bodies had no role in the conduct of the research or the manuscript. This study received infrastructure support from the Murdoch Children's Research Institute.
Open access publishing facilitated by The University of Melbourne, as part of the Wiley ‐ The University of Melbourne agreement via the Council of Australian University Librarians.
Pynaker C, Norris F, Hui L, Halliday J. Perinatal outcomes and genomic characteristics of fetal copy number variants: an individual record linkage study of 713 pregnancies. Prenat Diagn. 2023;43(4):516‐526. 10.1002/pd.6305
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request, subject to approval from the Victorian Perinatal Data Collection data custodian at the Victorian Department of Health and the Royal Children's Hospital health research ethics committee.
REFERENCES
- 1. Hillman SC, McMullan DJ, Hall G, et al. Use of prenatal chromosomal microarray: prospective cohort study and systematic review and meta‐analysis. Ultrasound Obstet Gynecol. 2013;41(6):610‐620. 10.1002/uog.12464 [DOI] [PubMed] [Google Scholar]
- 2. Oneda B, Rauch A. Microarrays in prenatal diagnosis. Best Pract Res Clin Obstet Gynaecol. 2017;42:53‐63. 10.1016/j.bpobgyn.2017.01.003 [DOI] [PubMed] [Google Scholar]
- 3. Wapner RJ, Martin CL, Levy B, et al. Chromosomal microarray versus karyotyping for prenatal diagnosis. N Engl J Med. 2012;367(23):2175‐2184. 10.1056/nejmoa1203382 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Committee Opinion No. 581: the use of chromosomal microarray analysis in prenatal diagnosis. Obstet Gynecol. 2013;122(6):1374‐1377. [DOI] [PubMed] [Google Scholar]
- 5. American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins—Obstetrics; Committee on Genetics; Society for Maternal‐Fetal Medicine . Screening for fetal chromosomal abnormalities: ACOG Practice Bulletin, Number 226. Obstet Gynecol. 2020;136(4):e48‐e69. 10.1097/aog.0000000000004084 [DOI] [PubMed] [Google Scholar]
- 6. Committee Opinion No. 682: microarrays and next‐generation sequencing technology: the use of advanced genetic diagnostic tools in obstetrics and gynecology. Obstet Gynecol. 2016;128(6):e262‐e8. [DOI] [PubMed] [Google Scholar]
- 7. Westerfield L, Darilek S, van den Veyver IB. Counseling challenges with variants of uncertain significance and incidental findings in prenatal genetic screening and diagnosis. J Clin Med. 2014;3(3):1018‐1032. 10.3390/jcm3031018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405‐424. 10.1038/gim.2015.30 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Shkedi‐Rafid S, Horton R, Lucassen A. What is the meaning of a ‘genomic result’ in the context of pregnancy? Eur J Hum Genet. 2021;29(2):225‐230. 10.1038/s41431-020-00722-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Li Y, Yan H, Chen J, et al. The application of late amniocentesis: a retrospective study in a tertiary fetal medicine center in China. BMC Pregnancy Childbirth. 2021;21(1):266. 10.1186/s12884-021-03723-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Stern S, Hacohen N, Meiner V, et al. Universal chromosomal microarray analysis reveals high proportion of copy‐number variants in low‐risk pregnancies. Ultrasound Obstet Gynecol. 2021;57(5):813‐820. 10.1002/uog.22026 [DOI] [PubMed] [Google Scholar]
- 12. Wang J, Chen L, Zhou C, et al. Prospective chromosome analysis of 3429 amniocentesis samples in China using copy number variation sequencing. Am J Obstet Gynecol. 2018;219(3):287.e1‐287.e18. 10.1016/j.ajog.2018.05.030 [DOI] [PubMed] [Google Scholar]
- 13. Egloff M, Hervé B, Quibel T, et al. Diagnostic yield of chromosomal microarray analysis in fetuses with isolated increased nuchal translucency: a French multicenter study. Ultrasound Obstet Gynecol. 2018;52(6):715‐721. 10.1002/uog.18928 [DOI] [PubMed] [Google Scholar]
- 14. Li L, Fu F, Li R, Liu Z, Liao C. Prenatal diagnosis and pregnancy outcome analysis of thickened nuchal fold in the second trimester. Medicine (Baltim). 2018;97(46):e13334. 10.1097/md.0000000000013334 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Cai M, Lin N, Su L, et al. Copy number variations in ultrasonically abnormal late pregnancy fetuses with normal karyotypes. Sci Rep. 2020;10(1):15094. 10.1038/s41598-020-72157-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Brabbing‐Goldstein D, Reches A, Svirsky R, Bar‐Shira A, Yaron Y. Dilemmas in genetic counseling for low‐penetrance neuro‐susceptibility loci detected on prenatal chromosomal microarray analysis. Am J Obstet Gynecol. 2018;218(2):247.e1‐247.e12. 10.1016/j.ajog.2017.11.559 [DOI] [PubMed] [Google Scholar]
- 17. Chen L, Wang L, Yin D, Zeng Y, Tang F, Wang J. Influence of the detection of parent‐of‐origin on the pregnancy outcomes of fetuses with copy number variation of unknown significance. Sci Rep. 2020;10(1):8864. 10.1038/s41598-020-65904-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Muys J, Jacquemyn Y, Blaumeiser B, et al. Prenatally detected copy number variants in a national cohort: a postnatal follow‐up study. Prenat Diagn. 2020;40(10):1272‐1283. 10.1002/pd.5751 [DOI] [PubMed] [Google Scholar]
- 19. Mardy AH, Wiita AP, Wayman BV, Drexler K, Sparks TN, Norton ME. Variants of uncertain significance in prenatal microarrays: a retrospective cohort study. BJOG. 2021;128(2):431‐438. 10.1111/1471-0528.16427 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Australian Bureau of Statistics . Births, Australia. Accessed January 14, 2023. https://www.abs.gov.au/statistics/people/population/births‐australia/latest‐release#states‐and‐territories
- 21. Lindquist A, Hui L, Poulton A, et al. State‐wide utilization and performance of traditional and cell‐free DNA‐based prenatal testing pathways: the Victorian Perinatal Record Linkage (PeRL) study. Ultrasound Obstet Gynecol. 2020;56(2):215‐224. 10.1002/uog.21899 [DOI] [PubMed] [Google Scholar]
- 22. Pynaker C, Hui L, Halliday J. The Annual Report on Prenatal Diagnostic Testing in Victoria, 2021. Murdoch Childrens Research Institute. 2022. Accessed January 14, 2023. https://mcri.figshare.com/articles/report/The_Annual_Report_on_Prenatal_Diagnostic_Testing_in_Victoria_2021/20479491/1 [Google Scholar]
- 23. Kearney HM, Thorland EC, Brown KK, Quintero‐Rivera F, South ST. American College of Medical Genetics standards and guidelines for interpretation and reporting of postnatal constitutional copy number variants. Genet Med. 2011;13(7):680‐685. 10.1097/gim.0b013e3182217a3a [DOI] [PubMed] [Google Scholar]
- 24. South ST, Lee C, Lamb AN, Higgins AW, Kearney HM. ACMG Standards and Guidelines for constitutional cytogenomic microarray analysis, including postnatal and prenatal applications: revision 2013. Genet Med. 2013;15(11):901‐909. 10.1038/gim.2013.129 [DOI] [PubMed] [Google Scholar]
- 25. Australian Bureau of Statistics . Net Internal and Overseas Migration Estimates by Region (SA2 and above) and Age [Internet]. Australian Bureau of Statistics; 2018. [Google Scholar]
- 26. StataCorp . Stata Statistical Software: Release 17. StataCorp LLC. 2021. [Google Scholar]
- 27. Sergeant, ESG . Epitools Epidemiological Calculators. Ausvet. 2018. http://epitools.ausvet.com.au [Google Scholar]
- 28. Morbidity CCoOaPMa . Victoria’s Mothers Babies and Children 2019 Report and Presentations. CCOPMM; 2019. [Google Scholar]
- 29. Chau MHK, Cao Y, Kwok YKY, et al. Characteristics and mode of inheritance of pathogenic copy number variants in prenatal diagnosis. Am J Obstet Gynecol. 2019;221(5):493.e1‐493.e11. 10.1016/j.ajog.2019.06.007 [DOI] [PubMed] [Google Scholar]
- 30. Del Gobbo GF, Yin Y, Choufani S, et al. Genomic imbalances in the placenta are associated with poor fetal growth. Mol Med. 2021;27(1):3. 10.1186/s10020-020-00253-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Lo KK, Karampetsou E, Boustred C, et al. Limited clinical utility of non‐invasive prenatal testing for subchromosomal abnormalities. Am J Hum Genet. 2016;98(1):34‐44. 10.1016/j.ajhg.2015.11.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Li R, Wan J, Zhang Y, et al. Detection of fetal copy number variants by non‐invasive prenatal testing for common aneuploidies. Ultrasound Obstet Gynecol. 2016;47(1):53‐57. 10.1002/uog.14911 [DOI] [PubMed] [Google Scholar]
- 33. Muys J, Blaumeiser B, Jacquemyn Y, et al. The Belgian MicroArray Prenatal (BEMAPRE) database: a systematic nationwide repository of fetal genomic aberrations. Prenat Diagn. 2018;38(13):1120‐1128. 10.1002/pd.5373 [DOI] [PubMed] [Google Scholar]
- 34. Breman A, Pursley AN, Hixson P, et al. Prenatal chromosomal microarray analysis in a diagnostic laboratory; experience with >1000 cases and review of the literature. Prenat Diagn. 2012;32(4):351‐361. 10.1002/pd.3861 [DOI] [PubMed] [Google Scholar]
- 35. Picinelli C, Lintas C, Piras IS, et al. Recurrent 15q11.2 BP1‐BP2 microdeletions and microduplications in the etiology of neurodevelopmental disorders. Am J Med Genet B Neuropsychiatr Genet. 2016;171(8):1088‐1098. 10.1002/ajmg.b.32480 [DOI] [PubMed] [Google Scholar]
- 36. Desai P, Haber H, Bulafka J, et al. Impacts of variants of uncertain significance on parental perceptions of children after prenatal chromosome microarray testing. Prenat Diagn. 2018;38(10):740‐747. 10.1002/pd.5323 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Halliday JL, Muller C, Charles T, et al. Offering pregnant women different levels of genetic information from prenatal chromosome microarray: a prospective study. Eur J Hum Genet. 2018;26(4):485‐494. 10.1038/s41431-017-0084-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Chaubey A, Shenoy S, Mathur A, et al. Low‐pass genome sequencing: validation and diagnostic utility from 409 clinical cases of low‐pass genome sequencing for the detection of copy number variants to replace constitutional microarray. J Mol Diagn. 2020;22(6):823‐840. 10.1016/j.jmoldx.2020.03.008 [DOI] [PubMed] [Google Scholar]
- 39. Chau MHK, Qian J, Chen Z, et al. Trio‐based low‐pass genome sequencing reveals characteristics and significance of rare copy number variants in prenatal diagnosis. Front Genet. 2021;12:742325. 10.3389/fgene.2021.742325 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Gross AM, Ajay SS, Rajan V, et al. Copy‐number variants in clinical genome sequencing: deployment and interpretation for rare and undiagnosed disease. Genet Med. 2019;21(5):1121‐1130. 10.1038/s41436-018-0295-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Royer‐Bertrand B, Cisarova K, Niel‐Butschi F, Mittaz‐Crettol L, Fodstad H, Superti‐Furga A. CNV detection from exome sequencing data in routine diagnostics of rare genetic disorders: opportunities and limitations. Genes (Basel). 2021;12(9):1427. 10.3390/genes12091427 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Congenital Anomalies in Victoria, 2015–2016. Consultative Council on Obstetric and Paediatric Mortality and Morbidity; 2018. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request, subject to approval from the Victorian Perinatal Data Collection data custodian at the Victorian Department of Health and the Royal Children's Hospital health research ethics committee.