Combined first-trimester screening and invasive diagnostics for atypical chromosomal aberrations: Danish nationwide data on prenatal profiles and detection compared with NIPT

OBJECTIVE

To examine the prenatal profiles of pregnancies affected by an atypical chromosomal aberration, focusing on pathogenic copy number variants (pCNVs). Further, we wanted to quantify the performance of combined first-trimester screening (cFTS) and a second-trimester anomaly scan in detecting these conditions. Finally, we aimed to estimate the consequences of a policy of using non-invasive prenatal testing (NIPT) rather than invasive testing with chromosomal microarray (CMA) to manage pregnancies identified as high risk from cFTS.

METHODS

A retrospective review of the Danish fetal medicine database identified all pregnant women who had cFTS and a trisomy 21 risk-assessment between January 1, 2008, and December 31, 2018. Chromosomal aberrations diagnosed prenatally, postnatally, or from fetal tissue following pregnancy loss or termination of pregnancy (TOP) were identified. Chromosomal aberrations were grouped into one of six categories: 1) Triploidy; 2) Common trisomies (trisomies 21, 18, and 13); 3) Monosomy X; 4) Other sex chromosome aberrations (SCAs); 5) pCNVs; and 6) Rare autosomal trisomies (RATs) and mosaicisms. The prevalence of each aberration-category was stratified by the individual cFTS markers and risk estimate, and the size of each pCNV diagnosed from CMA was calculated.

RESULTS

We included data on 565,708 pregnancies of which 3,982 were diagnosed with a fetal chromosomal aberration (0.70%). cFTS performed well in identifying triploidies (86%), monosomy X (92%), atypical SCAs (58%), and RATs and mosaicisms (70%). pCNVs comprised 28% (n = 1,091) of the chromosomal aberrations diagnosed overall, and the prevalence increased during the study period with more prenatal chromosomal microarray analysis being performed. In pregnancies with maternal age <30 years, NT <95th percentile, PAPP-A MoM ≥ 1, or trisomy 21 risk ≥1 in 1000, the prevalence of pCNVs significantly exceeded the prevalence of trisomies 21, 18, and 13. Pregnancies affected by a pCNV had significantly increased nuchal translucency thickness (NT) and decreased maternal biomarkers pregnancy associated plasma protein-A (PAPP-A) and β-human chorionic gonadotropin (β-hCG) compared with unaffected pregnancies. However, only 23% of these pregnancies screened positive from cFTS and 51% were not detected until after birth. Amongst high-risk pregnancies diagnosed with a chromosomal aberration, pCNVs comprised 14% and when other atypical aberrations were considered, conventional NIPT (screening for trisomies 21, 18, and 13, and monosomy X) would miss 28% of all pathogenic aberrations diagnosed following a high-risk cFTS result. Thus, 1 in 26 pregnancies at high-risk following cFTS would be affected by a chromosomal aberration despite a normal conventional NIPT result. In a contingent screening model with NIPT provided for the "intermediate" risk group (T21 risk of 1 in 100-300), 50% of the aberrations would be missed. In our cohort, 80% of the pCNVs diagnosed were <5Mb and therefore not detectable using current forms of "genome wide" NIPT.

CONCLUSION

As a by-product to screening for trisomies 21, 18, and 13, most triploidies and the majority of atypical SCAs, RATs, and mosaicisms are detected before birth. However, only 23% of pCNVs are high-risk from cFTS and only half are diagnosed before birth. Replacing invasive testing with NIPT for high-risk pregnancies would substantially decrease the first-trimester detection of pathogenic chromosomal anomalies. This article is protected by copyright. All rights reserved.

Decoding 22q11.2: prenatal profiling and first-trimester risk assessment in Danish nationwide cohort

OBJECTIVES

To examine the distribution of nuchal translucency thickness (NT), free β-human chorionic gonadotropin (β-hCG) and pregnancy-associated plasma protein-A (PAPP-A) in pregnancies with a fetal 22q11.2 aberration. Furthermore, the performance of combined first-trimester screening (cFTS) and a new risk algorithm targeting 22q11.2 deletions in detecting affected pregnancies was evaluated. Finally, prenatal malformations and pregnancy outcome were assessed.

METHODS

This was a nationwide registry-based cohort study of all pregnancies that underwent prenatal screening with a due date between January 2008 and December 2018 in Denmark. All cases with a fetal 22q11.2 deletion or duplication (hg19 chr22:18.9mio-25.0mio) diagnosed pre- or postnatally or following pregnancy loss or termination of pregnancy were retrieved from the Danish Cytogenetic Central Register and linked with pregnancy data from the Danish Fetal Medicine Database. Fetal and maternal characteristics, including cFTS results and pregnancy outcome, of pregnancies with any 22q11.2 deletion or duplication (LCR22-A to -H) and pregnancies with a classic deletion or duplication (LCR22-A to -D) diagnosed by chromosomal microarray were compared with those of a chromosomally normal reference group. A risk algorithm was developed for assessing patient-specific risks for classic 22q11.2 deletions based on NT, PAPP-A and β-hCG. Detection rates and false-positive rates at different risk cut-offs were calculated.

RESULTS

We included data on 143 pregnancies with a fetal 22q11.2 aberration, of which 97 were deletions (54 classic) and 46 were duplications (32 classic). NT was significantly increased in fetuses with a classic deletion (mean, 1.89 mm), those with any deletion (mean, 1.78 mm) and those with any duplication (mean, 1.86 mm) compared to the reference group (mean, 1.65 mm). β-hCG multiples of the median (MoM) was decreased in all 22q11.2 subgroups compared with the reference group (mean, 1.02) and reached significance in pregnancies with a classic deletion and those with any deletion (mean, 0.77 and 0.71, respectively). PAPP-A MoM was significantly decreased in pregnancies with a classic duplication and those with any duplication (mean, 0.57 and 0.63, respectively), and was significantly increased in pregnancies with a classic deletion and those with any deletion (mean, 1.34 and 1.16, respectively), compared to reference pregnancies (mean, 1.01). The screen-positive rate by cFTS was significantly increased in pregnancies with a classic deletion (13.7%), any deletion (12.5%), a classic duplication (46.9%) or any duplication (37.8%) compared to the reference group (4.5%). A risk algorithm targeting classic 22q11.2 deletions more than doubled the prenatal detection rate of classic 22q11.2 deletions, but with a substantial increase in the false-positive rate. Structural malformations were detected in 41%, 35%, 17% and 25% of the pregnancies with a classic deletion, any deletion, classic duplication or any duplication, respectively. Pregnancy loss occurred in 40% of pregnancies with a classic deletion and 5% of those with a classic duplication diagnosed prenatally or following pregnancy loss.

CONCLUSIONS

The distribution of cFTS markers in pregnancies with a classic 22q11.2 duplication resembles that of the common trisomies, with decreased levels of PAPP-A. However, classic 22q11.2 deletions are associated with increased levels of PAPP-A, which likely limits early prenatal detection using the current cFTS risk algorithm. The scope for improving early detection of classic 22q11.2 deletions using targeted risk algorithms based on NT, PAPP-A and β-hCG is limited. This demonstrates the capability, but also the limitations, of cFTS markers in detecting atypical chromosomal anomalies, which is important knowledge when designing new prenatal screening programs. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Atypicality index as an add-on to combined first-trimester screening for chromosomal aberrations

OBJECTIVES

To compute a set of atypicality indices from combined first-trimester screening (cFTS) markers and second-trimester estimated fetal weight (EFW), and to demonstrate their potential in identifying pregnancies at either reduced or increased risks of chromosomal aberrations following a low-risk cFTS result.

METHODS

The atypicality index quantifies the unusualness of an individual set of measurements relative to a reference distribution and can be computed from any variables or measurements available. A score of 0% on the atypicality index represents the most typical profiles, while a score of 100% indicates the highest level of atypicality. From the Danish Fetal Medicine Database, we retrieved data on all pregnant women seen for cFTS in Central Denmark Region between January 2008 and December 2018. All pregnancies with a cytogenetic or molecular analysis obtained prenatally, postnatally, or following pregnancy loss or termination of pregnancy were identified. A first-trimester atypicality index (AcFTS ) was computed from nuchal translucency (NT) thickness, maternal serum free β-human chorionic gonadotropin (β-hCG) and pregnancy-associated plasma protein-A (PAPP-A). Further, a second trimester index (AcFTS+EFW ) was computed from cFTS markers and estimated fetal weight (EFW) estimated at a routine second-trimester anomaly scan. All pregnancies were stratified into subgroups based on their atypicality levels and their cFTS risk estimates. The risk of chromosomal aberrations in each subgroup was then compared with the overall prevalence, and a graphical presentation of multivariate measurement profiles was introduced.

RESULTS

We retrieved data on 145,955 singleton pregnancies, of which 9824 (6.7%) were genetically examined. Overall, 1 in 122 of all pregnancies seen for cFTS (0.82% [95% CI 0.77-0.87%]) were affected by a fetal chromosomal aberration and in screen-negative pregnancies (cFTS T21 risk <1 in 100 and/or T18/13 risk <1 in 50), 0.41% [95% CI 0.38-0.44%] were affected. In screen-negative pregnancies with a typical first-trimester profile (AcFTS <80% ), the risk of chromosomal aberrations was significantly reduced (0.28%) compared to the overall risk. The risk of chromosomal aberrations increased with higher atypicality index to 0.49% (AcFTS [80-90%) ), 1.52% (AcFTS [90-99%) ), and 4.44% (AcFTS [>99%) ) and was significantly increased in the two most atypical subgroups. The same applied for the second trimester atypicality index (AcFTS+EFW ) with risks of chromosomal aberrations of 0.76% and 4.16% in the two most atypical subgroups (AcFTS+EFW [90-99%) and AcFTS+EFW >99% , respectively).

CONCLUSIONS

As an add-on to cFTS, the atypicality index identifies women with typical measurement profiles which may act as reassurance, whereas atypical profiles may warrant specialist referral and further investigations. In pregnancies at low risk from cFTS but with a highly atypical distribution of NT, PAPP-A, and β-hCG, the risk of a chromosomal aberration is substantially increased. The atypicality index optimizes the interpretation of pre-existing prenatal screening profiles and is not limited to cFTS markers or EFW. This article is protected by copyright. All rights reserved.