Impact of high‐risk prenatal screening results for 22q11.2 deletion syndrome on obstetric and neonatal management: Secondary analysis from the SMART study

One goal of prenatal genetic screening is to optimize perinatal care and improve infant outcomes. We sought to determine whether high‐risk cfDNA screening for 22q11.2 deletion syndrome (22q11.2DS) affected prenatal or neonatal management.

� The 22q11.2 deletion syndrome occurs in approximately 1 in 2000 pregnancies, unrelated to maternal age, with 5%-10% inherited from a parent.� Although variability in the neurocognitive and structural phenotype exists, early interventions in the neonatal period are available to improve many outcomes.

What does this study add?
� Confirmatory genetic testing, as well as obstetrical and neonatal evaluation and interventions, are inconsistently performed after high-risk cfDNA screening results for 22q11.2DS� Approximately half of the infants at risk for 22q11.2DSbased on high-risk cfDNA screening results are discharged from the hospital without confirmatory genetic testing or evaluation for 22q11.2DS-relatedabnormalities.
� To optimize outcomes for individuals with 22q11.2DS,prenatal and neonatal evaluation after high-risk cfDNA screening for 22q11.2DS is needed.

| INTRODUCTION
The 22q11.2 deletion syndrome (22q11.2DS),also known as DiGeorge or velo-cardio-facial syndrome, occurs in approximately 1 in 2000 live births. 1,2Most cases arise as a de novo abnormality unrelated to maternal age, while about 5%-10% are inherited from a parent. 3There are more than 50 genes in the 22q11.2critical region, 4 leading to a complex and variable phenotype that can include congenital heart defects, a compromised immune system, orofacial clefts, palatal insufficiency, developmental delay, and schizophrenia.
In an era prior to prenatal screening and comprehensive testing, the average age of diagnosis of 22q11.2DSwas 4.7 years 5 Historically, 22q11.2DShas been suspected in the pre-and perinatal period when ultrasound abnormalities, particularly conotruncal cardiac anomalies, were identified, prompting prenatal or postnatal diagnostic testing.However, some 22q11.2DSaffected fetuses and neonates do not present with a cardiac anomaly, reducing the sensitivity of prenatal ultrasound for the detection of 22q11.2DS.The performance of cell-free DNA (cfDNA) screening for 22q11.2DS[8][9] The SMART study enrolled over 20,000 women who had prenatal screening for 22q11.2DSusing SNP-based cfDNA and who consented to confirmatory genetic testing of the infant. 6[12][13][14][15][16][17]

| METHODS
This study was a secondary analysis from the SMART study, wherein the population of pregnant individuals receiving a high-risk cfDNA result for 22q11.2DSalong with a matched low-risk cohort were included.Supplementary data collection was performed to assess the differences in obstetrical and neonatal care based on cfDNA results for 22q11.2DS.Patients receiving a high-risk cfDNA result for 22q11.2DSwere compared with the full low-risk cohort for pregnancy characteristics and obstetrical management.To assess differences in neonatal care based on cfDNA results for 22q11.2DS,we compared the relevant high-risk cohort with a control low-risk cohort with the goal of 1:1 matching.
The performance of cfDNA screening, data collection protocols and testing methodology for 22q11.2DSused in the SMART study have been described in detail. 6Briefly, SMART study participants who received SNP-based cfDNA results for 22q11.2DSand had confirmatory genetic testing were eligible for participation in this study.
Those with a high-risk result for autosomal aneuploidy were excluded.
A risk of ≥1% for 22q11.2DSbased on cfDNA analysis was reported as high-risk.Chromosomal microarray analysis (CMA) was requested for all newborns, regardless of clinical prenatal diagnostic testing.If All participants enrolled in the SMART study consented to collect data related to the performance and utility of cfDNA testing for 22q11.2DS.This included confirmatory genetic testing, prenatal ultrasound examinations, and pregnancy outcome data, such as delivery information and complications in the newborn period prior to hospital discharge.Data on cfDNA and confirmatory genetic testing results, ultrasound data, and pregnancy outcome were collected for all SMART participants by trained research coordinators at each study site.Demographics, number of obstetrical ultrasounds, diagnostic prenatal testing results and pregnancy outcomes were compared for all SMART participants who received a cfDNA result that was either high-risk for 22q11.2DSor low-risk for all conditions screened, including trisomies 13, 18, 21, monosomy X and 22q11.2DS,and for whom both pregnancy outcome and genetic confirmation results were available.
To specifically address the frequency with which newborns underwent evaluations targeted to known 22q11.2DS-relatedphenotypic differences, an ancillary study was designed and received IRB approval in 2017 from all participating institutions.The supplemental data form is provided for review (Supplementary Figure S1) and included: fetal echocardiography, lymphocyte count, serum calcium level, postnatal echocardiography, palatal assessment, and diagnostic genetic testing for 22q11.2DS.The completion of the supplemental data form was requested for all pregnancies receiving a high-risk cfDNA result for 22q11.2DSthat resulted in a live-born infant surviving the immediate neonatal period, that is, prior to hospital discharge.To determine the baseline frequency with which these neonatal evaluations were performed for each participant with a high-risk 22q11.2DScfDNA result, coordinators collected these same supplemental outcomes for one study participant receiving a low-risk cfDNA 22q screening result, matched for maternal age, and date and site of delivery.No data were collected after discharge from the hospital.Incomplete data fields on the supplemental data form were excluded from both the numerator and denominator for those specific analyses.

| Statistical analysis
Continuous variables were compared using the Wilcoxon test and categorical variables using the chi-square test or Fisher's exact test, as appropriate.A nominal p value of <0.05 was considered to indicate statistical significance.SAS Studio 9.04 software (SAS Institute) was used for analysis.

| RESULTS
A total of 18,020 SMART study participants, enrolled between 2015 and 2019, met the criteria for analysis; 17,982 (99.79%) received low-risk and 38 (0.21%) high-risk cfDNA results for 22q11.2DS.In total, 12 cases were affected with 22q11.2DS, 3 had low-risk cfDNA screening results while 9 had high-risk results.S1.The mean maternal age was 31.9 years (range 20-41).Four of the infants were male and 8 were female.The three cases who had high-risk cfDNA screening and prenatal confirmation by diagnostic testing chose to proceed with pregnancy termination; two of these had cardiac defects identified after cfDNA results.Of the 9 liveborn neonates with 22q11.2DS,three were low-risk by cfDNA, none of whom had pre-or postnatal clinical genetic testing.The remaining six were high-risk by cfDNA, three of whom did not have clinical pre-or postnatal diagnostic genetic testing before hospital discharge.The results of the research CMA were not available prior to discharge.All three surviving infants with postnatal diagnostic testing prior to hospital discharge had abnormal postnatal echocardiograms, and two had cardiac surgery; additionally, all were hypocalcemic, and two had calcium administered.Supplemental data on neonatal management and outcomes were collected for 29/32 (90.6%) live born infants with high-risk 22q11.2DScfDNA results, including three with confirmed 22q11.2DS(Supplementary Table S1; cases 2,4,9), as well as 35 lowrisk control pregnancies (Figure 1).Of the 29 infants with high-risk 22q11.2DScfDNA results, 16 had prenatal diagnostic testing and all were confirmed negative for 22q11.2DS.Table 2   The clinical utility of prenatal genetic screening for 22q11.2DScannot be fully realized unless high-risk cfDNA screening results that were not addressed prenatally are communicated to the pediatric team, allowing appropriate evaluation for, and management of, associated phenotypic differences. 19While our study did not determine why 22q11.2DS-relatedevaluations were not performed for more than half of the high-risk newborns, it suggests that there is a need for improved communication between obstetric and neonatal care providers regarding the results of prenatal genetic screening tests and the implementation of protocols for evaluation of these infants.

| DISCUSSION
There are limitations to this study that should be acknowledged.
The study cohort was relatively small due to the frequency of the condition itself, 1 in 1524 pregnancies in the SMART cohort, and due to the low rate of high-risk cfDNA screening results (0.21%). 6though we found that women with high-risk cfDNA results underwent more prenatal ultrasound examinations, we do not know why these ultrasound examinations were ordered.We also do not know why postnatal evaluations and/or diagnostic genetic testing were not pursued.Specifically, we do not know if these evaluations were offered and declined, not offered, or if they were planned to be performed after hospital discharge.A concern would be a failure to offer testing by the pediatric provider either due to their not knowing the prenatal screening results or a false assumption that apparently normal infants without cardiac anomalies are at very low risk for 22q11.2DS.Finally, the study was not designed to assess long-term management and outcomes and it is possible that some of the 22q11.2DSrelevant tests were pursued in an outpatient setting after discharge.

| CONCLUSIONS
This study reports comprehensive data regarding prenatal care in a cohort of women who chose cfDNA screening for 22q11.2DS,along with data regarding neonatal testing and management in liveborn infants from pregnancies receiving high-risk cfDNA results.The data indicate that while more prenatal and neonatal testing was pursued after a high-risk cfDNA screening for 22q11.2DS,approximately half of the high-risk infants did not undergo diagnostic genetic testing or evaluation for 22q11.2DS-relatedcomplications before hospital discharge.These data underscore the need for the development of guidelines for immediate postnatal care when a high-risk cfDNA screening result for 22q11.2DS is received, but prenatal diagnostic genetic testing is not conducted.These guidelines will contribute to improved communication between obstetric and pediatric healthcare providers regarding high-risk prenatal genetic screening results.
postnatal CMA was not available, results from clinical testing with prenatal or postnatal CMA, fluorescence in situ hybridization, bacterial artificial chromosomes (BACs)-on-beads or multiplex ligationdependent probe amplification were used for genetic confirmation if available.A deletion of ≥0.5 Mb in the 22q11.2low copy repeat A-D region on diagnostic confirmatory testing was considered a positive test result.If the SMART study research CMA analysis identified a 22q11.2deletion, and prenatal or postnatal confirmatory genetic testing had not been performed, the study site principal investigator was notified as soon as study outcomes were unmasked.

Table 1
summarizes the demographics, prenatal evaluations and pregnancy outcomes for the SMART cohort, stratified by whether participants had received low-or high-risk cfDNA screening results for 22q11.2DS.Overall, 18.4% (7/38) of participants with high-risk cfDNA results had abnormalities identified on prenatal ultrasound as compared to 3.0% (522/17,982, p < 0.001) of those with a low-risk result.Those with

= 35 Infants high-risk by cfDNA without prenatal diagnostic testing.N = 13 includes 4 affected P value
Demographic and pregnancy clinical data for the low-risk and high-risk cfDNA result groups.Data for pregnancies with low-risk and high-risk and cfDNA results that were still considered low-risk and high-risk at birth.Flow diagram of patients in analysis.IUFD, intrauterine fetal demise; NND, neonatal death; SAB, spontaneous abortion; TAB, therapeutic abortion.evaluations such as CBC and serum calcium-level assessments.
29breviations: GA, gestational age; HR, high-risk; IUFD, intrauterine fetal demise; LB, livebirth; LR, low-risk; NND, neonatal death; SAB, spontaneous abortion; TAB, therapeutic abortion.aFisher'sExactcomparing LR and HR cfDNA screening result groups.b22q11.2duplicationconfirmedbyQFPCR n = 1.MARTIN ET AL.-1577 cfDNA result were significantly more likely to undergo pre-or postnatal confirmatory diagnostic testing and fetal and newborn assessments.However, our findings also suggest that these diagnostic evaluations were applied inconsistently.The proportion of high-risk infants who received a 22q11.2DS-specificclinicalandlaboratory assessment before discharge ranged from 30.8% to 53.9%.Of patients who received a high-risk 22q11.2DScfDNAresult,only 55% chose to pursue prenatal genetic testing and the 3 pregnancies confirmed to have 22q11.2DSresulted in pregnancy termination.29isimportant to optimize outcomes; an infant may have 22q11.2DSeven if the physical examination is normal.However, we acknowledge that the timing of diagnostic genetic testing for an infant at high risk for 22q11.2DSmay be tailored according to local differences in clinical practice.Before hospital discharge, while awaiting genetic testing results, evaluation for phenotypic (such as could result in missed opportunities for intervention.Unfortunately, our data indicate that only 46.2% of high-risk infants have undergone the necessary diagnostic testing, showcasing a prominent gap in healthcare delivery.Despite the promising positive predictive value of 52% in identifying 22q11.2DSdemonstrated by the SNP-based cfDNA using an updated algorithm, 6 many infants did not receive targeted T A B L E 2