Pregnancy outcomes of rare autosomal trisomies results in non‐invasive prenatal screening: clinical follow‐up data from a single tertiary centre

Abstract This study was performed to assess the association between detection of rare autosomal trisomies (RATs) by non‐invasive prenatal screening (NIPS) and adverse pregnancy outcomes. We retrospectively analyzed women with high‐risk RATs results from January 2014 to December 2020. The women's clinical information was collected, and their pregnancy outcomes were compared with those of women with low‐risk results. In total, 151 (0.24%) RATs results were reported among 62,752 NIPS examinations. Sixty‐five women chose to undergo amniocentesis for confirmation, which revealed 3 cases of true fetal mosaicism for RATs and a positive predictive value of 4.6% (3/65). Among the 139 women with available outcomes, 26 (18.7%) had a preterm birth, 10 (7.2%) underwent pregnancy termination because of fetal defects and 5 (3.6%) had miscarriages. Interestingly, compared with the control group, pregnancies in which NIPS revealed trisomy 16 (T16), T22, T9 and T2 were at higher risk of adverse outcomes, including preterm birth, miscarriage and ultrasound abnormalities. However, the risk of adverse outcomes was comparable between the control group and pregnancies with positive results of T7, T3, T8 and T20. In summary, the risk of adverse pregnancy outcomes was higher in women with specific RATs‐positive NIPS results. Pregnancies with T16, T22, T9 and T2 results, even if false‐positive, should be considered high‐risk pregnancies.

sensitivity, specificity, false-positive rate and pregnancy outcome information of the RATs results obtained by NIPS are still limited. Therefore, the clinical benefits of these results remain unclear.
Several previous researchers have reported the RATs results obtained by NIPS, and their data showed that most of these results were false-positive because of confined placental mosaicism. 3,5,8 In addition, different conclusions have been drawn regarding the clinical significance of RATs results. In some studies, RATs were associated with fetal growth abnormalities as well as higher miscarriage rates. 3,5 However, data from other studies showed that RATs were mostly associated with normal pregnancy outcomes. 8 Although the data regarding RATs are still limited, it has become necessary to clarify the clinical significance of RATs detected by NIPS because of the broadening acceptance of NIPS in clinical practice.
In this study, we retrospectively analyzed 151 RATs results from 62,752 NIPS examinations performed in our centre from January 2014 to December 2020. We reviewed the NIPS results as well as the follow-up records and compared the pregnancy outcomes with 10,397 low-risk NIPS results to evaluate the clinical significance of RATs results and to determine whether these results are associated with adverse pregnancy outcomes.

| Recruitment criteria
This study involved pregnant women who underwent NIPS testing for genome-wide cell-free DNA screening in the Medical Genetics

Center of the Affiliated Obstetrics and Gynecology Hospital with
Nanjing Medical University from January 2014 to December 2020.
The inclusion criteria for this study were a pregnancy with a highrisk RATs report and complete clinical information, and the exclusion criterion was loss to follow-up. This study was approved by the institutional review board of the Nanjing Maternity and Child Health Care Hospital (approval number NFLZ2019-KY-004).

| NIPS examinations
Five millilitres of maternal peripheral blood was collected from each participant using EDTA anticoagulant tubes and centrifuged within 8 hours to extract the plasma. The details of the procedure were similar to those described in our previous report. 2,9 A library was constructed using the BGI protocol and sequenced using the BGISEQ-500 platform (BGI Group). The cut-off fetal fraction was set at 3.5%. Fetal chromosomal trisomies for chromosomes 13, 18, 21, X and Y as well as other genomewide RATs and subchromosome copy number variants were analyzed.

| Outcome collection
We retrospectively analyzed the data of women with high-risk RATs results. The following data regarding these women's maternal and pregnancy characteristics were collected from the laboratory database and clinical records: maternal age, gestational age (GA) at the time of NIPS, NIPS results and pregnancy outcomes (cytogenetic results, ultrasound examination results, miscarriage, termination of pregnancy, GA at the time of delivery and newborn physical examination results). Birthweight percentile was calculated according to NICHD (national institute of child health and human development) (Asian) charts. 10 Preterm was defined as delivery at <37 weeks of gestation. The pregnancy outcomes for women with low-risk NIPS results in 2019 were used as comparative controls. Women who had pregnancies with RATs detected by NIPS were recommended to undergo confirmatory invasive prenatal diagnosis using amniocentesis F I G U R E 1 Flow of study design. NIPS, non-invasive prenatal screening; RATs, rare autosomal trisomies followed by chromosomal microarray analysis. A detailed flow diagram of the study design is shown in Figure 1.

| Statistical analysis
Continuous variables such as maternal age and GA are expressed as mean ± standard deviation. The incidence of RATs and pregnancy outcome results are expressed as number and percentage.
Differences in outcomes between the RATs group and the low-risk NIPS group were compared using the χ 2 test or Fisher's exact test.
A p value of <0.05 was considered statistically significant. Statistical analysis was performed using SPSS 19.0 (IBM Corp).

| Detection of RATs by NIPS
We retrospectively reviewed the data of 62,752 NIPS examinations performed in our centre from January 2014 to December 2020. The mean age of these pregnant women was 31.6 ± 4.8 years, and the mean GA was 17.7 ± 2.5 weeks. In total, 151 RATs results were reported (0.24%). Trisomy 7 (T7), T16, T3, T8, T22, T20, T9 and T2 were the most frequently detected RATs, whereas T4, T12, T17, and T19 were not detected ( Figure 2).

| Confirmation of RATs results
Among the 151 RATs cases, 65 women chose to undergo amniocentesis. The results showed three cases of confirmed true fetal mosaicism (TFM) for RATs, including one case of mosaic T9 and two cases of mosaic T22. Therefore, the positive predictive value of RATs was 4.6% (3/65). We also found six cases of uniparental disomy (UPD), including two cases involving chromosome 2 and four cases involving chromosome 16. There was an incidental finding of one case with confirmed fetal 22q11.2 microduplication. The other 55 (84.6%) women showed a normal karyotype in their amniocentesis results.
The detailed data are shown in Table 1.

| Pregnancy outcome results
Among the 151 RATs cases, 147 (97.4%) women were successfully followed up to retrieve the pregnancy outcome. The results showed  Table 1 and Table 2. Birthweight information was available in 107 of the 124 newborns. We found that 13 newborns were recorded with birthweight below 3rd percentile (13/107, 12.1%), including 7 cases of T16, 2 cases of T2 and single case of T3, T15, T20, T22, respectively (Table S1).

| Comparison of pregnancy outcomes between RATs results and low-risk results
To further investigate the clinical significance of the RATs results, we compared the pregnancy outcomes of women with RATs results F I G U R E 2 Distribution of rare autosomal trisomies. T, trisomy were used as the control group ( Figure 1 summarizes the study design). We found that the risk of adverse outcomes, including miscarriage, ultrasound abnormalities and preterm birth, was significantly higher in the RATs group than in the control group (p < 0.05). We next focused on T7, T16, T3, T8, T22, T9, T20 and T2, which had higher detection rates than the other RATs in our cohort ( Figure 2).
Compared with the control group, the occurrence of miscarriage was significantly higher in women with T16 results; ultrasound abnormalities were significantly higher in women with T16, T22 and T9 results; and preterm birth was significantly higher in women with T16, T22 and T2 results. However, no significant difference in the occurrence of adverse outcomes was found between the control group and women with T7, T3, T8 and T20 results (Table 3).  Trisomy 1  0  0  0  0  0  1  1  2   Trisomy 2  0  2 a  1  0  0  3  2  7   Trisomy 3  1  0  0  0  1  1  14  17 Trisomy 5   and 54 cases were confirmed to have confined placental mosaicism.

| DISCUSS ION
Pregnancy follow-up showed that 95% of the women in their study had an uncomplicated pregnancy with the exception of two cases of intrauterine growth restriction (one case of T16 and one case of compound T7/8). 8 In our study, 29.5% of RATs cases were associated with preterm birth, miscarriage, ultrasound abnormalities, TFM and UPD, and the other 70.5% of pregnancies ended in normal full-term live births. These differences in RATs outcomes may be associated with the distribution of the GA at the time of NIPS. In the studies by Pertile et al. 5 and Scott et al., 13 most samples were collected in the first trimester, which was earlier than in the study by He et al. 8  analyzed 24 cases of T7 and reported that all had normal pregnancy outcomes, and T7 was not confirmed in any cases. Our results also showed no significant difference between the T7 group and control group. Additionally, like T7, the pregnancy outcomes of T3, T8 and T20 were generally satisfactory in our study. Therefore, caution is needed before performing any invasive procedures in pregnancies with these RATs, which may be associated with a favourable outcome.
The main strength of this study is the relatively large number of RATs results with pregnancy outcomes from a single tertiary centre.
We systematically and comprehensively expounded the relationship between various RATs detected by NIPS and adverse pregnancy outcomes for the first time, and our data showed that the associations and relevance for each chromosome were different. The results could be informative in clinical counselling and pregnancy management.

| CON CLUS IONS
In this study, we retrospectively analyzed 151 RATs detected by NIPS and found that the chances of adverse outcomes, including miscarriage, ultrasound abnormalities and preterm birth, were significantly higher in the RATs group than the control group. Among the most frequent RATs, T16, T22, T9 and T2 were associated with adverse outcomes, and more intensive pregnancy monitoring should

ACK N OWLED G EM ENTS
This work was supported by the Nanjing Outstanding Youth Grant for Medical Science and Technology (JQX18008).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data described in this study are available upon request from the corresponding authors.