Prenatal diagnosis and pregnancy determination of sex chromosome aneuploidy screened by noninvasive prenatal testing from 52,453 unselected singleton pregnancies: a retrospective analysis of 4-year experience

Objective To estimate the positive predictive value (PPV) of fetal sex chromosome aneuploidy (SCA) screened in non-selective population and explore the rate of pregnancy ultrasound findings for fetal SCA and the factors influencing parents’ decisions in South China. Methods This is a large-scale retrospective cohort of positive SCA screened from unselected singleton pregnancies by non-invasive prenatal testing (NIPT) from a single prenatal center of a tertiary hospital, from January 2016 to November 2019. Clinical information, indications, diagnostic results, ultrasound findings, pregnancy determinations, and follow-up were reviewed and analyzed. Results 248 cases of SCA positive were screened out of 52453, giving a positive detection rate of 0.47%. The majority of indications (42.7%) were low-risk pregnancies. After genetic counseling, 43 pregnancies (17.3%) declined to prenatal diagnosis, the rest of 205 cases (82.7%) conducted with amniocentesis to detect fetal chromosome, of which 95 were confirmed as true positive of SCA with PPV of 46.3% (95/205). The SCA consisted of 68 sex chromosomal trisomies (26 cases of 47,XXY, 20 cases of 47,XXX and 22 cases of 47,XYY), 17 cases of monosomy X (45,X), three cases of 48,XXYY, three cases of mosaicisms (45,X/46,XX), four cases with sex chromosomal deletions, included two cases of 46,X,del(Y)(q11.21), one case of 46,X,del(X)(p11) and one case of 46,X,i(X)(q10). Of the 95 cases confirmed as true positive SCA, 50 cases (52.6%)chose to terminate the pregnancy (82.6%, 64.3%, 17.6% and 33.3% for 45,X, 47,XXY, 47,XXX and 47,XYY, respectively), 45 cases (47.4%) elected to continue the pregnancy. Ninety-three pregnant were also continued pregnancy after the exclusion of SCA. a first-tier routine method for screening autosomal aneuploidies, also could play an important role in screening SCA. Low-risk pregnant women are the main indication for the detection of SCA as NIPT test provides to non-selective population. The trend of overall termination rates of SCA-affected pregnancies is decreased. For 47,XXX and 47,XYY with mild phenotype, couples would like to continue the pregnancy. But for 45,X and 47,XXY, parents apt to terminate pregnancy no matter ultrasound abnormalities were found or not. pregnancy determinations are affected by types of SCA, sonographic findings, maternal age, and presence of infertility.


Introduction
Sex chromosomal aneuploidy (SCA), the incidence of newborn ranged from 1/1200 to 1/400. The incidence of fetal period is as high as 1/435 [1,2]. Common types are sex chromosome monosomy (45,X), sex chromosome trisomy(47,XXX, 47,XXY and 47,XYY), other aneuploidies and Various forms of mosaic chromosomal abnormalities [3]. Unlike autosomal chromosome aneuploidies ( trisomy 21, trisomy 13 and trisomy 18), most of the SCA patients have very mild phenotype and relatively better clinical prognosis [4]. They can be manifested as abnormal height development, abnormal organ structure, abnormal behavior, dysplasia of secondary sexual characteristics and so on. The phenotype of SCA patients are diversities, and some patients have no abnormal phenotype throughout their lifetimes (47,XXX and 47,XYY) [5]. Therefore, These make prenatal counseling and parental decisionmaking following the prenatal diagnosis of SCA more complicated and challenging. Its pregnancy determination of SCA is often influenced by many factors, such as types of SCA, ultrasound findings, family's socioeconomic level, education level, religions, couple's understanding and acceptance of SCA. Therefore, Prenatal screening or diagnosis of SCA is not only challenged clinical genetic counseling but also puts pregnancy determination in a dilemma [6].
With the development of second-generation sequencing technology, non-invasive prenatal testing (NIPT) has been rapidly and widely used worldwide [7]. NIPT has higher sensitivity and specificity in screening trisomy 21, trisomy 18 and trisomy 13 syndromes [8]. It has been recommended by many professional associations of Obstetrics and gynecology as the most accurate prenatal screening method for aneuploidy [9,10]. However, the effectiveness of NIPT in detecting SCA and whether SCA should be included in the NIPT report as a routine report are still controversial. In addition to sensitivity and specificity, the positive predictive value (PPV) is also an important index in evaluating the detection ability of a screening method [11]. Previous reports suggested that the sensitivity of NIPT for SCA screening could reach 88.6%~93.8%, but PPV could only reach approximately 50% [4,6,12]. The incidence of different types of SCA is very variable, and most of the previous studies have limited sample size, pregnancies with limited screening indications and the fluctuation range of PPV is various [6,12]. Hence, In the present retrospective study, we reviewed SCA cases that a largescale and found in non-selective pregnancies who underwent NIPT. The results of prenatal diagnosis, ultrasound findings, distribution of indications, pregnancy determination and follow-up of SCA cases were analyzed, the PPV of positive SCA was calculated. NIPT testing: Five mL of maternal peripheral blood was collected into an EDTA tube. Plasma was isolated from blood within 6 hours by two-step centrifugation, whole blood was centrifuged at 1600g for 10 min at 4°C , and supernatants were transferred to 2.0 mL tube continued to centrifuged at 16000g for 10 min at 4°C. The plasma was immediately stored at −80°C until DNA extraction.

Materials And Methods
Afterward, cffDNA extraction, library construction, quality control, and pooling were performed according to the manufacturer's protocol, Fetal Chromosome Aneuploidy Testing Kits (BGI-Wuhan, Hubei, China) was used to conduct large-scale parallel genome sequencing by BGIS EQ-500 sequencing platform ( BGI-Wuhan, Hubei, China). Sequencing data were compared with the human genome reference sequence (HG19, NCBI build 36). The z-score for each chromosome was calculated to judge abnormality referencing to the normalized chromosome representation. Samples failing the quality criteria of cffDNA extraction, library construction, and sequencing as well as fetal DNA concentration (< 4%)were kicked out.

Invasive Prenatal diagnosis:
Pregnancies associated with positive SCA were offered invasive prenatal diagnosis, fetal chromosomal karyotyping were analyzed. Amniotic Fluid samples were obtained according to the prenatal procedure protocol. Amniocytes were cultured with Gibco medium (Life Technologies Corporation, New York, United States). The cultured amniocytes were analyzed by routine cytogenetic analysis using G-banding techniques at a resolution of 400-500 bands. The number of metaphases examined varied between 20 and 30. The results of cultured amniocytes are available within 14 to 20 days. Pregnancies who declined to have prenatal diagnosis were followed-up until delivered for at least 3 months.
Statistical methods: Descriptive statistics were applied, using the Chi-squared test when comparing discrete categorical variables. SPSS 20.0 software (SPSS Inc., Chicago, USA) was used to analyze all data for statistical significance. Data are presented as mean±SDs. Analysis of variance was used to compare the differences between different groups. A P-value <0.05 was considered statistically significant.

Results
General information: From January 2016 to November 2019, 52,661 singleton pregnancies who underwent NIPT screening, 52,453 pregnancies were successfully got available results and 208 pregnancies were failed due to inadequate fetal fragments. Figure 1 shows the brief flow of this study.
The median maternal age of the blood collection was 29.3±4.69 years, ranging from 16 to 48 years.
The median gestational weeks were 16 Table 1 and Table 2.  Pregnancies decline to conduct prenatal diagnosis: Among 248 cases with positive SCA by NIPT, 43 pregnancies declined amniocentesis to confirm fetal chromosome. 31 cases had a live birth and no visible abnormal phenotype was found in clinical follow-up, 3 cases were still continued pregnancy with normal ultrasound findings. Two cases with positive results for 45,X were terminated the pregnancy due to ultrasound abnormalities (one for Cystic hygroma and one for pleural effusion).

Indications of positive SCA Cases
Three cases with positive results (two cases for 45,X and one case for 47,XXX) had spontaneous abortions in the second trimester. Besides, four cases lost to follow up after NIPT screening. The follow-up information was described in Table 4.

Discussion
NIPT as a prenatal screening method for autosomal aneuploidy has been widely applied clinical practice with superior detection and lower false-positive rates than conventional screening tests [13,14]. But the NIPT used to screening for SCA has been controversial because of less accurate compare to trisomies of 21, 13 and 18 [15]. Few studies reported that some of SCA may with normal  [3,6,12,18]. The reasons why the PPV for 45,X was lower compared to other types of SCA, firstly, it could partially be caused by a highly variable amplification of chromosome X because of its lower Guanosine-Cytosine content [19]. Secondly, study showed that there is an age-related loss of chromosome X in white cells in normal pregnancies, which influenced the effectiveness in predicting fetal 45,X. [20]. In our cohort, pregnancies with positive screening for a SCA who conducted prenatal diagnosis were higher (82.7%) than other reports [3,21], This difference may be caused by parental concerns about having a child with infertility and high risks of lack secondary sexual characteristics and gynecomastia in mainland China [12]. In contrary to Aarti er al reported that pregnancies with positive NIPT for a SCA elect prenatal diagnosis was only 34.3% (46/134), because of the relatively mild or moderate prognoses associated with liveborn infants with SCA compared to the invasive procedural-related risks and anxiety, the parental attitude is more likely to accept an SCA fetus [3].
Before NIPT applied to screen fetal SCA, that the detection of SCA is usually incidental during screening for Down syndrome by conventional screening methods [22,23]. Previous studies had shown the distribution of indications for prenatal diagnosis for each SCA-affected pregnancy included ultrasonography findings( such as cystic hygroma and hydrops), advanced maternal age(≥35 years), positive of serum screening test, increased nuchal translucency and so on [6,24]. For prenatal diagnosis of 45,X (Turner Syndrome), the majority of indications were fetal ultrasound abnormalities [6,12]. For the other types of SCA, the main indications for prenatal testing were advanced maternal age and high-risk serum screening tests [6,12]. In the present study, our databases indicated that after NIPT was applied to screening for all enrolled pregnancies as a first-tier routine method with cost-effective by public healthcare service and insurance coverage, the charge for NIPT screening was almost free for every pregnant woman. Therefore, The distribution of indications for screening a positive SCA were sharply changed, the majority of indications (42.7%) were low-risk population who before NIPT screening, no matter for screening 45,X or other types of SCA. Furthermore, the prenatal diagnosis of SCA had become more common and the demand for adequate pre-test and post-test genetic counseling was increased considerably.
In this study, The majority of the abnormal ultrasound findings were found in cases of 45,X (70.6%, 12/17), included cystic hygromas, ascites, pleural effusion, heart defects, and ambiguous genitalia.
For the other types of SCA(47,XXY, 47,XXX and 47,XYY) which ultrasound findings were not typically expected (7.0%, 5/71) and relatively manifested mild phenotypes, such as NT thickened and heart defects. Demonstrated for 45,X were more likely to be found ultrasound abnormalities than sex chromosomal trisomies. The previous study showed that obvious fetal ultrasound abnormalities had a highly significant impact on the parental decision, pregnancies prefer to terminate than those with normal ultrasound findings [3,6,24,25]. This was also shown in our study, that especially for 45,X with abnormal ultrasound and poor prognosis associated who were all elected to terminated pregnancy.
However, based on our data, ultrasound findings were not the only reason for couples to decide whether to continue or terminate pregnancy, younger maternal age, absence of history infertility also significantly associated with termination decisions. For cases with 45,X and 47,XXY would like to terminate pregnancy even those with normal ultrasound findings. For 47,XXX and 47,XYY would like to continue pregnancy even those with ultrasound abnormalities, such as ventricular septal defect and aortic coarctation [12].
The termination rates of SCA were varied across different countries and the possible reasons are different types of SCA, absence or presence of ultrasound abnormality, cultural background, education level, religious and counseling practices [6,26]. According to our data, the overall This study we collected the NIPT results, distribution of indications, results of prenatal diagnosis, ultrasound findings and determination of pregnancies based on a large size data from non-selective pregnancy women. We found out the higher rate of prenatal diagnosis, the change of indications prior to NIPT, the declining trend of termination rate as well as PPV for each SCA by NIPT, However, the limitation of this study is that can not follow-up all false-positive cases, the existence of CPM can not be ruled out. Most of SCA cases can not be diagnosed until childbearing age, and some of them can not be diagnosed throughout lifetimes [32]. Therefore, the sensitivity, false-positive rate and falsenegative rate of NIPT screening for SCA can not be accurately calculated. Furthermore, There was a insufficient of postnatal data for those who elected to continue pregnancies but delivered at normal phenotype without karyotyping results. Due to the majority of pregnancies may feel that confirmatory testing is unnecessary if no abnormalities are detected at birth. The limitations of NIPT should be carefully explained before and after testing.

Conclusion
This study based on a large-scale sample screened for SCA from non-selective singleton pregnancies, that NIPT provided as a first-tier routine screening method for all enrolled pregnancies, the demand for prenatal counseling and diagnosis of SCA were increased sharply. Our data indicated that SCA can effectively screened by NIPT, especially in predicting sex chromosome trisomies. ultrasound abnormalities were closely associated with 45,X. Pregnancies with 45,X and 47,XXY were more likely elected to terminated than those with 47,XXX and 47,XYY. Our data also highlighted the necessity for pre-and post-test counseling and prenatal diagnosis before chose to terminate pregnancy. Early screening and diagnosis of SCA may provide an opportunity for early intervention and comprehensive postnatal management, improving the life quality of the affected child. Our findings may provide more references for clinical genetic counseling and pregnancy management.

Declarations
Funding Figure 1 Study flow chart showing prenatal diagnosis and pregnancy determination of SCA positive screened by NIPT