How Are SNP‐Array, Karyotyping, and FISH Applied in Prenatal Practice? A Focus on Diagnosis of Mosaicism Involving the Sex Chromosomes

ABSTRACT

Objective

This study conducted a comparative analysis of chromosomal karyotyping, SNP‐array, and FISH results in 82 prenatal cases of mosaicism involving the sex chromosomes.

Methods

A retrospective analysis was conducted on 82 cases of mosaicism involving the sex chromosomes; the analysis focused on the concordance and discrepancies among karyotyping, SNP‐array, and FISH results.

Results

Among 10,825 high‐risk pregnancies undergoing amniocentesis, 82 cases (0.76%) of mosaicism involving the sex chromosomes were detected. These included 60 cases (73.2%) of X or Y chromosome aneuploidy mosaicism and 22 cases (26.8%) with mosaicism accompanied by structural abnormalities. Of the 82 cases, 64 showed consistent results across karyotyping, SNP‐array, and FISH, while 18 cases exhibited discrepancies between methods. Among the 18 discordant cases, 12 were diagnosed as mosaicism by karyotyping and FISH but showed normal results on SNP‐array. SNP‐array identified three additional cases of mosaic X chromosome copy number variations (CNVs). In three cases, Turner mosaicism was diagnosed solely by FISH; both karyotype and SNP‐array findings were normal.

Conclusions

This study demonstrates that the combination of SNP‐array, karyotyping, and FISH can effectively diagnose mosaicism involving the sex chromosomes. Karyotype analysis can effectively identify aneuploidy and balanced translocation mosaicism when X chromosome dosage shows no significant change, whereas SNP array is more effective in detecting mosaic copy number variations < 10 Mb. In cases of discrepant results between the two methods, FISH should be utilized for further verification.

Key Points

• What is already known about this topic?

  • °Sex chromosome mosaicism exhibits particularly complex and highly variable clinical manifestations. Currently, no single testing method is sufficient to fully identify sex chromosome mosaicism.

• What does this study add?

  • °This study conducted a comparative analysis of karyotyping, SNP‐array, and FISH results from 82 prenatal cases of fetal sex chromosome mosaicism, and explored the consistency and inconsistency among the three methods, thereby providing clinical reference data for the prenatal detection of sex chromosome mosaicism.

1. Introduction

Chromosomal mosaicism is a relatively common chromosomal abnormality identified in prenatal diagnosis and can be categorized into autosomal mosaicism and sex chromosome mosaicism. Among the latter, the Turner mosaicism spectrum is the most frequently observed. Common karyotypes of Turner mosaicism include 45,X/46,XX; 45,X/46,XY; and 45,X/46,X,i(X)(q10), among others. The clinical manifestations of mosaicism involving the sex chromosomes are highly variable. When structural abnormalities of the X chromosome are present—such as isochromosomes, ring chromosomes, dicentric chromosomes, deletions, or marker chromosomes—the resulting phenotypic presentation can become even more complex [1]. The severity of the clinical phenotype primarily depends on the proportion of abnormal cell lines, the type of mosaicism, and their tissue distribution. Prenatally diagnosed mosaicism involving the sex chromosomes may lead to fetal gonadal dysgenesis and can be associated with structural or functional abnormalities in other organ systems. Some cases may also present with intellectual developmental impairment, neurobehavioral abnormalities, or neurological dysfunction [2]. Due to variations in the proportion of abnormal cells, the type of mosaicism, and the distribution of affected tissues, the impact on fetal prognosis can vary significantly. Therefore, during genetic counseling, a comprehensive assessment based on karyotyping and clinical data is crucial to help pregnant individuals and their families make well‐informed decisions regarding pregnancy management [1, 3].

Most Turner syndrome patients who survive to birth are likely to have mosaicism. Conventional karyotype analysis of blood lymphocytes—typically assessing only 20 to 30 metaphase cells—may fail to detect low‐level mosaicism due to its limited sensitivity, resulting in misdiagnosis as a “pure” 45,X karyotype [4]. Currently, the primary techniques for detecting mosaicism in prenatal diagnosis are karyotyping, SNP‐array, and FISH. These methods differ in detection range and resolution because of variations in their underlying principles and technical characteristics. In clinical practice, they frequently complement and validate each other [5]. However, inconsistencies between their results can complicate the diagnostic process, increasing the challenges of prenatal genetic counseling and imposing a significant psychological burden on pregnant women and their families [6]. Based on a relatively large prenatal cohort of fetuses with mosaicism involving the sex chromosomes, this study systematically analyzes the concordance and discrepancies among karyotyping, SNP‐array, and FISH results, aiming to provide more comprehensive clinical evidence for prenatal diagnostic counseling.

2. Subjects and Methods

2.1. Subjects

This study is a retrospective analysis that included 82 fetuses diagnosed with mosaicism involving the sex chromosomes through amniocentesis at the Prenatal Diagnosis Center of Longgang District Maternal and Child Health Hospital in Shenzhen from January 2017 to October 2025. All cases underwent prenatal diagnosis due to high‐risk pregnancy indications, including non‐invasive prenatal testing (NIPT) suggestive of sex chromosome abnormalities, fetal ultrasound‐detected structural anomalies, advanced maternal age, high‐risk serological screening results, or a history of adverse pregnancy outcomes. The diagnostic approach was as follows: cases with NIPT results indicating a sex chromosome abnormality were evaluated using karyotyping, SNP‐array, and FISH simultaneously. In other high‐risk cases, karyotyping and SNP‐array were performed initially, with FISH confirmation used if mosaicism was detected. Comprehensive counseling regarding the study's purpose, methodology, and potential risks was provided to all pregnant women prior to the procedure, and written informed consent was obtained from each participant. Telephone‐based follow‐up was conducted to assess the growth and development of newborns from continued pregnancies. The study inclusion was strictly limited to mosaicism involving the sex chromosomes, with cases of other chromosomal variants being excluded. Approval for this study was granted by the Ethics Review Committee of Shenzhen Longgang District Maternal and Child Health Hospital, and all procedures were conducted in compliance with the ethical standards of the Declaration of Helsinki.

2.2. Methods

2.2.1. Karyotype Analysis

Karyotyping was conducted on amniotic fluid samples using standard cytogenetic protocols. Briefly, cells were cultured in AmnioMAX‐II Complete Medium (Gibco, USA) to promote proliferation and maintained in an incubator for 6–8 days, with the duration adjusted based on growth. Chromosome preparations were obtained using conventional methods. Metaphase chromosomes were scanned using an automated system (Leica GSL120, USA), with a minimum of 10 metaphase cells analyzed per specimen at a resolution of 400–550 bands per haploid set. Chromosome classification and reporting adhered to the International System for Human Cytogenetic Nomenclature (ISCN 2020) guidelines [7].

2.2.2. SNP‐Array Analysis

Genomic DNA was extracted directly from uncultured amniotic fluid samples using the QIAGEN kit (QIAGEN, Germany), and experimental procedures such as digestion, PCR amplification and labeling were performed in accordance with the manufacturer's instructions. The processed DNA was subsequently hybridized to the Affymetrix CytoScan 750K array (Thermo Fisher, USA), which contains approximately 200,000 single nucleotide polymorphism (SNP) probes and 550,000 copy number variation (CNV) probes. The array was scanned with an Affymetrix GeneChip Scanner 3000 (Thermo Fisher, USA), and the data were processed and analyzed using the Chromosome Analysis Suite (ChAS) software (Thermo Fisher, USA). All detected CNVs were annotated against the GRCh37 (hg19) reference genome and classified into five categories according to the guidelines established by the American College of Medical Genetics and Genomics (ACMG): pathogenic, likely pathogenic, variant of uncertain significance (VUS), likely benign, and benign [8].

2.2.3. FISH Testing

Fluorescence in situ hybridization (FISH) was performed on 5 mL of uncultured amniotic fluid samples. The prenatal multicolor DNA probe kit from Anbiping (Guangzhou, China) was used, which includes the probe set GLP13/GLP21/CSP18/CSPX/CSPY. Hybridization, washing, and counterstaining were carried out strictly according to the manufacturer's protocols, and the slides were examined under a fluorescence microscope. A minimum of 50 cells were randomly counted for each sample. A sample was considered normal if over 90% of the cells showed normal signals. If abnormal signals were detected, the cell count was increased to 100, and an abnormal cell proportion between 10% and 60% indicated mosaicism (If NIPT indicates a reduction in the X chromosome or the presence of associated clinical phenotypes, the findings may be reported to clinicians for reference, even when the mosaic level below 10%.). When karyotyping or SNP‐array suggested mosaicism in specific genomic regions, custom‐targeted FISH probes were applied to metaphase spreads for further confirmation.

2.2.4. Follow‐Up

Telephone follow‐ups were conducted within 2 weeks before or after the provided estimated due date. The follow‐up records included any fetal ultrasound abnormalities, pregnancy continuation status, mode of delivery, and—after confirmation of parental origin—the phenotypic characteristics of carrier parents. After birth confirmation, regular follow‐up calls were made every 6 months to monitor postnatal growth and development, with particular attention to language expression. The most recent follow‐up was completed on September 15, 2025. As of that date, the oldest child was 4 years and 5 months old, and the youngest was 18 months old.

3. Results

3.1. General Information

This study analyzed 10,825 high‐risk pregnancies that underwent amniocentesis between January 2017 and October 2025. Mosaicism involving the sex chromosomes was identified in 82 cases, yielding a detection rate of 0.76% (82/10,825). Diagnostic concordance among karyotyping, SNP‐array, and FISH was observed in 64 cases (78.0%), while discrepancies occurred in 18 cases (22.0%). These discordant cases fell into three distinct patterns: (1) Twelve cases were detected by both karyotyping and FISH, but not by SNP‐array; (2) Three cases were identified by SNP‐array and FISH, with karyotyping showing normal results; and (3) Three cases were detected exclusively by FISH, whereas both SNP‐array and karyotyping showed normal findings (Figure 1).

FIGURE 1.

The brief flowchart of this study shows that 82 cases with mosaicism involving the sex chromosomes were detected among 10,825 high‐risk pregnant women. Concordant results from karyotyping, SNP‐array, and FISH were observed in 64 cases of mosaicism, whereas discordant results were found in 18 cases.

3.2. Karyotype Distribution and Frequency in 82 Fetuses With Mosaicism Involving the Sex Chromosomes

Karyotypic analysis of 82 fetuses with Mosaicism involving the sex chromosomes revealed the following distribution patterns. X chromosome aneuploidy mosaicism was the most prevalent, identified in 52 cases (63.4%), comprising 45,X/46,XX (n = 45), 45,X/47,XXX/46,XX (n = 4), and 45,X/47,XXX (n = 3). Mosaicism associated with structural abnormalities of the X chromosome was observed in 15 cases (18.3%), with the most common karyotype being 45,X/46,X,i(X)(q10) (n = 7), followed by 46,X,i(X)(q10)/46,XX (n = 3), and 45,X/46,X,r(X)(p11.2q28) (n = 2). Three cases of X‐chromosome deletions and duplication mosaicism were identified, yielding a detection rate of 3.6%. Additionally, there were eight cases of mosaicism involving Y chromosome aneuploidy and four cases with structural aberrations of the Y chromosome (Table 1).

TABLE 1.

Karyotype distribution of 82 cases with mosaicism involving the sex chromosomes.

Karyotyping	n (%)
X aneuploidy mosaics	52 (63.4%)
45,X/46,XX	45 (54.9%)
45,X/47,XXX/46,XX	4 (4.9%)
45,X/47,XXX	3 (3.6%)
X structural abnormality mosaics	15 (18.3%)
45,X/46,X,i(X) (q10)	7 (8.6%)
46,X,i(X) (q10)/46,XX	3 (3.7%)
45,X/46,X,r(X)(p11.2q28)	2 (2.4%)
45,X/46,X,+mar	2 (2.4%)
46,X,t(X;5) (q11.2;p13.1)/46,XX	1 (1.2%)
X chromosome deletions and duplications mosaics	3 (3.6%)
46,XX,dup(X) (p22.33p22.31)/46,XX	1 (1.2%)
46,XX,del(X) (p22.33p22.31)/46,XX	1 (1.2%)
46,XX,del(X) (q27.3q28)/46,XX	1 (1.2%)
Y aneuploidy mosaics	8 (9.8%)
45,X/46,XY	4 (4.9%)
46,XY/47,XYY	2 (2.4%)
45,X/47,XYY/46,XY	2 (2.4%)
Y structural abnormality mosaics	4 4.9%
45, psu dic(X;Y) (p22.3;p11.3)/46,XY	1 (1.2%)
45,X/45,X, psu dic(14;Y) (p11.2;p13)	1 (1.2%)
45,X/46,X,t(Y;7) (q11.23;p23)	1 (1.2%)
45,X/46,X, psu idic(Y) (p13)	1 (1.2%)
Total	82

3.3. Discordant Findings in the Analysis of 18 Cases of Mosaicism Involving the Sex Chromosomes

Among 82 fetuses with mosaicism involving the sex chromosomes, 18 showed discordant results among karyotyping, SNP‐array, and FISH analyses. Twelve cases (Cases 1–12) were diagnosed as mosaicism by both karyotyping and FISH, while SNP‐array results were normal. These included two cases of 45,X/46,XX, two of 45,X/47,XXX, four of 45,X/47,XXX/46,XX, and four cases (Cases 9–12) involving mosaicism with balanced structural abnormalities of the X or Y chromosome. Three cases (Cases 13–15) were identified as mosaicism by both SNP‐array and FISH, with normal karyotyping results: one case displayed a 6.7‐Mb mosaic duplication at Xp22.33‐p22.31 (48% mosaicism), one had a 6.0‐Mb mosaic deletion in the same region (61% mosaicism), and one presented a 6.2‐Mb mosaic deletion at Xq27.3‐q28 (65% mosaicism). The remaining three cases (Cases 16–18) were detected exclusively by FISH, with abnormal signal proportions of 8%, 13%, and 15%, respectively, while both karyotyping and SNP‐array yielded normal findings (Table 2). Regarding pregnancy outcomes, aside from Cases 3, 4, 7, and 15—which were terminated—pregnancy was continued in the remaining 14 cases, and no abnormalities were observed during follow‐up.

TABLE 2.

Analysis of 18 cases exhibiting discordant results in mosaicism involving the sex chromosomes.

Case	High risk/ultrasound findings	Karyotyping	SNP‐array	FISH	Pregnancy outcome
1	NIPT indicates X chromosome reduction	45,X[12]/46XX[64]	arr[hg19] (X) × 2	Nuc ish (DXZ1) × 1[15] (DXZ1) × 2[85]	LB, no visible abnormalities
2	Adverse pregnancy history (child with trisomy 21)	45,X[18]/46XX[82]	arr[hg19] (X) × 2	Nuc ish (DXZ1) × 1[12]/ (DXZ1) × 2[88]	LB, no visible abnormalities
3	AMA, Increased NT (3.0 mm)	45,X[54]/47,XXX[34]	arr[hg19] (X) × 2	Nuc ish (DXZ1) × 1[78] (DXZ1) × 3[22]	TOP
4	FGR, VSD	45,X[36]/47,XXX[44]	arr[hg19] (X) × 2	Nuc ish (DXZ1) × 1[52] (DXZ1) × 3[48]	TOP, VSD was found on autopsy
5	AMA	45,X[10]/47,XXX[12]/46,XX[78]	arr[hg19] (X) × 2	Nuc ish (DXZ1) × 1[15], (DXZ1) × 3[12],(X) × 2[73]	LB, no visible abnormalities
6	AMA, choroid plexus cysts	45,X[10]/47,XXX[12]/46,XX[78]	arr[hg19] (X) × 2	Nuc ish (DXZ1) × 1[15], (DXZ1) × 3[12],(DXZ1) × 2[73]	LB, no visible abnormalities
7	NIPT indicates X chromosome reduction, oligohydramnios	45,X[38]/47,XXX[42]/46,XX[20]	arr[hg19] (X) × 2	Nuc ish(DXZ1) × 1[42], (DXZ1) × 3[43],(DXZ1) × 2[15]	TOP
8	NIPT indicates X chromosome reduction	45,X[26]/47,XXX[38]/46,XX[36]	arr[hg19] (X) × 2	Nuc ish(DXZ1) × 1[42], (DXZ1) × 3[25],(DXZ1) × 2[33]	LB, no visible abnormalities
9	NIPT indicates X chromosome reduction	45, psu dic(X;Y)(p22.3;p11.3)[15]/46,XY[85]	arr[hg19] (X,Y) × 1	45.ish psu dic(X;Y)(p22.3;p11.3)[12]/46,XY[88]	LB, no visible abnormalities
10	History of adverse pregnancy, AMA	45,X[3]/45,X, psu dic(14;Y)(p11.2;p13)[97]	arr[hg19] (X,Y) × 1	45,X[5]/45,X.ish psu dic(14;Y)(p11.2;p13)[95]	LB, no visible abnormalities
11	NIPT indicates X chromosome reduction	45,X[5]/46,X,t(Y;7)(q11.23;p23)[95]	arr[hg19] (X,Y) × 1	45,X[2]/46.X,ish t(Y;7)(q11.23;p23)[98]	LB, no visible abnormalities
12	High risk of serum screening	46,X,t(X;5)(q11.2;p13.1)[25]/46,XX[75]	arr[hg19] (X) × 2	46,X.ish t(X;5)(q11.2;p13.1)[21]/46,XX[79]	LB, no visible abnormalities
13	AMA, High risk of serum screening	46,XX	arr[hg19]Xp22.33p22.31(168,552_6,867,414) × 2–3 (48%) 6.7 Mb	46,XX.ish dup(X)(p22.33p22.31)[55]/46,XX[45]	LB, no visible abnormalities
14	NIPT indicates X chromosome reduction	46,XX	arr[hg19]Xp22.33p22.31(168,552_6,197,423) × 1–2 (61%) 6.0 Mb	46,XX.ish del(X)(p22.33p22.31)[72]/46,XX[28]	LB, no visible abnormalities
15	AMA	46,XX	arr[hg19]Xq27.3q28(146,626,296‐152,824,069) × 1–2 (65%) 6.2 Mb	46,XX.ish del(X)q27.3q28[82]/46,XX[18]	TOP
16	NIPT indicates X chromosome reduction	46,XY	arr[hg19] (X,Y) × 1	Nuc ish (DYZ3) × 0[8] (DYZ3) × 1[92]	LB, no visible abnormalities
17	Increased NT (3.0 mm)	46,XX	arr[hg19] (X) × 2	Nuc ish(DXZ1) × 1[13] (DXZ1) × 2[87]	LB, no visible abnormalities
18	NIPT indicates X chromosome reduction	46,XX	arr[hg19] (X) × 2	Nuc ish(DXZ1) × 1[15] (DXZ1) × 2[85]	LB, no visible abnormalities

Abbreviations: AMA, advanced maternal age; FGR, fetal growth retardation; FISH, fluorescence in situ hybridization; LB, Live birth; NIPT, non‐invasive prenatal testing; NT, nuchal translucency; SNP, single nucleotide polymorphism; TOP, termination of pregnancy; VSD, ventricular septal defect.

3.4. Mosaicism Detected by Karyotyping and FISH, With Normal SNP‐Array Results

Among the 18 cases with discordant results, 12 were identified as mosaicism involving the sex chromosomes by both karyotyping and FISH, whereas SNP‐array failed to detect the abnormality. This group included two cases of 45,X/46,XX, two cases of 45,X/47,XXX, four cases of 45,X/47,XXX/46,XX, and four cases (Cases 9–12) involving mosaicism with balanced structural abnormalities of the X or Y chromosome. In Case 9 (Figure 2A), the karyotype was 45,psu dic(X;Y)(p22.3;p11.3)[15]/46,XY[85]. Peripheral blood chromosome analysis of the father revealed a normal 46,XY karyotype, indicating that the abnormal cell line in the fetus was de novo. In Case 10 (Figure 2B), the karyotype was 45,X[2]/45,X,psu dic(14;Y)(p11.2;p13)[98], and the father's peripheral blood result was 45,X,psu dic(14;Y)(p11.2;p13), confirming paternal inheritance of the abnormal cell line. In Case 11 (Figure 2C), the karyotype was 45,X[4]/46,X,t(Y;7)(q11.23;p23)[96], and the father also carried 46,X,t(Y;7)(q11.23;p23), indicating paternal origin of the abnormal cell line. In Case 12 (Figure 2D), the karyotype was 46,X,t(X;5)(q11.2;p13.1)[25]/46,XX[75], while both parents had normal peripheral blood chromosomes, suggesting that the abnormal cell line occurred de novo. In Cases 9–12, interphase FISH using X/Y centromere probes (CSPX p11.1‐q11.1/CSPY p11.1‐q11.1) did not detect chromosome mosaicism. However, after karyotyping revealed mosaicism, custom‐designed region‐specific FISH probes were employed and confirmed the diagnosis of mosaicism involving structural abnormalities of the sex chromosomes (Figure 3A–D).

FIGURE 2.

Karyotyping of Cases 9–12. (A) Case 9 karyotype showing 45,psu dic(X;Y)(p22.3;p11.3)[15]/46,XY[85] mosaicism. Arrows indicate the derivative X and Y chromosomes. (B) Case 10 karyotype showing 45,X[2]/45,X,psu dic(14;Y)(p11.2;p13)[98] mosaicism. Arrows indicate the derivative chromosome 14 and Y. (C) Case 11 karyotype showing 45,X[4]/46,X,t(Y;7)(q11.23;p23)[96] mosaicism. Arrows indicate derivative chromosomes 7 and Y. (D) Case 12 karyotype showing 46,X,t(X;5)(q11.2;p13.1)[25]/46,XX[75] mosaicism. Arrows indicate derivative chromosomes 5 and Y.

FIGURE 3.

FISH results with targeted probes in Cases 9–12. (A) Case 9, the Yp terminal probe (red signal) is located on the short arm of the X chromosome (white arrows). (B) Case 10, the Yp terminal probe (red signal) is located on the short arm of the chromosome 14 (white arrows). (C) Case 11, the 7p terminal probe (red signal) and the Yq terminal probe (green signal) have swapped positions. (D) Case 12, the Xq11.2 probe (green signal) and the 5p terminal probe (red signal) have swapped positions.

3.5. SNP Array and FISH Detected Mosaic CNVs, Karyotyping Result Was Normal

Among the 18 cases with discordant results, three (Cases 13–15) were identified as mosaic CNVs by both SNP‐array and FISH, while karyotyping yielded normal results. In Case 13, SNP‐array revealed arr[hg19] Xp22.33p22.31(168,552_6,867,414) × 2–3 (Figure 4A), indicating a mosaic duplication of approximately 6.7 Mb with a mosaicism level of 48%. In Case 14, the‐array showed arr[hg19] Xp22.33p22.31(168,552_6,197,423) × 1–2 (Figure 4B), corresponding to a mosaic deletion of approximately 6.0 Mb with 61% mosaicism. In Case 15, an SNP‐array identified arr[hg19] Xq27.3q28(146,626,296_152,824,069) × 1–2 (Figure 4C), representing a mosaic deletion of around 6.2 Mb with 65% mosaicism. Interphase FISH using X/Y centromere probes (CSPX p11.1–q11.1/CSPY p11.1–q11.1) failed to detect mosaicism in these cases. Subsequently, custom‐designed FISH probes targeting the regions specified by the SNP‐array findings were used for verification. The FISH results confirmed the Turner mosaicism as follows: Case 13: 46,XX.ish dup(X) (p22.33p22.31)[55]/46,XX[45] (Figure 5A); Case 14: 46,XX.ish del(X) (p22.33p22.31)[72]/46,XX[28] (Figure 5B); Case 15: 46,XX.ish del(X)(q27.3q28)[82]/46,XX[18] (Figure 5C).

FIGURE 4.

SNP‐array results of Cases 13–15. (A) SNP‐array result of case 13 shows a mosaic duplication at Xp22.33p22.31 with a mosaicism level of 48% (red arrow). (B) SNP‐array result of Case 14 shows a mosaic deletion at Xp22.33p22.31 with a mosaicism level of 61% (red arrow). (C) SNP‐array result of Case 15 shows a mosaic deletion at Xq27.3q28 with a mosaicism level of 65% (red arrow).

FIGURE 5.

FISH results with targeted probes in Cases 13–15. (A) Case 13 metaphase FISH with duplicated signals (white arrows) at Xp22.33p22.31. (B) Case 14 metaphase FISH with signal absence (white arrows) at Xp22.33p22.31. (C) Case 15 metaphase FISH with signal absence (white arrows) at Xq27.3q28.

4. Discussion

The formation of chromosomal mosaicism primarily arises from chromosomal instability during early embryonic development. Key mechanisms include non‐disjunction, anaphase lagging or chromosome loss, and structural rearrangements occurring during meiosis or mitosis [9]. In sex chromosome mosaicism, the predominant cause is delayed separation or loss of chromatids during mitotic anaphase in the early cleavage divisions of the fertilized egg, resulting in numerical abnormalities of the X chromosomes. Individuals with sex chromosome mosaicism typically harbor one or more abnormal cell lines. The complete or partial loss of the X chromosome in these cell lines leads to haploinsufficiency and altered gene dosage effects [10]. Importantly, genes that escape X‐chromosome inactivation are expressed at approximately half the normal level, and many of these genes are essential for height attainment, skeletal development, and maintenance of ovarian function [11]. The influence of mosaicism on clinical phenotype may display a “threshold effect,” whereby a critical proportion of abnormal cells is necessary to elicit significant clinical manifestations; conversely, low‐level mosaicism may preserve relatively normal physiological function through epigenetic compensatory mechanisms [12]. Therefore, in prenatal diagnosis, it is essential to select reliable detection methods to accurately assess the degree of fetal mosaicism and, when necessary, to combine multiple techniques for mutual verification. This approach is critical for providing accurate genetic counseling and guiding pregnancy management decisions. Currently, the main methods for detecting fetal chromosomal abnormalities in prenatal diagnosis include karyotyping, SNP‐array, and FISH. Due to differences in their detection principles, these techniques vary in resolution and detection scope [13]. In this study, 82 cases of mosaicism involving the sex chromosomes were identified through the combined use of karyotyping, SNP‐array, and FISH. Of these, 52 cases (63.4%, 52/82) involved X chromosome aneuploidy mosaicism, 15 cases (18.3%, 15/82) were associated with structural abnormalities of the X chromosome. Concordant results across all three methods were observed in 64 cases, while discrepancies were noted in 18 cases (21.9%, 18/82). These findings demonstrate that prenatal mosaicism involving the sex chromosomes has a significant clinical incidence and shows substantial heterogeneity in its karyotypic composition. Furthermore, the results highlight the risk of underdiagnosis or misclassification when relying on a single detection method. Therefore, the application of rational, effective, and accurate genetic approaches to confirm both the presence and extent of fetal mosaicism is essential for informing pregnancy management decisions, evaluating fetal prognosis, and devising individualized management and treatment strategies.

Karyotyping, as a classical technique with decades of application, features well‐established and standardized laboratory protocols and interpretation criteria. It has long been regarded as the “gold standard” for diagnosing chromosomal disorders—including mosaicism—in numerous genetic diagnostic guidelines [14]. This method enables direct visualization of prenatal mosaicism involving both numerical and structural anomalies, which is essential for elucidating the mechanisms underlying mosaicism formation and for predicting clinical phenotypes [15]. However, with the widespread incorporation of molecular diagnostic technologies into clinical practice, the limitations of karyotyping have become increasingly evident. Its relatively low chromosomal resolution (approximately 5–10 Mb) may result in missed detection of low‐level mosaicism or mosaicism involving small segments of chromosomal abnormality. Consequently, a comprehensive prenatal diagnostic protocol has been established, typically combining QF‐PCR, karyotyping, and SNP‐array. In cases where NIPT suggests a possible sex chromosome abnormality, FISH is additionally performed as a supplementary diagnostic procedure [16]. Among the 18 cases with discordant results, 12 were identified as mosaicism by karyotyping, whereas SNP‐array results were normal. The specific karyotypes included two cases of 45,X/46,XX; two of 45,X/47,XXX; four of 45,X/47,XXX/46,XX; and four cases involving mosaicism with balanced structural abnormalities of the X or Y chromosome. Among them, cases with the karyotypes 45,psu dic(X;Y)(p22.3;p11.3)[15]/46,XY[85] and 46,X,t(X;5)(q11.2;p13.1)[25]/46,XX[75] showed, upon parental peripheral blood chromosome analysis, that the abnormal mosaic cell lines in the fetuses were due to de novo variants arising during embryonic development. In contrast, for cases with the karyotypes 45,X[2]/45,X,psu dic(14;Y)(p11.2;p13)[98] and 45,X[4]/46,X,t(Y;7)(q11.23;p23)[96], paternal peripheral blood verification confirmed that the abnormal mosaic cell lines in the fetuses were inherited from the father. Our findings confirm that karyotyping retains a unique clinical value in the initial diagnosis of mosaicism involving sex chromosome, especially in cases involving complex balanced structural rearrangements of the sex chromosomes or when the overall DNA dosage of the X chromosomes remains largely unchanged. Additionally, karyotyping enables the detection of fetal sex chromosome balanced translocation mosaicism, which facilitates tracing the origin of these abnormalities and provides important guidance for the couple's future reproductive planning [17].

SNP‐array is a highly effective molecular diagnostic tool for detecting mosaicism. With its superior resolution, it surpasses karyotyping in identifying copy number variations (CNVs) greater than 100 kb, thus aiding in the clarification of atypical prenatal clinical findings [18]. By analyzing the B‐allele frequency (BAF) of single‐nucleotide polymorphisms (SNPs), SNP‐array can detect mosaicism and provide reasonably accurate estimates of mosaic proportions through algorithmic analysis [19]. For instance, SNP‐array can definitively differentiate a 30% 45,X/70% 46,XX mosaic, whereas karyotyping can only approximate mosaic levels based on limited cell counts. Additionally, SNP‐array allows for the detection of loss of heterozygosity (LOH) and uniparental disomy (UPD) by examining allele‐specific information at SNP sites [20]. This technique analyzes DNA directly, independent of cell culture, thereby eliminating reliance on viable amniocytes and active cell division and reducing the risk of clonal bias associated with culture‐based methods. Despite these advantages, the SNP‐array has inherent limitations, including its inability to reliably identify balanced chromosomal rearrangements and reduced sensitivity in detecting very low‐level mosaicism [21]. Consequently, its integration with complementary techniques such as karyotyping and FISH is widely advocated in clinical practice to ensure a thorough and precise genetic diagnosis. Among the 18 discordant cases, three were identified as mosaicism by SNP‐array with normal karyotyping results. These comprised one case of a 6.7 Mb mosaic duplication at Xp22.33p22.31 (48% mosaicism), one case of a 6.0 Mb mosaic deletion in the same region (61% mosaicism), and one case of a 6.2 Mb mosaic deletion at Xq27.3q28 (65% mosaicism). These results further validate the unique advantage of SNP‐array in detecting small‐fragment copy number variation mosaicism involving sex chromosomes.

Fluorescence in situ hybridization (FISH) utilizes fluorescently labeled DNA probes that specifically bind to targeted sequences on sex chromosomes, such as the centromeric region of the X chromosome or the SRY gene on the Y chromosome [22]. This technique allows for precise quantification of abnormal cells in cases of sex chromosome mosaicism by accurately distinguishing the proportions of normal and abnormal karyotypes (e.g., 45,X; 47,XXY; 47,XXX). Unlike conventional karyotyping, FISH does not require metaphase chromosome preparation and can be performed directly on interphase nuclei, eliminating the need for cell culture [23]. This feature significantly enhances the sensitivity for detecting low‐level mosaicism, with a detection threshold as low as 5%–10%. However, the primary limitation of FISH is that its detection scope depends entirely on the chromosomal regions targeted by the probes. If chromosomal abnormalities lie outside the regions covered by the probes, they cannot be effectively identified, potentially resulting in false‐negative findings [24]. Therefore, in prenatal diagnosis, FISH is generally regarded as a rapid initial screening or auxiliary verification tool, and it should be combined with techniques such as karyotyping and SNP‐array to achieve a comprehensive diagnostic workflow [25]. When karyotyping or SNP‐array suggests the possible presence of subtle chromosomal structural variation mosaicism, targeted FISH probes can be specifically designed for further verification. This approach not only confirms the diagnosis but also enables accurate assessment of the mosaicism ratio [26]. Among the 18 cases with discordant results, SNP‐array failed to detect abnormalities in 2 cases of 45,X/46,XX mosaicism, 2 cases of 45,X/47,XXX mosaicism, and 4 cases of 45,X/47,XXX/46,XX mosaicism, whereas FISH verification produced results consistent with karyotyping. In another 4 cases of mosaicism involving balanced X/Y translocations, and 3 cases of copy number variation mosaicism detected by SNP‐array, customized targeted FISH probes provided findings concordant with those from karyotyping and SNP‐array. Additionally, FISH independently identified three cases of low‐level 45,X/46,XX mosaicism. Collectively, these results highlight the unique diagnostic value of FISH in the detection of complex abnormal cell lines, low‐level sex chromosome mosaicism, and abnormalities involving targeted chromosomal regions.

The principal strength of this study lies in the systematic analysis of 82 prenatally diagnosed cases of mosaicism involving the sex chromosomes in a comparatively large fetal cohort. We comprehensively assessed the concordance and discordance among karyotyping, SNP‐array, and FISH results, offering a detailed comparison of their technical strengths and limitations in detecting mosaicism. However, several limitations should be acknowledged. Firstly, in cases of continued pregnancy resulting in phenotypically normal live births, some families declined postnatal sampling, precluding direct comparison of prenatal and postnatal genetic data. Secondly, the children in the follow‐up cohort are still relatively young, while key clinical features of sex chromosome mosaicism—such as the development of secondary sexual characteristics—can only be fully evaluated after puberty. Thirdly, mosaic forms of male sex chromosome aneuploidy (e.g., 46,XY/47,XXY or 46,XY/47,XYY) are relatively uncommon and may be under detected in prenatal cohorts, which likely contributes to their limited representation in this study. Additionally, although our sample size is larger than those in previous studies that were primarily case‐based, the retrospective design and limited cohort size may still impact the generalizability of our conclusions.

5. Conclusion

This study demonstrates that the combination of SNP‐array, karyotyping, and FISH can effectively diagnose mosaicism involving the sex chromosomes prenatally, particularly when NIPT findings indicate fetal sex chromosome abnormalities. In prenatal diagnosis, Turner mosaicism is the most common type of sex chromosome mosaicism identified, representing a wide spectrum of mosaic patterns. Karyotyping is particularly effective for identifying X chromosome aneuploidy mosaicism without significant alteration in DNA dosage, as well as for detecting cases involving balanced chromosome translocations. In contrast, SNP‐array offers high sensitivity in detecting mosaic copy number variations smaller than 10 Mb on X chromosomes. When mosaicism is suggested by karyotyping or SNP‐array, FISH should be employed for confirmation. Additionally, FISH allows for the sensitive detection of low‐level Turner mosaicism directly in uncultured interphase cells from amniotic fluid.

Funding

The study was supported by the Shenzhen Science and Technology Plan Project (Grant No. JCYJ20220530162412029), the Key Clinical Projects of Peking University Third Hospital (Grant No. BYSYZD2024010), the National Clinical Research Center for Obstetrics and Gynecology (Peking University Third Hospital) (Grant No. NCRC2025B06), and the Key Laboratory of Assisted Reproduction (Peking University), Ministry of Education (Grant No. BYSYSZKF2023003).

Ethics Statement

This study was approved (No. LGFYYXLL‐028) by the Ethics Committee of Longgang District Maternity & Child Healthcare Hospital of Shenzhen City.

Consent

All patients provided written informed consent for the use of their biological samples for research, in accordance with Institutional Review Board requirements and the Helsinki's declaration. Consent for publication has been obtained from the patient and/or their legal guardians.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.