A Challenging Prenatal QF-PCR Rapid Aneuploidy Test Result Caused by a Maternally Inherited Triplication within Chromosome Xq26.2

In pregnancies associated with an elevated risk of aneuploidy, genetic prenatal diagnosis (PND) may be performed through chorionic villus sampling or amniocentesis, and diagnostic genetic results can allow for informed management of the pregnancy. Common indications for PND include a positive multiple biochemical marker screen, a high-risk result on noninvasive prenatal testing, and abnormal ultrasound findings.

Several genetic techniques are routinely employed in PND and are selected on the basis of resolution and scope of genomic information sought to manage a particular case. Common cytogenetic tests used in PND include conventional chromosome karyotyping, fluorescence in situ hybridization (FISH), quantitative fluorescence PCR (QF-PCR), and array comparative genomic hybridization (array-CGH). If a result obtained by one method is inconclusive or unexpected, reflex testing by other techniques can be utilized to clarify the issue.

QF-PCR is a molecular-based technique useful for rapid aneuploidy detection (RAD), and is increasingly being used in PND due to its ability to provide information in a cost-effective and timely manner [Badenas et al., 2010]. Due to its accuracy and rapid turnaround time, QF-PCR has been suggested as the most appropriate first line test in PND of pregnancies at increased risk of aneuploidy of chromosomes 13, 18, 21, X, or Y [Langlois and Duncan, 2011].

In certain cases, QF-PCR results may be ambiguous or uninformative, requiring further analysis. One cause of an ambiguous result is when there is a single abnormal marker that is discordant from the other markers on the same chromosome that are normal. This can represent either a benign or pathogenic copy number variant (CNV) overlapping the marker or partial aneuploidy. Additional testing to resolve the source of discordance is often required.

In the case presented herein, the QF-PCR prenatal RAD result was confounded by the presence of a biallelic trisomic pattern for the X chromosome DXS1187 marker with an otherwise normal male amplification pattern. To clarify the fetal QF-PCR result, which was reported as uninformative for the X chromosome, additional analyses by array-CGH and metaphase FISH were performed on maternal and fetal DNA. This case represents a unique occurrence involving a maternally inherited 304-kb copy number triplication within chromosome Xq26.2 overlapping the DXS1187 marker which confounded the prenatal QF-PCR RAD result.

A 35-year-old woman presented with a positive first trimester screen (1/101 risk for trisomy 21) and a subsequent low-risk result for trisomy 21 on noninvasive prenatal testing. Family history revealed this couple had a prior pregnancy with trisomy 18 and 1 healthy son who had mild microtia but was otherwise healthy. The couple opted for prenatal diagnostic testing with an amniocentesis performed at 15 weeks and 2 days gestation.

RAD using QF-PCR was performed on DNA extracted from uncultured amniotic fluid collected at 15 weeks and 2 days gestation according to the manufacturer's instructions (Elucigene® QST*R, vs2, Hologic Gen-Probe Inc.). The same RAD kit was used on the maternal DNA sample which was extracted from peripheral blood. Metaphase FISH analysis was performed using the BAC probes RP11-427M2, which maps to the region of interest containing the DXS1187 marker within chromosome Xq26.2, and RP11-800K15 (X chromosome control probe mapping within chromosome Xp22.33). Both FISH probes were purchased from The Centre for Applied Genomics (TCAG, The Hospital for Sick Children, Toronto, ON).

Array-CGH was performed on DNA extracted from cultured amniotic fluid and from peripheral blood (maternal and maternal grandmother samples) hybridized with same-sex reference DNA from 100 healthy normal individuals (Kreatech). The Oxford Gene Technology CytoSure™ Constitutional v3 (8×60 k) oligonucleotide array platform was used. The array design was derived from the Deciphering Developmental Disorders (DDD) study, as well as ClinGen (formerly ISCA/ICCG) and Decipher data. Data were analyzed using CytoSure Interpret v4.9.40 with Circular Binary Segmentation detection algorithm [Venkatraman and Olshen, 2007].

RAD by QF-PCR analysis of the fetal DNA showed a normal diploid complement for chromosomes 13, 18, and 21. With the exception of the DXS1187 marker located at chromosome Xq26.2, the other fetal sex chromosome markers were consistent with a typical male amplification pattern. The DXS1187 marker showed a biallelic trisomic pattern with one peak migrating at 149.53 bp with an area of 2,848 units and a second peak migrating at 153.87 bp with an area of 4,736 units, corresponding to a 1:1.6 area ratio (Fig. 1A). Due to this discordant abnormal marker, the result was reported as uninformative for the sex chromosomes with additional studies required to complete the analysis.

Chromosome analysis performed on cultured amniotic fluid cells showed a normal male karyotype in all metaphases examined (data not shown).

Array-CGH analysis of fetal DNA was subsequently performed. Analysis of DNA extracted from cultured amniotic fluid cells identified a 358-kb copy number gain within the long arm of the X chromosome: arr[GRCh37] Xq26.2(130725010_131083764)×3. The gain had a log2 ratio of 1.48 which is indicative of a triplication (3 copies in a male) (Fig. 1B). In addition to encompassing the DXS1187 marker, this copy number gain includes 1 RefSeq gene, FIRRE, which encodes the firre intergenic repeating RNA element, a long noncoding RNA with no known disease association. Maternal array-CGH analysis identified a similar 304-kb copy number gain, however, the log2 ratio was 0.98 which is indicative of a triplication (mother has a total of 4 copies of this region) (Fig. 1C). The discrepant sizes of the maternal and fetal copy number gains are most likely artifactual due to noise, and the true size is represented in the maternal array as her array had a lower derivative log ratio spread (0.133 compared to 0.24). The origin of the maternal triplication could not be determined as array-CGH of the maternal grandmother was normal and the maternal grandfather was not available for testing. Thus, the copy number gain detected in the fetus is maternally inherited and was reported as likely benign in both mother and fetus along with an explanation for the discordant and ambiguous QF-PCR RAD result.

Postnatal follow-up in the Clinical Genetics clinic at 2 months of age reported a healthy male newborn with normal growth, development, and no dysmorphisms noted. Aside from recurrent miscarriages, there are no maternal health issues. She reports no history of learning disabilities in her or her siblings, all obtaining university degrees, and she is of average height for all the women in her family.

QF-PCR analysis on a maternal peripheral blood sample showed a normal female amplification pattern. Maternal peaks for the DXS1187 marker were identical in size to those detected in the fetal sample (149.53 bp and 153.87 bp), but occurred in peak areas of 12,304 and 11,834, respectively, corresponding to a 1:0.96 ratio, which is indicative of biallelic disomy (normal) in a female. However, with additional information derived from the array-CGH result, this roughly 1:1 ratio corresponds to 4 maternal copies of the DXS1187 marker. There are 2 copies of each size, representing biallelic tetrasomy for this region of the X chromosome. Molecular X chromosome inactivation studies using the CAG repeat within the AR gene showed no skewing of X inactivation with a ratio of 59:41 which falls within normal limits for random X inactivation (data not shown).

Metaphase FISH analysis on fetal amniotic cells was conducted to exclude an insertional translocation event that could have caused a disruption of another locus or gene. Fetal metaphase cells showed 1 target signal corresponding to the expected location on chromosome X containing the DXS1187 marker, indicative of a tandem copy number gain rather than a translocation. Maternal FISH analysis using the same BAC probe performed on peripheral blood showed 2 target signals, one on each X chromosome, also at the expected location; however, 1 signal was consistently enhanced relative to the other (data not shown).

Taken together, these results indicate that the mother carries 4 DXS1187 marker alleles, 3 of which are in cison the same X chromosome (149.53 bp, 153.87 bp, and 153.87 bp) and a single 149.87 bp allele in transon the other X chromosome. The male fetus inherited the maternal X chromosome harboring the 3 DXS1187 markers in cis (149.53 bp, 153.87 bp, and 153.87 bp).

To our knowledge, this is a unique case in which prenatal QF-PCR RAD led to the identification of a “benign” CNV encompassing the DXS1187 marker on the X chromosome.

Mechanistically, there are several scenarios that can be proposed to account for the unusual configuration of the maternal DXS1187 markers. At least 2 independent events are required to have occurred. Duplication of 1 allele in cis as well as an unequal recombination event to obtain a third copy of unequal size in ciswith the duplicated copies. Nonallelic homologous recombination events can potentially explain these complex rearrangements.

This case has informed the approach taken by our laboratory in response to ambiguous or discordant RAD results. In consultation with the referring clinician, array-CGH will be offered to clarify this further, with a discussion explaining the possibility of identifying variants of uncertain clinical significance and/or incidental findings that are unrelated to the initial indication for performing the analysis.

Stan et al. [2012] reported similar discordant RAD results caused by inherited and, thus, presumably benign CNVs overlapping select short tandem repeat (STR) loci. Given that these are not unique occurrences, the public CNV databases (Database of Genomic Variants, DECIPHER, ClinVar) are rich resources that should be reviewed when discordant RAD results are obtained. Given the lack of CNVs encompassing DXS1187, and potentially many other STR loci, a resource compiling copy number changes that encompass the common markers/loci interrogated in QF-PCR RAD panels is required. Such a resource would be helpful in the clinical cytogenetics laboratory as it may minimize additional studies that would otherwise be needed to clarify the clinical significance of discordant abnormal markers.

Informed consent to publish the findings of this case was obtained from the parents as per the mandate of the Research Ethics Board of the University of Calgary.

The authors declare no conflicts of interest.