Microarray-Based Prenatal Diagnosis for the Identification of Fetal Chromosome Abnormalities

Lisa G Shaffer; Jill A Rosenfeld

Disclosures

Expert Rev Mol Diagn. 2013;13(6):601-611. 

In This Article

Limitations of Microarray Analysis

Microarray analysis will not detect certain chromosome rearrangements, such as balanced translocations (reciprocal and Robertsonian translocations) and inversions because, although there has been an exchange of DNA, there is no net gain or loss of DNA detectable by microarrays. Therefore, traditional cytogenetic analysis still plays an important role in prenatal testing in the identification of balanced chromosomal rearrangements. When an apparently balanced rearrangement is seen by karyotyping, microarray analysis could be considered to exclude an imbalance (gain or loss) at one or both of the translocation breakpoints.[44,73,74] Recently, 239 cases of apparently balanced rearrangements detected by fetal karyotyping were examined by microarray.[44] Of these, 7.9% were found to have a gain or loss at one or both of the translocation breakpoints (Figure 2). Another 1.7% had a clinically significant gain or loss at another location in the genome that explained the prenatal phenotype (Figure 3). Thus, nearly 10% of cases were found to have a chromosomal imbalance at the translocation breakpoints or in another region unrelated to the karyotype results. However, some balanced rearrangements may result in an abnormal phenotype because of disruption at one or both translocation breakpoints. Because there is no net gain or loss of DNA, these gene disruptions will not be detectable by microarray. Recently, whole-genome sequencing was applied to a prenatal case that had an apparently balanced translocation by karyotype and microarray analyses.[75] Using whole-genome jumping libraries and sequencing, a disruption of the CHD7 gene that causes CHARGE syndrome was revealed. Although this technique shows great promise as an adjunct to prenatal microarrays, it requires the knowledge of the potential translocation breakpoint locations. Therefore, microarray analysis alone would not reveal a balanced translocation and would be missed. The use of both karyotyping and microarray analysis provides different viewpoints of the genome that allows for a comprehensive assessment of the fetal chromosomes that cannot be achieved with either method in isolation.[76] Therefore, laboratories may consider doing a limited karyotype on approximately five cells to exclude a visible aneuploidy or rearrangement with concurrent microarray analysis to assess the genome. If a balanced rearrangement is found and microarray analysis does not reveal any genomic imbalance, one could consider the methods of Talkowski et al. to further assess whether the translocation breakpoint disrupted a gene or genes in those de novo, apparently balanced rearrangements.[75]

Figure 2.

Microarray result of an apparently balanced translocation t(16;17)(q21;q24) by karyotyping that showed a 2.7 Mb deletion at the 17q24.3 translocation breakpoint.

Figure 3.

Microarray result of an apparently balanced translocation t(5;19)(q11.2;q13.1) by karyotyping showing no gain or loss at the translocation breakpoints. However, a 300 kb deletion of 18p11.31, including the TGIF1 gene, was identified and is likely the cause of the holoprosencephaly seen in the fetus.

In other situations, normal microarray analysis in the presence of an abnormal karyotype, such as marker chromosomes, is reassuring, as it indicates that the marker is lacking euchromatic material and unlikely to cause disease. In a recent series of 271 fetuses with apparently unbalanced karyotypes, 30.3% (including 43.2% of cases with marker or ring chromosomes) had normal array results, suggesting either variation involving only heterochromatic (gene-poor) material or misinterpretation of the karyotype.[44] Combining a limited karyotype with the more objective microarray data will allow for a clearer interpretation for any disease-causing chromosome abnormalities, if any, in the fetus undergoing testing.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....