A quartet of recent papers authored by geneticists at Birmingham Women's Hospital has shed greater light on the application of chromosomal microarrays in prenatal diagnostics in the UK.
The studies, which appeared in four journals over the past two months, compared array-based testing to conventional karyotyping, compared lower-resolution array platforms with higher-resolution chips, and the authors discussed issues related to variants of unclear significance, as well as patient counseling.
Sarah Hillman, a clinical research fellow at BWH, is the lead author on all four papers. She told BioArray News this week that the new studies are partly the result of a three-year project funded by Sparks, a London-based medical research charity.
Between 2009 and 2012, Hillman said that BWH together with colleagues at West Midlands Genetics Laboratory performed array CGH on 243 prenatal samples, when couples had opted for prenatal invasive testing and their baby had a structural anomaly on prenatal ultrasound scan. According to Hillman, the teams used BlueGnome's bacterial artificial chromosome arrays for most of the study. BlueGnome has since been acquired by Illumina. They also ran Agilent Technologies-manufactured oligonucleotide arrays on some of the samples after switching to an oligo array platform last year.
Hillman and the coauthors detailed this study in an Ultrasound in Obstetrics and Gynecology paper in June. In it, the team put their own comparison of CMA with conventional cytogenetics in the context of a review and meta-analysis of literature describing detection rates by both approaches.
Using CMA, they were able to detect 4.1 percent more abnormalities than with conventional cytogenetics, when the clinical indication for testing was an abnormal fetal ultrasound finding. The rate of detection for variants of unknown significance was 2.1 percent when the indication was an abnormal scan finding. The authors concluded that it is likely that CMA will replace karyotyping in high-risk pregnancies.
In a separate study, the same team compared CMA with conventional karyotyping when diagnosing chromosomal abnormalities in miscarriage. Published online in the journal BJOG last week, the authors relied on a system review and meta-analysis of nine studies.
Overall, they found agreement between CMA and karyotyping in 86 percent of cases, and reported that CMA detected 13 percent additional chromosome abnormalities over conventional full karyotyping. The incidence of variants of unknown significance was 2 percent.
Arrays are increasingly being adopted in stillbirth studies, as several presentations at last month's Association of Genetic Technologists meeting in Las Vegas demonstrated (BAN 6/18/2013).
Hillman said that the BWH researchers looked at stillbirth data as part of an exercise in comparing diagnostic yield in CMA versus conventional karyotyping, and that BWH does not plan to adopt arrays for testing products of conception at the moment.
A third study, published online last week in the Journal of Maternal-Fetal and Neonatal Medicine, examined how altering the resolution of chromosomal microarray analysis in the prenatal setting affects the rates of pathological and uncertain findings.
According to Hillman, BWH last year adopted an Agilent-made, 60,000-probe oligo array for CMA. The JMFNM study compared findings made with the BlueGnome BAC array with the Agilent chip.
As reported in the June UOG paper, the BAC array detected pathogenic copy number variants in 4.1 percent of cases. The 60K array, in comparison, had an additional pathogenic detection rate of 4.8 percent over the BAC array but also detected an additional 8 percent variants of unknown significance compared to a targeted BAC array variants of unknown significance detection rate of 0.4 percent. "As the CMA platform resolution increases detection rates increase but are associated with an increase in VOUS rates," the authors noted in their conclusion.
They also called for further large-scale studies to inform the UK's consensus on the resolution required and on reporting of variants of unknown clinical significance in the prenatal setting.
At BHW, Hillman said that they are still performing CMA when there is a structural anomaly of a scan. The team is also running arrays on prenatal samples as part of the UK's Evaluation of Array Comparative Genomic Hybridization in Prenatal Diagnosis of Fetal Anomalies, or EACH, study.
Commenced last year, EACH is a three-year study that is evaluating the potential of CMA for the prenatal diagnosis of fetal anomalies. Ultimately, the British researchers aim to determine whether CMA can detect more clinically significant abnormalities in fetuses compared to conventional karyotyping.
The fourth recent paper to come out of BWH's CMA studies appeared in the American Journal of Medical Genetics this month. In the paper, the authors discussed the results of 25 interviews with women and their partners after receiving the results of array-based testing.
According to Hillman, the BWH team found that there were a number of misconceptions about array technology, especially in terms of what couples expected from CMA.
"There can be many misconceptions, often because people retain little information at such a stressful period in their lives, but the most common that I came across was that people thought if the array was normal that their child could not have anything genetically wrong, although we did explain the limited resolution of the array," said Hillman.
Another issue concerned reporting variants of unknown clinical significance. Such findings present a "particularly difficult challenge," Hillman said, and "good clear communication by healthcare professionals is paramount."
Based on the study, Hillman and the coauthors made other suggestions. "When counseling women and their partners for fetal chromosomal testing it should be reinforced that the most common, trisomy 13, 18, and 21, only account for some of the chromosomal changes resulting in abnormal scan findings," Hillman said. Also, "couples should have literature to take home summarizing scan anomalies and reinforcing information about genetic testing."
As part of their work with EACH, the BWH team is not expected to report variants of unknown significance to couples, Hillman said. Instead, such findings are reported back to a panel of experts for analysis. Any variants identified by the panel as pathogenic are subsequently reported to parents.
In the prior, Sparks-funded study, she said that the use of a targeted BAC array meant that the team only had to discuss one such variant, the reporting of which was "handled sensitively with quick follow up with a fetal medicine specialist and clinical geneticist."
Given BWH's experiences with prenatal CMA, Hillman said that the adoption of next-generation sequencing in the clinical setting will "require serious consideration." Variants of unclear clinical significance that are uncovered using sequencing will need to go to a "panel of experts to review the information and decide whether the link to pathogenicity is strong enough to tell parents."
Additionally, Hillman said that couples undergoing analysis by sequencing should be informed that the technology might identify variants that "would not be reported if deemed to have tenuous link so as not to create unnecessary anxiety."
She noted that more information continues to be gathered on questionable variants, but said that, in regards to the implementation of next-generation technologies in prenatal diagnostics, "further large-scale qualitative research is required to assess the thoughts, feelings, and perceptions of women and their partners undergoing the prenatal testing."