This article was originally posted on Dec. 6.
Several studies published last week in the New England Journal of Medicine underscore the rapidly shifting landscape for genomic tools in the diagnostic setting.
One study, led by researchers at Columbia University Medical Center, indicated that chromosomal microarrays can provide more clinically relevant information than traditional karyotyping, and should become the standard approach in prenatal diagnostics going forward.
The chromosomal array community has anticipated this paper since February, when lead author Ronald Wapner presented preliminary data at the Society for Maternal-Fetal Medicine meeting in Dallas (BAN 2/14/12). Since then, several study participants have said they believe the results could influence decision makers to revisit guidelines concerning prenatal genetic testing.
However, another paper also appearing in NEJM last week, which used whole-genome sequencing for the clinical diagnosis of a prenatal sample, highlights how sequencing is being used increasingly for prenatal testing. Indeed, a number of firms, such as Sequenom and Ariosa, have already launched commercial tests that use sequencing to noninvasively detect aneuploidies in prenatal samples.
The Columbia-led study aimed to compare the amount of clinically relevant information that could be obtained from arrays versus karyotyping, which has been the standard of care since the 1970s. Using arrays manufactured by Affymetrix and Agilent Technologies as well as karyotyping, they conducted a blinded trial of 4,400 patients at 29 centers across the US over a four-year period. Amniotic fluid samples were collected from women with advanced maternal age and those whose fetuses were shown in early screening to be at heightened risk for Down syndrome, to have structural abnormalities as seen with ultrasound, or to have indications of other problems. Wapner, professor and vice chairman for research at the department of obstetrics and gynecology at CUMC, was the principal investigator on the study.
The study found that array-based analysis performed as well as karyotyping in identifying common aneuploidies that can cause genetic disorders such as Down syndrome and Edwards syndrome. Moreover, among fetuses in which a growth or structural anomaly had already been detected with ultrasound, microarrays were able to detect clinically relevant chromosomal deletions or duplications in 6 percent of cases, changes that were not observed with karyotyping the same samples. Additionally, in cases sampled for advanced maternal age or positive screening results, array analysis picked up an abnormality in one out of every 60 pregnancies, about 1.7 percent of all cases that had a normal karyotype.
Based on these results, Wapner said in a statement last week that arrays "will and should replace karyotyping as the standard for evaluating chromosomal abnormalities in fetuses," adding that chromosomal microdeletions and duplications found with microarray are "often associated with significant clinical problems."
Currently, samples used in array analyses require fetal cells obtained via invasive procedures, such as amniocentesis or chorionic villus sampling. Should a noninvasive means for testing fetuses for genetic abnormalities become available, Wapner said that all women, not just those in special risk categories, should have access to the information obtainable by array analysis.
"We hope that in the future — when microarray can be done non-invasively — every woman who wishes will be offered microarray, so that she can have as complete information as possible about her pregnancy," Wapner said in a statement.
Wapner did not respond to questions about the study in time for this publication. However, BioArray News spoke with him at length about the project two years ago (BAN 6/22/2012), and about a National Institutes of Health-funded follow-on study that commenced earlier this year (BAN 4/3/2012).
For the new project, researchers led by Wapner will use a five-year, $5 million grant from the National Institute of Child Health and Human Development to track pediatric patients who had an abnormal array result of unclear significance over a three-year period to obtain phenotypic information that may result from the abnormality. The team also intends to build an interactive online educational resource for test providers and patients.
"Unfortunately, it is sometimes difficult to predict the full spectrum of some diseases indicated by a particular deletion or duplication," Wapner said this week. "We are studying what these mean clinically, and science continues to catch up with our ability to obtain the information."
Chromosomal microarray advocates awaited the results of the prenatal study since Wapner first presented initial results in February.
"I think the study is large enough to influence current guidelines and to inform that process in an important way," David Ledbetter, executive vice president and chief scientific officer for Geisinger Health System and a co-author on the study, told BioArray News in an interview earlier this year. "Because of the size of the study and its design, the data should be very useful in updating current thinking and guidelines," Ledbetter added.
The American College of Medical Genetics and Genomics updated its guidelines two years ago to recommend microarrays as a first-tier diagnostic test for the postnatal evaluation of patients with developmental delay or intellectual disability, autism spectrum disorders, or multiple congenital anomalies. However, the organization's guidelines continue to state that arrays should only be used as an adjunctive test to karyotyping in a prenatal setting.
Despite this, many institutions and companies have been offering array-based prenatal testing for years. Baylor College of Medicine was arguably first in the US to begin offering array-based prenatal testing in 2004, and others, such as Signature Genomic Laboratories, now part of PerkinElmer, and CombiMatrix Molecular Diagnostics began offering similar services in following years. In these services, the assays are performed as laboratory-developed tests in Clinical Laboratory Improvement Amendment-compliant facilities, and findings are confirmed using other technologies, such as fluorescence in situ hybridization.
Nevertheless, companies and institutions that offer array-based prenatal testing expect that volumes will increase should ACMG revise its guidelines to make chromosomal microarray analysis the recommended, first-tier approach for prenatal diagnostics. CombiMatrix CEO Judd Jessup said as much during a recent earnings call. "The findings of [the] study prove the benefits of microarrays, and thought leaders are confident that the paper will usher in a paradigm shift in prenatal testing favoring microarray analysis over the standard methods," Jessup said ( BAN 8/14/2012).
Wapner became a member of CombiMatrix's scientific advisory board earlier this year. According to Jessup, CombiMatrix is working with Wapner and others to "make sure that CombiMatrix delivers a responsible message to the marketplace and is in the best position to take commercial advantage as the shift toward microarrays takes place."
Another paper in NEJM discussed the ability of chromsomal microarrays and karyotyping to detect clinically relevant results in stillbirth cases. According to the paper, genetic abnormalities have been associated with between six and 13 percent of stillbirths, but the true prevalence may be higher. Members of the NICHD-sponsored Stillbirth Collaborative Research Network used karyotyping and Affymetrix SNP 6.0 arrays to survey CNVs of at least 500 kilobases in placental or fetal tissue in 532 cases and then compared the results.
They found that array analysis yielded results more often than did karyotype analysis. Specifically, arrays provided results 87 percent of the time, versus karyotyping, which yielded results 71 percent of the time. As compared with karyotype analysis, microarray analysis provided a relative increase in the diagnosis of genetic abnormalities of 42 percent in all stillbirths, 35 percent in antepartum stillbirths, and 54 percent in stillbirths with anomalies.
The authors concluded that array-based analysis of stillbirth is more likely than karyotype to provide a genetic diagnosis, "primarily because of its success with nonviable tissue," as, unlike karyotyping, chromosomal microarray analysis does not require cultured cells and can be performed using DNA extracted directly from the tissue received.
Brynn Levy, director of the Clinical Cytogenetics Laboratory at New York-Presbyterian/Columbia and a coauthor on the paper, told BioArray News that karyotyping's higher rate of failure to return a result is tied to the viability of the cells that are received for cytogenetic analysis. "Remember that cytogenetic studies require cell culturing to study the chromosomes," said Levy. "If the lab receives dead or dying tissue, the likelihood of establishing a cell culture is significantly reduced," he said.
Levy noted that not being able to explain why a stillbirth occurred can be "very hard" for families, and that he can "only see a benefit" in using arrays as a first-tier approach to analyze stillbirth samples, the "primary benefit being the likelihood of actually yielding a result, which is often so important for the emotional closure of the grieving family." He added that the study's visibility in a "high impact journal" like NEJM will "hopefully be an impetus for greater utilization" going forward.
"Like most things in medicine, patients, clinicians and insurance companies want to see the evidence that justifies the clinical use of a technology," he said.
Looking ahead, Levy sees future studies for the network using genomic technologies. He said that the "big issue" raised by the present study is causality of smaller genomic imbalances in the etiology of stillbirths. The "natural next step" would be to investigate newly discovered genomic aberrations and discern their relationship to the stillbirth, Levy said. At the same time, the number of samples required to power such a study is "very significant" and until funds are available, such follow-up studies will "remain in limbo," he said.
Levy added that, should he and fellow researchers obtain the resources to pursue such studies, they will be "more likely to perform these follow-up studies using the latest technologies like next-generation sequencing."
... and Sequencing
NGS was in focus in the third prenatal diagnostics-focused paper featured in NEJM last week. The authors reported the use of whole-genome sequencing on the Illumina HiSeq 2000 to identify translocation breakpoints in a sample from a fetus with an isolated heart defect at 19 weeks of gestation and additional abnormalities that had been revealed via imaging studies performed throughout the third trimester.
According to the paper, after the child was delivered it was diagnosed with CHARGE syndrome, a rare genetic syndrome with multiple symptoms, including heart defects and developmental delay. Studies have indicated that mutations in the CHD7 gene are associated with the syndrome, but only in 60 percent of cases, meaning that in the case of the patient, the syndrome "could not have been unequivocally diagnosed on the basis of ultrasonography, original karyotyping, or subsequent array-based [comparative genomic hybridization] testing."
Instead, the authors, who included researchers from Harvard Medical School and Signature Genomics, sequenced the prenatal DNA sample and were able to identify precise translocation breakpoints that directly disrupted CHD7, the CHARGE-associated gene, and the LMBRD1 gene at a pathogenic locus in a recessive disorder of vitamin B12 metabolism.
The authors were therefore able to identify a pathogenic gene disruption by sequencing the DNA obtained from a prenatal sample with a balanced translocation, "providing a definitive sequence-based prenatal diagnosis that was consistent with the diagnosis based on postnatal clinical findings."
Based on this outcome, the authors argued in the paper that sequencing aimed specifically at detecting structural variations can offer a "rapid adjunct to cytogenetic techniques," because it "enables precise definition of individual disrupted genes," and provides information that can be used for outcome prediction, medical planning, and genetic counseling.
They noted that results obtained with karyotyping and array-based CGH were consistent with a balanced de novo translocation, but that these tests "did not identify the gene or genes responsible either for the isolated cardiac defect or for the additional fetal abnormalities that were subsequently detected."
In an editorial discussing the implications of the three papers, Lorraine Dugoff, associate professor of obstetrics and gynecology at the Hospital of the University of Pennsylvania, noted the various advantages of chromosomal microarrays and sequencing over karyotyping in making a clinical diagnosis, including a faster turnaround time, the elimination of the need to culture cells for analysis, and the higher yield of clinically relevant information. Still, she wrote that greater information yield can produce more findings of unclear clinical significance that can "cause considerable stress and anxiety for the parents, who may be considering termination of the pregnancy."
Indeed, Dugoff wrote that the "major disadvantage" of chromosomal microarrays and sequencing is this inability to interpret the clinical significance of an unreported CNV or to accurately predict the phenotype of pathogenic CNVs associated with variable expression and penetrance. She acknowledged international efforts to catalog genomic and clinical information, and wrote that it is "critical" to develop large databases of CNVs identified prenatally to "avoid the ascertainment bias associated with discovery in postnatal cases studied only because of existing abnormalities."
Additionally, Dugoff discussed some of the limitations of arrays, such as the technology's inability to identify balanced chromosomal arrangements like translocations or inversions, or to differentiate free trisomies from unbalanced Robertsonian translocations. "Although microarrays are useful in detecting copy-number variants in apparent new balanced translocations or other structural rearrangements, microarray analysis may fail to detect mutations in some cases," she wrote.
More specifically, Dugoff addressed the Columbia-led trial that compared arrays with karyotyping for prenatal diagnostics, stating that while the ability to detect more clinically useful CNVs is beneficial, it may be outweighed by the "counseling conundrum" that results in trying to translate such findings into information that can be relayed to clinicians and patients. She urged pretest and post-test counseling by trained genetics counselors and geneticists and wrote that pretest counseling must include a discussion of the genetic principles of uncertainty and variable expressivity, the lack of precise correlation between genotype and phenotype, and the possibility that genetic variants that cause adult-onset disorders may be identified in a fetus or a parent. Post-test counseling should include interpretation of the results and an explanation of indicated follow-up studies and the possible implications for the family.
A number of studies are underway or have been conducted to better understand the impact of unclear findings in the clinical setting. BioArray News spoke with researchers at the University of Pennsylvania earlier this year about a National Institutes of Health-funded study that seeks to determine how chromosomal microarray test results are communicated to the parents of children with autism spectrum disorder (BAN 3/27/2012). A special issue of Prenatal Diagnosis in April also looked at the ethical issues raised by array-based prenatal tests, with two papers specifically focused on genetic counseling (BAN 4/10/2012).
With regards to the stillbirth and sequencing papers, Dugoff wrote that the former provides a "good rationale for performing microarray analysis in cases of stillbirth, particularly when congenital anomalies are present." She also agreed with Levy that the identification of an abnormal result may "provide comfort, end the search for a cause, and help with the assessment of risk and the development of a plan of care for future reproduction."
And the sequencing paper, in Dugoff's words, illustrates both the limitations of chromosomal microarray to detect mutations as well as the advantage of a sequencing approach.
"Given the rapid advances and application of whole-exome sequencing in pediatric practice, it will not be long before this technology is ushered into the field of prenatal diagnosis," Dugoff wrote. Still, whole-exome sequencing faces a number of "future challenges," Dugoff wrote. These include "determining how to interpret large data sets" and "how to appropriately apply this information in clinical practice."
A number of targeted noninvasive sequencing-based tests are already on the market for prenatal testing. For example, GATC Biotech subsidiary LifeCodexx offers PrenaTest, a noninvasive, sequencing-based test to detect fetal trisomy 21. Verinata Health's Verifi prenatal test also detects trisomy 21, as well as trisomies 18 and 13, and common sex chromosome abnormalities. And Sequenom's MaterniT21 Plus test detects trisomies 21, 18 and 13.