Now in its ninth month, Baylor College of Medicine's Prenatal Chromosomal Microarray Analysis service — the first of its kind — is just beginning to gain momentum, according to Arthur Beaudet, director of Baylor's cytogenetics lab.
Beaudet, who also chairs Baylor's Department of Molecular and Human Genetics, told BioArray News that demand so far has been "relatively minimal, [but] it's just starting to increase. I think we are doing something between two to four [analyses] per week right now."
Baylor's service makes use of array comparative genomic hybridization using bacterial artificial chromosomes to interrogate a sample of amniotic fluid for over 65 genetic disorders. While many in the field agree that array CGH is well-suited for prenatal screening, some of Baylor's commercial rivals believe its service is too much too soon. The test has also underscored general appeals to federal regulatory agencies to reconsider how they evaluate the safety of genetic tests.
According to Mansoor Mohammed, CSO and executive vice president of CombiMatrix Molecular Diagnostics, "prenatal testing is an injustice to the rest of the [array CGH] community."
"One of the areas we feel there has been a lag in regulation has been with CLIA. Genetic testing, with the exception of traditional cytogenetics, has in particular not been focused on."
CMDX has its own plans to offer a diagnostic service using array CGH, and will begin screening samples for constitutional changes in adult patients with mental retardation later this month. However, Mohammed told BioArray News last month that there is not enough data available to support comprehensive array CGH-based prenatal testing.
"More genotype-phenotype associations need to be [made] before [array CGH] can be used in the prenatal arena," Mohammed said. "We need to have a wider corpus of data so we can gain appreciation of genotype-phenotype, phenotype-genotype associations. A lot more maturity needs to be put into this before it is used [for prenatal analysis]," he added. He said that it is too soon to begin using an emerging technology like array CGH to counsel patients that could decide to terminate a pregnancy based on the analysis.
Others have voiced concern over Baylor's CMA service, and the issue prompted a Nature editorial on the topic in the Dec. 8, 2005 issue of the journal.
The Nature editorial used Baylor's case to specifically call for an increased regulatory presence in the area of microarray-based diagnostics.
"Researchers at Baylor College of Medicine in Houston, Texas, are offering pregnant women a genome-scanning test that checks for abnormalities in developing fetuses, but because of the way this test was developed, it is not currently subject to FDA review," the editorial stated.
According to the editorial, "the introduction of prenatal microarray tests into this picture creates additional urgency for the regulators to act. If misdiagnoses occur, children may be born with an unexpected disease, or fetuses may be terminated on the basis of false information."
CMS and the FDA
The issues at play in the debate over array CGH prenatal testing actually have the most impact on two different federal regulatory bodies, the US Food and Drug Administration's Office of In Vitro Diagnostics, which regulates the use of devices in genetic testing, and the Centers for Medicare and Medicaid Services, which enforce the 18-year-old Clinical Laboratory Improvement Amendments.
Of the two, the latter has the most oversight over Baylor's offering, according to Beaudet, who said his lab is CLIA-approved and has been evaluated by CMS since it introduced its prenatal service.
"We have had a routine visit from CLIA as part of the fact that we are a CLIA-approved lab," he said. "I don't think there was any indication they were coming specifically over the issue of [prenatal testing] although they did get into that as part of their visit," according to Beaudet.
CMS did not return phone calls seeking comment on how it evaluates genetic testing. However, according to Gail Javitt, the law and policy director at the independent Washington, DC-based Genetics and Public Policy Center, Baylor's service underscores a blind spot in CLIA regulation.
"One of the areas [where] we feel there has been a lag in regulation has been with CLIA," Javitt told BioArray News this week. "CLIA was supposed to give the government more teeth over the quality of laboratories and different laboratory testing specialties have been addressed over time," she continued, but "genetic testing, with the exception of traditional cytogenetics, has in particular not been focused on."
"So while genetic testing labs are subject to the general laboratory standards, there is no genetic testing specialty area that addresses things that are specific to molecular and biochemical testing and in particular proficiency testing," she said.
Javitt said GPPC believes that "the absence of a genetic testing specialty area is a gap in oversight," and that it called on CMS director Mark McClellan in November "asking that CMS issue a proposed rule for a genetic testing specialty, which is something that the agency has said for years and years that it intends to do, but has not done."
In addition, she said that the GPPC has "prepared a white paper analyzing comments that the agency has received to the recommendations around a genetic testing specialty."
The FDA's OIVD could also have an impact on array CGH diagnostic development, considering the strides microarray technology has made towards the clinic since the evaluation and approval of Roche's cytochrome P450 AmpliChip in December 2004 (see BAN 1/5/2005).
As Mohammed pointed out last month, that approval, coupled with recent agency guidances, has given the microarray sector clear roadmaps for diagnostic development. However, BAC arrays, like the ones used in Baylor's service, as well as in services provided by Signature Genomic Laboratories and CombiMatrix Molecular Diagnostics, have yet to be touched upon in any OIVD document.
"The context within which arrays were initially brought forth to the FDA was for expression analyses and SNP genotyping," Mohammed said. "Guidelines for either would not be particularly valuable when applied to BAC array CGH," he added. BACs are artificially constructed chromosomes, and behave differently from oligonucleotides in a microarray experiment.
Though the FDA has not addressed BAC array CGH yet, OIVD officials said they are taking an active interest in the technology. According to Zivana Tezak, a scientific reviewer at OIVD, the FDA is "very interested in helping to bring to the market diagnostic tests based on array CGH technology."
Tezak said that the FDA "encourages companies developing these tests to come and interact with us under a free informal consultation process." She said that FDA does not currently have any guidances specific to array CGH, but recommended array CGH developers look at a recent peer-reviewed publication discussing "mostly scientific issues pertaining to different kinds of microarrays, including CGH arrays," [FDA perspectives on potential microarray-based clinical diagnostics. Hum Genomics. 2006 Jan;2(4):236-43.].
OIVD is currently in discussions with Spectral Genomics over clearing its Constitution Chip for diagnostic use in screening adults and children for 42 genetic disorders, but Spectral did not return an e-mail seeking comment on that process, and in January declined to discuss its interaction with the FDA (see BAN 1/3/2006).
Baylor's Beaudet sits on Spectral's scientific advisory board, and uses a newer version of the Constitution Chip that was developed at Baylor, but he said he has had no contact with the FDA thus far.
"I think we have a lot of experience introducing new genetic diagnostic testing into the clinic," he said. "We do require an informed consent, [and] I don't see any problem as long as one is careful, gives people informed consent — and it's how we learn to go forward," he said. Beaudet discussed the ethics surrounding the service in a December interview with BioArray News (see BAN 12/14/2005).
GPPC's Javitt, who describes the current regulatory environment as "fragmented," said that Baylor's situation is common. "Unless the test is marketed as a free-standing test kit, there is no independent, third-party review before a test goes into clinical use to say, 'Yes, this test meets the standards we think it should meet.' The laboratory directors make that decision," Javitt said.
'The equivalent of snake oil?'
Another attempt to steer regulatory agencies toward updating the way they address emerging genetic tests, like array CGH diagnostics, is a project sponsored by the Wellcome Trust Sanger Institute and the University of Cambridge and co-led by Cambridge research associate Stuart Hogarth.
The project, entitled "Policy issues in the evaluation of clinical genetic testing for common complex conditions," was initiated in 2004 and is scheduled to be completed by March 2007, according to Hogarth, and will feature recommendations for policy makers in Europe and the US based on various roundtables and interviews the project has held. Some of the participants in the project included FDA officials, and Hogarth said that he hopes that the report will have an impact on US and European policies.
"The FDA has expressed a strong interest in this report's outcome," he said. He added that the report would not be technology-specific, but would rather be on a general level concerning processes of evaluating technologies.
Javitt commented that the need for a third-party review for genetic tests remains high. "There is no equivalent for pre-market review [like they have] for drugs," she said.
Javitt declined to comment specifically on Baylor's prenatal testing but said that it, or similar possible controversies, are a "clear consequence of the lack of an oversight system."
"It may be the case that this test is revolutionary, but without a third-party system of review and agreed-upon standards, how are the patient and the provider supposed to know the difference between the revolutionary technology and the equivalent of snake oil?" she asked.
— Justin Petrone ([email protected])