Changes in mutation panel recommendations for cystic fibrosis molecular diagnostics from a leading organization is causing some vendors to alter their assays.
The changes outlined for this pioneering general-population screen — made by the American College of Medical Genetics , and possibly followed by the American College of Obstetricians and Gynecologists — may serve as a model for future genetic panel development.
With an expanded data set and 5 years of experience with the CF screen, the ACMG reported in a policy statement recently that the new version of its CF carrier screening panel shrunk by two SNPs, from 25 to 23.
The change so far is reflected in assays made by Tm Bioscience, and soon those of Nanogen and likely other vendors will follow, according to company officials and Michael Watson, executive director of the ACMG in Bethesda, Md.
The ACMG statement, which appears in the September/October issue of Genetics in Medicine, said that before adjusting the number of SNPs in such a panel, their frequencies in affected and unaffected individuals should be evaluated. In addition, adequate informed consent is needed because many laboratories report “any and all” results, the statement concluded.
But will most CF assay vendors change their products to reflect the revision? “Boy, they followed suit when we did it the first time,” said Watson, who is lead author of the statement.
Much of CF carrier screening in the United States is conducted by obstetricians, and since ACOG may release a screening panel of its own, some technology vendors are choosing to remain cautious. However, an ACOG spokeswoman said the organization has no distinct plans to release an updated panel of its own.
Nanogen, which sells its own cystic fibrosis analyte-specific reagent, isn’t going to jump the gun. “We’ve been waiting and watching and we anticipate that we will” adjust the company’s panel to fit the ACMG and ACOG recommendations, said Graham Lidgard, Nanogen senior vice president of research and development.
“Most of the customers that we’ve spoken to are intending to keep [the two dropped SNPs] in until there’s a more formal recommendation in from ACOG,” said Lidgard. If ACOG makes its own recommendation soon, and continues to follow the ACMG’s lead, Nanogen will likely retool its ASR “sometime next year,” he added.
For Roche Diagnostics, the world’s biggest diagnostics company, changes to its panel will have to wait a couple of weeks. “Historically, we have followed the ACMG and ACOG in those recommendations, which is why our panel is 25 mutations today,” said Glenn Martin, Roche marketing manager for genomics and oncology. But any changes in Roche’s products are not going to happen until the Nov. 10-13 Association for Molecular Pathology meeting in Los Angeles, where company representatives can discuss the panel with customers, said Martin.
One of the two ill-fated ACMG SNPs, I148T, was found not to be a functional mutation after the group studied its prevalence in a sample of the general population. “It’s serious in the sense that I148T was not really a mutation, but it was in all the databases and was part of mutation panels for many years,” said Watson.
The second SNP, meantime, “turned out to be incredibly rare” because of the sampling bias of a contributing group, he added. To be included in the panel, the ACMG requires SNPs to have a frequency over 0.1 percent.
There is reason to believe that not all producers of CF screening assays will adopt either organization’s panel. Among companies and laboratories, there are CF assays offering “50 or 87 or some number well above 25” SNPs, said Martin. “There’s a drive within industry that ‘more is better,’ even though the recommendation has been 25 mutations for the last four years. It’s become a marketing attempt to give a product differentiation from what everyone else has,” he said.
As for the difficulty of altering assay products: “It’s not that difficult to change; it’s just process time,” said Nanogen’s Lidgard.
Moreover, it is “far easier” to drop a few SNPs from an assay than it is to add them, because assay conditions have already been worked out for those already included, said Martin.
In the end, it is possible that costs to the consumer will stay fairly stable. Testing reimbursement is “fairly well established, so those things are probably not going to shift that much,” said Martin.
The DNA Diagnostic Laboratory at the Baylor College of Medicine, for example, uses a homebrew assay that will be “dropping” at least the non-mutant SNP, said Ben Roa, the lab’s director. “Why would you want to screen people for something that is not really a mutation? That needs to come off,” he said.
The CF screen experience can inform similar ventures, the ACMG wrote in its statement. Problems can arise when providing informed consent when tests include additional SNPs. For instance, some labs reported the presence of an allele that should have only been reported when found in conjunction with another particular mutation, said the statement. Moreover, reporting the detection of the many alleles found in CF individuals at low frequencies complicates the situation further.
Superfluous SNPs pose a parallel difficulty. “If you find that, what do you tell that person? ‘You’re positive for something, but it’s not going to give you disease?’” said Roa. “It becomes more of an issue of interpretation and reporting. Then it’s up to the lab to address that appropriately, so that’s why these genetic tests really require genetic counseling.”
Testing for CF is widespread in the United States, but a market size is difficult to ascertain. “As far as genetic parameters are concerned, CF testing is by far probably the greatest of them,” said Roche’s Martin. “The number of people being screened is well less than 50 percent, even if you just look at pregnant women,” Martin said. “You’re talking about something that — depending on the population you’re looking at — probably has a carrier frequency of one in 20 or 25,” he added.
Facilities that perform the tests include “university hospitals that have their own testing facilities, regional reference labs, and national reference labs — it’s spread across the board,” said Martin. He was unable to provide an estimate of Roche’s market share.
The US National Institutes of Health held a consensus conference in 1997 that concluded carrier screening should be conducted, said Watson of the ACMG. Arguing that no tests should be conducted before groundwork had been laid in standardization and education, the ACMG and ACOG objected, he said.
By April 2001, the ACMG Cystic Fibrosis Carrier Screening Working Group produced its first panel of 31 CFTR gene SNPs for frequency testing of the general population, according to the ACMG statement.
“Nanogen and Third Wave and all those other companies who had targeted products […] none of them were brave enough to go to the [US Food and Drug Administration] with a clinical validation database,” said Watson.
“Last March we made manufacturers and laboratories aware of the pending change in the mutation panel, because these are often perceived as establishing a standard of care, and it’s necessary that the laboratories have considerable lead time to modify their assays and validate them for use,” said Watson.