The Food and Drug Administration has issued a first draft of guiding principles for the development and marketing of multi-analyte-based diagnostic tests.
The document, which is open for public comment until July 21, is found on the FDA site at: http://www.fda.gov/cdrh/oivd/guidance/1210.pdf.
It is an initial attempt at creating a framework that may lead to government regulation of multiplexed microarray-based diagnostic products, and introduces at least two novel accommodations for industry that do not exist for other life-sciences devices or technologies.
The draft may unnerve diagnostic and pharmaceutical companies who decry regulation that may result in greater costs for developing, and thus selling, these technologies — a consequence that might ultimately impact the overall cost of health care in the US.
Some experts contend it’s simply too early for industry anxiety. “I wish that I could tell you that this draft guidance is the key to all future questions, but it ain’t,” said David Link, executive vice president of Expertech of Boston, a consulting firm that deals with medical devices and FDA regulatory issues. Link worked at the FDA for 10 years and helped establish regulatory guidelines for medical devices.
“While the full impact of this new technology is uncertain,” the agency concedes it has “little experience” with multiplex or array device submission — the FDA is anxious to provide clear guidance to assist sponsors in developing submissions that will support marketing of safe and effective products using this technology,” the draft reads.
The guidance primarily reviews nucleic-acid analyses based on oligonucleotides and cDNAs, but said “many of the principles” also apply to protein and tissue arrays as well, said Elizabeth Mansfield, an official at the Office of In Vitro Diagnostic Device Evaluation and Safety, who wrote the draft.
A few of these principles may eventually become useful to industry, some experts believe. For example, it is understood that guidance documents are designed to be laundry lists of what the FDA currently knows about a particiular topic, as well as what it might ask industry. In this case, however, the draft departs from existing guidance by suggesting that multiplexed microarray-based diagnostics can fit into one of three submission categories — pre-market approval, a 510(k), or a de novo 510(k). The pre-market approval, or PMA, process attracts the highest level of regulatory scrutiny because the products it covers traditionally are life-sustaining devices like heart valves; a 510(k) is a less-stringent application for a device that is equivalent to an existing product and that is not a critical device, like a diagnostic test; and a de novo 510(k) is a new category for devices that do not have a marketed equivalent and that do not deserve the rigorous PMA regulatory requirements.
“We recommend that the sponsor or manufacturer consults with the FDA to determine the appropriate type of submission,” the draft submission reads.
According to Link, the agency is “just staking out its territory when it comes to these kinds of tests and saying to the outside world, “Well, OK, guys, before you submit something, you ought to take it to us so we can discuss it. We can’t tell you in advance because the purpose and function of these tests will range over a wide spectrum.’
“As you get into [the document], it is so broad that it is very difficult to say what would apply, and what the nature of the submission is,” he said. “Up till now, guidances have always been confined to 510(k)s. This one hits everything, depending on what it is. If you were to submit a test that says ‘I will detect 100 percent of a very specific type of cancer with 100 percent specificity and sensitivity,’ there’s no doubt in mind that this would be a PMA and not a 510(k). But if it is a test that may have some comparability to something in the marketplace but it does it with a different technology, it is conceivable it will be a 510(k)” he said.
“I’m sure it’s useful in that it makes people think about the kinds of information they would possibly have to collect, and it strongly encourages them to go talk to the FDA at the oustet so you don’t go down one path, and the FDA says, ‘That’s nice, but that’s not what we want.’ “
In writing the draft, Mansfield said her department, which was moved from the Division of Clinical Laboratory Devices last November, had “a lot” of communication with industry and “some” with academia, as well as with various FDA scientists. It took around four months to write the guidance.
Today, there are two routes that diagnostic companies take to bring their products to market without FDA approval: the so-called home-brew test, which are products developed and validated in the laboratory for use only in the laboratory; and analyte-specific reagents.
However, though many in the industry categorize multiplexed microarrays as ASRs, the FDA disagrees: “Our opinion is that multiplexed tests and microarrays are not ASRs, though some are being sold with that appellation anyway,” Mansfield said. “These are required to be regulated, and we’re just trying to give guidance for companies.
“We’ve recognized for several years now that these new multi-analyte tests were coming down the road, and we were not seeing anybody coming [to the FDA],” Mansfield said. “We think it’s because companies didn’t know what we want.”
Partly because it is so convinced that multi-analyte tests will become an important component in disease diagnosis, and partly because it is so inexperienced with the technologies involved, the FDA altered its review process in two ways. Typically, the agency writes a guidance document after it has reviewed “several versions of tests so we know what to look for,” said Mansfield. In this case, however, the agency “felt it was necessary to write this before we’d ever received anything.”
Another change comes within the FDA’s quiet decision to offer industry the chance to comment on two consecutive draft versions of the document (the second draft guidance will be published around four months after the agency reviews comments submitted for the first draft). Normally, the agency formalizes a guidance after considering just one round of comments.
“That shows how uncertain they are,” said Link, the consultant, “that also shows some enlightenment on the part of the agency because people there are not in touch with what’s going on in the outside world. They don’t know what the companies are working on, they don’t have a good feel for the marketplace. So for them to write a guidance … exceeds anybody’s expectations. They’re going to take two shots at it to refine it and get it reasonably right in the third cut.”
Jenny Butler, chief of the FDA’s documents management branch, said that as of May 19, her office had not received a single comment from industry. This is normal, she said: Companies wait until days before the deadline to submit their comments — which immediately become public property — fearing that competitors might glean marketing secrets from them.
None of the 10 pharmaceutical or diagnostic-tool companies asked to comment on the draft guidance would do so, even on a background-information-only basis.
To be sure, while details in most approved guidance eventually make their way into regulation, it is extremely unlikely that the multi-analyte draft will go that route. In fact, insiders like Link speculate that the agency, implying its own experience on the subject, will likely let the guidance stand alone as a map for industry rather than impose a rigid regulation that will be difficult to amend.
By its definition, a guidance is not binding, and can be amended relatively easily as new technologies or methodologies emerge.
“A regulation is ‘thou shalt,’” said Link. “A guidance is ‘thou should.’”