Diagnostics & Imaging Week Washington Editor

GAITHERSBURG, Maryland — Makers of in vitro diagnostics for in-house use have faced little FDA oversight in the past, and despite the existence of regulations written by the Centers for Medicare & Medicaid Services has written regulations covering this sector.

Nevertheless, the agency published guidance last year for in vitro diagnostic multivariate index assays (IVDMIAs or MIAs).

According to the draft guidance, much of the agency’s concern was based on the notion that MIAs are not among the “common elements of in-house tests” and because of the complexity of these high-tech assays, the FDA is seeing more reasons to be concerned about safety and effectiveness.

At a meeting on the topic here last week, the agency admitted that the current guidance, when overlaid on the 1997 guidance for analyte-specific reagents, has created some confusion. The FDA is motivated by a proliferation of assays thanks to the surge in interest in personalized medicine, especially since modern assays often employ algorithms used by computer software, which the agency has asserted jurisdiction over in a wide range of devices. Up to now, the agency has seen many assays as a “very narrow niche of devices” that have benefited from the regulatory discretion that the agency has heretofore exercised.

Courtney Harper, PhD, the lead biologist at the Division of Chemistry and Toxicology Devices at the Office of In-Vitro Diagnostics, insisted that despite the reagent guidance, clinical labs that develop in-house tests are “acting as manufacturers of medical devices and are subject to FDA jurisdiction.” The rationale here is that the reagent guidance does not “actually extend to the tests that use” those reagents.

However, the first draft of the MIA guidance probably generated as much confusion as it resolved.

“We realized after we published the guidance that a lot of the questions that we had been hearing have been related to clarification of the definition” of a device, Harper said, adding that a device can use an algorithm similar to those commonly employed in MIAs to detect protein structure and not qualify as a device, as an example. “We would like to see comment from the public on the definition” of a device for this class of assays, she noted.

Paul Redensky, a partner in the Miami, Florida office of McDermott, Will & Emery (Chicago), stated that the guidance has “raised a large number of issues, including . . . FDA’s authority” to regulate these assays. However, he focused on “questions that laboratories need to have answered,” which include the definition of an MIA and which elements within a lab’s operation are subject to FDA regulations. Other issues, he noted, include how the quality system regulations (QSRs) mesh with the Clinical Laboratory Improvement Amendments (CLIA) standards, which are overseen by the CMS.

“Those who fund research . . . will not find it acceptable to find out late [in the development process] that the FDA issued a guidance document” that will impose a new regulatory burden,” Redensky observed.

Redensky said that one area of ambiguity is that of the difference between a prognostic claim and a predictive claim with respect to a given assay. “Generally, we will have a claim that is talking about the likelihood of some outcome,” with survival of a disease vs. the recurrence of a disease as an example. “These things are key because” the answer to that question plays a large role in whether the device is eligible for a 510(k) filing, which is vital information when lining up finances.

Richard Samp, chief counsel with the Washington Legal Foundation (WLF; Washington), said that he is “convinced that any attempt by FDA to impose significant regulation on laboratory-developed tests will be a setback for public health” and is “contrary to law.”

Samp commented that the Sept. 28 citizen petition filed by WLF requested that the agency avoid imposing regulations on assays for in-house use, which he said “are developed by the thousands each year,” and that “there is no evidence that LDTs [lab-developed tests] are inaccurate.” Because of CLIA, “it is difficult to understand why FDA wants to fix” a system that is not broken. Samp further charged that the guidance would “have a crippling effect” on availability of LDTs.

Thomas Tsakeris, the president of the Coalition for 21st Century Medicine (Arlington, Virginia), indicated that the coalition “is concerned that in its current form, the draft guidance will have adverse, unintended consequences,” including impedance of innovation and preclusion of improvements to current MIAs. He said that the effort to get up to speed on the quality systems regulations “could take years, would be prohibitively costly, and would drive up healthcare costs.” He also said that the guidance would create reimbursement problems and potentially discourage future investment.

Carolyn Jones, the vice president for technology and regulatory affairs at the Advanced Medical Technology Association (AdvaMed; Washington) said that the association “represents a diverse group of interests” and that “the vast majority of AdvaMed’s membership has concluded that laboratory-developed tests . . . meet the definition of a medical device and should be subject to a reasonable risk-based approach.”

However, Jones also said that AdvaMed members “believe the process would have been better served if FDA had issued a concept paper and held a public meeting before issuing the guidance.”

“The clinical lab community does not understand which types of medical algorithms that FDA intends to regulate,” Jones asserted, adding that because some algorithms have been in use for an extended period of time, “FDA should provide more detailed information on which products would be subject to regulation.”

Daniel Schultz, the director of CDRH, said at the end of the meeting that “we’re on the cusp of a fundamental change in how medicine is practiced” thanks in part to diagnostics. He pointed out that the “divergent opinions regarding the scope of the guidance,” suggested that FDA “may not be able to satisfy everyone.”

The FDA needs to “do a better job of associating technology with risk,” he remarked, a reference to comments that the proposal seems to operate more from a technological innovation standpoint rather than a risk-based approach, but Schultz nonetheless insisted that there is a very real “link between the changes in technology and the level of risk.”

“The idea that we can simply go back to where we were several months ago . . . is unacceptable,” Schultz noted, but promised that the guidance will “not be the last piece of this discussion.”