Heated Public Meeting Illustrates Vexing FDA/CMS Tussle to Regulate IVDMIAs

GAITHERSBURG, Md. – At a public meeting held here last week to discuss the US Food and Drug Administration’s proposition to regulate a subset of complex laboratory-developed genetic tests, stakeholders questioned how the FDA’s regulatory authority would dovetail with the existing clearance pathway laid out by Centers for Medicare and Medicaid Services’ Clinical Laboratory Improvement Amendments guidelines.

 

 

The draft guidance raises “concerns over FDA’s legal authority to regulate clinical laboratories,” is unclear over the definition of in vitro diagnostic multivariate index assays, and does not clearly identify the elements of an IVDMIA that comprise a medical device subject to FDA regulation versus those regulated under CLIA, Paul Radensky, a health law attorney with McDermott, Will, & Emery, said during the meeting.

 

The guidance also does not define pathways for premarket review, fails to explain how sponsors can comply with the FDA’s Quality System Regulations, and doesn’t identify conflicts between the agency’s limitations on labeling and promotional statements versus CLIA requirements for laboratory reporting, said Radensky, whose firm represents many of the labs that would be required to follow the IVDMIA guidance.

 

Several speakers said that the guidelines, by imposing costly and time-consuming regulatory hurdles, may hobble innovation and make reimbursement even more difficult (see related story), while a minority view held that without the FDA regulating IVDMIAs patient safety would be jeopardized.

 

Amid the volley of pros and cons during the nearly eight hour meeting, presenters were largely unified in the charge that the FDA must provide more specificity and clarity in the final version of the guidelines.

 

The agency ultimately conceded that it can better clarify the scope and intent of the guidance. Steven Gutman, director of the Office of In Vitro Diagnostic Device Evaluation, maintained that it is not the agency’s intent to “surprise or confuse people” or issue regulations that would delay the entry of innovative products to the market.

 

Gutman urged presenters to offer constructive comments on the draft guidance by March 5. The deadline for the public to comment on the document was extended from a Jan. 5 deadline after the agency sensed growing interest in the issue.

 

 

In its proposed guidelines, issued last fall, the FDA defines IVDMIAs as tests that use mathematical formulas to interpret gene and protein data to guide medical decision-making.

 

Specifically, the FDA views IVDMIAs as “test systems” that use clinical data to empirically identify variables in an algorithm; employ the algorithm to integrate these variables to calculate a patient-specific score, which cannot be independently confirmed without access to the proprietary information; and report the test result, which cannot be interpreted by the well-trained health care practitioner without information from the test developer regarding its clinical performance and effectiveness.

 

According to the agency, such tests, ordinarily overseen by CLIA regulations, must instead be cleared by the agency due to their complexity [see PGx Reporter 9-13-2006].

However, the FDA acknowledges in the draft that its definition of a “test system” doesn’t match CLIA’s definition. In a footnote within the draft guidance, the agency notes that it considers a product a “test system when it has some or all of the products needed to conduct a particular test, such as reagents, controls, equipment, software, etc., and/or it is one of these products and has instructions for use in a test.” In the preamble to the ASR rule, FDA describes a test system as having a proposed intended use, indications for use, instructions for use, and performance characteristics.

 

Some presenters wanted to know how the guidance would align CLIA regulations, which are designed to assure the reliability and accuracy of test results, with the FDA’s Quality System Regulations, which are intended to assure that medical devices are manufactured effectively and safely.

 

According to Radensky, CLIA-certified labs could incur extensive costs to come into compliance with FDA QSRs.

 

The FDA told attendees that it plans to issue a Special Control Guidance document that will describe the types of information that should be submitted for IVDMIA clearance.

 

 

 

“These rules are in conflict,” Radensky said. The FDA and CMS “must work through leadership at HHS to set out clearly how laboratories can meet these conflicting requirements among the regulations of sister HHS agencies.”

 

He added that when the FDA formalizes the guidelines, test developers seeking 510(k) clearance and pre-market approval should receive a transition period of two years and four years, respectively. 

 

Additionally, while some speakers took the “if it isn’t broken, don’t fix it” stance in favor of maintaining CLIA jurisdiction, others noted that CLIA needed to strengthen its oversight of lab tests.

 

This issue is ongoing, as last summer Johns Hopkins University’s Genetics and Public Policy Center together with the Genetic Alliance filed a Citizen Petition with CMS requesting that CLIA regs be strengthened by creating a genetic specialty. “We have not received a response nearly six months later,” said Gail Javitt, director of law and policy at the Genetics and Public Policy Center.

 

Without improving the way CLIA regulates clinical and analytic validity for genetic tests, the GPPC said that the FDA should step in to ensure that IVDMIAs are safe.

 

“While CLIA requires laboratories to independently establish analytic validity of tests, there is insufficient oversight to ensure that laboratories do so,” Javitt said. “CLIA hasn’t been interpreted to require that laboratories demonstrate clinical validity. But clinical validity is profoundly important when considering under what circumstances a genetic test should be made commercially available.

 

“Offering tests without adequate evidence of clinical validity endangers the public pocketbook and moreover the public’s health,” she added.

 

 

Several presenters suggested that the FDA has no legal authority to regulate IVDMIAs, a notion FDA officials tried to dispel.

 

“There is a lot of confusion about the regulation of diagnostic tests that are developed by, and used in, a single laboratory,” Courtney Harper, associate director of OIVD’s Division of Chemistry and Toxicology Devices, acknowledged at the start of the meeting.

 

She said the agency believes this confusion “derives in part” from the way it regulates lab-developed tests that use commercially available analyte specific reagents and other commercially available, FDA-regulated components.

 

In the preamble to its final ASR rule, the FDA said that “clinical laboratories that develop [in-house] tests are acting as manufacturers of medical devices and are subject to FDA jurisdiction under the act.” However, the agency has historically held off regulating lab-developed ASRs and tests developed in single labs.

 

But in its draft guidance, the agency decided that it, and not CLIA, should regulate IVDMIAs since the “do not fall within the scope of laboratory-developed tests over which FDA has generally exercised enforcement discretion.”

 

“IVDMIAs, a growing category of tests, include elements that are not standard primary ingredients of in-house tests that raise safety and effectiveness concerns,” Harper said.

However, Richard Samp, chief counsel of the Washington Legal Foundation, countered that the Medical Device Amendments of 1976, under which the FDA claims its authority, “do not have anything at all to say about laboratories or lab testing.” Rather, it grants regulatory authority to CMS under CLIA.

 

Samp was referring to the 1976 Medical Device Amendments to the Food, Drug & Cosmetic Act, which created three classes of devices defined by the level of regulatory control issued, with Class III devices being the most regulated.

 

According to the draft guidance, IVDMIAs will likely be Class II devices, requiring premarket notification in the form of a 510(k) submission, or fall in the Class III category, requiring premarket approval. 

 

Samp argued during the meeting that Congress never intended to grant the FDA the authority to regulate IVDMIAs under the 1976 amendments. “It was not until the 1990s that FDA first began to suggest that it might possess regulatory authority over labs offering [laboratory-derived tests], and it is only in the past year that FDA has sought to exercise its alleged authority,” Samp said.

 

“In the absence of authority from Congress, FDA’s recent attempts to regulate lab tests are improper and should cease,” he said.

 

 

Confounding this is the observation that over the past year the FDA has taken a piecemeal approach toward regulating IVDMIAs by sending letters to some test developers, including Genomic Health, Correlogic, LabCorp, Agendia, and InterGenetics, suggesting that CLIA certification may not be enough to bring their products to market. The companies may need to file for FDA approval, the agency wrote.

 

Guido Brink, director of regulatory affairs at the Dutch diagnostic company Agendia, noted that “all IVDMIAs must be held to the same standard. Once the first IVDMIA has been cleared or approved, all comparable IVDMIAs must be regulated to create a level playing field.”

 

Brink’s support of the draft guidance is understandable: Two days before the Gaithersburg meeting, the FDA cleared the first IVDMIA, a breast cancer-recurrence test called MammaPrint made by Agendia [see PGx Reporter 02-07-07].

 

OIVD’s Gutman has stressed that by clearing MammaPrint so close to the meeting, the agency is not hinting how it intends to regulate IVDMIAs.

 

Exagen is another of a small number of diagnostic developers that favor the draft guidance. The Albuquerque, NM-based company said it plans to file its first diagnostic, an IVDMIA for breast cancer recurrence, with the FDA within the next two months.

 

Exagen CEO James McClintic told Pharmacogenomics Reporter sister publication BioInform last week that it favors the draft guidance because it aligns with its business model. The company’s technology platform allows it to identify much smaller sets of genes than its competitors — between three and five genes, as opposed to 70 genes for Agendia’s MammaPrint – which allows it to sell the gene sets as reagent kits for standard lab diagnosis platforms (see related story).

 

“Obviously, for a company like ourselves, these guidelines are good because we know how to build the tests with very few bets, so therefore our approach, which for right now — except for the Agendia test, which was just approved — we don’t think that there is going to be any other competition here for quite awhile,” McClintic said.

 

He acknowledged that Exagen is in the minority and that at the public meeting the majority of industry stakeholders were “dead set against any kind of regulation.”

 

“Nobody likes regulation, but we spoke entirely for it because it fits our business model like a glove. Everything that was in that FDA guidance, we’re already doing,” McClinitic said. “We have had several conversations with the FDA … and we are 100 percent in favor of what they’re doing, but we are in the minority, I can assure you.”

 

Others, like the Genetics and Public Policy Center’s Javitt, challenged the FDA’s notion that it needs to regulate IVDMIAs because physicians are unable to independently interpret test scores without proprietary information.

 

“Numerous studies have documented that healthcare providers lack education generally to interpret the results of genetic tests,” Javitt said. “Clinician confidence would appear to be an insufficient basis for distinguishing between IVDMIAs and other laboratory tests.

 

Additionally, presenters criticized that the draft guidance was “technology based,” and not “risk based.”

 

“FDA seems to be operating under the assumption that IVDMIAs as a class are inherently more risky than other laboratory tests,” Javitt said. “This is certainly true in some cases. But we are concerned that FDA’s piecemeal approach overlooks other high-risk tests that do not fall within the IVDMIA framework, at the same time putting all IVDMIAs in the same high risk class when compared to other diagnostic tests.”

 

Daniel Schultz, director of FDA’s Center for Device and Radiological Health, suggested that presenters had misinterpreted the agency’s intent. “At the end of the day, what we are very, very interested in is providing a regulatory oversight framework that does in fact reflect the risk of the product,” Schultz said.

 

“I will say to some extent — and perhaps we can do a better job of defining this — but there is a link between the changes in technology and the level of risk. We need to explain that. If you don’t understand it then we haven’t succeeded.”

 

Despite her objections of the guidance, Javitt said that the FDA has a “critical role” to play in “in ensuring the safety, effectiveness, and availability of genetic tests.”

 

“The status quo leaves the public health insufficiently protected,” she said. “Effective stewardship by FDA is needed to develop and implement a coherent and equitable system of oversight.”

 

Javitt said the draft is “an important first step in articulating what FDA’s role will be” in ensuring the quality of IVDMIAs.

 

FDA officials acknowledged that while the guidance needs to be refined, the agency would likely move forward with its plans to regulate IVDMIAs in some form.  

 

CDRH’s Schultz held that “the idea that we can simply go back to the way we were several months ago – that to me is the one notion that would be unacceptable.

 

“The field is moving very quickly, the technology is moving very quickly, expectations in this room and throughout the country are moving very quickly in terms of us being able to get a handle on this type of technology,” he said. “I think we need to keep moving forward based on today’s discussion.”

 

At the end of the meeting, OIVD’s Gutman pointed out that the agency has been clear in expressing its discomfort over the lack of federal oversight of IVDMIAs. He noted that the FDA “always expected that the document was not perfect and I guess you have suggested the same.

 

“As we were interacting with people in the early life of the document, I thought that people were either over-reading or misreading the document,” Gutman said in his closing comments. “So, I was blaming you, not us, but I will take part of the blame perhaps for not having crafted language that clearly. …

 

“We didn’t think that this was going to be easy,” he added. “And you reinforced that.”