AMP Proposal Frames a CLIA Alternative to LDT Regulation
Interview by Steve Halasey
In October 2014, FDA published a long-anticipated draft guidance document on the regulation of laboratory-developed tests (LDTs).1 The agency followed that publication at the beginning of this year by hosting a 2-day workshop on LDTs that gathered a wide range of stakeholders from the clinical laboratory community.2
But if the agency was hoping that the workshop would generate a consensus sufficient to propel the laboratory community toward implementation of its proposed LDT regulations, it’s safe to say those hopes were quickly dashed. Far from building consensus, the workshop surfaced many stakeholders with widely disparate views about the regulation of LDTs, including some who pronounced themselves ready to file suit if the agency implemented the regulations as proposed.
In the months since the workshop, many stakeholders have continued working on the issue of LDT regulation. In April, FDA moved to bolster its own proposal by partnering with the Centers for Medicare and Medicaid Services (CMS) to create a joint task force on LDT quality requirements.3 And at the beginning of August, the Association for Molecular Pathology (AMP) released its own regulatory proposal, framed as a modernization of the Clinical Laboratory Improvement Amendments of 1988 (CLIA) specifically for the class of services that AMP terms “laboratory-developed testing procedures,” or LDPs.4
To find out more about AMP’s proposal for modernizing the regulation of LDPs, CLP talked with Roger D. Klein, MD, JD, FCAP, medical director for molecular oncology at the Cleveland Clinic and chair of AMP’s professional relations committee.
CLP: Why did AMP consider it necessary to create a proposal for regulating laboratory testing separate from the effort being guided by FDA and CMS?
Roger D. Klein, MD, JD, FCAP: The Senate Committee on Health, Education, Labor, and Pensions (HELP) has been looking at the issue of laboratory-developed test oversight. As experts in the field, we wanted to provide information that would assist the committee.
Of course, we’re very concerned about the specter of FDA regulation in this area, for several reasons. First, we haven’t been provided with any evidence of a large-scale or systemic problem with laboratory testing that would justify such broad regulations as FDA has proposed. Second, the burden and costs of complying with FDA’s proposed regulations could easily run into the hundreds of thousands or even millions of dollars, far exceeding what the vast majority of laboratories, including our members, could afford to pay.
Today, payments for hospital laboratory tests are generally either bundled into an inpatient payment system where they are part of a diagnosis-related group (DRG), or they are made through the outpatient prospective payment system. For most lab tests, there really isn’t an independent payment.
Within these payment structures, most laboratories could never economically justify the costs of submitting their tests for FDA approval. And although molecular tests are excluded from those payment structures and have a separately challenging reimbursement process, the same is true for molecular pathology labs.
CLP: How has AMP’s proposal been received, either by the HELP Committee or elsewhere?
Klein: I can’t speak specifically about the committee, but the feedback we have received from a number of quarters has been generally positive.
CLP: What are the key features that distinguish AMP’s proposal from FDA’s proposed regulations? Are there particular differences that make AMP’s proposal more palatable to laboratories and ultimately more doable?
Klein: FDA’s approach proposes to treat laboratory-developed procedures—our professional activities—as though they are medical devices. The agency’s proposed regulations would impose the same kind of review and other requirements as it currently applies to an artificial hip or an in vitro diagnostic test kit that’s designed, developed, validated, manufactured, boxed, and shipped to users around the country over whom the supplier has no control.
By contrast, AMP’s proposal represents an extension or modernization of the CLIA program for laboratory oversight. Laboratories are already familiar with the CLIA program, so it makes sense to build on that base in order to account for the many changes that have occurred in laboratory testing since the CLIA regulations were written.
The AMP proposal includes the potential for reviewing some tests before they are made available to the public, as is already being done under the CLIA program in the state of New York. But our approach would be to validate the clinical applications of a biomarker and test by reference to published literature and the community of experienced laboratory users. This approach would impose far fewer costs on laboratories, while helping them to maintain very high quality in the testing they perform and the services they provide to patients.
CLP: FDA’s proposed regulations and AMP’s recent proposal each incorporate a risk-based, tiered structure for reviewing lab testing. How are the two systems different, and how do you expect that risk levels will be assigned under the AMP proposal?
Klein: First, it’s important to understand that all of the tests under discussion are categorized as high complexity tests under the CLIA regulations. Consequently, all of these tests are already subject to the highest performance and personnel standards imposed by CLIA.
FDA’s risk structure is focused on the “intended use” of a product. That is, the level of risk associated with a product is determined mostly by how the product is used.
In our proposal, we looked at risk in a very different way, taking into account the broader context of testing as well as the methodologies used to perform a test. So a test that uses well-characterized and proven methods might be considered to have only moderate risk, even if its intended use might cause FDA to consider it a high-risk test.
In our proposal, the group of tests considered to be high risk would be much smaller, and would include essentially those tests for which formal review by FDA would be an option. That group would include a small number of tests with characteristics that would make them very difficult for a CLIA reviewer to assess, such as multianalyte assays with algorithmic analysis (MAAAs), which employ algorithms that have typically remained proprietary.
Those types of tests could be reviewed under CLIA, but that would require the provider to submit to the CLIA reviewing entity all information necessary for evaluation of its validation—including the test’s algorithm. The test would likely require a statistical review that would be outside the ability of most clinical users. Another issue with tests that make use of proprietary algorithms is that there’s no way to perform interlaboratory comparisons.
In our view, these two issues would push such a test into a high-risk category. Labs would then have the option of submitting the test for FDA review or providing all the information necessary to conduct the review under CLIA.
So, in our proposal, risk has less to do with intended use, and more to do with how readily a user or reviewer can assess the performance of a test or evaluate it through interlaboratory comparison.
CLP: The AMP proposal would require labs to submit their test validation information 30 days prior to public availability for moderate-risk testing, and 90 days prior to availability for high-risk testing. Do most labs have experience in making these kinds of submissions?
Klein: This proposal would provide for a formal CLIA program review of validation information in advance of an onsite inspection. So before a test could be put into service, there would be a review under the CLIA program.
Most labs are not used to performing this step, but it is required by the New York State Department of Health, so any lab that serves patients in New York would have experience in compiling and submitting their validation information. There is a subset of labs that currently undergo these types of reviews for laboratory-developed tests, but most laboratories do not.
CLP: The AMP proposal works hard to preserve a structure of third-party review of proficiency and periodic accreditation, which labs seem comfortable with. Would the proposal change anything about who would conduct such reviews, or the scope of their activities?
Klein: CMS makes extensive use of third-party reviewers now, and we anticipate that they’d continue to do so for reviewing tests prior to introduction. The reviewers who would take this on could be current or new reviewers, including perhaps some third-party accreditors or exempt state agencies, and they might or might not be the same folks who are currently accrediting labs. Given their experience, one would anticipate that a number of current accreditors could have an interest.
CLP: With FDA and CMS already developing regulations under the current legislative structure, what are the procedural possibilities for advancing AMP’s proposal?
Klein: There are a number of possibilities ranging from complete adoption of our proposal to just taking and using certain ideas that seem to meet a perceived need. We put forward ideas that we believe are workable and would help to modernize the CLIA program. We think they would also go a long way toward addressing some stakeholder concerns, even though there is little evidence of systemic problems with laboratory-developed tests.
We’re certainly not saying that there aren’t any problems at all. But it is an extreme measure to shift jurisdiction over clinical laboratories from CMS to an agency like FDA, which is not really accustomed to regulating clinical services. By contrast, AMP’s proposal would build on the existing program.
CLP: How does AMP’s proposal make use of proficiency testing to update CLIA requirements for laboratories?
Klein: Under the current CLIA regulations, formal proficiency testing is required for only 83 analytes. But it is important to understand that the laboratory accrediting organizations—including the College of American Pathologists, which accredits most labs that conduct significant molecular testing—require proficiency testing for many more analytes than are required under CLIA. Yet the regulations themselves don’t mandate this.
The AMP proposal extends these much broader requirements for proficiency testing into the regulations and adds a mechanism for making regular additions, so that even new analytes may have an explicit regulatory requirement for proficiency testing.
Most laboratories that are conducting significant molecular testing must meet standards that exceed those of the actual CLIA regulations.. For example, they need to meet the requirements of their accrediting organization, or of the New York State Department of Health. So, in effect, the AMP proposal would raise the formal requirements under CLIA to match the levels already being met by CAP-accredited labs.
The same is true with regard to demonstrating the clinical validity of a test. Although not required under current CLIA regulations, both CAP and the State of New York require labs to provide a demonstration of a molecular test’s clinical validity. The AMP proposal would extend this requirement also to the CLIA program regulations.
Another element that we considered was how to handle adverse event reporting, which FDA requires whenever there is a death or serious injury caused by a medical device. But those kinds of adverse events are quite uncommon for laboratory tests, so FDA’s requirement isn’t really suited for this purpose. Instead, we built upon existing CLIA language meant to identify failures that would cause a laboratory to stop testing, and we applied that language to individual lab tests. An incident that resulted in significant danger to public health, for instance, would constitute a reportable instance that would cause a lab to cease testing. We applied that same approach to individual tests, and created a vehicle for reporting problems that would require a lab to stop performing those tests.
CLP: Is some of the need for ratcheting up proficiency testing a response to the fact that so much has changed since the CLIA regulations were written—especially in the field of molecular testing?
Klein: I think there is general recognition that proficiency testing is an important method—but by no means the only method—for assessing laboratory performance and functioning. It’s what’s done by most laboratories that are accredited by the College of American Pathologists. The AMP proposal would extend this standard good practice by making it a part of the regulations themselves.
CLP: One of the desired outcomes of AMP’s proposal is to ensure that CMS is not burdened with a lot of additional duties that it won’t be able to handle. Do you feel the final proposal accomplishes that goal?
Klein: We were very cognizant of limitations within CMS, particularly in today’s world where resources are scarce, and we wanted to ensure that CMS would be able to provide the functions described in the proposal. So we have provisions for user fees, which build upon the current system of the CLIA program and are proportionate to the size and volume of testing handled by each laboratory. And just as the CLIA program does today, the proposal makes extensive use of third-party reviewers, so that CMS’ most important role would be to oversee the performance of the program reviewers—not to conduct the reviews themselves.
Steve Halasey is chief editor of CLP.
- Draft Guidance for Industry, Food and Drug Administration Staff, and Clinical Laboratories: Framework for Regulatory Oversight of Laboratory Developed Tests (LDTs) [Framework Guidance]. Silver Spring, Md: Office of In Vitro Diagnostics and Radiological Health, Center for Devices and Radiological Health, FDA, 2014. Available at: www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM416685.pdf. Accessed February 19, 2015.
- Hoffmeister DM, Norviel V, Andres CJ. FDA regulation in clinical labs. Clinical Lab Products. 2015; 45(3):12–15. Available at: http://clpmag.com/2015/03/fda-regulation-clinical-labs-2. Accessed August 25, 2015.
- FDA and CMS form task force on LDT quality requirements. FDA Voice [online]. Silver Spring, Md: FDA, 2015. Available at: www.blogs.fda.gov/fdavoice/index.php/2015/04/fda-and-cms-form-task-force-on-ldt-quality-requirements. Accessed August 25, 2015.
- Proposal for modernization of CLIA regulations for laboratory-developed testing procedures (LDPs). Washington, DC: Association for Molecular Pathology, 2015. Available at: www.amp.org/advocacy/documents/AMPCLIAmodernizationproposalFINAL8.14.15.pdf. Accessed August 25, 2015.