FDA Plans to Regulate Laboratory Developed Tests; Many Questions Remain as to Details of Regulatory Scheme

July 27, 2010

By Jamie K. Wolszon & Jeffrey N. Gibbs

FDA officials, at a July 19-20, 2010 workshop, stated that the agency intends to end the “exercise enforcement” authority over some laboratory-developed tests (“LDTs”).  Put more plainly, FDA now plans to regulate some LDTs as medical devices.  However, while FDA was clear that it plans to regulate LDTs, key FDA officials, including Center for Devices and Radiological Health (“CDRH”) director Jeffrey Shuren emphasized that the agency had not decided on the details of their regulatory scheme. 

LDTs are diagnostic tests developed and performed by a single laboratory.  They are widely used; virtually all genetic tests are LDTs, as are many tests for emerging diagnosis and for rare conditions.  Regulating these tests will raise many policy, regulatory, legal, and public health questions.

Many attendees, panelists and presenters at the meeting accepted the idea of FDA regulation of LDTs.  However, many speakers stressed that FDA must proceed carefully or the regulation of LDTs could have unintended negative consequences. 

Those potential unintended consequences include reduced innovation, delay in developing tests, fewer improvements in tests, and limiting patient access to tests.  Many speakers were particularly concerned about the effect on tests that are developed in response to new and emerging threats, and tests for rare disorders.  According to many meeting attendees, patient care could suffer and advances into personalized medicine could slow if LDT regulation is not implemented in a balanced manner.

One of the FDA speakers, Elizabeth Mansfield, Director for Personalized Medicine, Office of In Vitro Diagnostic Device Evaluation and Safety (“OIVD”) at CDRH, said that the framework for regulation of LDTs still needs to be written.  Elements that need to be outlined by the agency include: risk categorization, a phase-in period for premarket review and quality systems requirements for new LDTs; registration and listing; and inspections of laboratories.  Other speakers identified additional details that would need to be addressed.

The End of Enforcement Discretion?

Starting in 1992, FDA asserted that all LDTs are devices subject to regulation under the Federal Food, Drug, and Cosmetic Act.  Since then, the agency said it was exercising its enforcement discretion and not regulating LDTs.  Thus, the primary federal regulation of laboratories has been under the Clinical Laboratory Improvement Amendments of 1988 (CLIA).  As a number of speakers noted, LDTs are also regulated by the states (notably New York) and other bodies (notably the College of American Pathologists) (“CAP”).

Until recently, FDA has departed from this position of enforcement discretion in relatively few instances.  For example, FDA has asserted that a test was not a true LDT and advanced its controversial and now-defunct proposal to regulate a subset of LDTs, known as In Vitro Diagnostic Multivariate Index Assays (“IVDMIAs”).  (IVDMIAs are tests where the results of multiple markers are combined to generate an “index score.”)

We previously reported that FDA announced last month that it is revisiting its years-long policy of exercising enforcement discretion over LDTs, and considering adopting a risk-based framework.  The agency announced that it was holding the July 19-20 workshop as a first step to gather comment on how to more actively regulate LDTs.

FDA officials at the meeting reiterated that they intend to regulate at least for some LDTs.  FDA is still grappling with many unanswered questions as to the details of a regulatory scheme.  As shown by the debate over the less sweeping IVDMIA proposal, extending FDA regulation to LDTs will raise many questions.  Issues discussed at the meeting include the following.

Notice and Comment Rulemaking

The agency took the position that FDA does not need to pursue notice-and-comment rulemaking to implement this policy shift.  Dr. Shuren stated that because the laws which permits FDA to regulate laboratories are already in effect; the agency merely has been exercising its enforcement discretion, and therefore can implement the new regulatory framework through guidance. 

Notice-and-comment rulemaking is a more demanding process than issuing a guidance.  For example, in notice and comment rulemaking, FDA must consider and respond to each of the primary comments.  In a guidance, the agency does not have to address comments.  Rulemaking must explicitly address other factors, such as economic impact.  When FDA proposed to regulate IVDMIAs, it also did so through guidance.

A few attendees urged the agency to proceed through notice-and-comment rulemaking.  For example, Anna Longwell, of Longwell and Associates, questioned the approach of moving forward through guidance instead of rulemaking.  Ms. Longwell said there should be a formal notice of proposed rulemaking.  If FDA does proceed to regulate LDTs through a guidance document, any legal challenge is likely to assert that FDA violated the Administrative Procedure Act by skipping the rulemaking process. 

Risk Categorization and Premarket Review

The agency must determine how it will categorize risk.  Meeting presenters and attendees provided examples of possible categorization schemes.

Examples of tests that the agency might consider high risk, according to Dr. Mansfield, include companion diagnostics, such as an LDT used to help determine whether the drug is used properly.  Other examples of high-risk tests include those where an undetected false result could seriously affect the patient; tests that diagnose cancer; and tests that manage serious diseases.

Examples of medium-risk tests provided by Dr. Mansfield include those that could lead to non-serious injury; those where it is easy to detect false results; and adjunctive tests.  Medium-risk tests also could include those where it could lead to psychosocial issues and genetic tests where the assessment is based on a phenotype and the genetic tests are merely confirmatory.  They also include those tests intended to monitor patients with a known disease. 

Examples of lowest-risk tests are those that are highly adjunctive, and for medical knowledge only and evaluation without directed management.  Examples of low-risk tests also include those that have little clinical impact.

Asked whether a test intended for use in evaluating cancer recurrence would be high-risk, Dr. Mansfield responded that currently tests that monitor the cancer of someone who already has been diagnosed generally is not high-risk, although it would depend on the specific claims made.

William Clarke, representing the American Association for Clinical Chemistry (AACC), advocated a classification scheme that includes three categories: high risk, moderate risk, and low risk.  He said that IVDMIAs should be high risk under the upcoming scheme and that high-risk tests should be subject to FDA oversight.  

Gail Vance, representing CAP, proposed that low-risk LDTs be validated and placed into service.  An accredited organization would inspect the low-risk LDTs.  For moderate-risk LDTs, the laboratory would validate the test and then give the data to an accredited party to review.  For high-risk LDTs, there would be validation and the laboratory then would give the data to FDA to review.

Other speakers offered their own ideas.  A key element of any regulatory scheme will be to appropriately and precisely define which tests are subject to FDA regulation.  Ambiguous criteria will create confusion and uncertainty.  In follow-up discussions with clinicians regarding the representative classification of well-known markers, we found that they often sharply disagreed about the correct classification.  Characterizing the risk can get even more difficult with new intended uses.   
Quality Systems Regulation (“QSR”)

Several presenters said it will be difficult for laboratories to comply with the QSRs.  (21 C.F.R. Part 820.)  Of particular concern is the design control requirement.  There appears to be agreement that how QSRs would be applied is an important practical issue, but there was no consensus on that topic.  One suggestion was that FDA rely on inspections by third parties conducted pursuant to CLIA.

Advisory Panel IVD Classification

Some attendees noted that the agency will have to consider how to conserve resources, including staff, as it embarks on such a massive undertaking of regulating LDTs.  Depending on what LDTs are subject to regulation, FDA could receive many new pre-Investigational Device Exemption (“IDE”) requests and marketing applications.  Moreover, because of their novelty, many of these tests will place greater demands on OIVD.  Dr. Mansfield suggested that the agency is considering using advisory panels for down classifications of current IVDs, which would free up staff.  Several meeting attendees, including Elaine Lyon at the Association for Molecular Pathology, recommended use of advisory committees for risk categorization of LDTs.

Phase-In Period for new LDTs

FDA is considering phasing in new premarket review and quality systems requirements over time to help facilitate predictability and planning, according to Dr. Mansfield.  FDA could first require compliance for high-risk tests, and later implement requirements for other tests, she suggested.

Applying New Scheme to Currently Marketed LDTs

When the agency proposed to regulate IVDMIAs, it proposed a grandfather period – albeit a comparatively short one – during which companies already marketing those tests would have a certain amount of time to submit premarket review applications.  Asked how the agency would treat tests already offered, Dr. Mansfield responded that the agency does not yet know the answer to this question.  She did state, however, that the agency has no intention of disrupting clinical testing. 

One presenter suggested that FDA grandfather in all LDTs approved by the New York State Department of Health.

Registration and Listing

FDA also seeks a way to determine a list of who offers what tests.  Dr. Mansfield noted that the National Institutes of Health is in the process of establishing a voluntary genetic testing registry.  Dr. Mansfield noted that FDA probably will need to expand registration and listing to include LDTs.  A number of speakers thought that a mandatory registry could play a constructive role.

Utilizing CLIA Inspectors
Dr. Mansfield also noted that the agency is considering piloting third-party accreditation for inspection of LDTs.  Currently, the CLIA scheme for regulating laboratories includes an inspectional requirement.  In addition, the CAP inspects laboratories for accreditation.  FDA could use those same CLIA or CAP inspectors to conduct inspections on FDA’s behalf. 

Rare Disorders

Attendees almost universally recognized tests for rare disorders as needing special protection under the future scheme.  Some, including William Clarke, representing the AACC, recommended exempting tests for rare disorders from the future regulatory scheme.  While it was suggested by attendees that the Humanitarian Use Device/Humanitarian Device Exemption program could help, these mechanisms can play only a limited role.  The HUD/HDE process is unavailable if the number of patients to be tested exceeds 4000 per year.

Emergencies Including Infectious Diseases

Several attendees noted that when an emerging disease arises, LDTs are often the first test to meet that previously unmet need.  Delaying the introduction of these tests could hamper public health agencies in responding in the case of an emergency. 

Steve Gutman, the former Director of OIVD, noted that the agency can process with emergency use authorizations (“EUA”) rapidly; however, some attendees appeared unconvinced that the EUA program would be sufficiently timely or responsive, particularly for less common emerging diseases.

Modifications and Personalized Medicine

Attendees also warned that an unintended consequence could be to reduce modifications made to tests and to slow advances in personalized medicine.  Once a product is cleared or approved, even relatively small modifications can trigger the need for a new submission.  Thus, the ability to rapidly incorporate new information into assays could be impeded.


The expansion of FDA’s device regulatory scheme to LDTs – even a small subset of LDTs – will raise significant questions.  Many speakers expressed their hope that this will be a collaborative process.  The FDA’s release of series of letters issued to laboratories on the first day of the meeting did prompt some concerns as to whether this hope would be realized fully.  Given the complexity of LDT regulation and its potential consequences – both expected and unexpected – it is critical that this wish for active collaboration be granted. 

Categories: Medical Devices