Part one of this article, published in the November/December 2009 issue of IVD Technology, offered reasons why FDA should regulate laboratory-developed tests (LDT). Part two examines some reasons frequently given for not regulating LDTs and explains why each argument falls short of justifying FDA's inaction.
Imposing FDA regulation on LDTs would disrupt the use of important diagnostic tests
One rationale the agency has asserted is that, while it has the authority to regulate LDTs, the use of LDTs has contributed to enhanced standards of medical care and that significant regulatory changes in this area could have negative effects on the public health.
FDA's focus on the patient is of course well placed. FDA should never disregard the impact on the patient, and always should consider the pros and cons of any regulatory action. But this hurdle is not necessarily difficult to overcome. Doing so requires planning, notice to affected parties, and organization. Furthermore, because many LDT makers assert they are already doing substantial evidence collection, there should not be much of an additional cost or delay, particularly for a marketing submission. The point is that FDA is not being asked to act precipitously, but simply to act. FDA can give notice of intent and allow a reasonable time for compliance. However, simply because the regulation would need to be planned and rolled out in an organized way should not prevent the agency from acting, as it largely has done so far.
FDA should direct its regulatory energies to higher risk products
In general, risk-based regulation is a fundamental concept of the FDA regulatory framework. The problem is that regulation according to risk does not in any way explain the differentiation between LDTs and IVDs. Risk, in this area, is influenced primarily by three factors: the intended use of the test, the technology involved, and the quantity of a given test on the market. Consider the risk presented by LDTs and IVDs in each of those three areas.
Intended Use. The intended uses for LDTs are very often exactly the same as they are for IVDs. With the exception of the IVDMIA proposal, there is nothing structural in the agency's current LDT policy that prevents LDTs from being used for the most risky of diseases and conditions. Therefore, the risk associated with intended use cannot be a justification for leaving LDTs unregulated.
Technology. As with intended use, the technologies used in LDTs may often be the exact same technologies as are used in IVDs. Again, with the exception of IVDMIAs, there is nothing in the LDT policy that would prevent an LDT from comprising the most cutting edge—and riskiest—technologies known to science.
Quantity. Some might argue that by limiting the availability of LDTs to the institution in which they are developed, the quantities will be inherently less than those of IVDs, so the risk will be less. That proposition, though, has no empirical support. Many LDTs achieve quantities comparable to IVDs in at least two senses. First, a single LDT may be produced in enormous quantities when it is used by a major national laboratory company that has multiple sites throughout the United States. Some of these clinical laboratories have revenues greater than $1 billion, and the tests they produce and use can be performed on thousands of patients.
Second, quantity should not only be examined on an individual, proprietary test basis. The aggregate of these tests must also be examined. By analogy, there might be literally a million small-scale wheat farmers in the United States, none of which is producing an amount of wheat that is individually significant. However, the total amount of wheat produced is extraordinary, and the potential aggregate risk is high. FDA would not omit wheat from its regulatory oversight simply because it is produced by many different farmers.
In total, the evidence does not seem to exist to support a conclusion that LDTs are somehow safer than IVDs. Indeed, systemic data on that point are lacking, in large part because the agency has thus far declined to regulate LDTs and make them subject to the same adverse-event reporting requirements that apply to IVDs.
FDA lacks the legal authority to regulate LDTs
In more recent times, people have questioned whether FDA has legal authority to regulate LDTs. If the agency did not, that would indeed be a reason for the differential; however, for years, FDA has asserted dozens of times that it has the legal authority to regulate LDTs. The authors agree with that assertion.
FDA should encourage innovation, and LDTs can be innovative
Again, FDA should encourage innovation. But its current LDT policy is not at all tailored to that objective. Innovation can come from nearly any part of society, be it academia, laboratories, manufacturers, or something else. The LDT policy is not directed at giving more freedom or latitude specifically to those involved in innovation. Indeed, the LDT policy is both over inclusive and under inclusive when it comes to innovation. The policy is under inclusive because there is plenty of innovation being done within FDA-regulated IVD manufacturing that needs to be encouraged, but FDA's LDT policy does not encourage that innovation. In fact, FDA's policy discourages and puts at a disadvantage the innovation efforts of manufacturers. The policy is also over inclusive because many of the LDTs being developed are not at all innovative. Currently, certain clinical laboratories are buying products from manufacturers, reverse-engineering them, and trying to produce them at commodity prices without being burdened by all of the research and quality controls that were necessary to develop the product. So the LDT policy actually encourages simple generic diagnostic test development, not innovation.
CLIA regulation precludes FDA regulation
This argument seems to have two underlying points. First, that CLIA legislation legally precludes additional regulation under the Federal Food, Drug, and Cosmetic (FD&C) Act. Second, that as a practical matter, CLIA makes additional regulation by FDA unnecessary. The first point is wrong on its face. Nothing in the CLIA statute precludes applying a co-equal federal statute such as the FD&C Act. Indeed, FDA obviously already applies the act to IVDs which, like LDTs, are used in CLIA-regulated laboratories. In this way, IVD tests are doubly regulated in that all non-waived tests must also be performed in CLIA labs. With regard to the second argument, that as a practical matter FDA regulation is unnecessary, there are only three logical scenarios to describe whether there is a need for FDA regulation on top of CLIA. Policymakers must believe one of the following scenarios:
1. FDA regulation adds enough value (in the form of safer and more-effective products) to warrant applying it to LDTs. The cost of complying with FDA regulation is outweighed by the benefit of having the risk-based principles of such regulation in addition to CLIA, in the form of safer and more-effective products than would otherwise exist. In this regard, FDA regulation has much to do with risk management and establishing efficacy and performance, while CLIA does not.
2. FDA regulation does not add enough value to warrant applying it to LDTs. The cost of complying with FDA regulation is not outweighed by the benefit of having risk-based oversight of LDTs. In other words, FDA should not regulate IVDs because CLIA is sufficient and risk-based regulation is not needed.
3. No one knows whether FDA regulation adds enough value or not. This could be because no one is collecting systematic data on the safety of LDTs comparable to the data collected on the safety of IVDs. As discussed elsewhere in this article, there are postmarket reporting obligations that apply to IVD manufacturers, which provide oversight of the safety and risk profile of those devices. However, no corresponding reporting requirement for LDTs exists. Policymakers in this category simply do not know whether risk-based FDA regulation adds value because the data is not there.
The most important point here, though, is that, again, FDA has no reason based on available data to justify regulating IVDs and not LDTs. Either it makes sense for FDA regulation to be layered over CLIA to regulate both IVDs and LDTs, or it makes sense to exempt both IVDs and LDTs from FDA regulation. But CLIA provides no basis for treating LDTs and IVDs differently.
At most, FDA simply lacks data on whether LDTs possess comparable safety and effectiveness to their IVD counterparts, and that ignorance cannot be the justification for leaving LDTs unregulated. While the agency figures that out, FDA should treat both the same. IVDs and LDTs must be presumed to have the same risk profile unless and until there are data to distinguish them.
FDA lacks the resources to regulate LDTs
Another rationale FDA has advanced in the past is that, while it has the authority to regulate LDTs, the number of LDTs would exceed the agency's review capacity. There are two fairly obvious reasons why this explanation does not justify the regulatory distinction between IVDs and LDTs.
On its face, this rationale does not have anything to do with differentiating between manufacturer-produced IVDs versus LDTs made by laboratories. If it were true that the agency did in fact lack the resources to enforce the law in this area, such a circumstance would demand that the agency prioritize the risk of the various categories of products and regulate those categories that are the highest risk. In contrast, a lack of resources does not in any way mean that LDTs should be unregulated and IVDs should be regulated. This argument does not speak at all to the distinction between those categories. Why regulate one and not the other, just because FDA cannot regulate both? That argument would only make sense if it were impossible to subdivide the categories any other way, but of course there is an almost endless number of ways to subdivide them.
This is an important point, because it means that FDA's decision to regulate one category and not the other is simply a product of an historical anomaly. The agency simply chose to regulate one before considering the other. But while that historical sequence offers an explanation, it offers no justification for why today FDA would choose its current direction. FDA could easily subdivide the whole category of diagnostic tests in ways that would be far more rational than using the historical sequence of the order in which they were thought of by the agency. The second reason this justification fails is that it is no longer accurate. True, the agency is chronically under funded, and nearly everyone supports additional funding. But there are new sources of funding that would be available for the agency to tap should it decide to regulate LDTs.
In particular, on the product review side, the user fee program allows the agency to pay for additional workload. Although there may be some transitional challenges involved to obtain the necessary amount of review expertise, that transitional challenge hardly seems a basis for avoiding the agency's regulatory duties.
Furthermore, on the field oversight front, there are both sources of funding and potential efficiencies that make the challenge far less daunting than it seems. Right now, CMS routinely sends surveyors to clinical laboratories to assess compliance with requirements under CLIA. With some training from FDA, the same surveyors could also assess compliance with FDA's quality system requirements. Moreover, there are mechanisms available to charge clinical laboratories fees for inspection services.
In sum, resources are available without any act of Congress that would allow FDA to equalize the regulatory treatment of LDTs and IVDs. Moreover, even if resources were an issue, resources do not explain the "all or nothing" approach FDA has taken to the regulation of LDTs.
None of the major arguments advanced to justify exempting LDTs from regulation holds water. There is no compelling reason to justify not regulating them. FDA regulation of LDTs is needed; there are good reasons for such regulation. As a matter of patient safety, encouraging innovation, and historical precedent, it makes sense to regulate LDTs.