By Rachel Sachs
Over the past several months, I’ve been blogging (here, here, and here) about the FDA’s recent forays into regulating laboratory-developed tests (LDTs). Since the release of the draft guidance framework in October, serious arguments have been made on opposing sides of the issue, and industry groups have made additional moves in opposition to the proposed regulation. And now, today (and tomorrow), the FDA is holding a public workshop on their draft guidance framework, focusing on a wide range of issues.
Today’s workshop featured sessions on three main issues: 1) labeling considerations, 2) clinical validity and intended use, and 3) categories for continued enforcement discretion. Many commenters simply presented the unique concerns of their organization and urged the FDA to consider them in finalizing the guidelines, which was helpful when it did seem as if the draft guidance may have insufficiently considered the needs of a particular set of laboratories, such as public health laboratories that focus on testing for infectious diseases like Ebola and chikungunya (about which I’ve also blogged, here and here).
More helpful, though (at least in my view), were the comments of those who sought to provide concrete recommendations for the FDA on the basis of 1) the policy concerns they saw underlying the guidance and 2) the practical effects of implementation that they foresaw. I’ll illustrate with an example, which hopefully will display the complexity inherent in even the simplest questions that the FDA must answer here.
Historically, the FDA did not regulate LDTs because they were 1) relatively simple tests, 2) involved well-known scientific principles, and 3) were performed by doctors or pathologists responsible for taking care of the patient. The FDA calls these “traditional LDTs,” and they have in mind tests like a Pap smear or a complete blood count. But many new LDTs do not seem to match this model, like complex genomic tests. The FDA would like to subject the latter to premarket review requirements, while allowing continued enforcement discretion over the former.
The challenge for the FDA, though, is to define specific criteria delineating these two groups. Much of the lifting is done by its criterion that a “traditional LDT” must be “interpreted by qualified laboratory professionals without the use of automated instrumentation or software for interpretation.” But at the hearing today, there was extensive debate over another the delineating factor the FDA has proposed: “whether [the test] is manufactured and used by a health care facility lab … for a patient that is being diagnosed and/or treated at that same health care facility or within the facility’s healthcare system.” This criterion is clearly related to the FDA’s historical conception of LDTs as described above. But several commenters noted the potential for this criterion to affect the way in which laboratories and hospitals choose to structure their testing facilities, offering examples of ways in which “health care facility” might be defined either narrowly or broadly.
The valence of this argument is not apparent. A narrow definition of “health care facility” that encourages in-house cultivation of laboratory capacity might be praised for the connection it fosters between pathologists and patients, or it might be criticized for creating fragmentation and inefficiency. A broad definition of “health care facility” that incorporates seemingly disparate entities might be praised for the way in which it permits the cultivation of expertise, or criticized for leaving too many LDTs insufficiently regulated. And because the organization of healthcare is changing independently of the FDA’s actions, it might be prudent to stay out of the issue entirely. As much as I hate to admit it, the views of one panelist seem to sum up the situation: “Whatever you come up with for a definition, you’re going to be criticized. This is impossible.”
On that cheerful tone, stay tuned for Part II of this blog post, recapping the highlights of Friday’s panels!