These days there are two topics that dominate legal conferences, presentations, and CLEs: artificial intelligence (AI) and Loper Bright.  You will doubtless see us frequently bloviate about the former, but today’s case – American Clinical Laboratory Ass’n v. Food and Drug Administration, 2025 U.S. Dist. LEXIS 59869, 2025 WL 964236 (E.D. Tex. March 31, 2025 – is about the latter.  In fact, American Clinical is the first FDA-specific application of Loper Bright.  Spoiler alert: the FDA lost.  

The lawsuit was brought by various plaintiffs, including laboratory associations, challenging an FDA final rule that would regulate laboratory testing as medical devices.  (The court had no difficulty holding that these plaintiffs had standing to challenge the FDA’s proposed rule because compliance would cost them plenty.) The plaintiffs argued that the final rule exceeded the FDA’s authority, and that it must be vacated under the Administrative Procedure Act.  The parties filed cross motions for summary judgment. 

We have heard a lot lately about the text-history-tradition method of constitutional and statutory interpretation.  In American Clinical, the court framed its analysis in terms of “text, structure, and history.”  The “tradition” component has always puzzled us a bit, and we are not sure how “structure” differs from text, but the American Clinical court’s approach ends up being fairly easy to follow.  

The court starts, as most do these days, with the text. Put simply, the court held that it could not stretch the statutory definition of “medical device” to include laboratory-developed tests.  The Food, Drug, and Cosmetic Act (FDCA) defines “device” as a “instrument, implement, machine, contrivance, implant, in vitro, or other similar related article, including any component, part, or accessory” that diagnoses or treats disease. By contrast, the lab tests at issue here constituted professional medical care, not the sort of “tangible, physical products” that the statute – applying “ordinary, contemporary, common meaning” – includes as medical devices.  Moreover – and maybe this is a “structure” point – several of the FDCA’s device regulations can be “understood to apply only to manufactured products.”

The history at issue involved FDA’s prior positions being inconsistent.  The court observed that “[a]lthough FDA’s view of its authority to regulate laboratory-developed test services as devices has been a moving target for decades, the relevant statutory framework has not changed since 1988.” The FDA’s inconsistency is normal – after all, new presidents mean new FDA people and policies – and was one of the things driving SCOTUS in Loper Bright to inter Chevron deference.  

Congress passed statutes regarding medical devices in 1938 and 1976, but the FDA did not endeavor to regulate laboratory services until 1992, and even then the FDA’s moves amounted to wavering gestures, some of which “were not well received by Congress.”  (But what did Congress actually do to set the FDA straight? Not much. Par for the course, right?  For a while now, Congress has been more performative than functional. The failure of Congress to do its job is part of the Chevron-Loper Bright story.)

After years of vague mutterings about regulating lab tests as devices, the FDA took concrete steps in 2023 when it announced “its intent to move forward with regulating virtually all laboratory-developed test services as medical devices.”  The “FDA acknowledged that the costs of the proposed rule would be significant.”  It also recognized that the rule would “impose major burdens on laboratories.” 

But the FDA’s effort to regulate lab services as devices ignored the Congressional assignment of oversight of such tests to a different agency (the Centers for Medicare and Medicaid Services) governed by a different statute (the Clinical Laboratory Improvement Act of 1967 (CLIA), which was expanded/amended in 1988). The sequence of legislative enactments reflected that “Congress viewed (1) ensuring medical device safety and effectiveness, and (2) ensuring laboratory-testing accuracy, as distinct problems requiring different regulatory solutions.”  And, after all, Congessional intent, rather than agency aspirations, is what matters when it comes to statutory interpretation.

When it issued its final rule in 2024, the FDA stated that the FDCA “has always classified laboratory test services as medical devices and that the entire history of the medical device requirements, FDA has declined to apply them to laboratory-developed tests simply as a matter of ‘enforcement discretion.’”

Pre Loper Bright, a court applying Chevron deference might well have felt constrained to adopt the FDA’s interpretation and assented to the assertion of regulatory power over lab tests. But Loper Bright is the new sheriff in town, and Chevron is on Boot Hill. Thus, the court in American Clinical concluded that the FDA could not rely on non-reviewable “enforcement discretion” to pound the square peg of therapeutic tests conducted by medical professionals into the round hole of medical device manufacturing requirements. The FDA’s “creative attempt to expand its jurisdiction under the FDCA” failed because it had no authority to expand the definition of “device,” and it contravened “the FDCA’s limits on FDA’s jurisdiction.”  Neither the CLIA’s “statutory text nor its legislative history make any reference to FDA’s allegedly preexisting authority to regulate laboratory-developed test services as ‘devices.’”

Under Fifth Circuit precedent, district courts should generally “nullify and revoke” illegal agency action. The American Clinical court concluded that, in light of the profound compliance burdens that the FDA’s proposed rule would have visited upon labs, the rule had to be vacated. The case was remanded to the FDA for “further consideration in light of this opinion.”  

As a practical matter, decisions such as American Clinical, which apply Loper Bright to curtail FDA authority, might not matter all that much in this moment. The current FDA, which seems to be, er, retrenching, does not appear to have the resources to expand into anything new.  We almost said that the current FDA also does not appear to have the ambitions to expand into anything new, but who can really say? Let’s face it: the executive branch is now through the looking glass.  Any talk about abench inconsistency right now sounds like a comical understatement. American Clinical reminds us that the judiciary – the “least dangerous branch” – appears to be the one branch of government that actually does its job with more logic than lunacy.