A startup's guide to LDT development.

On March 31, 2025, Judge Sean D. Jordan of the U.S. District Court for the Eastern District of Texas vacated and set aside the FDA's Final Rule on Laboratory Developed Tests. The rule, issued in May 2024, had intended to bring LDTs under the agency's medical device authority. The court ruled that the FDA had exceeded its authority. The matter was remanded to the Department of Health and Human Services. Clinical laboratories breathed out.

Most of the commentary that followed was a sigh of relief and a return to business as usual. We think that is the wrong reading. The rule is dead — the operational pressure that produced it is not. For diagnostic startups in particular, the regulatory calculus has just gotten more interesting, not less, and the conventional advice we have been giving founders for the last eighteen months no longer applies.

Through 2024, the conventional advice to a startup with an early-stage assay was straightforward: launch as an LDT to generate evidence and revenue, then sequence the 510(k) pathway around real-world performance data. The Final Rule disrupted that sequence by adding a clock — the rule's phased enforcement schedule meant that an LDT launched in 2024 was a regulated medical device on a defined timeline, whether or not the underlying assay was ready. The right answer for many startups in early 2025 was to push toward 510(k) clearance ahead of the schedule, even at the cost of slower launches and weaker evidence packages.

That answer is now wrong. With the Final Rule vacated, the LDT pathway is again what it was before May 2024 — a path to market that does not require pre-market FDA clearance, governed primarily by CLIA and state-level oversight. A startup launching an LDT today is not on the 510(k) clock the rule had created. The right move is to slow down, generate the evidence the assay deserves, and sequence the regulatory path against the data rather than against the rule.

That is the obvious read. The less obvious read — and the one we want diagnostic founders to think about — is what the FDA does next.

The court did not tell the FDA to stop caring about LDTs. It told the FDA that the path it took to regulate them was wrong. The matter was remanded to HHS. The agency now has to decide whether to pursue new rulemaking, push for legislation, or pivot to a different enforcement angle. Of those three, the third is the fastest, and it is the one we believe the agency is most likely to pursue.

The pivot we expect targets specimen collection kits. The FDA already regulates specimen collection devices that move across state lines or are used outside a single laboratory facility — they are medical devices under the existing statute, and the agency has issued warning letters on this basis before. During COVID, direct-to-consumer collection kits drew exactly this kind of enforcement. For any clinical laboratory that operates outside a single facility — accepting specimens from a wide range of collection locations, or running a home-collection program — the FDA could begin enforcing medical device regulations on the kits without any new rulemaking. The authority is already there.

This is not speculation about a hypothetical worst case. It is the most direct path the FDA has to address the same concerns the LDT rule was attempting to address — laboratories operating at scale, beyond their original facility footprint, on assays the agency has not reviewed. Specimen collection kit enforcement reaches the same operating model through a different door.

For a diagnostic startup, this changes the design question that comes after "how do we go to market." It is no longer enough to confirm that the assay is launching as an LDT. The follow-on question is whether the operating model relies on specimen collection kits that ship to patients or to non-laboratory collection sites. If it does, the founder should assume that the kit itself will eventually need a regulatory pathway, even if the LDT does not. The kit is a medical device, full stop. Building the quality system around that assumption now — and building the assay validation evidence to support an eventual 510(k) on the kit — is materially different from building toward an LDT-only regulatory profile.

The flipside, for laboratories whose operating model is genuinely contained within a single facility, is that the LDT pathway is meaningfully cleaner today than it was eighteen months ago. The race-to-510(k) advice that made sense in 2024 was a response to a clock that no longer exists. Slow down. Validate properly. Generate the evidence. Use the time the court bought.

Two things are true at once. The conventional wisdom about LDTs has just become obsolete, and the path the FDA takes from here is not the one the industry was preparing for. The startups that adapt to both — without overcorrecting in either direction — will be the ones whose regulatory profile holds up when the next pivot lands.