$5.7 billion in new biopharma facilities. A wave of gene therapy approvals on the horizon. And now, the most significant shift in FDA approval policy since 1962.

On February 18, 2026, FDA Commissioner Marty Makary and CBER Director Vinay Prasad published a commentary in the New England Journal of Medicine announcing that one adequate, well-controlled clinical trial, combined with confirmatory evidence, is now the FDA’s default standard for marketing authorization of novel products. The two-trial requirement, a safeguard that had been in place since Congress passed the Kefauver-Harris Amendment in the wake of the thalidomide disaster, is officially over.

Everyone is celebrating the speed. Nobody is talking about what this does to your floor.

What Actually Changed

The FDA has had the legal authority to approve drugs on a single pivotal trial since the FDA Modernization Act of 1997. This isn’t new law. What changed is the default posture. Previously, sponsors seeking single-trial approval had to justify the deviation. Going forward, two trials requires justification. The burden of proof flipped.

Makary and Prasad framed it plainly: modern drug development establishes credibility in multiple ways, relying on both statistical and biologic inferences. Two trials, they argued, should be seen as just one of many interlocking facets of clinical credibility. Biomarker data, mechanistic plausibility, effect size, and sophisticated trial design have made the duplicate-confirmation model a relic of a less precise era.

In practice, this affects common disease indications most acutely. Oncology and rare disease programs have already been operating under single-trial norms for years. But now a sponsor developing a cardiovascular drug, a metabolic disorder therapy, or an autoimmune biologic, areas where two trials was historically expected, gets the same flexibility.

The FDA also confirmed that this standard applies across all expedited pathways: Breakthrough Therapy designation, Accelerated Approval, Priority Review, and standard NDA/BLA submissions alike.

The Part Nobody Is Printing

Here is the sentence that should be on the wall in every CDMO operations office in the country:

The FDA is not eliminating data requirements. It is resequencing them.

What that means operationally: the evidence burden that used to sit in a second Phase 3 trial, running 18 to 36 months before approval, now sits in your post-market surveillance infrastructure, running indefinitely after it. Some analysts are already calling it the “RWE mortgage.” In exchange for the speed of a single-trial approval, the FDA is demanding registry-grade real-world evidence to confirm durability over the drug’s lifecycle.

This matters enormously for manufacturing, because the second trial wasn’t just a clinical exercise. It was also time. Time to stabilize your process. Time to close deviation trends before PAI. Time to mature your training program from Phase 2 operator habits into commercial-grade competency. That buffer is gone now.

The Cell and Gene Therapy Problem Is Larger

For conventional pharma, the single-trial shift compresses timelines. For cell and gene therapy companies, it compounds problems that were already structural.

CGT manufacturing has never benefited from the repetition that conventional biologics programs take for granted. An autologous CAR-T program might process 50 to 100 patients across a Phase 1/2 study before pivotal. By the time a company reaches BLA submission, the manufacturing team may have run the full process fewer than 200 times, across multiple operators, at multiple sites, with material variability that begins at the patient’s apheresis.

The second pivotal trial, for programs that had one, wasn’t just clinical evidence. It was also 50 more process runs. 50 more batch records. 50 more opportunities to identify training gaps before a PAI investigator does.

Consider the specific risks that get compressed under a single-trial model for CGT programs:

Process Validation

Under FDA’s 2011 Process Validation Guidance, Stage 2 process qualification requires demonstration that the process design operates reproducibly. For small-batch autologous products, the statistical basis for PPQ is already difficult to establish. With a single trial, sponsors may submit PPQ data derived from clinical manufacturing campaigns, a technically acceptable approach but one that FDA has historically scrutinized heavily for autologous products where patient-to-patient variability complicates reproducibility claims.

Operator Qualification

In CGT manufacturing, operators are part of the process in a way that is difficult to overstate. Manual steps in cell activation, transduction, and formulation introduce variability that equipment qualification does not capture. A training program that was adequate for a Phase 2 operator running the process quarterly is not the same as a commercial qualification program for operators running the process weekly. The second trial gave companies time to make that transition. That time is now gone.

Equipment and Facility Qualification

Many CGT sponsors bring new commercial-scale equipment online between Phase 2 and BLA filing. Bioreactors, closed system processors, automated fill-finish systems. Under the old model, the second trial provided data to support commissioning and qualification of that new equipment at meaningful commercial scale. Under the single-trial model, the IQ/OQ/PQ for commercial equipment may rest on a smaller data set than FDA inspectors are accustomed to seeing.

Critical Raw Material Qualification

Plasmid DNA, viral vector components, GMP-grade reagents, and cell culture media suppliers all require qualification with meaningful batch data. A single trial may not generate sufficient batches to complete supplier qualification to the standard required for commercial operations. This is one of the most underappreciated risks in the single-trial framework for CGT specifically.

What the PreCheck Signal Tells You

The single-trial announcement did not land in a vacuum. Three weeks earlier, on February 1, the FDA launched its PreCheck Pilot program, an initiative designed to frontload facility evaluation so that manufacturing shortfalls don’t derail approvals at the finish line.

The traditional PAI model has always been a source of operational risk: you build the facility, run the clinical trials, submit the BLA, and then an investigator shows up and potentially finds something that collapses the approval. PreCheck inverts that sequence by creating a communication channel between FDA and manufacturers during facility construction or expansion, before the BLA clock is running.

Read those two signals together. The FDA is telling you two things simultaneously:

  1. We are going to approve your drug faster.
  2. We expect your manufacturing operation to be ready when we do.

For companies that have not yet internalized that message, the math is unforgiving. A BLA filed after a single pivotal trial may reach FDA review 18 to 24 months ahead of where it would have been under the old model. If your commercial manufacturing infrastructure, your quality system, your training program, your supplier qualification packages, are sized for the old timeline, you now have a readiness problem that no amount of regulatory enthusiasm will fix.

The Training System Failure Nobody Wants to Discuss

Here is the uncomfortable version of this story.

Most biopharma training programs are designed around the assumption that they have time. Time to build the SOP library. Time to qualify operators. Time to run mock audits. Time to close the gaps that show up in deviation investigations. That assumption has always been partially wrong, but it was survivable under a two-trial model because the second trial bought you the time you needed to catch up.

Under a single-trial model, the training system has to be commercially ready at the same time the clinical trial data is. Not six months later. Not after the first PAI observation. Simultaneously.

The companies most at risk are not the large pharma organizations with dedicated training infrastructure. They are the mid-size biotechs and CDMOs that have been running lean, treating training as a compliance function rather than an operational readiness function, and assuming that any gaps would surface during the second trial and get fixed before BLA filing.

The FDA PreCheck program is useful precisely because it forces this conversation earlier. But PreCheck is voluntary and still in pilot. Companies that do not engage with it proactively will get their training program reviewed during PAI, under time pressure, with a BLA clock ticking.

The Five Moves That Matter Now

This is not a call to panic. It is a call to sequence correctly.

1. Treat your pivotal trial as your commercial manufacturing readiness window, not just your clinical data generation window. If the pivotal trial starts and your training program is still in draft, you are already behind. The clinical trial is the last substantive opportunity to collect batch data, qualify operators, and mature your quality system before FDA scrutiny.

2. Build your post-approval surveillance infrastructure before BLA filing, not after. The “RWE mortgage” is real. Sponsors who get to approval without a registry design, a pharmacovigilance plan, and a long-term follow-up framework for CGT products will face immediate post-approval compliance pressure. That pressure lands on operations, not just medical affairs.

3. Review your PPQ strategy with single-trial data assumptions. If your PPQ plan assumed commercial-scale runs from two Phase 3 campaigns, that plan needs revision. This is a CMC/QA conversation that needs to happen now, not at BLA writing.

4. Accelerate your supplier qualification program. For CGT companies especially, critical raw material suppliers cannot be adequately qualified on Phase 2 volume alone. Build your commercial supplier qualification program in parallel with the pivotal trial, not sequentially.

5. If you are a CDMO, this is your differentiator. Sponsors who do not have the internal manufacturing infrastructure to execute single-trial readiness will look for CDMOs that do. The CDMOs who can credibly demonstrate commercial-ready training programs, qualification frameworks, and QMS infrastructure, aligned with a compressed BLA timeline, will win the business that the single-trial wave generates.

The Bottom Line

The FDA’s single-trial shift is not a gift to industry. It is a stress test with a faster clock.

For conventional biopharma, it compresses the manufacturing readiness window by 18 to 24 months. For cell and gene therapy companies, it compounds structural process validation and training challenges that were already difficult under the old model. For CDMOs, it creates competitive separation between organizations that run training as a compliance function and those that run it as an operational readiness capability.

The companies that come out ahead will be the ones who recognized early that faster approval does not mean lower standards. It means the same standards, with less time to reach them.

Your training program is not a regulatory checkbox. Under the single-trial model, it is part of your BLA package whether it appears in Section 3.2.P or not.

References

  1. Prasad V, Makary MA. One pivotal trial, the new default option for FDA approval: ending the two-trial dogma. N Engl J Med. 2026;394(8):815-817.
  2. Anderson J. The “one-trial” trap: why 2026’s FDA efficiency push actually doubles your operational risk. Clinical Leader. 2026.
  3. US Food and Drug Administration. Guidance for Industry: Process Validation: General Principles and Practices. January 2011.
  4. US Food and Drug Administration. PreCheck Pilot Program. BioSpace Policy Tracker 2026.
  5. US Food and Drug Administration. Considerations for the Development of Chimeric Antigen Receptor (CAR) T Cell Products. March 2022.

That’s it for Issue #3. If this was useful, forward it to someone on your quality team. If you want to talk about any of this in more detail, I’m at bdrapeau@gxpframe.com.

Brian Drapeau
Founder, GxP Frame
gxpframe.com

See you next week!