On a date now circulating across regulatory affairs teams, FDA announced that one adequate and well-controlled clinical trial, combined with confirmatory evidence, will serve as the new default standard for all drug approvals. The announcement lands while a substantial share of Phase 3 programs in active development were scoped, powered, and budgeted against a two-trial assumption that FDA has now formally displaced. Sponsors who set that assumption in their development plans last year did not make an error in judgment at the time. They may, however, be carrying excess trial infrastructure that the current regulatory posture no longer requires.
The Nature Reviews Drug Discovery analysis published under the title “Two trials or not two trials? That should not be the question” frames the dual-trial debate as a category error. The article’s central argument is that the field has been optimizing for a structural requirement, two adequate and well-controlled studies, rather than for the evidentiary threshold those two trials were designed to satisfy: replicable demonstration of a treatment effect with controlled error rates. When the question shifts from trial count to evidentiary sufficiency, single-trial approval supported by mechanistic, biomarker, or real-world confirmatory data becomes a coherent regulatory position rather than an exception requiring extensive justification.
What FDA’s New Posture Actually Requires
The practical content of FDA’s single-trial default warrants careful reading, because the announcement changes the burden structure without eliminating the evidentiary bar. A single pivotal trial satisfying the “adequate and well-controlled” standard under 21 CFR § 314.126 remains necessary. What changes is that sponsors no longer face an automatic expectation of a second independent replication study before NDA submission. The confirmatory evidence requirement fills that gap, and FDA has paired the default with a “plausible mechanism framework” intended to govern how mechanistic or biological data can serve confirmatory functions for individualized therapies specifically.
That framework distinction matters operationally. Sponsors in oncology, rare disease, and precision medicine, where patient populations constrain enrollment and biological rationale is frequently robust, gain the most direct benefit. Sponsors in broad-population indications, where biological plausibility is harder to specify and regulatory reviewers have historically demanded replication as protection against type I error inflation, face a more nuanced calculus. FDA had not, at the time of the announcement, published a guidance document specifying exactly which confirmatory evidence types satisfy the new default across indication classes. The absence of that specification is itself a design constraint sponsors need to account for in their next Type B or Type C meeting request.
The integrated safety evidence question compounds this. FDA’s March 2024 webinar on integrated safety analyses in drug marketing applications emphasized that early planning for the integrated safety analysis is a prerequisite for adequate review, and identified Type C meetings as the appropriate venue for sponsors to align the integrated safety analysis plan with the agency before data lock. Under a single-trial architecture, the safety database accumulated from one pivotal study is thinner than the combined dataset from two trials. Sponsors transitioning to single-trial designs without revisiting their integrated safety analysis plan inherit a structural gap between the new trial count and the old safety evidence expectations.
The EMA Divergence Sponsors Cannot Ignore
FDA’s new default does not travel automatically across the Atlantic. The EMA’s CHMP has maintained that randomised controlled evidence remains the standard for demonstrating efficacy in marketing authorisation applications, and its guidance on single-arm trials submitted as pivotal evidence positions that pathway as an exception with specific evidentiary conditions, not a general alternative to controlled replication. Sponsors building a global registration strategy around FDA’s single-trial default must map whether their confirmatory evidence package, however persuasive to an FDA reviewer, meets CHMP’s distinct expectations for the same application cycle.
The EMA’s conditional marketing authorisation framework provides a parallel but structurally different pathway. Under the conditional marketing authorisation framework, EMA can approve on incomplete data subject to post-authorisation obligations, including specific studies the applicant commits to completing. The key distinction from FDA’s new default is timing and obligation structure: EMA’s conditional route defers confirmatory evidence to post-approval with binding commitment, whereas FDA’s framework appears to require confirmatory evidence as part of the NDA package, not as a post-market obligation. Sponsors targeting simultaneous FDA and EMA submissions under a single-trial design face a sequencing problem that the available published guidance from either agency does not clearly resolve.
A 2022 economic evaluation published in JAMA Network Open provides relevant cost framing for this divergence. The analysis found that a single platform trial evaluating 10 interventions cost substantially less and completed faster than a series of conventional two-group trials covering equivalent ground. Translated to the dual-agency context: sponsors who design the single FDA pivotal trial with CHMP conditional authorisation requirements built in, specifying post-authorisation study commitments at the design stage rather than retrofitting them after NDA submission, preserve cost efficiency while maintaining a viable EMA pathway. Sponsors who treat FDA’s new default as license to reduce total trial investment without adjusting the EMA strategy are optimizing for one regulator at the cost of the other.
The Design Implications for Programs in Progress
Sponsors with Phase 2 readouts expected in Q4 2026 or Q1 2027 face the most acute version of this decision. The protocol for the Phase 3 pivotal study is being finalized now, and the assumption embedded in that protocol, either one trial or two, will be extremely difficult to revise once FDA has reviewed and commented on the IND amendment containing it. The Nature Reviews Drug Discovery analysis explicitly identifies trial count as the wrong optimization target; the right optimization target is pre-specification of the evidentiary chain that will satisfy both the primary endpoint and the confirmatory evidence requirement FDA has attached to its new default.
For sponsors with programs already in Phase 3 under a two-trial design, the publication and FDA’s announcement together create a legitimate question for the next Type B meeting: whether the second trial, if it has not yet enrolled, could be restructured as a confirmatory study with a different design burden than a full independent replication. That restructuring requires FDA agreement and a protocol amendment, carries its own timeline risk, and should be modeled against the cost of completing the second trial as originally designed before any decision is made.
Site networks and CROs supporting multi-trial programs in rare disease and oncology face a distinct implication. If sponsors move toward single-trial Phase 3 designs at scale, total enrollment demand per program decreases, which affects site activation timelines, per-site patient flow projections, and the revenue models CROs use to staff therapeutic area units. The platform trial data from JAMA Network Open is instructive here: consolidated trial designs generated efficiency gains at the program level, but they concentrate enrollment risk in a single protocol. A single-trial strategy that encounters an operational disruption, a safety signal requiring protocol amendment, or an enrollment shortfall has no second trial to absorb the delay.
The next regulatory artifact to track is FDA’s anticipated guidance specifying which confirmatory evidence types satisfy the new default standard across indication classes. Until that document is published, sponsors should use Type C meetings to document FDA’s indication-specific expectations in writing, and they should treat any informal agreement reached in those meetings as binding only to the extent it is reflected in the written meeting minutes FDA issues following the meeting. The plausible mechanism framework announced alongside the single-trial default applies explicitly to individualized therapies; its scope outside that category remains an open question that the forthcoming guidance will need to resolve.
References
- Nature Reviews Drug Discovery, “Two trials or not two trials? That should not be the question”
- Troutman Pepper, “FDA Announces a Single Pivotal Trial as the New Default Standard for All Drug Approvals and Unveils a Plausible Mechanism Framework for Individualized Therapies”
- EMA CHMP, “Establishing efficacy based on single-arm trials submitted as pivotal evidence for marketing authorisation”
- FDA, “Integrated Safety Analyses in Drug Marketing Applications: Avoiding Common Mistakes” (March 2024)
- JAMA Network Open / PMC, Economic evaluation of platform trial versus conventional trials for 10 interventions
- EMA, “Conditional marketing authorisation”
Moe Alsumidaie is Chief Editor of The Clinical Trial Vanguard. Moe holds decades of experience in the clinical trials industry. Moe also serves as Head of Research at CliniBiz and Chief Data Scientist at Annex Clinical Corporation.
