U.S. Representation in Multiregional Trials: FDA Approval Trends

U.S. representation in multiregional clinical trials: is it becoming a harder requirement for FDA approval?

Short answer

Yes. U.S. representation in multiregional clinical trials (MRCTs) is moving from an implicit expectation toward an explicit, planned-for requirement, most sharply in oncology. FDA has not created a new legal standard: the applicability of foreign data to the U.S. population and U.S. medical practice has been codified for decades. What has changed is enforcement posture and specificity. FDA now (1) publicly flags the declining share of U.S. participants in oncology MRCTs as a threat to interpretability, (2) tells sponsors to prospectively engineer site distribution and regional allocation to secure an adequate U.S. subgroup, and (3) has begun issuing CRLs that name limited U.S./Western generalizability as a reason for non-approval. The direction of travel is toward earlier, more quantitative agreement on U.S. enrollment as a condition of approvability.

The legal baseline has always required U.S. applicability

FDA regulations have long permitted foreign clinical data to support an NDA or BLA. Under 21 CFR 312.120, FDA will accept a well-designed, well-conducted foreign study not conducted under an IND if it met good clinical practice and FDA can validate the data, including by on-site inspection if necessary ⁸⁶. But reliance solely on foreign data triggers a higher bar: 21 CFR 314.106(b) requires that the data be applicable to the U.S. population and U.S. medical practice ⁸⁶. That applicability test, not the mere acceptability of foreign data, is the pressure point.

The principle recurs across therapeutic-area guidance. In chronic hepatitis B, FDA states that development programs should include a sufficient number of U.S. patients to ensure prevalent U.S. genotypes are represented ⁹¹. In hepatitis D, FDA notes that even where foreign data may be acceptable as a sole basis for approval, it encourages sponsors to include U.S. patients to provide experience relevant to the U.S. population ⁸⁹. For medical devices, FDA's 2018 FAQ guidance says out-of-U.S. data may be used only if the sponsor explains how they are applicable to the U.S. population and U.S. medical practice ³⁹⁴. In other words, the "U.S. applicability" gate is a cross-center, long-standing feature, not an oncology novelty.

Oncology is where the requirement is being tightened

The clearest signal is FDA's draft guidance "Considerations for Generating Clinical Evidence from Oncology Multiregional Clinical Development Programs" (draft, September 16, 2024) ⁸⁷. It does three things that materially raise the bar on U.S. representation.

First, it names the trend directly. FDA writes that "in oncology MRCTs, there has been decreasing proportion of U.S. participants included in these trials; this can limit the assessment of treatment effect consistency between U.S. enrolled participants and the effect observed for the overall study population in the MRCT" ⁷⁹. FDA frames applicability as the decisive question: "the paramount consideration for FDA when evaluating such oncology trials is whether the results are applicable to the intended use population in the U.S., and to U.S. standard oncological care" ⁷⁹.

Second, it tells sponsors to design for U.S. representation rather than hope for it. FDA recommends enrolling an adequately representative subgroup of U.S. participants so that safety and efficacy can be robustly assessed relative to the overall trial population ⁸¹, and to "plan to enroll a sufficient number of U.S. participants" to support a robust assessment in U.S. patients in the context of U.S. standard of care ²⁹⁶. Sponsors should "prospectively plan the distribution of clinical sites" to achieve adequate U.S. representation ⁸¹. On allocation, FDA is now prescriptive: a strategic approach based partly on U.S. incidence or prevalence, with regions defined as major geographies (Africa, Asia, Europe, North America) rather than single countries ⁸¹. For cancers common in the U.S. such as colorectal or breast cancer, FDA recommends equal allocation across selected major regions, including North America ⁸¹; for cancers much less common in the U.S. (for example squamous cell esophageal cancer), it recommends proportional allocation ⁸¹.

Third, it closes the "single-country" pathway that had become common for programs run predominantly in one region. FDA recommends that a trial "be conducted across major geographical regions (e.g., across several continents) rather than predominantly in a single country or in a single geographical region (e.g., Asia)" ²⁹⁷, and warns that early-study data showing no observed differences in factors such as pharmacokinetics, diet, pharmacogenetics, race, age, ethnicity and sex should not be considered adequate justification to limit a trial to a single non-U.S. country or region ⁸¹. Where a sponsor nonetheless plans substantial single-region foreign enrollment, that "approach should be discussed with FDA in early clinical development, preferably prior to initiating any of the pivotal studies" ²⁹⁷, and sponsors conducting any part of the development program outside the U.S. should consult FDA early on how to obtain data with sufficient U.S. representation ³⁰⁰.

FDA also makes clear the U.S. subgroup gets special scrutiny: "the subgroup of patients enrolled in the U.S. will be of particular interest in FDA's assessment of the results of an MRCT," with pooling available when subgroup size is limited, and regional differences potentially rendering results "not applicable to the U.S. population or to U.S. medical practice" ²⁹⁹.

This sits on top of the ICH E17 framework and the statutory diversity mandate

The oncology guidance operationalizes principles already in ICH E17, "General Principles for Planning and Design of Multi-Regional Clinical Trials" (adopted 2018-07-19) ⁵¹. E17 recommends a balance between proportional and equal allocation across regions so recruitment is feasible but the drug can still be evaluated in its regional context, and it discourages arbitrary fixed regional minimums absent scientific justification ⁵³. It requires a pre-specified strategy to evaluate consistency of treatment effects across regions, defining consistency as the lack of clinically relevant differences between regions ⁵⁴, assessed with credibility considerations spanning biological plausibility, internal and external consistency, and statistical uncertainty ⁵¹. Notably, E17 treats region as a surrogate for intrinsic and extrinsic factors and offers the example that pooling Canada and the United States into a North American region is often justified given similar medical practice and concomitant medication use ⁵³. FDA's older Integrated Summary of Effectiveness guidance (final, 2015) reinforces the point: where effectiveness data come from trials both within and outside the U.S., there should be comparisons of results in U.S. and non-U.S. populations ³⁸⁷, and data should be pooled where individual studies have too few subjects to support meaningful conclusions ³⁸⁸.

Running in parallel is the statutory diversity mandate. FDORA sections 505(z) and 520(g)(9) of the FD&C Act require Diversity Action Plans for relevant drug and device studies, and FDA's draft guidance (June 26, 2024) requires enrollment goals disaggregated by race, ethnicity, sex and age, informed by U.S. disease prevalence or incidence, plus a plan to meet them ⁶⁰⁶¹. FDA's stated expectation is that "sponsors of investigational new drugs and investigational devices should enroll participants who reflect the population that will use the medical product if approved" ⁶³. The requirement attaches to studies whose enrollment commences 180 days after the final guidance publishes ⁶³. Diversity Action Plans address demographic representativeness within a population; the MRCT applicability test addresses geographic and medical-practice representativeness. Together they push in the same direction: the trial population, including its U.S. component, must look like the intended U.S. user.

The requirement is already appearing in non-approvals

The applicability principle is not theoretical. Recent CRLs cite limited U.S. or Western generalizability as a deficiency, and the pattern reaches beyond oncology:

• Atropine sulfate ophthalmic solution 0.01% (SYDNEXIS Inc., NDA 219694, CRL dated 2025-10-22): FDA wrote that most studies were of insufficient duration and "were conducted in populations (generally Asian populations) with different genetics, environments, and healthcare settings that limit generalizability to children in the United States," and that "population-specific variation raises questions about the generalizability of atropine 0.01% efficacy to diverse populations, particularly in the United States where the intended patient population includes significant non-Asian demographics" ⁴¹³³¹⁸.
• NDA 217711 (Vyluma, Inc., CRL dated 2025-08-27): FDA cited studies "conducted in populations (generally Asian populations) with different genetics, environments, and healthcare settings that limits generalizability to United States children" ³²⁵.
• NDA 208352 (Evofem, Inc., CRL dated 2016-04-28): FDA stated that "the majority of the data should come from the US population and any foreign data submitted should be generalizable to the US population" ³³³, showing the expectation predates the current guidance wave.

These decisions show FDA applying the 314.106(b) applicability test as a live approvability filter, and doing so with increasing explicitness in 2025.

How this is likely to evolve

Several trajectories are supportable from the current record:

1. Finalization and hardening. The September 2024 oncology MRCT draft guidance is likely to be finalized with its core expectations intact: prospective site planning, region-level (not single-country) allocation, equal allocation including North America for U.S.-common cancers, and early FDA engagement before pivotal trials ^81297300. Once final, its allocation logic becomes the default review yardstick.

2. Quantitative expectations negotiated early. FDA repeatedly directs sponsors to agree on U.S. representation in early development, "preferably prior to initiating any of the pivotal studies" ²⁹⁷³⁰⁰. Expect U.S. enrollment targets to become a standard end-of-Phase-2 / pre-pivotal discussion item, with the U.S. subgroup pre-specified as an analysis of interest ²⁹⁹.

3. Convergence with the diversity mandate. As Diversity Action Plans move from draft to enforced requirement ⁶⁰⁶³, demographic representativeness and geographic representativeness will be assessed together, so a numerically adequate but demographically skewed U.S. cohort will not necessarily satisfy reviewers.

4. Single-region programs become high-risk. Programs run predominantly in one country or region, historically a fast, low-cost route, now carry documented approvability risk. FDA has said absence of observed differences in early studies is not adequate justification to confine a trial to a single non-U.S. region ⁸¹, and CRLs are already penalizing limited U.S. generalizability ³¹⁸³²⁵. Sponsors relying on such data should expect to defend applicability directly or face a bridging requirement.

Bottom line for regulatory strategy

Treat adequate U.S. representation as a design input, not a post hoc argument. For oncology MRCTs, align site distribution and regional allocation to the U.S. incidence/prevalence logic in the 2024 draft guidance, secure a pre-specified and analyzable U.S. subgroup, and lock the plan with FDA before the pivotal trial ^81296297. For any program leaning on ex-U.S. data, build the 314.106(b) applicability case prospectively, addressing demographics, disease biology, and U.S. standard of care ⁸⁶²⁹⁹. The requirement is not formally new, but the evidentiary and procedural expectations around it are tightening, and the cost of ignoring them is now visible in the CRL record.

Deeper follow-ups a reader may want to ask Rhizome directly: the full text and vote record of specific ODAC meetings on single-country oncology data; a side-by-side of U.S. enrollment fractions across recently approved vs. refused oncology MRCTs; and how EMA and other ICH regulators are treating regional applicability under E17.