Every gene therapy program moving toward global approval eventually has to reconcile two regulatory frameworks that share the same scientific foundation but diverge in structure, emphasis, and specificity. The FDA works through a family of stage-specific guidance documents; the EMA works from a single comprehensive guideline written for the marketing application. Neither framework is more rigorous across the board — each sets the higher bar in different areas — which means the practical question for any dual-jurisdiction program is not which agency to satisfy first, but where the two frameworks actually diverge and what a single program has to be built to in order to clear both.
Organizations pursuing gene therapy approvals in both the United States and the European Union face a regulatory landscape that is similar enough to be deceptive and different enough to be costly if the differences are discovered late. The FDA and EMA share the same foundational scientific principles — product quality must be demonstrated, manufacturing must be controlled, nonclinical safety must be established before human exposure, and long-term follow-up is required for products that permanently modify human cells. But the two agencies organize those principles differently, apply them at different stages of development, and in several areas set meaningfully different expectations about what must be demonstrated and when.
The organizations that navigate this landscape most efficiently are not those who build one program for the FDA and a second for the EMA. They are those who understand where the two frameworks diverge, identify the more conservative standard in each area, and design a single program capable of satisfying both simultaneously. This is not a theoretical efficiency. It is a practical strategy that reduces late-stage rework, eliminates duplicative studies, and produces regulatory submissions in both jurisdictions grounded in the same data.
Understanding the Two Documents
The foundational EMA guidance for gene therapy is the Guideline on the Quality, Non-Clinical and Clinical Aspects of Gene Therapy Medicinal Products, adopted by the Committee for Advanced Therapies in 2018. It is a broad, 46-page document covering all gene therapy medicinal products — viral vectors, plasmid DNA, bacterial vectors — across quality, nonclinical, and clinical development, written for Marketing Authorisation Applications (MAA).
The FDA does not have a single equivalent document. Instead, it has built a family of guidance documents organized by product type and development stage: the Gene Therapy CMC Guidance for IND-stage manufacturing, the CAR-T Cell Product Development Guidance finalized in January 2024, the Genome Editing Guidance also finalized in January 2024, the Long-Term Follow-Up Guidance, and others. Together these documents cover similar ground to the EMA guideline, but they are more operationally specific, more product-focused, and more explicit about what is expected at each development stage from IND through BLA.
This structural difference is the first thing any organization needs to understand. The EMA guideline is primarily an MAA-level document — it specifies what the marketing application must contain. The FDA guidance family is stage-specific — it tells sponsors what is expected at IND, what can be developed progressively, and what must be in place before BLA submission. For early-stage programs, the FDA’s staged expectations are more immediately useful as a development guide. For late-stage programs building toward simultaneous global filings, the EMA’s comprehensive MAA framework becomes the organizing structure.
The table below summarizes the most operationally significant differences between the two frameworks. Gray shading indicates the more conservative or restrictive standard in each area — the standard that a dual-jurisdiction program should be designed to meet.
FDA vs. EMA: Key Differences at a Glance
| Topic | FDA | EMA | More Conservative Standard |
|---|---|---|---|
| Document structure | Family of product-specific, stage-specific guidance documents (IND through BLA) | Single comprehensive guideline covering all GTMPs at MAA level | Comparable — different architecture, not directly comparable |
| Vector design documentation | Addressed across multiple guidance documents; no single consolidated checklist | Enumerated checklist of nine required design considerations in Section 4.1.1 | EMA |
| Genome sequencing of viral vectors | Sequence integrity verification; whole-genome NGS increasingly applied | Entire genome must be sequenced at production batch level unless justified (Section 4.1.2) | EMA |
| Full/empty capsid characterization | Detailed analytical specification: AUC or equivalent, lot release criteria for full capsid %, patient safety framing | Particle-to-infectivity ratio required on harvests with minimum acceptable titres (Section 4.2.1.1) | Both — FDA is more analytically specific; both require characterization and control |
| Shedding studies | Addressed in separate 2015 Shedding Guidance; treated as standalone safety assessment | Integrated as clinical PK requirement in Section 6.2.1 with defined collection, timepoints, monitoring duration | EMA — its integration into the clinical PK section sets a higher design standard |
| Risk Management Plan | No formal RMP required; REMS required only if FDA determines necessary | Formal RMP required as component of MAA (Section 6.8); includes pharmacovigilance, risk minimization measures | EMA |
| Long-term follow-up | 15-year follow-up recommended (LTFU Guidance); specific protocol required in IND | Long-term follow-up required; duration determined by product risk profile; pharmacovigilance plan required | FDA — its 15-year explicit requirement is more specific |
| Genome editing off-target characterization | Genome-wide empirical methods required; multi-donor evaluation for ex vivo; chromosomal integrity assessment (Jan 2024 Guidance) | 2018 guideline predates CRISPR clinical maturity; supplemented by 2020 reflection paper; less operationally specified | FDA — its 2024 standard is substantially more developed |
| Genetically modified cells (CAR-T, ex vivo) | Dedicated CAR-T Guidance (Jan 2024): VCN per CAR+ cell, chain of identity, comparability study design, cryopreservation recommendation | Covered under separate guideline (EMA/CAT/GTWP/671639/2008); less operationally specific for CAR-T | FDA — its 2024 CAR-T guidance is more prescriptive |
| Raw materials and adventitious agents | Addressed in GT CMC Guidance; risk-based approach to raw material qualification | More detailed treatment of biological raw materials, TSE/BSE risk, regulatory gap in raw material frameworks (Sections 4.2.2, 4.7) | EMA |
| Comparability studies | Statistical analysis with pre-specified criteria required; same starting material requirement for cell therapy (CAR-T Guidance) | References ICH Q5E; gene-therapy-specific elaboration less detailed | FDA — its statistical and same-starting-material requirements are more specific |
| IND/CTA stage expectations | Explicit staged requirements by phase; defines what can be deferred to later development | MAA-focused; separate investigational ATMP guidance referenced but less developed | Comparable — FDA is more useful for early-stage planning; EMA is better for MAA preparation |
Where the Frameworks Diverge: What It Means in Practice
Vector design documentation. The EMA’s Section 4.1.1 provides an explicit enumerated checklist for vector design considerations that must be documented in the MAA: pathogenicity and virulence of the parental organism, minimization of non-essential accessory components, packaging cell line sequence homology, RCV risk minimization, tissue tropism, transduction efficiency in dividing versus non-dividing target cells, persistence of viral sequences relevant to antiviral therapy, tissue specificity of replication, and germline transmission. The FDA addresses these topics across multiple guidance documents without a single comparable checklist. The EMA standard is more systematically documented. Build to it.
Full capsid characterization. The EMA’s manufacturing guidance requires determination of titre and particle-to-infectivity ratio on harvests, with minimum acceptable titres established. The FDA has developed this requirement in more analytical detail in recent guidance — specifying AUC or equivalent methods for full/empty capsid separation, lot release criteria for full capsid percentage, and the connection between empty capsid content and total patient particle burden. Both agencies converge on the same underlying requirement: the full/empty capsid ratio must be measured, characterized, and controlled. The FDA’s analytical specification is more developed; the EMA’s lot release framing is simpler but equally demanding. Build the full capsid characterization method early — before IND if possible — and establish the lot release specification before Phase 3.
Shedding studies. The EMA integrates shedding studies as a clinical pharmacokinetic requirement in Section 6.2.1 of its main guideline, treating them as part of the clinical PK characterization package with defined collection timepoints, sample types, monitoring duration, and reporting structure. The FDA addressed shedding in a separate 2015 guidance document and does not integrate it as prominently into the primary clinical development framework. The EMA’s treatment sets a more systematic expectation for shedding study design. Designing the shedding study protocol to satisfy the EMA’s clinical PK integration — defined collection, defined monitoring window, results reported within the PK section — produces data that satisfies both agencies and avoids the reformatting and potential data gap issues that arise when shedding is treated as a standalone safety test rather than a PK study.
Risk Management Plan. The EMA requires a formal Risk Management Plan as a component of the MAA under Section 6.8 of the guideline. The RMP is a structured document identifying the product’s safety concerns, the pharmacovigilance activities planned to monitor them, and the risk minimization measures to be implemented — including long-term follow-up protocols, patient registries, and restricted distribution systems appropriate to products with permanent genetic effects. The FDA does not require an RMP in the same explicit structural form, though the BLA must include a Risk Evaluation and Mitigation Strategy if the agency determines one is needed. The EMA requirement is the more conservative standard. Building an RMP-equivalent document early — even before it is formally required by either agency — as a living document that grows with the program, serves both regulatory purposes and the internal analytical function of forcing systematic identification of the product’s safety profile and monitoring strategy.
Genome editing coverage. The 2018 EMA guideline predates the clinical maturity of CRISPR-based approaches and provides limited guidance on off-target characterization methodology, therapeutic editing thresholds, or the ex vivo versus in vivo regulatory classification. The FDA’s January 2024 Genome Editing Guidance fills this gap in detail: genome-wide off-target analysis using empirical methods, chromosomal integrity assessment, multi-donor characterization for ex vivo products, and potency assay requirements that measure downstream biological effect rather than editing efficiency alone. For genome editing programs filing in both jurisdictions, the FDA’s 2024 standard is more developed and more demanding. Build the off-target characterization program to the FDA’s 2024 standard — genome-wide empirical methods, multi-donor evaluation, chromosomal integrity assessment — and the EMA will have no basis for requesting additional characterization. Build to a lower standard and the FDA review will identify the gap.
The Conservative Standard Strategy
The operational principle that resolves most FDA/EMA divergence questions is straightforward: in any area where the two agencies set different standards, design to the more conservative one. This is not overcaution. It is efficiency. The cost of designing to the higher standard at program inception is almost always lower than the cost of upgrading a program when the more demanding jurisdiction identifies the gap.
Applied systematically, this principle produces a program where manufacturing documentation meets the EMA’s MAA-level vector design and development genetics requirements from the beginning of process development, not assembled retrospectively at submission. Analytical methods are developed and qualified on an IND-stage timeline to support both early submissions and the eventual MAA/BLA, avoiding the late-stage analytical build that delays global submissions. Shedding studies are integrated into Phase 1 or Phase 2 clinical protocols as PK studies, not added as standalone studies after the primary clinical dataset is complete. The RMP equivalent is initiated early and maintained as a living document that is ready when the EMA submission requires it, without the cost of building it from scratch under MAA preparation timelines.
Clinical Trial Design as a Simultaneous Filing Enabler
The clinical trial design decisions that create the most regulatory friction for simultaneous US/EU submissions are those made to optimize for one agency’s expectations at the expense of the other’s.
On endpoints: the EMA’s guidance emphasizes that the same principles applying to conventional medicinal products in specific therapeutic areas apply equally to gene therapies in those areas — meaning disease-area endpoint guidance matters as much as gene therapy-specific guidance. The FDA has moved increasingly toward patient-reported outcomes and functional endpoints in rare disease gene therapy programs, supported by accelerated approval pathways that allow surrogate endpoints when clinical benefit is anticipated. Designing the efficacy endpoint strategy to satisfy both agencies requires understanding the therapeutic area endpoint guidance in both jurisdictions and identifying endpoints acceptable in both markets. This analysis belongs at protocol design, not after the Phase 3 data is collected.
On comparators: for indications with existing approved therapies, the EMA’s efficacy requirements may be more demanding about comparative evidence than the FDA’s rare disease accelerated approval framework. This is an area where the more conservative standard may require a study design element — a concurrent observational comparator or an additional cohort — that satisfies EMA comparative efficacy expectations without undermining the statistical power of the primary FDA endpoint analysis. These decisions must be made before the Phase 3 protocol is finalized.
On study population eligibility: for products with immunogenicity concerns, both agencies require pre-specified exclusion criteria based on serological testing with analytically validated assays. For autologous products with manufacturing failure risk, both the FDA’s CAR-T guidance and the EMA’s expectations require the protocol to address contingency planning for manufacturing failure — what the sponsor does for the patient, whether re-manufacturing is attempted, how bridging therapy affects the analysis. Aligning eligibility criteria and contingency plans across both frameworks before enrollment begins prevents the situation where a patient population eligible under one agency’s criteria is partially ineligible under the other’s.
The Practical Starting Point
For programs that are early in design, the most efficient entry point for a dual-jurisdiction strategy is a systematic gap analysis against both the EMA guideline and the relevant FDA guidance documents for the product type. Map each development activity — vector design documentation, manufacturing process development, analytical method development, nonclinical study design, clinical trial design, long-term follow-up planning, risk management — against the requirements of both agencies simultaneously. Where the two frameworks converge, one activity satisfies both. Where they diverge, the more conservative standard sets the program requirement.
This analysis is most valuable before significant development investment has been committed — before the vector design is locked, the manufacturing process established, or the nonclinical study plan finalized. The earlier the dual-jurisdiction requirements are mapped, the lower the cost of designing to them.
For programs that are mid-development and have built significant history under one agency’s framework, the gap analysis identifies what is missing relative to the other agency’s requirements and produces a prioritized remediation plan. Not all gaps are equal: some require new studies, some require additional analytical characterization of existing samples, and some require documentation of work that was conducted but not recorded in the form the second agency expects.
The gene therapy regulatory landscape reflects two agencies navigating the same underlying scientific uncertainty about a class of products that are more variable, more biologically complex, and more long-term in their effects than any therapeutic category that preceded them. Their guidance documents represent the current best effort to define standards in a field that is still generating the data those standards should eventually be based on. Understanding both frameworks — and knowing where each sets the higher bar — is the analytical foundation of a global regulatory strategy that does not have to be rebuilt at each jurisdiction’s request.
MKA Insights brings cross-market regulatory intelligence and strategic perspective to life sciences organizations navigating complex global development landscapes. If you are evaluating a gene therapy program’s regulatory pathway or assessing readiness for multi-jurisdiction submissions, we provide the analytical framework and market understanding that inform better strategic decisions.