Field Notes: APEC RHSC's Workshop On ATMP Development And Evaluation

By Eddie H.P. Tan and James W.Y. Leong, Centre of Regulatory Excellence, Duke-NUS Medical School, National University of Singapore

Advanced therapeutic medicinal products (ATMPs) represent a transformative frontier in modern medicine, offering unprecedented promise for patients with previously incurable diseases. Yet realizing the potential of ATMPs demands navigating a complex landscape of scientific, regulatory, ethical, and logistical challenges — a collection of obstacles that no single jurisdiction can address alone.

Recognizing this, the Asia Pacific Economic Cooperation Regulatory Harmonization Steering Committee (APEC RHSC) has prioritized ATMPs and biotherapeutic products as a key work area, with efforts centered on building regulatory science capacity, fostering regulatory and standards convergence, and strengthening technical competence across the region. Through its Centers of Excellence (CoE), including the Centre of Regulatory Excellence (CoRE) at Duke-NUS Medical School, the RHSC connects regulators, industry, and academia in shared dialogue.

In February, the two-day virtual RHSC CoE workshop brought together experts to exchange practical approaches, real-world experiences, and forward-looking strategies addressing the clinical development, non-clinical evaluation, trial design, and safety monitoring of ATMPs. The workshop curriculum focused on clinical and safety assessment of ATMPs, complementing previous workshops on regulatory frameworks and clinical trials. The workshop welcomed 218 registrants from nine APEC economies, including participants from national regulatory authorities, industry, and academia. The program committee, consisting of international regulatory experts from government, industry, and academic institutions, provided curriculum advice to ensure relevance.

The workshop was organized around four thematic sessions:

• ATMP Clinical product development and regulatory frameworks
• Clinical data evaluation and statistical approaches
• Safety monitoring and pharmacovigilance frameworks
• Regulatory innovations, trends, and challenges impacting access to ATMPs

ATMP Clinical Product Development And Regulatory Frameworks

Session 1 outlined ATMP clinical development opportunities and challenges, discussing how success depends on understanding mechanism of action (MoA) and building a sound strategy for clinical product development of ATMPs.

• It was highlighted that living and often personalized products depend on complex, time-sensitive manufacturing, with little to no redundancy, so batch failures can directly affect patient treatment timelines. Therefore, quality-by-design (QBD), clear target product profiles (TPPs), and well-defined critical quality attributes (CQAs) — such as viability, transduction efficiency, purity, potency, and vector safety — are essential.
• It was also emphasized that operational trial risks span chain of identity (COI), chain of custody (COC), cold chain, cross-border logistics, site capability, and long-term follow-up.
• Regulatory talks presented the current, fragmented state of APAC regulatory pathways, the need for growing regulatory reliance/harmonization initiatives, and an introduction to the WHO’s risk-based approach towards the development of global regulatory framework to support safe, scalable access.

For ATMP developers, the landscape is shifting from one-off problem-solving to building repeatable playbooks. Companies are starting to embed quality by design (QbD), lifecycle thinking, and fit-for-purpose analytics into their ATMP programs from day one, rather than retrofitting control late in development. In parallel, APAC regulatory agencies and global bodies (e.g., WHO, ICH, PIC/S) are gradually turning down the regulatory friction through clearer frameworks, convergence, and reliance tools. Sponsors who engage early and strategically have a window to move faster across multiple markets without compromising quality or safety.

Clinical Data Evaluation And Statistical Approaches

Session 2 highlighted why ATMP evidence evaluation should utilize flexible, risk-based methods across non-clinical, clinical, and statistical domains.

• First, cell/tissue/gene therapies have complex, long-lasting biological effects and often lack pharmacologically relevant animal models, so regulators increasingly prioritize mechanism-based studies, biodistribution/persistence assessments (including germline risk rationale), and scientifically justified waivers aligned to clinical risk management.
• Next, two speakers from the European regulatory authorities reviewed ATMP trial barriers and operational concerns, suggesting opportunities for regulatory science progression: patient-only early phases, ethical limits on placebo, dependence on single-arm or seamless designs, and long-term follow-up with structured benefit/risk assessment and conditional approvals when data are incomplete.
• The speakers also touched on statistical challenges of ATMP trials, including dose finding, defining analysis populations and estimands, interpreting single-arm thresholds, and cautious use of external/RWE and Bayesian methods.

Across the ATMP clinical development stages, regulators are gradually adopting more flexible tools in their evaluation tool kit. These tools include seamless phases, master protocols, historical controls, conditional approvals, and structured post‑authorization follow‑up. As a result, developers can bring high‑impact therapies forward despite small patient population sizes and ethical constraints on randomized trials.

For industry, the goal used to be running a perfect Phase 3, like in traditional biologics. For ATMPs, it has become a multi-factorial consideration of designing smart, risk‑proportionate trials. In this new era, you must plan for long-term data from the outset and use early scientific advice to align your evidence package with evolving expectations on N‑of‑1, gene editing, and ultra‑rare indications.

Safety Monitoring And Pharmacovigilance Frameworks

Session 3 examined how authorities in the region, including Indonesia and Thailand, are building ATMP‑specific pharmacovigilance (PV) systems to manage delayed, complex risks after approval.

• Indonesia’s BPOM described a tiered legal framework in which ATMPs are defined by more-than-minimal manipulation and/or non‑homologous use, with mandatory risk management plans (RMPs), long‑term follow‑up, inspections, outsourcing controls, and strong donor–product–patient traceability; hospitals can also act as marketing authorization holders.
• Thailand FDA outlined a lifecycle approach starting at IND stage, now anchored in an internationally harmonized RMP model with tailored risk minimization (education and controlled distribution to trained centers) and long‑term follow‑up expectations (often up to 15 years).
• The session concluded with an industry perspective that highlighted practical barriers to safety monitoring and pharmacovigilance, including fragmented global requirements, patient/site retention over five to 15 years, data burden, and the need to leverage registries, real‑world data, and digital tools.

The global approach to ATMP safety is evolving from product‑specific, prescriptive controls toward structured, lifecycle‑based pharmacovigilance architectures that integrate mandatory long‑term follow‑up, formalized RMPs and, increasingly, real‑world data. Registries and digital tools are treated as core regulatory evidence streams rather than optional add‑ons. For industry, this signals that sustained market presence for ATMPs will hinge less on satisfying procedural checklists and more on demonstrating mature safety governance: harmonized global RMP strategies, interoperable registry and data infrastructures, and calibrated risk-minimization measures that can be credibly de-escalated as cumulative post‑marketing evidence strengthens the benefit-risk profile.

Regulatory Innovations, Trends, And Challenges Impacting Access To ATMPs.

Session 4 showcased regulatory innovations that can be incorporated to speed ATMP access while maintaining safeguards to patient health.

• The concept of regulatory reliance was introduced to participants as a way of using trusted authorities’ assessments to reduce duplication, strengthen efficiency, and focus local review on context-specific risks. The concept of unilateral versus mutual reliance was more deeply explained, as was the importance of maintaining sovereignty and accountability within the relying authority.
• Workshop participants explored how artificial intelligence (AI) is increasingly used in ATMP submissions (e.g., manufacturing control, biomarker/endpoints, trial design, and safety monitoring) and why regulators emphasize context of use and credibility evidence, plus lifecycle management of models.
• Real‑world evidence (RWE) was presented as an essential component for considerations related to trial design, external comparators, and long-term effectiveness/safety and reimbursement.
• The EU Substances of Human Origin (SoHO) perspective highlighted the importance of leveraging existing donor/traceability systems, risk-based oversight, and hospital exemption/decentralized manufacturing to support equitable access of ATMPs.

Reliance, AI, and RWE are moving from nice-to-have concepts to practical levers that regulators and sponsors are beginning to operationalize for ATMPs, thereby shifting the task from building everything alone, from scratch, toward credible ways of borrowing, adapting, and stress‑testing evidence generated elsewhere.

In the context of reliance, unilateral reliance (e.g., CoGenT and Project Orbis, where smaller agencies draw on the assessments of a lead agency) is already being used to shorten timelines and reduce duplication. More ambitious mutual reliance and work‑sharing models are emerging in regional clusters (e.g. ACCESS Consortium), offering sponsors the prospect of coordinated reviews and more efficient multi‑market launches if their dossiers and post‑approval plans are aligned with those templates. This signifies a shift from the regulators asking: “Who did the original assessment?” toward questions like, “Is this evidence fit for our context of use, and under what conditions can we responsibly rely on it?”

For industry, the key is to design evidence that travels: AI tools, RWE packages, and local clinical data will be judged less by their novelty than by how clearly their context of use is defined and how robust their credibility evidence is — for example, whether the data, methods, and performance are demonstrably fit for the specific regulatory decision they are intended to support.

Strategically, ATMP developers who align trial designs, registries, and long‑term safety programs to common data standards and reliance templates will be best positioned to convert national approvals into regional portfolios, turning today’s fragmented requirements into an integrated, lifecycle evidence strategy rather than a series of one‑off, country‑by‑country negotiations.

Key Takeaways From The APEC-RHSC-CoE Workshop:

• Unique biological complexity of ATMPS: ATMPs are often living drugs or personalized batches-of-one, requiring rigorous COI and COC protocols.
• Importance of flexible regulatory pathways: Regulators are increasingly adopting risk-based lifecycle approaches, utilizing conditional marketing authorizations (CMA) and regulatory reliance to help accelerate and broaden patient access without compromising safety.
• Innovations around clinical trial design and RWE: Traditional randomized controlled trials (RCTs) are often unfeasible due to small patient populations, ethical concerns, and other limitations, leading to an increased use of single-arm trials, master protocols, and RWE.
• Requirement for long-term oversight for ATMPs is here to stay: Given the potential for delayed risks, such as insertional mutagenesis, long-term follow-up of up to 15 years is becoming a global standard for integrating vectors, reflecting a growing global consensus around vigilance for delayed risks.
• Regulators are integrating technological tools into their workflows to become more efficient: AI and machine learning (ML) are being integrated into manufacturing, personalization, and safety monitoring, necessitating new regulatory frameworks to assess model credibility.

Assessed together, these takeaways reinforce that advancing ATMPs requires sustained international collaboration and continued investment in regulatory science capacity.

About The Authors:

Eddie H.P. Tan is an assistant professor at the Centre of Regulatory Excellence (CoRE) at Duke-NUS Medical School, National University of Singapore (NUS). He co-leads a collaborative project with the Ministry of Health and Health Sciences Authority in Singapore, to study regulatory innovation of emerging healthcare technologies. He has more than 10 years of experience in process and analytical development, validation and qualification and quality management systems for cell and gene therapy clinical manufacturing.

James W.Y. Leong is an assistant professor at the Centre of Regulatory Excellence (CoRE) at Duke-NUS Medical School, National University of Singapore (NUS). He is the head of health products and regulatory science at CoRE and leads the identification of educational needs for the various stakeholders involved in regulatory affairs in the Asia Pacific region, and establishing education roadmaps, priorities and deliverables.