MHRA opens a risk-based route for non-OECD MAD non-clinical data in UK trial applications

What happened

On 4 November 2026 the MHRA published guidance on using non-clinical data generated outside the OECD Mutual Acceptance of Data (MAD) system to support UK clinical trial authorisation applications. According to the agency's summary, the guidance describes a risk-proportionate framework for applications that rely on such data.

The published summary is short, and this article does not go beyond it. It does not say which studies are covered, what evidence the MHRA will ask for, or how risk will be tiered. Those details sit in the guidance itself, and teams should read it in full before changing any submission plan.

Why OECD MAD matters to trial applications

Under OECD MAD, safety data generated in a participating country's test facilities, following OECD test guidelines and Good Laboratory Practice principles, should be accepted by other participating countries. In practice, this spares sponsors from repeating animal studies just to satisfy a second regulator.

Non-clinical packages do not always fit that model. Studies may be run at facilities outside the system, or by contract labs whose GLP status cannot be shown through MAD. Before a first-in-human or early-phase trial, the regulator still has to be confident that the toxicology and pharmacology data support the proposed dose, duration and population.

A framework that is explicitly risk-proportionate suggests the MHRA intends to assess such data case by case, with scrutiny scaled to the risk of the proposed trial and the data. That is an inference from the wording of the summary, not a confirmed description of the process.

What we do not yet know

The records do not say how the framework will affect review timelines, whether extra documentation such as GLP statements or audit evidence will be expected, or whether certain trial types will face a higher bar. They also do not say how the approach interacts with scientific advice or with the combined review process for UK trial applications.

It is also not yet known how the MHRA will treat data from facilities with no inspection history. Until the full guidance is parsed, sponsors should avoid assuming that non-MAD data will be accepted simply because a framework now exists.

Who is affected

The most directly affected are sponsors whose non-clinical programmes were run partly or wholly outside OECD MAD. That includes smaller developers using lower-cost contract research capacity, and companies moving assets between regions. Regulatory affairs and non-clinical safety teams will need to reconcile where each pivotal study was done with what the UK now expects.

The wider pipeline shows why UK positioning matters to globally minded developers. In the same news cycle, uniQure reported that the MHRA had validated a marketing authorisation application for its Huntington's disease gene therapy, in parallel with an FDA priority review. Nuvectis cited a Chinese approval of its oral Factor B inhibitor for PNH, with US discussions continuing. Neither record says where the underlying non-clinical work was done, or whether non-OECD MAD data are involved. They are context for how multi-region development works, not examples of the new framework in use.

Other records from the same day, including FDA action dates for Summit's ivonescimab (14 November 2026) and Mineralys's lorundrostat (22 December 2026), underline how crowded the US calendar is. For sponsors weighing where to start clinical work, a predictable UK route for trial authorisation is one more variable. These records say nothing about UK trial applications, so they carry no implication for this guidance.

What to watch

The first practical signals will be how the MHRA handles early applications under the framework, and whether it publishes worked examples, checklists or clarifications. Teams should also watch for any related updates to the agency's trial authorisation guidance, and for feedback from industry bodies on how workable the risk criteria are.

In the meantime, the sensible step is a gap assessment of existing and planned non-clinical studies against the new document. That means identifying which studies fall outside OECD MAD, what quality evidence can be supplied for them, and whether a bridging or confirmatory study would still be the safer course.