One Patient, One File: Britain's Regulatory Route for Bespoke CRISPR
Photo by Photo by Igor Sporynin on Unsplash on Unsplash
A CRISPR therapy built for one person poses an awkward question for any regulator: what counts as evidence when the cohort is a single named patient? For decades, UK medicines regulation has been organised around populations. Trials recruit groups, statistics compare arms, and licensing assumes that what works for many can be prescribed for the next patient who fits a label. Bespoke gene editing unsettles that logic. In 2026, the debate has moved from whether single-patient CRISPR therapies are imaginable to how Britain might assess them without abandoning the safeguards that make clinical research trustworthy.
The UK is not starting from zero. The Medicines and Healthcare products Regulatory Agency (MHRA) already handles advanced therapies, rare-disease programmes and unlicensed medicines. The NHS has long experience of named-patient arrangements, and the Health Research Authority oversees a research ethics system accustomed to difficult risk-benefit judgments. What is missing is a clear, public account of how those pieces fit together when the edit, the delivery strategy and the manufacturing batch are all tied to one individual.
Personalised is not the same as bespoke
Much of what is called personalised medicine is still population medicine with better sorting. A biomarker identifies who is likely to respond; the medicine itself remains a standardised product. A bespoke CRISPR therapy is different. The editing strategy may be designed around a particular variant, the guide RNA may be patient-specific, and the manufacturing run may serve exactly one person. That distinction matters for regulators because it changes what can be tested before use and what can be learned afterwards.
It also creates a labelling problem. A single-patient therapy is not simply an unlicensed medicine supplied under a familiar exemption. Nor is it a conventional clinical trial with a sample size of one appended for appearance. If Britain blurs those categories, patients may gain faster access at the cost of weaker evidence; if it insists on population-style proof, the therapy may never reach the bedside. The regulatory task is to define a middle route with explicit limits.
What the MHRA can actually assess
The MHRA's central question will be familiar: do the benefits justify the risks? In an n-of-1 CRISPR study, that judgment cannot rest on statistical power. It will depend on the strength of the biological rationale, the quality of preclinical work, the specificity of the edit, the likelihood of off-target effects, and the plausibility of delivery to the right tissue. The regulator can also ask for prospective monitoring, independent oversight and pre-specified stopping rules. Those are not decorative additions. They are the substitute for the protections that randomisation and large numbers normally provide.
A credible single-patient pathway would need to distinguish between platform data and patient-specific data. If a delivery system or editing platform has already been tested, regulators may be able to rely on that accumulated evidence while scrutinising the novel components more closely. But each new edit can create new risks, and each patient may have a different underlying condition. The MHRA will therefore need expert advice that is both agile and transparent. A closed expert committee may be appropriate for commercial confidentiality, but the reasoning behind approval or refusal should be publishable in some form.
Manufacturing a batch of one
Bespoke therapy turns manufacturing into a regulatory issue at the smallest possible scale. Batch release, sterility, identity and potency testing still matter when the batch serves one patient. Yet comparability exercises of the kind used for conventional products are largely unavailable. Regulators may need to accept modular approaches: a common platform with patient-specific elements, supported by validated controls and a clear account of what changes from case to case.
There is also a risk of confusion with hospital exemption or named-patient supply. Those routes can be legitimate for unlicensed advanced therapies prepared on a non-routine basis, but they are not designed to generate generalisable evidence. If a bespoke CRISPR therapy is supplied under an exemption, the patient may be treated but the wider system may learn little. A formal n-of-1 trial, by contrast, should produce data that can be reviewed, published and used to improve future decisions, even when the result is negative.
Consent, ethics and the NHS
Research ethics committees will face a familiar set of questions in an unfamiliar setting. Can a patient give meaningful consent when the evidence base is thin? How can confidentiality be protected when the patient may be identifiable from the condition, the edit and the clinical team? What happens if the therapy fails, or if the trial stops early? These are not reasons to refuse the work. They are reasons to design it carefully, with independent monitoring and a clear aftercare plan.
The NHS adds a further layer. A bespoke therapy may be authorised for research but not commissioned for routine care. Clinicians and patients will want to know who pays for manufacture, administration and follow-up, and whether access depends on postcode or on the willingness of a specialist centre to take on financial risk. Devolved health systems add another dimension: the MHRA operates across the UK, but NHS commissioning and hospital practice differ in England, Scotland, Wales and Northern Ireland. A single-patient pathway should not become a single-nation pathway by accident.
A pathway, not a precedent
The strongest argument for a bespoke CRISPR route is not that every rare disease deserves a custom edit. It is that a small number of patients may have no satisfactory authorised alternative, a strong mechanistic case and a measurable clinical need. Britain can regulate such cases without embracing a general right to experimental treatment. The criteria should be narrow, the oversight independent, the consent rigorous and the results public.
A time-limited pathway, reviewed after a set number of cases, would allow the MHRA, NHS and ethics system to learn without making irreversible promises. It would also give patients and clinicians a realistic account of what a trial of one can and cannot prove. The trial of one is ultimately a test of governance. If Britain can regulate it transparently, the country will have shown that precision medicine can be precise about its own limits.
