Inside the UK's Push to Turn Cannabis Into Real Medicine

Photo by Thirdman via Pexels.

Three cannabis-derived medicines currently hold marketing authorisation in the UK. That's it. Sativex, Epidyolex, and Nabilone -- a short list for a country that rescheduled medical cannabis back in 2018 and has spent years being cited internationally as a case study in cautious reform. What doesn't make the headlines as often is the machinery running quietly behind that short list: a set of NHS- and NIHR-backed trials at UCL, Great Ormond Street, Oxford, and Leeds that are testing cannabinoids against some of the hardest problems in medicine, from drug-resistant childhood epilepsy to glioblastoma to psychosis risk.

The flagship is an epilepsy program out of University College London and Great Ormond Street Hospital, using two formulations from Ananda Developments -- MRX2 and MRX2T -- with an eventual target of 500 patients. It's not a small pilot. It sits alongside a Dravet Syndrome drug moving toward Phase Three, a chemotherapy-combination cancer trial running across 14 hospitals, and two Oxford studies probing whether CBD can blunt the progression to psychosis. This is serious, funded, peer-reviewed science, not the kind of research you find written up in a wellness newsletter.

And yet rescheduling didn't open any floodgates. NICE guidance still keeps most cannabis-based medicines off NHS prescription pads outside narrow circumstances, and funding gaps continue to slow trials that would otherwise move faster. The story of UK cannabinoid research right now is really two stories at once: genuinely ambitious science, and a persistent lag between what that science finds and what patients can actually get.

The Epilepsy Trials Everyone's Waiting On

Photo by Gustavo Fring via Pexels.

The trial patients and parents ask about most is the one being led by Professor Finbar O'Callaghan and Professor Helen Cross, both associated with UCL and Great Ormond Street Hospital NHS Foundation Trust. It's jointly funded by NHS England and the National Institute for Health and Care Research, which matters -- this isn't industry-only money chasing a commercial outcome, it's the health service itself deciding the question is worth answering properly.

The program actually splits into two distinct trials. One targets early-onset epilepsies in children under three, a population where treatment options are thin and the stakes of getting dosing wrong are high. The other focuses on genetic generalised epilepsies (GGE) that have proven resistant to standard anti-seizure medication -- patients who've typically already tried and failed several conventional drugs before qualifying. Both trials use Ananda Developments' formulations: MRX2, a CBD-only product, and MRX2T, which combines CBD with THC.

Recruitment across both trials is expected to reach 500 patients in total, with enrolment beginning in 2025. Coverage of the program's progress has varied in its specifics -- some reporting stretches back to October 2024 -- so anyone tracking eligibility or current recruitment numbers should go directly to UCL or GOSH rather than relying on secondhand summaries, since trial timelines shift for reasons that rarely make the news.

What makes this trial worth watching isn't just its size. If the results come back positive, they're designed to support a formal submission to the MHRA and potentially other regulators, which would be a meaningfully different outcome than the current situation where clinicians can prescribe unlicensed cannabis-based products under specialist supervision but without the backing of a full marketing authorisation. That distinction -- unlicensed access versus licensed medicine -- is exactly the gap this trial is built to close.

A Related Breakthrough Nearby: Zorevunersen and the GOSH Pipeline

Photo by LEDC via Unsplash.

In March 2026, UCL and Great Ormond Street put out an announcement that, on its face, had nothing to do with cannabinoids at all. It concerned zorevunersen, a treatment for Dravet Syndrome that doesn't involve CBD or THC in any form, moving toward Phase Three trials on the strength of encouraging early results.

Galia Wilson, chair of Dravet Syndrome UK, described the results as encouraging news for a community that's spent decades watching treatment-resistant epilepsy get comparatively little dedicated drug development. That reaction is worth sitting with -- families dealing with Dravet Syndrome have often ended up as some of the most visible advocates for cannabis-derived epilepsy treatments precisely because so little else has worked, so a non-cannabinoid option advancing through trials is genuinely significant on its own terms.

What ties zorevunersen back to the cannabinoid story is infrastructure rather than chemistry. It's running through the same NIHR and GOSH clinical trial apparatus as the MRX2 and MRX2T epilepsy studies -- the same institutional review processes, some of the same clinical teams, the same hospital wards and research nurses. That overlap isn't incidental. It reflects the fact that Great Ormond Street has become something like a national hub for treatment-resistant epilepsy research broadly, regardless of whether the compound under investigation happens to be a cannabinoid or something else entirely.

For patients and families following the cannabinoid trials specifically, zorevunersen's progress is a useful reminder that the epilepsy research ecosystem in the UK is bigger than any one drug class. GOSH isn't betting everything on cannabinoids working out -- it's running a portfolio, and that portfolio approach is arguably why the institution has built the credibility to run large, well-funded trials in the first place. The cannabinoid epilepsy program benefits from sitting inside that broader machine, not from standing apart from it.

Cannabis Against Cancer: The ARISTOCRAT Glioblastoma Trial

Photo by محمد عزام الشيخ يوسف via Pexels.

Glioblastoma is about as unforgiving a cancer diagnosis as exists in oncology -- aggressive, fast-growing, and notoriously resistant to durable treatment response. It's also the target of one of the UK's more closely watched cannabinoid trials, ARISTOCRAT, led by Professor Susan Short at the University of Leeds and coordinated through the Cancer Research UK Clinical Trials Unit in Birmingham.

The trial tests nabiximols -- marketed under the brand name Sativex -- in combination with standard chemotherapy, rather than as a standalone treatment. That combination approach matters clinically: nobody involved is suggesting a cannabinoid spray replaces chemotherapy for glioblastoma, but there's a real scientific question about whether it can improve outcomes or tolerability when layered onto the existing standard of care.

ARISTOCRAT is running across 14 NHS hospitals nationwide, which is a substantial geographic footprint for a trial of this kind and speaks to how seriously it's being resourced. Enrolment has been extended through the end of April 2026, and that extension tells its own story -- recruiting glioblastoma patients into trials is genuinely difficult, given how quickly the disease progresses and how narrow the window can be between diagnosis and a patient being well enough to participate in a demanding research protocol.

There's a certain logic to Sativex specifically being the compound in this trial. It's already one of only three cannabis-derived medicines with UK marketing authorisation, originally approved for spasticity in multiple sclerosis. Repurposing an already-licensed medicine for a new indication is generally a faster, better-understood regulatory path than starting from scratch with a novel formulation, since the safety profile and manufacturing standards are already established. If ARISTOCRAT reads out positively, it would represent one of the more concrete oncology applications for a cannabis-derived medicine anywhere in the world -- not a symptom-management role, but a potential role in the treatment regimen itself.

Oxford's CBD Psychosis Studies: STEP-PROMOTE and STEP-ENHANCE

Photo by Ivan Babydov via Pexels.

Oxford's Primary Care Clinical Trials Unit is running the UK arms of two international studies looking at CBD's relationship to psychosis, and the two trials ask genuinely different questions using genuinely different timeframes -- which is worth understanding before assuming they're variations on the same experiment.

STEP-PROMOTE is the larger and longer of the two, targeting 376 participants who are at clinical high risk of psychosis -- meaning they've shown early warning signs or risk factors but haven't yet experienced a first psychotic episode. Participants are randomised to CBD or placebo and followed over 104 weeks, with recruitment running from May 2026 through November 2028. That's a two-year follow-up window on people who may or may not go on to develop psychosis at all, which gives some sense of how patient this kind of prevention research has to be.

STEP-ENHANCE asks a different question entirely. It enrolls 250 people who've already had a first episode of psychosis, testing CBD against placebo over a much shorter six-week period. The comparison here isn't about prevention -- it's about whether CBD offers any measurable benefit once psychosis has already emerged, layered alongside whatever standard treatment those patients are already receiving.

The gap in trial length -- two years versus six weeks -- isn't a design inconsistency, it's the whole point. Preventing progression to a first episode is a slow-motion question that requires watching people over time; treating an existing episode is a nearer-term question that can be answered on a shorter clock. Together the two studies represent one of the more ambitious and methodologically serious attempts yet to test CBD's antipsychotic potential, a claim that's circulated in cannabis research circles for years but has rarely been put through trials of this scale. Whatever STEP-PROMOTE and STEP-ENHANCE find, they'll carry real weight simply because so little comparably rigorous data currently exists on this specific question.

Why So Few Cannabis Medicines Actually Reach Patients

The UK currently has three cannabis-derived medicines with marketing authorisation, each approved for a distinct condition: Sativex for MS spasticity, Epidyolex for epilepsy, and Nabilone for chemotherapy-induced nausea.

Here's the number worth sitting with: three. That's how many cannabis-derived medicines currently hold UK marketing authorisation -- Sativex (nabiximols), Epidyolex (purified CBD), and Nabilone. Everything else prescribed clinically in the UK falls into the unlicensed category, available only through specialist routes and rarely covered by NHS funding.

Rescheduling medical cannabis under UK law in 2018 changed what was legally possible, not what was practically accessible. NICE, the body that determines what the NHS will actually fund and recommend, has continued to withhold recommendation for cannabis-based medicines outside a narrow set of specific circumstances. A doctor can be legally permitted to prescribe something and still have no realistic NHS pathway to do so, and that gap between legal permission and funded access is where most UK patients seeking cannabis-based treatment currently sit.

Industry voices covered in March 2026 kept returning to the same underlying issue: trial funding. Running a properly powered clinical trial -- the kind that could actually change NICE's guidance -- costs money that most cannabis companies, and even most public research bodies, are reluctant to commit without stronger preliminary signals. That's part of why the epilepsy, glioblastoma, and psychosis trials described above matter so much: they're some of the few programs with the funding and institutional backing to actually generate NICE-grade evidence.

And the evidence doesn't always cooperate. A 2026 Cochrane Review examining 21 studies covering 2,187 participants found no clear evidence that CBD-dominant medicines meaningfully relieve neuropathic pain -- one of the most commonly cited hoped-for uses for CBD in patient communities. It's a useful corrective. Serious research doesn't guarantee the answer everyone's hoping for, and neuropathic pain now joins a growing list of conditions where enthusiasm has outpaced what controlled trials actually support.

Put the pieces together and the UK's cannabinoid research infrastructure looks like something few other countries can currently match. NHS England, the NIHR, UCL, Great Ormond Street, Oxford's Primary Care Clinical Trials Unit, the University of Leeds, Cancer Research UK -- that's not a scattered handful of small studies, it's a genuinely institutional research effort with public funding and serious academic leadership behind it. Plenty of countries have looser cannabis laws. Very few have this depth of trial infrastructure asking rigorous questions about what cannabinoids actually do.

The frustration patients and advocates feel isn't really about scientific ambition falling short -- it's about time. Even a trial that reads out cleanly positive results doesn't translate into NICE guidance changes overnight; that process typically takes years, layered with its own reviews, consultations, and cost-effectiveness assessments. A family waiting on the O'Callaghan-Cross epilepsy trial results isn't just waiting for a data readout, they're waiting for everything that happens after it, and that second wait is often longer than the trial itself.

If you want two honest signals of where UK cannabinoid medicine is actually headed, rather than where the hype has pointed, keep an eye on the epilepsy trial results out of UCL and GOSH, and on how future reviews build on that 2026 Cochrane pain finding. One points toward where cannabinoids might genuinely earn a formal place in NHS treatment. The other is a reminder that not every promising idea survives contact with a properly controlled trial -- and that distinction, more than any single piece of legislation, is what will actually determine what UK patients can get prescribed five years from now.