Long-Term Follow-Up Shows Safety Signal for OXB-102 Gene Therapy in Parkinson Disease

In recent months, researchers published an independent, investigator-led account of the clinical course of the 6 patients treated in SUNRISE-PD (OXB-102-01; NCT03720418), a phase 1/2 trial of the dopamine gene therapy OXB-102 (AXO-Lenti-PD) for Parkinson disease (PD).¹

Early 6-month results from the trial's first 2 participants were announced in 2019, but the study was terminated in 2022 after its sponsor, Sio Gene Therapies, became insolvent.^1-3 Because access to the original study database was lost when the company folded, the authors manually compiled outcomes from investigator records at the 2 treating sites, Addenbrooke's Hospital in Cambridge and the National Hospital for Neurology and Neurosurgery in London.¹

Trial background and design

For context, OXB-102 is a gene therapy that delivers 3 genes involved in dopamine synthesis, tyrosine hydroxylase, GTP-cyclohydrolase 1, and aromatic L-amino acid decarboxylase, via a lentiviral vector injected directly into the putamen. The agent was developed as a more potent successor to ProSavin, an earlier lentiviral gene therapy from the same research group that had shown safety and motor improvement in an open-label phase 1/2 trial.^1,4

SUNRISE-PD's part A dose-escalation phase intended to enroll up to 30 patients, but only 6 had been treated by the time of termination: patients A and B received the lowest dose (2.64 × 10⁶ transducing units) and patients C through F received a higher dose (9.0 × 10⁶ transducing units); a planned third, highest dose cohort was never enrolled.¹

Safety findings

Across the 6 patients, 8 serious adverse events (AEs) occurred in 3 patients; only 1, a surgical site infection, was considered definitely related to the intervention, while 3 others (headache, confusion, anxiety) were deemed possibly related and 4 (gastrointestinal hemorrhage, fall, delirium, psychotic depression) were considered unrelated.

All serious events were transient and resolved without lasting effects. Investigators also recorded 130 mild-to-moderate adverse events, most commonly autonomic symptoms such as constipation and postural hypotension, motor fluctuations or dyskinesias, and neuropsychiatric symptoms including worsening anxiety or suicidal ideation. Postoperative MRI confirmed correct vector delivery to the posterior putamen in all cases, and no unexplained abnormalities emerged on brain imaging, labs, or physical exam through up to 3 years of follow-up.

Efficacy signals and limitations

At 6 months, 5 of 6 patients had reduced their levodopa equivalent daily dose (LEDD) by 8% to 30%, and all 5 patients who consented to off-medication testing showed improvement on the MDS-UPDRS Part III motor score, with no apparent difference between the 2 dose groups. Durability was more mixed at 2- and 3-year follow-up: LEDD remained unchanged in 3 patients, increased in 2, and was reduced in 1 patient due to neuropsychiatric side effects, while UPDRS Part III scores showed persistent improvement in 3 patients and worsening in 2, with 1 patient not assessed.¹

The authors were explicit about the limits of what this small, uncontrolled, and prematurely halted dataset can support. "No formal statistical analysis was undertaken due to the small, open-label design and premature termination of the study, recruiting only 25% of the planned sample," the authors, led by Roger A. Barker, PhD, of the University of Cambridge, wrote.¹ They added that the trial's early end also prevented completion of planned biodistribution and PET-based target engagement studies, leaving true dose-response relationships undefined.

The authors situated the episode within a broader pattern they linked to "neuro-abandonment," in which patients enrolled in device or biotech trials are left without continuity of care or scientific follow-through when a sponsoring company fails. "This study highlights the dangers of small biotech companies undertaking trials that then become insolvent," the authors wrote, noting the phenomenon has been more widely documented in the neural device field.¹

REFERENCES
1. Rowe S, Evans A, Kayhanian S, et al. Observations on an Open-Label Phase 1/2 Dopamine Gene Therapy Trial (OXB-102/Axo-Lenti-PD) in People with Parkinson's Disease. Mov Disord. Published online May 28, 2026. doi:10.1002/mds.70378. https://movementdisorders.onlinelibrary.wiley.com/doi/10.1002/mds.70378
2. 2 Parkinson's Patients Improve on AXO-Lenti-PD in Phase 1/2 Trial. Parkinson's News Today. Accessed August 5, 2026. https://parkinsonsnewstoday.com/news/axo-lenti-pd-gene-therapy-shows-benefits-2-patients-phase-1-2-trial/
3. Bryson S. Development Halted on Parkinson's Therapy Candidate AXO-Lenti-PD. Parkinson's News Today. February 2, 2022. Accessed August 5, 2026. https://parkinsonsnewstoday.com/news/sio-gene-therapies-oxford-biomedica-therapy-candidate-axo-lenti-pd-parkinsons/
4. Palfi S, Gurruchaga JM, Ralph GS, et al. Long-term safety and tolerability of ProSavin, a lentiviral vector-based gene therapy for Parkinson's disease: a dose escalation, open-label, phase 1/2 trial. Lancet. 2014;383(9923):1138-1146. doi:10.1016/S0140-6736(13)61939-X