Retinitis pigmentosa (RP) has historically offered patients few therapeutic options beyond low vision aids, with photoreceptor loss considered largely irreversible. Among the research highlighted at the Association for Research in Vision and Ophthalmology (ARVO) 2026 annual meeting, May 3 to 7 in Denver, Colorado, two presentations reported 3-year outcome data from clinical trials evaluating distinct biological approaches to preserving or restoring visual function in RP: an optogenetic gene therapy and a neuroprotective lentiviral vector.

MCO-010 optogenetic therapy: Three-year BCVA improvements in the RESTORE/REMAIN program

Vinit B. Mahajan, MD, PhD, and colleagues reported 3-year outcomes from the REMAIN long-term follow-up study of MCO-010, an optogenetic gene therapy developed by Nanoscope Therapeutics, in patients with advanced RP.¹ MCO-010 delivers a multi-characteristic opsin (MCO) via intravitreal adeno-associated virus serotype 2 (AAV2) to target bipolar cells, aiming to restore light sensitivity by bypassing damaged photoreceptors. The approach is described by the authors as disease-agnostic, meaning it is not dependent on the specific genetic mutation underlying a patient’s RP.

RESTORE was a randomized, controlled phase 2b/3 trial in which patients with advanced RP received a single intravitreal administration of MCO-010 at high dose (1.2 × 10₁₁ gc/eye; n = 9), low dose (0.9 × 10₁₁ gc/eye; n = 9) or sham control (n = 9). Patients who completed RESTORE were eligible to continue into the REMAIN long-term follow-up study at week 100. The primary endpoint of RESTORE was best-corrected visual acuity (BCVA) change from baseline at week 52.

At week 52, mean BCVA change from baseline was 0.337 ± 0.083 LogMAR in the high-dose group (vs sham, P = .021) and 0.382 ± 0.124 LogMAR in the low-dose group (vs sham, P = .029). At week 152 of REMAIN, mean BCVA improvements from baseline were 0.264 ± 0.112 LogMAR in the high-dose group and 0.453 ± 0.140 LogMAR in the low-dose group. The authors reported that these improvements correspond to approximately 3 ETDRS lines of vision gain. No treatment-related serious adverse events were reported.

Post hoc multivariate analyses identified several baseline factors associated with greater treatment response at week 52: MCO-010 expression as measured by fundus autofluorescence (FAF) imaging, baseline central subfield thickness greater than 100 ?m, and disease duration of less than 30 years. BCVA gains correlated with retinal transduction and MCO protein expression on FAF imaging, less retinal thinning on optical coherence tomography and shorter duration of disease.

The authors described this as the first randomized controlled trial in RP to demonstrate clinically meaningful vision improvement sustained over 3 years, and noted that MCO-010 expression, baseline anatomy and disease duration may influence treatment outcomes.

DVC1-0401 neuroprotective gene therapy: 36-month safety and functional outcomes

Yusuke Murakami, MD, PhD, and colleagues reported 36-month outcomes from a phase 1/2a investigator-initiated clinical trial of DVC1-0401, a third-generation, non-replicating lentiviral vector encoding pigment epithelium-derived factor (PEDF), a neuroprotective protein expressed in the retina.² Unlike MCO-010, which targets residual bipolar cells to restore light sensitivity, DVC1-0401 aims to slow photoreceptor degeneration through neuroprotection rather than restoration of vision. The trial was conducted in Japan and was not registered with a public trial registry number in the abstract.

The single-arm, open-label, dose-escalation trial enrolled 12 patients with typical RP, regardless of genotype, with BCVA between 20/200 and 20/20. Each participant received a subretinal injection of DVC1-0401 in the worse-seeing eye at one of three dose levels (low, middle or high; n = 4 per cohort). A 12-month primary observation period was followed by a long-term follow-up phase extending to 36 months.

No serious adverse events (SAEs) related to DVC1-0401 were identified across all 12 participants. Five SAEs occurring during the primary trial period—including cataracts (n = 3), one rhegmatogenous retinal detachment and one case of persistent subretinal fluid—were attributed to surgical procedures rather than the vector. One additional SAE (macular hole) during the follow-up period and one SAE (cataract) in an untreated fellow eye were also reported.

Using linear mixed-effects models, the investigators found that treated eyes showed a significantly slower rate of decline in both BCVA and Humphrey 10-2 foveal retinal sensitivity compared with fellow eyes (P < .05). The authors noted that the reduced rate of functional decline may suggest a neuroprotective effect of DVC1-0401, while cautioning that the small sample size and single-arm design warrant validation in future pivotal trials.

The genotype-independent design of the trial—enrolling patients regardless of the causative RP mutation—is notable, as it reflects the potential for a neuroprotective approach to be applicable across the heterogeneous RP patient population without requiring mutation-specific targeting.

The road ahead

Together, the two presentations reported 3-year durability data for distinct therapeutic strategies in RP: an optogenetic approach demonstrating sustained BCVA improvement in a randomized controlled setting and a neuroprotective gene therapy showing a slower rate of functional decline compared with untreated fellow eyes. Both programs remain in early-stage development, and larger confirmatory trials will be required to establish their clinical roles.

References

1. Mahajan VB, Zak V, Bergstrom L, et al. Optogenetic REMAIN 3-year study data: lasting vision improvement with MCO-010 in the RESTORE phase 2b/3 trial for retinitis pigmentosa. ARVO 2026 Annual Meeting; May 3–7, 2026; Denver, Colorado. Presentation 2516.

2. Murakami Y, Hisai T, Mawatari G, et al. Long-term outcomes of lentivirus-based neuroprotective gene therapy for retinitis pigmentosa: a phase I/IIa investigator-initiated clinical trial of DVC1-0401. ARVO 2026 Annual Meeting; May 3–7, 2026; Denver, Colorado. Presentation 2517.