REGENXBIO Inc. (Nasdaq: RGNX) reported on July 18 that investigational surabgene lomparvovec, also known as sura-vec or ABBV-RGX-314, produced encouraging long-term findings in wet age-related macular degeneration and diabetic retinopathy. The company presented the results at the 44th Annual Meeting of the American Society of Retina Specialists in Montreal, including five-year follow-up from a Phase I/IIa wet AMD study and 2.5-year interim findings from the Phase II ALTITUDE diabetic retinopathy trial.
The new data strengthen the durability argument behind sura-vec, which is being developed with AbbVie Inc. as a potential one-time retinal gene therapy. They do not, however, establish regulatory-level efficacy. The wet AMD findings come from small early-stage cohorts, while the diabetic retinopathy analysis involves a limited number of evaluable patients and no continuing randomized control group at 2.5 years.
That distinction matters because the programme is now approaching a much more consequential test. More than 1,200 participants have been enrolled across the randomized ATMOSPHERE and ASCENT wet AMD pivotal studies, with topline results expected during the fourth quarter of 2026. Those trials, rather than the supportive long-term datasets presented in Montreal, will determine whether sura-vec can preserve vision against active anti-vascular endothelial growth factor therapy while materially reducing injection burden.
What do the five-year wet AMD results establish about sura-vec’s treatment durability?
Surabgene lomparvovec uses an adeno-associated virus 8 vector to deliver genetic instructions that enable retinal cells to produce an antibody fragment designed to inhibit vascular endothelial growth factor. The strategy seeks to provide sustained anti-vascular endothelial growth factor activity inside the eye after a single administration, instead of relying entirely on repeated intravitreal injections.
The Phase I/IIa wet AMD study originally treated 42 patients across five dose cohorts. Thirty-seven entered long-term follow-up, including six patients from Cohort 3 and 12 from Cohort 4. Those two cohorts received doses similar to the doses being studied in the pivotal ATMOSPHERE and ASCENT trials.
These were not treatment-naive patients with short disease histories. Participants were required to have received at least four anti-vascular endothelial growth factor injections during the eight months before entering the study, and the overall population had received an average of approximately nine injections during the preceding year. That makes sustained reduction in treatment burden particularly relevant, although it also makes the study less representative of newly diagnosed patients.
REGENXBIO reported stable to improved visual acuity through five years in the pivotal-dose cohorts, apart from one Cohort 4 participant with polypoidal choroidal vasculopathy that was refractory to anti-vascular endothelial growth factor treatment. Cohorts 3 and 4 achieved reductions of more than 64% in annualized supplemental injection burden during the fifth year compared with the year before sura-vec administration.
The finding is clinically interesting because it suggests that gene-mediated protein expression may persist for several years. A treatment that reduces the need for injections without sacrificing visual acuity could address a central weakness of chronic retinal care, where missed appointments and undertreatment can gradually erode outcomes.
Durability alone is not sufficient, however. The pivotal programme must show that the reduction in injections is accompanied by vision preservation, anatomical disease control and an acceptable long-term safety profile in a much larger and more diverse population.
Why does the small early-stage sample leave the pivotal wet AMD readout as the decisive test?
The five-year follow-up provides valuable evidence of persistence, but the underlying sample is too small to establish how reliably the outcome would reproduce across routine clinical populations. Early-phase studies can also be influenced by patient selection, investigator discretion over supplemental injections, attrition and the absence of a concurrent randomized comparator during long-term follow-up.
REGENXBIO also presented an external-control analysis comparing sura-vec recipients with matched patients receiving real-world intravitreal therapy. At one year, the company reported that sura-vec participants preserved vision, experienced less fluctuation in central retinal thickness and received significantly fewer injections than the external-control patients.
External matching can add context, particularly when long-term randomized controls are unavailable, but it cannot eliminate every difference between trial participants and real-world patients. Treatment adherence, disease severity, imaging schedules, rescue-treatment decisions and unmeasured clinical factors can influence the comparison. The analysis therefore supports the development hypothesis without replacing randomized evidence.
ATMOSPHERE and ASCENT are designed to provide that evidence. ATMOSPHERE compares subretinal sura-vec with ranibizumab, while ASCENT compares the gene therapy with aflibercept. Their primary endpoints assess non-inferiority based on the change from baseline in best-corrected visual acuity at 54 weeks and one year, respectively.
Secondary measures include retinal thickness, safety, tolerability and supplemental injection use. If sura-vec preserves vision within the prespecified non-inferiority margins while substantially reducing rescue injections, the commercial and regulatory case would become much stronger. Failure to protect visual acuity would be difficult to offset with injection reduction alone.
How meaningful are the 2.5-year ALTITUDE results for diabetic retinopathy development?
The diabetic retinopathy programme uses suprachoroidal rather than subretinal administration. This distinction is strategically important because suprachoroidal administration is intended as an in-office procedure, potentially providing a more scalable route than the surgical subretinal delivery being evaluated in wet AMD.
In the Phase II ALTITUDE trial, the pivotal dose was 1.0 × 10¹² genome copies per eye, accompanied by a short course of prophylactic topical steroids. The latest efficacy analysis included 11 evaluable participants at that dose after 2.5 years.
REGENXBIO reported that 55% achieved an improvement of at least two steps on the Diabetic Retinopathy Severity Scale without additional diabetic retinopathy treatment. Approximately 70% experienced no vision-threatening event during the disclosed period. The company also observed that most participants who had improved by one step after one year subsequently achieved a two-step improvement by 2.5 years.
The direction of the findings is encouraging because it suggests that the retinal response may continue to develop after the first year. The result also uses the same dose selected for the Phase IIb/III NAAVIGATE programme, improving its relevance to the next stage of development.
The limitations remain material. The 2.5-year analysis is interim, the evaluable pivotal-dose group is small and observational control participants were eligible to cross over after one year. The Diabetic Retinopathy Severity Scale is a recognised anatomical grading measure, but an improvement on that scale does not by itself prove durable preservation of visual function, quality of life or avoidance of future treatment.
Claims that sura-vec may modify the underlying disease therefore remain a development hypothesis. The finding needs confirmation through the randomized, masked and sham-controlled NAAVIGATE trial, particularly its one-year primary endpoint and longer-term assessment of vision-threatening complications.
What does the disclosed safety record reveal across both administration routes?
REGENXBIO said no new safety signal emerged among patients completing the five-year wet AMD follow-up, and no postoperative intraocular inflammation event was considered related to sura-vec. Four deaths occurred during long-term follow-up, none of which was considered treatment-related.
The complete presentation nevertheless contained safety details that require context. Retinal pigmentary changes were reported in 26 of the 37 long-term follow-up participants across all dose cohorts. Five ocular serious adverse events were recorded, including two visual-impairment events in the higher-dose Cohort 5 that investigators considered related. Cohort 5 is not one of the two dose groups that advanced into the pivotal trials, which makes dose selection an important part of interpreting the programme.
For diabetic retinopathy, the cumulative safety analysis covered 50 patients with non-proliferative disease across three dose levels. Twenty-nine serious adverse events were recorded, but none was considered drug-related. The company reported no cases of vasculitis, vascular occlusion, hypotony or chorioretinitis.
No intraocular inflammation was observed among the 17 patients treated at the pivotal diabetic retinopathy dose with prophylactic topical steroids. Earlier lower-dose groups without the same steroid regimen experienced mild intraocular inflammation that resolved with topical corticosteroids.
The steroid requirement is therefore part of the treatment protocol, rather than a minor footnote. Late-stage studies must show that inflammation prevention, intraocular pressure monitoring and follow-up requirements remain manageable across a larger clinical network. Gene therapy can reduce repeated drug administration, but it introduces a different set of procedural and long-term surveillance responsibilities.
Can sura-vec compete as established anti-VEGF treatment becomes less frequent?
The commercial proposition rests on more than replacing monthly injections. Existing retinal therapy has continued to evolve, with longer-acting anti-vascular endothelial growth factor drugs and implant-based approaches extending treatment intervals for some patients.
Sura-vec must therefore demonstrate an advantage against a moving standard of care. The relevant test will be whether one administration can maintain vision and retinal anatomy for years, reduce rescue injections by a clinically meaningful amount and avoid risks or workflow demands that offset the convenience benefit.
The two delivery routes create different commercial opportunities. Subretinal administration in wet AMD requires specialist procedural capability and may initially concentrate treatment in experienced retinal centres. Suprachoroidal administration for diabetic retinopathy could potentially support broader office-based use, but its scalability will depend on training, reproducibility and confidence in managing inflammation and pressure changes.
Payers will also examine durability carefully. A one-time treatment may carry a substantial upfront cost, while the economic benefit arrives gradually through avoided injections, fewer visits and potentially fewer disease complications. Strong long-term data can support that value argument, but reimbursement decisions are likely to depend on pivotal efficacy, durability commitments and post-treatment management requirements.
Why do AbbVie’s partnership economics and REGENXBIO’s financing change the investor interpretation?
REGENXBIO’s latest completed trading session occurred before the July 18 ASRS release, meaning investors have not yet traded on the new clinical data. The shares closed at $9.89 on July 17, down 11.70% for the session and approximately 29% across five trading days. The stock remained about 25% above its level one month earlier and within a 52-week range of $5.46 to $16.19.
The immediate weakness coincided with financing rather than the retinal presentation. REGENXBIO priced 10,003,889 common shares at $9 each and 1,111,111 pre-funded warrants at $8.9999, targeting approximately $100 million in gross proceeds. The offering was expected to close on July 20, subject to customary conditions.
The company also expects a separate $100 million milestone payment from AbbVie following the first patient being dosed in the Phase IIb/III NAAVIGATE trial. Another $100 million would become payable when the first patient is dosed in a second Phase III diabetic retinopathy study.
REGENXBIO had estimated cash, cash equivalents and marketable securities of approximately $105.5 million at June 30, before the July AbbVie payment and the proposed equity proceeds. The combined inflows should strengthen its ability to fund late-stage development, although the share offering increases the common-equivalent count and diluted existing ownership.
Investor sentiment is consequently split. The financing has created a clear near-term dilution concern, but AbbVie’s continued funding commitment and the long-term retinal data provide a more constructive clinical signal. The next market session will offer the first indication of whether investors believe the five-year durability evidence offsets some of the pressure created by the capital raise.
Which milestones will determine whether durability becomes an approvable retinal therapy?
The most important catalyst remains the fourth-quarter 2026 topline readout from ATMOSPHERE and ASCENT. These trials must demonstrate non-inferior visual acuity against active therapy, while secondary results will show whether sura-vec can deliver the injection reduction and anatomical stability suggested by earlier studies.
For diabetic retinopathy, attention will shift to NAAVIGATE enrolment, its randomized Phase IIb comparison and the initiation of the planned second Phase III trial. The Phase IIb portion is expected to enrol approximately 135 patients with non-proliferative diabetic retinopathy without centre-involved diabetic macular oedema.
The ASRS presentation gives REGENXBIO and AbbVie a stronger durability narrative going into those milestones. It shows that treatment effects can remain observable years after a single administration and that the selected diabetic retinopathy dose continues to produce encouraging anatomical responses.
The programme’s value will ultimately depend on whether large controlled studies reproduce those signals without compromising vision or introducing an impractical safety burden. Five-year persistence is an important asset, but the pivotal trials must now establish that the persistence is both clinically reliable and commercially useful.
