The endgame of inherited retinal disease trials

The development of therapies targeting untreatable inherited retinal diseases (IRDs) is a research priority in ophthalmic innovation. Functionally devastating for patients, IRDs are estimated to cost more $15 billion per year in the US and Canada.¹ With more than 100 distinct diseases under the IRD umbrella, these conditions show significant variability in phenotype and prognosis. Up to 75% of patients have identifiable genetic defects that hypothetically could be targeted by gene therapy.² To date, however, there is only one such approved treatment.

The genetic heterogeneity of IRDs and their generally slow progression and inherently small patient populations make measuring the benefits of treatment—and then translating those into clinical trial end-points—a significant challenge. Typical outcome measures such as central visual acuity and optical coherence tomography imaging of the macula may not be adequate. Alternative, novel outcome measures are necessary to assess loss of peripheral vision, color vision, night vision, light and contrast sensitivity and changes to retinal structure.

The rise of functional end-points: Why mobility matters

IRDs dramatically impair patients’ ability to perform activities of daily life. Their unique nature has propelled functional outcomes measured by mobility or maze-based end-points to the forefront.

The breakthrough was voretigene neparvovec-rzyl (Luxturna; Spark Therapeutics, Inc) for RPE65-mediated Leber congenital amaurosis (LCA). Here, the multi-luminance mobility test (MLMT), a real-world maze performed under varying light conditions, provided compelling evidence of how treatment improved patients’ functional vision, ultimately setting a new global standard for regulatory approval.³

The MLMT, developed by researchers at Children’s Hospital of Philadelphia and the University of Pennsylvania and later licensed to Spark Therapeutics, was specifically engineered to capture meaningful improvements in daily function, not just retinal sensitivity. Unlike visual acuity tests that measure clarity at a single fixation point, MLMT tracks a patient’s ability to navigate a course dotted with obstacles under light conditions ranging from near darkness (1 lux) to well-lit environments (400 lux).

This design suited the natural history of RPE65-mediated disease, where significant vision loss manifests as reduced light sensitivity and legal blindness by the time patients reach adolescence.⁴ Results of a peer-reviewed validation study cemented MLMT as a robust measurement tool, ensuring the data would be acceptable to regulators and the scientific community.

The challenge of heterogeneity

The heterogeneity of IRDs renders drug development particularly complex. Even within a single mutation, such as RPE65, patients can exhibit a wide range of severity and degeneration patterns. This diversity affects the suitability of any assessment tool: one size rarely fits all. Although MLMT was tailored for diseases characterised by declining light sensitivity, other IRDs may require alternative approaches.

The Luminance Dependent Navigation Assessment (LDNA), used by Ocugen in a phase 3 clinical trial (NCT06388200) of its gene therapy with a broad retinitis pigmentosa (RP) indication, is one example of this evolution.⁵ The LDNA is a more sensitive and specific assessment tool, demonstrating uniform correlation between light intensity and lux levels. It is designed to optimise sensitivity and specificity for the targeted condition by employing a 10-level, logarithmically spaced scale, which allows for finer discrimination of changes across the full spectrum of disease progression. Importantly, LDNA designed for patients with early to advanced cases of RP has undergone validation, demonstrating high reliability, with participants showing minimal variability in their ability to successfully navigate the courses and the time required.

According to Ora Clinical, its specialised mobility assessment end-point, the visual navigation course, has been accepted by the FDA as an integral end-point in the United States for relevant clinical trials.^6,7 It is highly adaptable to various patterns of vision loss, such as central, light perception and peripheral vision issues, and can be tailored for specific disease characteristics.

Ray Therapeutics is using a multiluminance mobility measurement in its phase 1 ENVISION safety study (NCT06460844) of RTx-015 gene therapy in patients with RP or choroideremia.⁸ The company is also sponsoring an observational study (NCT06375239) to assess end-points and measurement in these patients.⁹

The evolution of maze testing

Physical mazes offer immediacy and realism, as patients are confronted with real-world obstacles and variable lighting. They allow for direct observation and patient feedback, which can help in their adaptation to diverse phenotypes seen in IRD populations. However, physical maze assessments may not be practical for all trial centers. They are logistically burdensome, require meticulous certification of light levels and cannot easily simulate every real-world scenario. Young children may excel at navigation due to their enhanced flexibility, whereas adults may struggle, introducing psychosocial confounders into the measure.

The advent of virtual reality (VR) mazes—such as the University of Pennsylvania’s VR orientation and mobility protocol and Streetlab’s Mobility Standardized Test—has introduced a new layer of flexibility and scalability.^10,11 These tests can be deployed broadly without physical infrastructure and potentially standardise obstacle presentation and lighting for reproducibility.

Still, VR mazes introduce their own challenges. Young children may struggle with headset interfaces, the devices can cause motion sickness and headaches and the underlying technology has yet to fully replicate the richness of real-world navigation. Currently, VR tools lack substantial validation, which may change as the technology evolves and is employed in more trials.

Regulatory perspectives and the search for meaningful change

As functional end-points gain prominence, regulatory bodies such as the US Food and Drug Administration (FDA) have begun to set expectations around what constitutes clinically meaningful improvement. Yet, many uncertainties remain when it comes to disease stabilisation, which may mean regulators will need to show increased flexibility. This is especially true of conditions associated with slower progression, such as RP.

Low luminance visual acuity (LLVA) has been recognised by the FDA as a viable primary end-point in pivotal trials, with its clinical significance viewed as analogous to the 15-letter gain on the Early Treatment Diabetic Retinopathy Study chart for best-corrected visual acuity (BCVA). At the same time, the use of LLVA remains debated, given the heterogeneity of RP across early and advanced stages and uncertainty over whether smaller changes, such as a 5-letter gain, carry the same clinical relevance as the 15-letter threshold.

Taken together, these considerations highlight the need for a flexible, multimodal approach, one that combines LLVA with complementary measures such as full-field stimulus threshold, patient-reported outcomes and novel assessments such as LDNA—to more comprehensively capture therapeutic benefit in RP trials.

One highlight of physical maze assessment, as evidenced by Luxturna’s approval and the MLMT experience, is its intuitive impact: any observer can appreciate how a patient is able to navigate the course. This ease of interpretation helps bridge gaps with regulators, payers, clinicians, and, most importantly, patients in a simple and meaningful way that transcends the complexity of the underlying test protocol.

The pitfalls

In designing appropriate mobility and maze assessments in varied IRD presentations, researchers must confront the ceiling effect. Many patients quickly max out the scale, making further improvements difficult to demonstrate. Moreover, in gene-agnostic therapies or trials addressing ultrarare mutations, end-points must balance broad applicability with specificity. It is rare to find a universal test.

There are also examples of trials in which mobility testing was not successful. Nanoscope Therapeutics’ RP gene therapy MCO-010 failed to meet its initial primary end-point of mean change from baseline in multi-luminance Y-mobility test scores at 52 weeks. It later met a new primary end-point of BCVA after a protocol change approved by regulators.¹² Similarly, Janssen’s gene therapy for X-linked RP did not meet its primary end-point using a vision-guided mobility assessment maze.¹³

Simplicity is the holy grail. If less labor-intensive end-points—such as LLVA and FST—could reliably correlate with meaningful mobility improvements, trials would both accelerate and broaden. Although some non-US regulatory agencies have accepted correlations between MLMT and FST, the FDA has so far declined to follow suit for the primary outcome measure. Patient-reported outcomes are emerging as important end-points, especially in late-stage or ultra-rare diseases where other measurements lose value.

Looking forward, the destiny of IRD trials may lie in fully customisable, virtual mobility assessments that can be tailored to the disease, genotype, and even individual patient severity. This would further align appropriate end-points with the therapy’s mechanism of action and, crucially, the real deficits that matter most to patients in their daily lives. Large, well-structured natural history studies will remain essential to demonstrate that intervention meaningfully alters the disease trajectory for each mutation.

Conclusion

All stakeholders—industry sponsors, academic researchers, regulators, payers and patient advocacy groups—must remain agile, receptive to new tools and open to evolving definitions of treatment success. Early and frequent engagement with regulatory agencies and payers is critical, as is introducing the patient voice into end-point selection. As the technology improves and emerging data sets support virtual or hybrid approaches, the field will converge on solutions that balance scientific rigor,
patient centricity, and feasibility.

Ultimately, data-driven innovation will guide the future and success of IRD trials. For now, maze-based mobility assessments remain at the center of IRD drug development, but the future promises that more precise tools and scalable, patient-centric measures will pave the way for transformative treatments for patients in need.

References

1. Gong J, Cheung S, Fasso-Opie A, et al. The impact of inherited retinal diseases in the United States of America (US) and Canada from a cost-of-illness perspective. Clin Ophthalmol. 2021;15:2855-2866. doi:10.2147/OPTH.S313719

2. Thirunavukarasu AJ, Raji S, Kapetanovic C. Visualizing treatment effects in low-vision settings: proven and potential endpoints for clinical trials of inherited retinal disease therapies. Gene Ther. Published online August 7, 2025. doi:10.1038/s41434-025-00552-7

3. Russell S, Bennett J, Wellman JA, et al. Efficacy and safety of voretigene neparvovec (AAV2-hRPE65v2) in patients with RPE65-mediated inherited retinal dystrophy: a randomised, controlled, open-label, phase 3 trial. Lancet. 2017;390(10097):849-860. doi:10.1016/S0140-6736(17)31868-8

4. Clinical Review Report: Voretigene Neparvovec (Luxturna): (Novartis Pharmaceuticals Canada Inc.): Indication: Vision Loss, Inherited Retinal Dystrophy. Canadian Agency for Drugs and Technologies in Health; 2021. Accessed October 2025. https://www.ncbi.nlm.nih.gov/books/NBK569039/

5. A phase 3 study of OCU400 gene therapy for the treatment of retinitis pigmentosa (liMeliGhT).ClinicalTrials.gov. Updated October 16, 2025. Accessed October 2025. https://clinicaltrials.gov/study/NCT06388200

6. Shapiro A, Corcoran P, Sundstrom C, et al. Development and validation of a portable visual navigation challenge for assessment of retinal disease in multi-centered clinical trials. Invest Ophthal Vis Sci. 2017;58:3290.

7. Retina, glaucoma, refractive & CED. Ora Clinical. Accessed October 2025.https://www.oraclinical.com/wp-content/uploads/2024/04/Ora_FactSheet_Posterior_Segment_042624_digital.pdf

8. Study to evaluate safety of RTx-015 injection in retinitis pigmentosa or choroideremia patients (ENVISION). ClinicalTrials.gov. Updated June 13, 2025. Accessed October 2025. https://clinicaltrials.gov/study/NCT06460844

9. Observational study to assess end-point operational feasibility & measurement properties in patients with retinal degeneration. ClinicalTrials.gov.
   Updated January 8, 2025. Accessed October 2025.https://clinicaltrials.gov/study/NCT06375239

10. Bennett J, Aleman EM, Maguire KH, et al.Optimization and validation of a virtual reality orientation and mobility test for inherited retinal degenerations. Transl Vis Sci Technol. 2023;12(1):28. doi:10.1167/tvst.12.1.28

11. Authié CN, Poujade M, Talebi A, et al. Development and validation of a novel mobility test for rod-cone dystrophies: from reality to virtual reality. Am J Ophthalmol. 2024;258:43-54. doi:10.1016/j.ajo.2023.06.028

12. Nanoscope Therapeutics provides regulatory update on MCO-010 for the treatment of retinitis pigmentosa. News release. Nanoscope Therapeutics. Accessed October 2025.https://www.prnewswire.com/news-releases/nanoscope-therapeutics-provides-regulatory-update-on-mco-010-for-the-treatment-of-retinitis-pigmentosa-302037671.html

13. Incorvaia D. J&J gene therapy fails to improve visual navigation in late-stage rare eye disease trial. Fierce Biotech. May 5, 2025. https://www.fiercebiotech.com/biotech/jj-gene-therapy-fails-improve-visual-navigation-late-stage-rare-eye-disease-trial