Gene Therapy for Leber’s Hereditary Optic Neuropathy Patients: The REFLECT Trial
July 1, 2023
By Marc Dinkin, MD
Director of Neuro-Ophthalmology; Associate Professor, Departments of Ophthalmology and Neurology, Weill Cornell Medical College
SYNOPSIS: Gene therapy for Leber’s hereditary optic neuropathy, a mitochondrial genetic disorder, using nuclear deoxyribonucleic acid that is transported into mitochondria, shows promising benefits in the treatment of this disease.
SOURCE: Newman N, Yu-Wai-Man P, Subramanian PS, et al. Randomized trial of bilateral gene therapy injection for m.11778G>A MT-ND4 Leber optic neuropathy. Brain 2023;146:1328-1341.
The consequences of Leber’s hereditary optic neuropathy (LHON) can be devastating, as central vision is lost painlessly — first in one eye and, weeks or months later, in the other. Mutations in one of three mitochondrial genes important to the electron transport chain (m.3460G>A in MT-ND1, m.11778G>A in MT-ND4, and m.14484T>C in MT-ND6) lead to most cases, causing retinal ganglion cell (RGC) dysfunction and eventual cell loss. Mutations in MT-ND4, which codes for an important subunit of complex I, cause the most common and severe form of the disease. With no inflammation or vascular event to target, prevention and treatment of ensuing vision loss has been a challenge. Yet, the sequential nature of the vision loss provides a unique window where action taken could help prevent the devastating effects on the second eye. In this context, the medication idebenone, which helps transport electrons in a crippled electron transport chain (ETC) in LHON, demonstrated trends toward improvement in visual acuity as compared with placebo, especially in patients with more discordant inter-eye visual acuity at baseline, but it did not meet its primary endpoints.
Intraocular gene therapy for retinal and optic nerve disease holds great promise, since the vitreous provides a medium through which genetic material can diffuse into adjacent neurons. In 2017, the Food and Drug Administration approved the use of Luxturna (voretigene neparvovec-rzyl), a gene therapy for biallelic RPE65 retinal dystrophy (Leber congenital amaurosis) after studies demonstrated patients’ improved ability to navigate an obstacle course in low light. However, treating mitochondrial disease poses additional challenges; specifically, the localization of replacement wild-type proteins through the outer and inner layers of the mitochondria. To solve this problem, allotopic therapy was developed, in which the mitochondrial deoxyribonucleic acid (mtDNA) is transcribed within the nucleus, but the resulting messenger ribonucleic acid (mRNA) is drawn to the mitochondrial surface by a mitochondrial targeting sequence (MTS), upon which it is translated into the wild-type protein, which then is imported into the mitochondria.
For LHON, a single-stranded DNA recombinant adeno-associated virus vector 2, serotype 2 (rAAV2/2), containing a hybrid version of the human wild-type ND4 protein, has been developed, with elements of the human cytochrome c oxidase 10 (COX10) mtDNA that serve as its MTS. This therapy, known as lenadogene nolparvovec (LN), first was demonstrated to preserve retinal ganglion cells (RGCs) and vision in rodents, followed by a Phase I/II trial (REVEAL) that demonstrated its safety. In the REVERSE and RESCUE trials, a unilateral intravitreal injection of LN did not prevent vision loss, but there was an unexpected improvement in vision at 96 weeks after vision loss, as compared with the nadir visual acuity, not only in the treated eye, but in the sham eye, suggesting a trans-chiasmal transport of rescue proteins to the fellow eye.
The REFLECT study was a randomized, double-blinded, placebo-controlled, Phase III trial conducted across 13 centers. The trial set out to compare the effect of bilateral injections of rAAV2/2 vs. unilateral injection in patients with LHON. In the first treatment arm (TARM1), 48 patients (mean age 32.4 years) received a treatment injection into both the first affected eye and the second eye (whether already affected or not yet affected), while in the second treatment arm (TARM2), 50 patients (mean age 31.9 years) received treatment in the first eye and a sham injection into the second eye.
The mean change in best corrected visual acuity (BCVA) from baseline to 1.5 years was -0.09 Logarithm of Minimum Angle of Resolution (LogMAR) in the bilaterally treated group and -0.04 LogMAR in the unilaterally treated group, a difference of only -0.05, which did not meet the primary endpoint of a difference between the two groups of -0.3 LogMAR (which would equate to a 15-letter difference on the Early Treatment Diabetic Retinopathy Study [EDTRS] chart). However, all eyes in the REFLECT study demonstrated a statistically significant and clinically relevant increase in BCVA at 1.5 years (compared with the visual nadir) superior to that seen in LHON natural history studies. As in the RESCUE and REVERSE studies, this effect was seen in contralateral, sham-injected eyes as well, consistent with trans-chiasmal spread of the therapy, as has been demonstrated in primate studies. Furthermore, bilaterally treated patients showed a trend toward better improvement in BCVA than unilaterally treated patients, and treated eyes demonstrated superior outcomes to placebo eyes. Additionally, first-affected eyes experienced better outcomes in BCVA than second-affected/not yet-affected eyes.
Secondary outcome analysis demonstrated a correspondence between improvement in Humphrey visual fields (HVF) and BCVA, as well as between the degree of ganglion cell layer (GCL) thickness preservation as measured by optical coherence tomography (OCT) and BCVA. Quality of life assessment using the validated Visual Function Questionnaire (VFQ-25) demonstrated an improvement of approximately +6 points at 1.5 years. Finally, the safety of LN was demonstrated, with no life-threatening systemic adverse events (AE) or AEs necessitating study cessation. Ocular AEs mostly were limited to intraocular inflammation that was treatable with topical (and, rarely, systemic) steroids.
COMMENTARY
Although REFLECT, similar to its predecessor studies, did not meet its primary endpoint, its results still added to the growing body of literature that treatment with allotopic gene therapy for ND4-mutated LHON patients leads to improvement in the outcome of BCVA, both in the treated and untreated eyes. Perhaps the most interesting finding was that first-affected eyes experienced better outcomes in BCVA than second-affected/not yet-affected eyes.
The results are counterintuitive, since the first-affected eyes are, by definition, treated later in their course relative to their vision loss and since, in most neurological disease, earlier treatment is more effective, as it prevents progression earlier in the course of neuronal degeneration. The explanation that acutely affected ganglion cells may be in a metabolic state unfavorable to transfection with LN is fascinating but remains unproven.
The most significant limitation of the study is related to one of its most interesting observations — that treatment effects are not limited to the injected eye. As such, the untreated eye cannot serve as a proper control. Since a prospective study with patients left completely untreated would be problematic, the authors are left to compare outcomes to retrospective natural history studies with varied methodology. Nevertheless, as a practicing neuro-ophthalmologist, based on the data of REFLECT and related studies, I would encourage my patients with LHON to be treated with intravitreous LN, and, based on the REFLECT results, preferably in both eyes to give them the best chance of safely enjoying some visual recovery. Hopefully, with refinements in the ideal timing of injection, even stronger improvements will be demonstrated in future studies of this promising treatment.