Cone photoreceptor preservation with laser photobiomodulation in murine and human retinal dystrophy

We used a novel slit lamp-delivered photobiomodulation (PBM) retinal laser to preserve cone photoreceptors in an animal model of retinitis pigmentosa (RP) and to improve visual acuity in individuals with advanced RP. RP refers to a genetically heterogenous group of blinding inherited retinal dystrophies (IRDs), and PBM refers to the treatment of tissue with light in the far red to near-infrared spectrum. Bioenergetic dysfunction and oxidative stress are implicated in the pathogenesis of secondary cone degeneration in RP;1, 2 and are mitigated by the photonic action of PBM on cytochrome c oxidase in the electron transport chain.3 To date, PBM research on the retina has almost invariably used light-emitting diode (LED) systems; however, this methodology suffers the disadvantage that the energy delivery at the level of the retina is uncontrolled. Our experimental 670 nm slit lamp-delivered retinal laser enables controlled delivery of a known intensity (irradiance) to the retina. The methods are described in the Supporting Information. We firstly assessed the effect of PBM on a mixed retinal cell culture preparation (including tau-immunoreactive neurons, rhodopsin-immunoreactive rods and S opsin-expressing S-cones), under conditions of oxidative stress and mitochondrial compromise.4, 5 Treatment with PBM alone did not detrimentally affect cells at exposures up to 100 mW/cm2 (Figure S1). Exposure to either stressor for 24 h resulted in dramatic reduction of rods, which was significantly mitigated by pre-treatment with PBM (100 mW/cm2; Figure 1A, B). Similar results were found with S-cones (Figure 1C, D) and neurons (Figure S2A). Subsequent investigations using MitoSOX Red and cytochrome oxidase enzyme histochemistry confirmed that PBM treatment (100 mW/cm2) also stimulated an immediate, robust, short-lived increase mitochondrial activity (Figure S2B, F). Finally, using qPCR and western blotting, we showed the elevated expression of photoreceptor genes as well as antioxidant genes in PBM-treated samples (Figure S2C). We also found that the striking induction in haemoxygenase-1 evoked by oxidative stress injury was mitigated by PBM treatment (Figure S2D). These data demonstrate that PBM influences mitochondrial function. We then progressed to in vivo investigation of the effects of PBM on retinal cones in rd1 mice. We assessed the effect of PBM on cone preservation at P60, using cone cell immunostaining density averaged across each retinal flatmount as the primary outcome. Mice received twice weekly PBM treatment to one eye commencing at P21. At P60, M/L cone density was significantly greater in rd1 mice treated with PBM at either 25 or 100 mW/cm2 compared to shams (Figure 2A). S-cone cell body density was similarly preserved (Figure 2A). We also assessed the survival of M/L cone outer segments, whose presence is required for detecting light. M/L cone outer segment survival was significantly prolonged by PBM (Figure 2A and Figure S3). Electroretinographic activity in the rd1 mouse is unrecordable by P30 due to the early onset of cone segment degeneration. Hence, we assessed residual vision by recording the optokinetic reflex at P35 and observed partial preservation of this visual reflex in the PBM-treated groups (Figure 2A). To determine whether the neuroprotective influence of PBM extended to longer durations, we then investigated cone survival at P90. Given the similarity of the effect at both 25 and 100 mW/cm2, to reduce animal numbers, we elected to treat at 100 mW/cm2W only. We again observed marked preservation of cones in PBM-irradiated eyes compared to untreated controls (Figure 2B). Motivated by the safety and efficacy in vivo, we rapidly translated this technology to a phase I trial of patients with advanced RP who had progressed to tunnel vision and impairment of cone-derived visual acuity: 12 patients were entered into the study and designated as Group 1 (receiving 25 mW/cm2) or Group 2 (receiving 100 mW/cm2) according to enrolment chronology (Figure S4). The procedure was well tolerated by all participants and there were no adverse reactions at either irradiance. There were no missing data. Combining groups, patients recovered a mean of 5.4 letters (SD 5.5) at the 8-week time point (4 weeks after completion of treatment, Figure 3A). The group receiving 100 mW/cm2 displayed a greater variance (Figure 3C). Cone-derived photopic flicker responses were almost completely abolished in all participants with a large within- subject and between-subject variance (Figure 3B). No significant change in electroretinogram amplitude was observed (Figure 3D, F), but responses were essentially unrecordable in these patients due to advanced disease. The procedure had no significant effect on any of the measured ophthalmic parameters. There was a suggestion that the effect on visual acuity was tapering by 6 months (Figure 3E). PBM is an attractive therapeutic modality for retinal diseases that comprise bioenergetic failure, oxidative stress and/or gliosis components.6 In 2013, Kirk et al. reported that 670 nm LED light (Quantum Devices, Barneveld, WI) attenuated the loss of retinal function and structure in a rat model of RP (the P23H rat).7 Although cones were not specifically investigated in this study, the results are consistent with the findings from the current study. The IRDs remain a significant visual health problem and have traditionally been recalcitrant to therapeutic intervention. However, based on the results of improved multi-luminance mobility testing in a phase III randomized controlled trial (RCT),8 the Food and Drug Administration recently approved voretigene neparvovec-rzyl for the treatment of patients with biallelic RPE65-mediated IRD.9 RPE65 mutations account for approximately .1–1% of IRD. Gene therapy for other recessive IRDs is an explosive research area. However, these genetic engineering techniques are disease specific, expensive and restricted to high socio-economic index populations. The development of artificial retinal implants is largely targeted at IRDs but has had limited clinical impact to date. Recently, Campochiaro et al. reported that oral N-acetyl-cysteine preserved 2–3 letters in each of the three cohorts of 10 patients receiving different doses of N-acetyl-cysteine over a 24-week period.10 Hence, the five-letter short-term improvement noted in the current study compares favourably with oral anti-oxidants, but it must be noted that this is within the range of the test-retest variability. A relatively low-cost strategy that preserves central vision irrespective of the rod gene defect would be a major medical breakthrough at the individual and community level. Our translational research strongly motivates further studies investigating PBM as a treatment for RP. Further work is required to determine the optimal frequency of delivery and to assess the effect in a suitably powered RCT. This research was sponsored in part by a National Health & Medical Research Council grant: (APP1102568) ‘Novel Photoreceptor Bioenergetics: Basic Science & Clinical Translation’ and a grant from the Ophthalmic Research Institute of Australia (ORIA): ‘Rescuing Cone Photoreceptors in Retinitis Pigmentosa with Photobiomodulation’. The authors have declared that no conflict of interest exists. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.