The inventor of the Eyerising International repeated low-level red light (RLRL) therapy for myopia, Professor Mingguang He, has outlined “notable design flaws” in a study suggesting reduced cone density in children who received the therapy and has provided a glimpse into upcoming studies he expects to show opposite findings.
Prof He, the company’s founder and chief medical officer, was responding to a study entitled ‘Cone density changes after repeated low-level red-light treatment in children with myopia’, authored by Liao et al. and published in JAMA Ophthalmology.
The study investigated the potential retinal effects of RLRL using adaptive optics scanning laser ophthalmoscopy (AOSLO). It concluded that long-term RLRL therapy may be associated with structural changes in cone photoreceptor density.
Specifically, the authors reported a reduction in cone density within 0.5 mm of the foveal centre in the RLRL-treated group compared to controls, particularly pronounced in the temporal region. In addition, some children exhibited hyperreflective signals on AOSLO imaging, and one case showed a small cystoid change on OCT.
Prof He said adhering to the highest standards of safety and clinical efficacy was a top priority for his company, which welcomes rigorous scientific investigations and appreciate the authors’ efforts to advance understanding of RLRL therapy.
However, he said there were several methodological and clinical issues warranting careful consideration “before drawing definitive conclusions”.
Some of these limitations were reinforced in an accompanying commentary in the same journal (JAMA Ophthalmology) authored by an independent researcher – Professor Jacque Duncan, a Professor of Ophthalmology at the University of California, San Francisco School of Medicine.
In a comprehensive response, Prof He outlined what he said were key limitations and concerns over the study design, the device and imaging used, and clinical considerations.
Each point and the rationale can be found in his official statement. Prof He’s biggest concerns relate to:
- The study’s retrospective and non-randomised design inherently carries a risk of selection bias. Baseline cone density measurements prior to RLRL therapy were not available, severely limiting the ability to establish causal relationships.
- AOSLO data acquisition is technically demanding for children to complete, requiring 15 to 20 minutes of capture time and sensitive to motion artifacts, which may raise difficulties with reproducibility.
- In retinal imaging, Prof He said it is impossible to define the boundary of individual cones in the densely packed fovea. In addition, a 6% change at 0.5 mm from the foveal centre is within known measurement error and far below thresholds considered clinically significant.
- The analysis involved 32 statistical comparisons (four regions × eight sites = 32 tests), greatly increasing the risk of type I errors (false positives).
- The study’s small sample size (RLRL: 52 versus non-users: 47) substantially limits the robustness and reliability of its conclusions, the company said. This is further evidenced by the extremely wide confidence interval (95% CI: 1–300), which indicates a high degree of uncertainty around the estimated effect and reduces confidence in the reported findings.
- Prof He noted the hyperreflective foci noted within 2.4° of the fovea was not corroborated by co-existing OCT changes or visual symptoms. These foci were located in a technically challenging imaging zone, and may instead reflect image artefacts rather than pathologies.
“For the general public and clinicians, it is important to understand that small changes in cone density observed in this recently published study do not necessarily indicate visual impairment or retinal damage, in particular when this study contains notable design flaws, including the lack of baseline measurements, retrospective data collection, and inadequate statistical adjustments, which limit the strength of its conclusions,” Prof He said.
“To our knowledge, the robustness and reliability of the findings reported in the JAMA Ophthalmology publication have been called into question by leading researchers and clinicians in the myopia field, as evidenced by the prompt submission of multiple letters to the editor and expert commentaries.”
Prof He said to accurately assess the impact of RLRL therapy on cone density, larger, well-designed randomised controlled trials were required.
“Several such trials are already ongoing and will provide more definitive evidence. At present, there is no clinical evidence indicating that RLRL treatment harms children’s visual health,” he said.
Studies still to come
Hinting at what’s to come, Prof He said he was aware of one independent randomised controlled trial and one cross-sectional study conducted in children in China that have “yielded findings that are completely opposite” to those reported by Liao et al.
He said both studies used cone density measured by AOSLO imaging as a key safety outcome – and both had confirmed higher levels of myopia are associated with lower cone density.
The randomised controlled trial was conducted in Beijing Tongren Hospital in China (Jie et al) involving 336 children, using cone density derived from AOSLO imaging as a key safety outcome.
“Their findings found no difference in cone density between the RLRL and control groups (36016 versus 37183 cells/mm2, P=0.553) after a 33-month follow-up. This study is currently under submission,” Prof He said.
Meanwhile, the cross-sectional study by Yue et al. from Chongqing Medical University compared cone density among three groups: children with myopia who underwent RLRL treatment for eight months (20 eyes), children with myopia who received no treatment (32 eyes), and non-myopic children (14 eyes).
“The preliminary results suggested a trend toward higher cone density in the myopia-RLRL group compared to the myopia-untreated group, although the difference did not reach statistical significance (41274 vs. 40202/mm², P=0.627),” he said.
“The highest cone density was observed in the non-myopic group (45649/mm²).”
This manuscript is currently under review by Ophthalmology, widely considered a top journal in the field.
“The results of these studies are expected to be published soon and will provide valuable additional evidence to further clarify the safety profile of RLRL therapy,” Prof He said.
More reading
Repeated low-level red-light therapy: a guide for clinicians (CPD)
Eyerising International responds to China myopia red light regulatory changes
Eyerising Myopia Management Device granted TGA approval for home-use red-light therapy
