Early Axial Length Changes Signal Red Light Therapy Success in Myopia Control

In a growing field of myopia interventions, repeated low-level red light (RLRL) therapy is emerging as a promising tool. However, identifying which patients are most likely to benefit remains an ongoing clinical challenge. A 2025 study by Chen et al. published in Photodiagnosis and Photodynamic Therapy provides evidence that early axial length (AL) dynamics—particularly within the first three months of treatment—may serve as a reliable predictive biomarker of therapeutic efficacy.

This analysis combined data from two randomized clinical trials, encompassing a total of 148 children with myopia. The primary cohort (n=119) originated from a multicenter trial across five tertiary hospitals in China, while a second group (n=29) was drawn from a single-center study in Shenzhen. All participants underwent red light retinal laser (RLRL) therapy following a standardized regimen: exposure to red light at 650 ± 10 nm for three minutes, twice daily. This treatment was administered alongside single-vision spectacles.

The study defined full myopia control as AL progression of less than 0.1 mm/year. By this measure, 54.2% of eyes in the primary cohort (training set) and 55.0% in the second cohort (validation set) achieved myopia control after one year.

Researchers also used machine learning models—specifically random forest algorithms—to analyze biometric data collected at baseline, one month, and three months. Among the variables evaluated, the rate of change in AL over three months emerged as the most robust predictor of long-term success, outperforming both static measures and other biomarkers. The best-performing model achieved an area under the curve (AUC) of 0.98 in the external validation set. In contrast, models relying on subfoveal choroidal thickness (sChT) performed modestly at best (AUC as low as 0.61).

Eyes that demonstrated AL shortening of approximately 0.05 mm in the first month were more likely to achieve full myopia control. In contrast, eyes with minimal early change tended to show renewed elongation by the third month. In non-responders, AL barely changed initially but increased again by month 3.

Univariate logistic regression reinforced these findings. Both the one-month and three-month changes in AL were significantly associated with treatment outcome. However, sChT did not translate into strong predictive performance across validation cohorts, despite the changes being statistically significant.

Although red light therapy is known to induce transient choroidal thickening, this study found that sChT was a weak long-term predictor of treatment efficacy. The correlation between choroidal thickening and AL change was notably weaker in eyes that achieved full control than in non-responders.

Balancing Efficacy with Safety and Rebound Risk

While RLRL demonstrates a high efficacy rate, safety concerns persist, particularly with laser-based devices that may exceed exposure thresholds. Cases of retinal damage have prompted regulatory scrutiny in China. These results, therefore, offer a valuable filter to reduce unnecessary exposure in children unlikely to benefit from long-term use.

Importantly, sustained benefit was seen in most initial responders at the two-year mark, but a rebound effect post-discontinuation remains a concern. AL progression surged to 0.5 mm/year after stopping therapy, nearly matching the rates seen in standard single-vision lens wearers.

Early AL tracking could inform whether to continue, escalate, or de-escalate treatment. For non-responders, less aggressive alternatives such as DIMS or diffusion optics spectacles may offer a safer balance of efficacy and risk.

Reference

Chen Y, Li M, Shang X, et al. Early changes in choroidal thickness and ocular biometry in predicting who will achieve full myopia control with repeated low-level red light therapy. Photodiagnosis and Photodynamic Therapy. 2025;54:104672. https://pubmed.ncbi.nlm.nih.gov/40532774/