Red Light Therapy for Dry AMD: The Science, the Evidence, and My Clinical Approach

A patient with intermediate dry AMD asks you: "I've read about red light therapy for macular degeneration, should I try it?" Two devices, Valeda and Eyelight, now have clinical trial data supporting their use. But the evidence is nuanced, the mechanisms are genuinely interesting, and knowing which patients to treat - and when - requires careful assessment. Here is my review.

Science of red light therapy

Red light and long wavelength light on mitochondrial function

It appears that the primary effect of red light and long wavelength light is on the mitochondria. Mitochondria are the organelles in the cells that are responsible for energy production, along with control of apoptosis (regulated cell death). In the mitochondria, red light and long wavelength light is absorbed primarily by cytochrome c oxidase and calcium ion channels. Furthermore, water within the mitochondria can also absorb red and long wavelength light, which reduces the viscosity of interfacial water within the mitochondria, and allow the F₀F₁ ATP synthase, which rotates as a nanomotor to turn faster. The net result is an increase in efficiency of the mitochondria and higher production of ATP.

Inflammation

Red light and long wavelength light modulate the mitochondrial membrane potential by its effect on cytochrome c oxidase. It has been shown that in healthy tissue, application of red light and long wavelength light can result in a transient increase in reactive oxygen species mediated by a change in mitochondrial membrane potential. However, in oxidative stressed cells and tissues, the mitochondrial membrane potential is lowered leading to reactive oxygen production, application of red and long wavelength light increases the mitochondrial membrane potential and reduces reactive oxygen species, leading to its therapeutic effect in inflammation.

Consistent with the above mechanism, red and long wavelength light has been shown to decrease inflammatory markers in activated inflammatory cells, reduce markers of M1 phenotype in activated macrophages, reductions in reactive nitrogen species and prostaglandins in various animal models, and overall reduction in inflammation.

Energy production

The effect of red and long wavelength light on cytochrome c and mitochondrial water viscosity translate to an upregulation of ATP production. Using 670nm long-wavelength light, Powner and Jeffery showed that a single 15-min exposure significantly reduces blood glucose using a standard oral glucose tolerance test.  Powner and Jeffery also summarised that 670 nm PBM increased ATP production by 20% in mouse retina, over 50% in mouse brain, and approximately 30% in whole flies. Because higher ATP output requires additional glucose and oxygen, the authors hypothesised and confirmed that a single 15-minute exposure could measurably reduce blood glucose on an oral glucose tolerance test.

It is worth noting that ATP production can decline due to both aging and disease state. Exposure to red and long wavelength light can offset some of the ATP decrease due to aging in experimental setting (30% increase in whole flies, 15% in mouse retina and 50% in mouse brain). Furthermore, application of 670nm light led to functional improvement in a mouse model of Parkinson disease and in a mouse model of diabetic retinopathy.

Red and long wavelength light penetration

Initially, it seems unusual for red and long wavelength light to have effects on the mitochondria inside the body. After all, the human body is opaque, and most of the body is covered by clothing. However, it has been demonstrated that long wavelength lights actually penetrate through the body even with clothing on, and in the same paper, the longer wavelengths in sunlight has a systemic effect in improving vision in terms of colour contrast sensitivity.

Spectrum of light in common light sources

One of the concerns is that, with modern light sources, especially LED, the emission spectrum is truncated at the longer wavelength with lower proportion of red and long wavelength light. Although this improves energy efficiency of the light source, it may have a detrimental effect on human health from the mechanisms above.

Furthermore, modern building with high energy efficiency blocks the transmission of red and long wavelength light to reduce the energy use for heating and cooling the building. Jeffery et al published a paper regarding the lack of red and long wavelength light in this setting and presented evidence that supplemental long wavelength light in the form of an incandescent lamp could improve colour contrast sensitivity. This paper supports the idea that a lack of red and long wavelength light could be detrimental to health.

Systemic conditions treated by photobiomodulation

Photobiomodulation is a commonly used term in medical literature to describe red light therapy for the treatment of diseases. Photobiomodulation has been shown to be safe and effective in these systemic condition in a major systematic review:

• several types of ulcers
• peripheral neuropathy
• acute radiation dermatitis
• androgenic alopecia
• cancer-therapy-related oral mucositis

Furthermore, there is animal model or clinical evidence on the effectiveness of these conditions:

• metabolic disease
• cardiovascular disease
• severe COVID-19 to reduce hospital stay
• Parkinson disease
• psychiatric and neurological disorders

‍

The use of photobiomodulation in age-related macular degeneration

Pilot study by Glen Jeffery

The team led by Glen Jeffery published a pilot study on the use of 670nm light in healthy aging participants and also in patients with intermediate age-related macular degeneration. The intervention is daily treatment to the treated eye in the morning with a 670 nm hand-held light source housed in a torch-like tube that emitted energy equivalent to 40 mW/cm² or 4.8J/ cm² for 2 min at the viewing aperture. Although there was an improvement in scotopic thresholds in healthy aging patients, no beneficial effect was observed in those with age-related macular degeneration.

Lightsite studies

At the same time, there is a separate effort to investigate the use of photobiomodulation in dry (non-neovascular) age-related macular degeneration led by Clark and Stephanie Tedford in Lumithera, resulted in the publications on the Lightsite I to III studies.

All three Lightsite trials share a common design:

• Lumithera light delivery system - three wavelengths, 590 nm, 660 nm, and 850 nm
• nine treatments over three to five weeks
• double-blinded, randomised, sham controlled, with sham arm exposed to light at 50–100× reduction in light fluence.

‍

Lightsite I

This single centre study involved 46 eyes in 30 patients with dry age-related macular degeneration. They are randomised 1:1 into PBM or sham. The treatment was two cycles of nine treatments, six months apart. The follow up period was one year.

In this study, most patients have advanced dry age-related macular degeneration (AREDS4 in 70% of eyes, geographic atrophy in 60% of eyes). Despite the presence of advanced disease, the study was able to show a statistical improvement in best-corrected visual acuity (BCVA) with a gain of 8 letters and improvements in contrast sensitivity, central drusen volume, central drusen thickness, and quality of life.

Lightsite II

This study was meant to be a larger study seeking regulatory approval. However, due to COVID at the time, recruitment for the study was interrupted, and this ended up to be a smaller exploratory study. The study involved 53 eyes from 44 patients with dry age-related macular degeneration. Subjects were randomized in a 2:1 fashion (PBM; sham). Treatment was 3 cycles of 9 sessions of PBM, with one cycle every four months. The follow up period was 10 months.

In this study, most patients had intermediate age-related macular degeneration with AREDS score of 2 or 3 (90%). PBM-treated eyes showed statistically significant improvement in BCVA at 9 months (n = 32 eyes, p = 0.02) with a 4-letter gain in the PBM-treated group versus a 0.5-letter gain in the sham-treated group (not statistically significant) for patients that received all 27 PBM treatments (n = 29 eyes). However, in the intention-to-treat analysis (i.e. including patients who did not received all 27 treatments), there was no difference between the PBM group and the sham group in terms of visual acuity outcome.

Macular drusen volume was not increased over time in the PBM treated group but did show increases in the sham-treated group. While PBM and sham groups both showed geographic atrophy lesion growth in the trial period, there was 20% less growth in the PBM group over 10 months, suggesting potential disease-modifying effects. However, there were only eight patients with geographic atrophy in each group. Furthermore, the baseline measurements of central subfield drusen thickness and geographic atrophy were not balanced between groups. No safety concerns or signs of phototoxicity were observed.

Lightsite III

This is the pivotal study leading to the Food and Drug Administration (FDA) approval of the Valeda light delivery system for dry age-related macular degeneration.

This study involved 148 eyes from 100 patients with dry age-related macular degeneration, randomised 2:1 PBM:sham. The PBM group received one cycle of treatment every four months, with a pre-planned analysis at month 13 after four cycles of treatment were completed.

86.9% of subjects were classified as AREDS3 and 72.3% were classified as intermediate age-related macular degeneration. 8% had geographic atrophy. At Month 13 (4 series of treatment), the average change from baseline in BCVA was an increase of 5.4 letters in PBM and 3.0 letters in sham-treated eyes, with this result reaching statistical significance. In the small number of eyes with geographic atrophy, there was a numerical trend showed an increase in GA lesion area in sham compared to PBM-treated eyes at Month 13. For the eyes without geographic atrophy at baseline, 10% of sham treated eye developed geographic atrophy, compared to 1.1% in PBM-treated eyes (p = 0.024). In terms of drusen volume, there was no change in drusen volume in PBM-treated eyes while an increase of 0.049mm³ was seen in the sham group.

Eyelight study

The treatment in the Eyelight study is different from the Lightsite studies. The Eyelight device uses two wavelengths - 590nm and 630nm. The study population involved 152 eyes from 76 patients. They were randomised 1:1 to PBM and sham. The sham treatment used the same device but emitted less than 30% of light output as compared to the PBM group. Each cycle of treatment involved six to eight sessions, with the sessions three to four days apart. This study reported the short-term outcome at month four.

Around 85% of all patients were classified as AREDS2 or 3. This study did not detect any adverse safety signals.

At month 4, PBM-treated eyes were significantly more likely to gain five or more letters than sham-treated eyes (20.3% vs 8.9%, p = 0.043). Drusen volume changes also favoured the PBM group (p = 0.013). No significant difference in central subfield thickness was observed between groups over the 4-month period.

Recommendations

There is biological plausibility for a beneficial effect of both red and long wavelength exposure, and PBM in age-related macular degeneration. Based on the strength of the evidence available so far, I recommend my patients to get adequate sunlight exposure, either by a morning walk or a walk in a sheltered leafy area (like a park with trees) on a sunny day. This lifestyle intervention is unlikely to be of harm and will also boost red and long wavelength exposure.

At this stage, there is no clear guidance on when to commence PBM treatment in dry age-related macular degeneration, and how to track improvement. My clinical approach is to track anatomical changes using OCT and AI (RetInsight GA algorithm summarised here) for all my patients with dry age-related macular degeneration. If I see progression on the OCT images or on the AI algorithm, I will recommend PBM treatment if there is no geographic atrophy. If there is already geographic atrophy, I will discuss PBM, stating that this technology is safe even in the presence of geographic atrophy but the strength of data on improvement is weaker in this situation.

Following PBM treatment, I will continue to track the anatomical outcome using AI to see if there is a benefit, i.e. slowing of progression on the GA algorithm or reduction / stabilisation in drusen volume. I will then make a further recommendation based on that outcome.

When to consider referral for PBM assessment

‍If your patient has intermediate dry AMD (AREDS 2–3) with drusen on OCT, I can perform a detailed assessment using AI-based retinal analysis to determine whether PBM treatment is appropriate and track response over time. Referrals via Healthlink (EDI: ret66ina) or email info@retinaspecialists.co.nz.

About Dr Leo Sheck

Book your appointment now to see Dr Sheck

Dr Sheck is a RANZCO-qualified, internationally trained ophthalmologist and runs a popular and respected practice in Auckland. He combined his initial training in New Zealand with a two-year advanced fellowship in Moorfields Eye Hospital, London. He also holds a Doctorate in Ocular Genetics from the University of Auckland and a Master of Business Administration from the University of Cambridge. He specialises in medical retina diseases (injection therapy), cataract surgery, ocular genetics, uveitis and electrodiagnostics.