Red Light Therapy for Macular Degeneration: What the 670nm Trials Show

Age-related macular degeneration (AMD) is the leading cause of vision loss in older adults, and for the "dry" form there have long been few treatment options. In the last few years a different idea has been tested in real clinical trials: shining specific wavelengths of red and near-infrared light, including 670nm, into the back of the eye to support the retina's energy-making machinery. This review walks through what those trials actually found, where the science is strong, and where the hype outruns the evidence.

A quick, honest summary

Red light therapy for macular degeneration is investigational and device-specific. It is not a cure, and it does not "restore" lost vision in any reliable way. The most studied system, LumiThera's Valeda Light Delivery System, received FDA marketing authorization for dry AMD in late 2024, but that decision rests on modest visual-acuity gains in a specific patient group, measured over months, not miracles. Read this as a sober status report, not a sales pitch.

If you or a family member has AMD, this article is a starting point for an informed conversation with a retina specialist. It is not medical advice and should not replace one.

What is dry age-related macular degeneration?

The macula is the small central part of the retina that gives you sharp, straight-ahead vision. You use it to read, recognize faces, and drive. AMD damages this area over time.

There are two main types:

• Dry (non-exudative) AMD is the most common, making up roughly 85 to 90 percent of cases. Yellow deposits called drusen build up under the retina, and the light-sensing cells slowly thin out. Late dry AMD is called geographic atrophy.
• Wet (exudative) AMD is less common but more aggressive. Abnormal blood vessels leak fluid and blood under the retina.

Wet AMD has well-established drug treatments (anti-VEGF injections). Dry AMD has historically had far fewer options, which is exactly why light-based therapies have drawn so much interest. You can read more about the condition from the National Eye Institute's AMD overview.

The mechanism: why would red light help the retina?

The retina is one of the most energy-hungry tissues in the body. Its photoreceptors burn enormous amounts of fuel to keep working, and they rely on tiny structures inside cells called mitochondria to make that fuel (a molecule called ATP).

Here is the core idea behind retinal photobiomodulation:

1. Mitochondria contain an enzyme called cytochrome c oxidase. This enzyme absorbs light strongly in the red and near-infrared range, including around 670nm and the 590 to 850nm window.
2. When that light is absorbed, it appears to nudge the mitochondria to work more efficiently, producing more ATP and reducing harmful oxidative stress.
3. In an aging or stressed retina, where mitochondria are running down, the theory is that this extra energy support could help the surviving cells function better and slow decline.

This is the same broad mechanism described for red light therapy elsewhere in the body. If you want the cell-level basics, see our explainer on the science of photobiomodulation and the deeper cellular science behind red light therapy.

A key point: the eye is unusual because light can reach the retina directly through the pupil. You do not have to push photons through skin and bone. That makes the retina an unusually clean target for light therapy, but it also raises safety questions, since the retina is delicate.

The Glen Jeffery 670nm work

Much of the foundational lab science comes from Professor Glen Jeffery's group at University College London. Their work in aged animals and then in older humans showed that brief exposure to 670nm light could measurably improve aspects of visual function tied to mitochondrial performance.

In a small 2020 human study, researchers reported that short daily exposure to deep red 670nm light improved color contrast detection in older participants, with the effect tied to better mitochondrial function in retinal cells. The authors framed this as "optically improved mitochondrial function" rescuing aged visual decline (Shinhmar et al., 2020, PubMed). This is an important proof-of-concept, but it was a small, short study in normally aging eyes, not a treatment trial in AMD patients.

The actual AMD trials: LIGHTSITE I, II, and III

The most rigorous clinical evidence in AMD specifically comes from the LIGHTSITE trial program, which tested LumiThera's Valeda device. Importantly, Valeda does not use 670nm alone. It delivers multiwavelength photobiomodulation, combining yellow (590nm), red (660nm), and near-infrared (850nm) light. So when people say "the 670nm trials" for AMD, the real-world device usually means this multiwavelength protocol in the same deep-red and near-infrared family.

Here is how the three trials compare.

Sources: LIGHTSITE I (Markowitz et al., 2020, PubMed), LIGHTSITE II (2023, PubMed), and LIGHTSITE III (Boyer et al., 2024, PubMed).

What the numbers actually mean

The headline result from LIGHTSITE III is roughly a 5-letter improvement in best-corrected visual acuity in treated eyes compared with sham over about 13 months. On an eye chart, 5 letters is one line. That is a real, measurable difference, and for people losing vision slowly it is meaningful. But it is not the same as "getting your vision back."

A few honest caveats:

• These are average gains. Some patients improved more, some did not improve, and outcomes depended on how advanced their AMD was at the start.
• The benefit was seen in intermediate dry AMD, not in eyes with significant central geographic atrophy already present. Late-stage atrophy did not reverse.
• The trials were sham-controlled and double-masked, which is the gold standard and a real strength here. That design helps rule out placebo effects. Even so, the sample sizes were small by drug-trial standards.

A closer look at the geographic atrophy signal

One of the more interesting findings in LIGHTSITE III was not about reading an eye chart at all. It was anatomical. Geographic atrophy is the dreaded endpoint of dry AMD, where patches of the retina die off and create permanent blind spots in central vision. Once those patches form, they do not come back.

In LIGHTSITE III, fewer treated eyes developed new geographic atrophy over the 13-month window compared with sham eyes. If that holds up in larger and longer studies, it would matter more than the letter gains, because preventing atrophy protects vision for years, not months. But two cautions apply. First, the numbers were small, so this could be a real effect or partly chance. Second, the trial was not primarily designed to prove atrophy prevention, so this is best read as a promising secondary signal that needs confirmation, not a settled conclusion.

The honest framing: light therapy may help the retina hold its ground, but we do not yet have the long, large, multi-year data that would let a retina specialist promise a patient that atrophy will be slowed.

The placebo and sham question

Vision is unusually vulnerable to expectation effects. People who believe a treatment is working often perform better on eye charts, concentrate harder, and report feeling sharper. That is exactly why the design of these trials matters so much.

The LIGHTSITE studies used double-masking and a sham device that looked and felt like the real treatment but delivered no therapeutic light. Neither the patients nor the people measuring their vision knew who got the active treatment. This is the strongest available defense against placebo effects, and it is a genuine credit to the program. It is also why this body of work deserves more respect than the typical wellness claim.

Still, masking is never perfect, sample sizes were modest, and a single company (LumiThera) ran the program around its own device. None of that invalidates the results. It simply means independent replication by other groups would strengthen confidence. For a broader look at how sham design separates real gains from placebo across this field, see our piece on red light therapy sham trials and the placebo effect.

Who might this help, and who probably not?

Based on the trial populations, the people most likely to be candidates are:

• Adults with intermediate dry AMD who still have measurable central vision worth protecting.
• People treated before large areas of geographic atrophy develop.

It is unlikely to help, and was not designed for:

• Advanced geographic atrophy with established central vision loss.
• Wet AMD (which needs anti-VEGF injections, not light).
• People expecting reversal of long-standing blindness or severe vision loss.

This is fundamentally a "slow the decline and modestly improve function" therapy, not a restorative one. Anyone promising vision restoration from a red light device is overselling.

What a treatment course actually looks like

This is not a one-and-done procedure, and it is not a daily home routine either. In the LIGHTSITE trials, treatment was delivered in series of short in-office sessions, then repeated periodically.

A rough picture of the protocol:

• Each session lasts only a few minutes per eye.
• Sessions are clustered into a series, often around nine sessions over a few weeks.
• Those series are then repeated on a schedule, roughly every four months in the longer trials, to maintain the effect.

The repeating cadence is important to understand. The early gains in LIGHTSITE I faded somewhat between treatment series, which is part of why later trials tested ongoing, periodic re-treatment rather than a single course. In practice this means a real commitment: travel to a clinic that has the device, repeat visits, and treatment that continues indefinitely rather than fixing the problem once.

For anyone weighing this, the practical question is not only "does it work" but "can I sustain a clinic-based treatment cadence for years." That is a very different commitment from buying a panel.

How it compares to other dry AMD options

Dry AMD has historically been a frustrating diagnosis precisely because there was little to offer beyond monitoring and lifestyle steps. Putting light therapy in context helps:

• AREDS2 nutritional supplements. A specific vitamin and mineral formula (based on the large Age-Related Eye Disease Studies) can reduce the risk of intermediate dry AMD progressing to advanced disease. This is well established, cheap, and low-risk, and it is usually the first thing a doctor recommends. Light therapy does not replace it.
• Geographic atrophy drugs. Newer injectable complement-inhibitor drugs can slow the growth of existing geographic atrophy, but they involve repeated eye injections and carry their own risks, and they slow damage rather than improve vision.
• Lifestyle steps. Not smoking, controlling blood pressure, eating leafy greens and fish, and protecting eyes from excessive UV are all part of standard AMD care.
• Photobiomodulation (Valeda). Sits as an additional, non-injection option for intermediate dry AMD, aiming to modestly improve function and possibly slow decline.

The realistic way to think about light therapy is as a complement to established care, not a replacement for supplements, lifestyle changes, or regular monitoring by an eye doctor.

Cost, access, and the practical reality

Because Valeda is a clinic-administered medical device, access depends on finding a retina or eye-care practice that has invested in the system. It is not yet widely available everywhere, and that footprint is still growing.

Cost is the other practical hurdle. Insurance coverage for newly authorized photobiomodulation is uneven, and patients may face significant out-of-pocket expense for repeated treatment series over time. This contrasts sharply with consumer red light therapy, where the main cost is buying a device once. Anyone considering retinal photobiomodulation should ask the clinic directly about total expected cost across multiple treatment series, not just the price of a single session.

None of this is a reason to dismiss the therapy. It is a reason to go in with clear eyes about the commitment.

Safety: what the trials report

Across the LIGHTSITE program, the Valeda treatments were generally well tolerated, with no signal of light-induced retinal damage at the doses and wavelengths used. That matters, because the retina can be harmed by the wrong kind or intensity of light. The whole approach depends on staying in a low-power, specific-wavelength range delivered by a controlled medical device under clinical supervision.

Two safety points deserve emphasis:

1. This is a clinic-delivered medical device. The trial results apply to the Valeda system used as studied. They do not transfer to consumer red light panels, masks, or handheld units pointed at your face.
2. Do not shine consumer red light devices into your eyes. General red light therapy guidance is to protect the eyes during treatment, not expose the retina deliberately. See our piece on red light therapy eye safety, goggles, and research. The macular work uses a calibrated ophthalmic device, not a wellness panel.

FDA status (as of 2026)

In November 2024, the FDA granted De Novo marketing authorization to LumiThera's Valeda Light Delivery System for the treatment of dry AMD, making it the first FDA-authorized photobiomodulation therapy for this condition. LumiThera describes Valeda as "the first and only FDA-authorized treatment for dry AMD to improve and maintain vision," based on the LIGHTSITE III data (LumiThera Valeda page).

What this status means in plain terms:

• Valeda is a prescription, clinic-administered device, used by eye-care professionals, not something you buy for home use.
• FDA authorization reflects a favorable benefit-risk judgment in the studied population. It is not a claim that the therapy works dramatically or for everyone.
• Coverage, availability, and out-of-pocket cost vary, and treatment involves repeated sessions over time.

You can track ongoing and future studies on ClinicalTrials.gov and the broader literature through this PubMed search on photobiomodulation and AMD.

How this differs from general red light therapy

It is easy to conflate the macular degeneration research with the red light panels marketed for skin, recovery, and pain. They share a mechanism (mitochondrial photobiomodulation) and an overlapping wavelength family (660nm and 850nm show up in both). But the differences are large:

• Target tissue: retina vs skin and muscle.
• Device: a controlled ophthalmic medical device vs a consumer panel or mask.
• Dose: carefully calibrated, low-power, eye-specific delivery vs whatever a wellness panel emits.
• Evidence: sham-controlled clinical trials with FDA review vs a much more mixed consumer market.

For why those wavelengths matter in the first place, see our breakdown of red light therapy wavelengths explained. For the emerging-evidence pattern in another nervous-system tissue, our review of photobiomodulation for traumatic brain injury shows the same "promising but early" theme.

The bottom line

The case for light therapy in dry AMD is more serious than most red light claims floating around online. There is a plausible mechanism rooted in retinal mitochondrial biology, real sham-controlled trials (LIGHTSITE I, II, and III), and an FDA authorization for one specific device. That is a meaningful bar that most red light therapy uses never reach.

But keep the limits front and center. The benefit is modest, measured in a line or so of eye-chart letters. It applies mainly to intermediate dry AMD, not advanced atrophy or wet AMD. It is delivered by a clinical device, not your home panel. And it is not vision restoration. Talk to a retina specialist before making any decisions.

Frequently Asked Questions

Can red light therapy cure or reverse macular degeneration?

No. The best clinical evidence shows modest improvements and a possible slowing of decline in intermediate dry AMD, not a cure. Treated eyes in the largest trial gained roughly one line of vision on average versus sham over about a year, and advanced vision loss did not reverse. Anyone promising restoration is overstating the science.

Is the Valeda device FDA approved?

The Valeda Light Delivery System received FDA De Novo marketing authorization for dry AMD in late 2024. It is a prescription, clinic-administered device used by eye-care professionals, not a consumer product, and authorization reflects a favorable benefit-risk judgment in the studied patients, not a guarantee of dramatic results.

Can I treat my macular degeneration with a home red light panel?

No, and you should not try. The AMD trials used a calibrated ophthalmic device with controlled wavelengths and very low power delivered directly to the retina. Consumer panels and masks are not tested or designed for retinal treatment, and shining them into your eyes risks harm. Standard red light therapy guidance is to protect the eyes during use.

What wavelengths are used in the macular degeneration trials?

The Valeda protocol uses multiwavelength photobiomodulation, combining yellow (590nm), red (660nm), and near-infrared (850nm) light. The foundational mechanism research from Glen Jeffery's lab focused on deep red 670nm light, which is absorbed strongly by mitochondria in retinal cells.

Who is the best candidate for retinal photobiomodulation?

Based on the trials, the strongest candidates are people with intermediate dry AMD who still have measurable central vision and who are treated before large areas of geographic atrophy develop. It was not designed for advanced central atrophy or for wet AMD, which is treated with anti-VEGF injections. A retina specialist should make the call.

Medical disclaimer

This article is for general education only and is not medical advice; talk to a qualified eye-care professional before making any decisions about AMD treatment.