Nutritional Interventions Against AMD in the AREDS2 Era

What the AMD patient eats really matters.

PAUL S. BERNSTEIN, MD, PhD

Retinal physicians have participated in a spectacular expansion of treatment options for age-related macular degeneration over the past two decades. In the 1990s, we had the unenviable choices of just observing AMD’s course of relentless progression to legal blindness or else performing macular photocoagulation with a potential immediate loss of several lines of vision in the hope of providing better visual acuity than the natural history of the disease a few years in the future.¹

Now, in the anti-VEGF era, we have effective treatments for patients with exudative AMD that can stabilize and even improve vision, but these advances come at a price to patients, physicians, and society because patients may have to return monthly for intravitreal injections of expensive medications.

Moreover, no treatments for geographic atrophy have passed the rigorous FDA approval process yet. Thus, there is considerable interest on the part of both physicians and patients to develop and optimize strategies to delay or prevent visual loss from AMD.

Over the past 30 years, diverse epidemiological studies have identified multiple risk factors for AMD. Some of these factors are unmodifiable, such as age, heredity, and light skin color, while other powerful, modifiable risk factors, such as cigarette smoking, apply to a minority of the US population. However, nutritional status occupies a unique position as a modifiable AMD risk factor that can be assessed and potentially improved in almost everyone.

The American public readily embraces diets and supplements because these interventions empower them to act against perceived health threats, but there is a wealth of contradictory and confusing information for them to sort through, and physicians often feel inadequately trained to field their questions and to offer advice.

This review offers some guidance for retinal specialists when they counsel their patients who wish to modify their risk of AMD through diet or supplements.

EARLY NUTRITIONAL INTERVENTIONS AGAINST RETINAL DISEASE

Retinal disorders were among the earliest human health conditions linked to nutrition. Ancient Egyptian papyri document recommendations to consume liver to treat night blindness, which we now recognize as a manifestation of vitamin A deficiency. Chinese traditional medicine has long advocated zeaxanthin-rich goji berries and their extracts for eye health, and during World War II, Royal Air Force pilots reportedly consumed bilberry anthocyanins in pies and jams to enhance their night vision.

Thus, it was not a great conceptual leap for clinicians and scientists to consider whether or not nutrition and antioxidants could play a role in the prevention and treatment of AMD, especially when there was a growing appreciation that chronic diseases of aging such as AMD could be precipitated and exacerbated by oxidative stress.

Large-scale epidemiological studies and small-scale clinical trials of the 1970s and 1980s provided intriguing leads and testable hypotheses that could guide the design of larger prospective clinical trials, but at that time, there was still very little scientific evidence to support clinicians’ recommendations for any nutritional interventions against AMD.

AREDS PROVIDES SOME ANSWERS

In response to growing interest in nutritional supplements for the prevention of visual loss from AMD and cataracts, the National Eye Institute initiated the Age-Related Eye Disease Study, the first adequately powered, randomized, placebo-controlled clinical trial to test whether defined nutritional supplements could modify the course of eye diseases associated with aging.

The AREDS investigators distilled the best nutritional knowledge of the 1980s to come up with a four-group clinical trial including: (1) zinc; (2) antioxidants; (3) zinc + antioxidants; and (4) placebo. They chose zinc oxide at a dose of 80 mg per day (along with 2 mg of cupric oxide per day to prevent zinc-induced anemia), based on an intriguing small clinical trial that showed a beneficial effect against intermediate AMD.²

The antioxidant arm tested 500 mg of vitamin C, 400 IU of vitamin E, and 25,000 IU (15 mg) of β-carotene after considering their general safety and efficacy in alleviating oxidative stress and after reviewing multiple epidemiological studies of dietary consumption of these antioxidants, linking them with lower incidence of AMD and cataracts.

The interventional portion of the study focused on high-risk intermediate AMD (large soft drusen and pigmentary changes) or advanced AMD in one eye and intermediate AMD in the fellow eye, but lower-risk individuals were also recruited for natural history investigations and risk factor analyses.

AREDS followed more than 3,500 subjects for at least five years at 11 sites and announced the primary endpoint results in 2001.³ All three active treatment groups exhibited a statistically significant 20% to 25% reduction in progression to advanced AMD, with the zinc + antioxidants group performing the best. There were no detectable benefits with respect to cataract prevention.

AREDS Formulations Appear

AREDS-compliant formulations rapidly became the standard of care in the United States and elsewhere because they were reasonably priced, well-tolerated interventions that could reduce the treatment burden and risk of blindness from a common public health problem affecting millions of people, but these recommendations were not without controversy.

Specifically, there was considerable concern about the dose of β-carotene, because halfway through the trial, several studies reported an increased risk of lung cancer in smokers who consumed high-dose β-carotene supplements.^4,5 Also, the doses of zinc oxide and vitamin E were considered too high by some researchers due to potential increases in urinary, gastrointestinal, and cardiovascular complications.

Moreover, nutritional science had progressed dramatically since the inception of the original AREDS study, so there was a strong perception that the formulation could be improved⁶ and that a new study, AREDS2, was warranted, with particular attention paid to other carotenoids and to omega-3 fatty acids.

AREDS2

AREDS2 was patterned after AREDS in size and duration, but it included many more clinical sites across the country to provide more diversity and generalizability. This time, the four main treatment arms were: (1) lutein/zeaxanthin (10 mg and 2 mg daily, respectively); (2) EPA/DHA (650 mg of eicosapentaenoic acid and 350 mg of docosahexaenoic acid daily from omega-3-rich fish oil); (3) lutein/zeaxanthin + EPA/DHA; and (4) placebo.

Because the original AREDS formulation had already been proved effective, all of the subjects were offered the original AREDS formula or assignment to a modified AREDS formula, in addition to their study randomization, and nearly all of the subjects accepted this secondary assignment or randomization.

Current smokers were assigned to a modified AREDS formula without β-carotene. Nonsmokers and past smokers were secondarily randomized into four categories: (1) original AREDS; (2) original AREDS without β-carotene; (3) original AREDS with low zinc (25 mg of zinc oxide per day); and (4) original AREDS with low zinc and no β-carotene.

Lutein and zeaxanthin were chosen because these xanthophyll carotenoids are specifically concentrated in naturally high amounts at the fovea, where they form the yellow macular pigment.⁷ They are exclusively derived from the diet, primarily from dark green leafy vegetables and orange and yellow fruits and vegetables.

In contrast, β-carotene cannot serve as a precursor for lutein or zeaxanthin, and it is not typically detectable in the human retina because, unlike lutein and zeaxanthin, it is rapidly converted to vitamin A by endogenous cleavage enzymes.⁸

Americans typically consume 1-2 mg of lutein and ~0.2 mg of zeaxanthin daily from the diet, but vegetarians may easily consume five to 10 times these amounts daily. These carotenoids are thought to act as localized, highly efficient blue-light absorbers and antioxidants in a region of the eye that is very vulnerable to light-induced oxidative stress,⁹ and they are remarkably safe, with no known link to elevated cancer risk.

The 1993 publication of Seddon et al’s analysis of dietary data from the Eye Disease Case-Control (EDCC) study particularly caught the attention of the AREDS2 planners because it reported that there was a 43% reduction in exudative AMD risk when the highest quintile of lutein and zeaxanthin consumption (>6 mg per day) was compared to the lowest quintile (<1 mg per day),¹⁰ and similar findings were reported when the AREDS investigators analyzed the dietary intakes of the AREDS cohort.¹¹ Based on these data and some in-house dose-ranging studies, 10 mg of lutein and 2 mg of zeaxanthin were chosen as the dosages to study in AREDS2.

Numerous epidemiological studies have noted that individuals who consume several servings of fish per week have a lower risk of AMD relative to those who do not consume fish regularly.^{12, 13} This finding makes sense physiologically because the retina has extraordinarily high levels of long-chain omega-3 fatty acids in the photoreceptor membranes, and many of these lipids are derived from the EPA and DHA found in abundance in fish oil. Many different formulations and dosage levels of omega-3 fatty acids were considered for AREDS2, but they ultimately chose 1,000 mg daily of EPA/DHA at a 2:1 ratio.

Bad News and Good News

When the AREDS2 results were announced in 2013, the investigators had to report that none of the primary endpoints had been reached because, when each active group was compared to placebo, none achieved >25% additional reduction in AMD risk.¹⁴ However there was a wealth of positive information directly relevant to clinical practice that could be discerned from preplanned secondary analyses.¹⁵

For example, main-effect analyses of lutein/zeaxanthin provided a statistically significant reduction of AMD risk when the cohorts receiving lutein/zeaxanthin were compared to those who did not. This positive effect was enhanced when the researchers compared the subgroups that received lutein/zeaxanthin without β-carotene to the subgroups that received β-carotene alone, in part because the β-carotene had interfered with absorption of lutein and zeaxanthin from the gut.

Moreover, lutein and zeaxanthin again appeared to be safer than β-carotene because past smokers exhibited increased risk of lung cancer when they received β-carotene, while the lutein/zeaxanthin groups had no such association with elevated risk of malignancy.

In contrast to the abundance of positive secondary results for lutein and zeaxanthin, there were no significant positive or negative effects for the fish oil group, and the investigators detected no significant safety or efficacy reasons to reduce zinc oxide from 80 mg to 25 mg daily.

Based on the preplanned secondary analyses, the AREDS2 investigators recommended that, when safety and efficacy are considered, 10 mg of lutein and 2 mg of zeaxanthin constitute an appropriate substitution for the β-carotene in the original AREDS formulation for all current and former smokers who meet the high-risk criteria for AMD required for entry into the AREDS2 study.

For simplicity’s sake, it is reasonable to recommend the modified AREDS2 formula to nonsmokers as well. The AREDS2 investigators did not recommend adding fish oil, and they were neutral on changing the zinc.

AREDS3?

Just like the original AREDS study, the results from AREDS2 have been rapidly incorporated into clinical practice, but will there be an AREDS3? In the current research funding environment, it is very unlikely that another large-scale clinical trial will be conducted in the foreseeable future, but it is interesting to speculate what could be next.

First, the current AREDS2 formula could be optimized with respect to lutein and zeaxanthin ratio and content, preferably in conjunction with noninvasive assessment and longitudinal monitoring of macular pigment optical density and distribution using optical or psychophysical methods.¹⁶

Also, there is greater awareness that a third carotenoid, meso-zeaxanthin, is present in equal amounts relative to lutein and zeaxanthin in the human macula. Meso-zeaxanthin is not ordinarily encountered in the human diet and is thought to be a metabolite of lutein, but it is now available commercially and certainly deserves consideration as a possible component of a potential AREDS3 supplement if research studies can demonstrate comparable safety and enhanced efficacy relative to lutein and zeaxanthin alone.¹⁷

Fish oil supplementation should be investigated further in different doses and formulations. Numerous nutraceuticals consumed by retina patients containing polyphenols, anthocyanins, and flavonoids derived from botanicals, such as bilberry extracts and red wine, could be rich sources of new compounds to test in randomized, clinical trials.

Finally, we now live in the genomic era, so more and more investigators are examining whether genetic risk factors for AMD influence response to nutritional supplementation. Recent retrospective genetic reanalyses of the AREDS subgroups have proved exceedingly controversial,¹⁸ so it will be interesting to see how similar studies of the AREDS2 population turn out.

CURRENT RECOMMENDATIONS FOR AMD

When retina specialists counsel their AMD patients, it is still important to emphasize diets rich in fruits, vegetables, and fish as first-line interventions because epidemiological studies consistently demonstrate their benefits, and such diets will have further value for the patient’s general health.

Realistically, however, changing dietary habits late in life is likely to be very difficult, so AREDS2 supplements can provide rational, scientifically based recommendations for patients at significant risk for vision loss from AMD (see box on page 28). When it comes to the “worried well,” such as the middle-aged children of AMD patients who do not yet show signs of AMD themselves, it is best to strongly emphasize healthy diets and lifestyles first and to then counsel that dietary supplements might be helpful, but that there are no definitive data to support such recommendations at this time.

We retina specialists are indeed fortunate to have straightforward, relatively inexpensive, evidence-based nutritional recommendations to offer our AMD patients to help delay or prevent a common but dreaded disorder. Many other medical specialties do not currently have such options. RP

REFERENCES