AREDS2: Lutein, zeaxanthin show beneficial effect for advanced AMD

The recently published results of the second Age-Related Eye Disease Study show that, perhaps most significantly, replacing beta carotene with lutein/zeaxanthin concomitantly improved the safety of the supplement formula and increased its effectiveness in reducing progression to advanced age-related macular degeneration. Surprisingly, the study also found that, in the ratios measured, omega-3 fish oils had no statistically significant effect.

According to Jeffrey Anshel, OD, FAAO, the first step in translating the results of this study into applicable terms is to understand exactly what the study was setting out to discover.

“The trial was designed to determine whether adding lutein and zeaxanthin, docosahexaenoic acid and eicosapentaenoic acid, or both to the original formulation decreases the risk of developing advanced AMD,” Anshel said in an interview with Primary Care Optometry News. “It was also designed to evaluate the effect of eliminating beta carotene, lowering zinc doses or both in the original formulation.

“The key here is ‘developing advanced AMD,” he said. “This study was not designed to address prevention or stopping or reversing macular degeneration.”

“AREDS2 was a very different study population than AREDS1,” study chair, Emily Y. Chew, MD, said in an interview. “We took patients who had much more severe AMD status than those in AREDS1. We know those patients are likely to progress.”

The study

Along with the nutrients of the original Age-Related Eye Disease Study (AREDS) formulation consisting of vitamins C, E, beta carotene, zinc and copper, three nutritional supplements were added: lutein, zeaxanthin and eicosapentaenoic acid (EPA)/docosahexaenoic acid (DHA) from fish oil. The AREDS2 participants consented to take either the original AREDS formulation or a randomly assigned variation of the AREDS formulation, according to the study recently published in JAMA.

The primary outcome measurement was progression to advanced AMD, neovascular or central geographic atrophy determined by centralized fundus photographic grading with a standardized protocol and by history of treatment.

Participants had comprehensive ocular exams annually, which included best-corrected visual acuity testing and ocular photography.

Progression to cataract surgery and progression of lens opacity, determined by centralized ocular photography using red reflex photos with a standardized protocol and obtained at annual study visits, was a secondary outcome.

Lutein and zeaxanthin (L/Z) plus the original AREDS formula reduced advanced AMD 10% over AREDS alone in the total cohort and reduced progression to neovascular AMD by 11% over AREDS in the total cohort.

The formulation also reduced the progression to advanced AMD by 26% in subjects with the lowest dietary intake of L/Z (all P < .05).

“This just points to the fact that you are what you eat,” Chew said. “You need to have a good diet. I think that’s the significance of the quintile analysis, more than anything else. We should recommend that people need to take care of themselves and eat well.”

L/Z also reduced risk of progression to both advanced AMD and legal blindness by 18% solely by substituting L/Z for beta carotene. Risk of neovascular AMD was also reduced by 22% as a result of this substitution.

Lutein/zeaxanthin

The primary change – the addition of L/Z – that was posited for AREDS2 over AREDS1 was made for two reasons, according to Chew. First, it was determined that beta carotene had an adverse effect in smokers, increasing their chance of developing lung cancer. Almost half of the AREDS population were past smokers, and 7% were current smokers.

Second, beta carotene and lutein have the same carriers, which causes competitive absorption, Chew said.

“We saw it in AREDS1,” she said. “When you elevated beta carotene, the lutein levels in the blood went down. So when you then take away the beta carotene and look at people who are only randomized to lutein with the AREDS formulation minus beta carotene vs. the AREDS formulation without lutein, you see a beneficial effect of about 20%.”

However, this is not an additive effect, she explained.

“If you look at the general population of AREDS1, we had a 25% reduction in the risk of advanced disease. It’s not 20% on top of the 25%. It’s probably actually another 5% more. Is that a huge change? No, but it’s sufficient. It does have significant public health implications,” Chew said.

Plus, it just makes things easier, she said.

“You put in L/Z and you have one simple formulation that you can use for everyone, regardless of their smoking history,” Chew said.

Zinc

The AREDS1 formulation has been established as being effective, Michael J. Tolentino, MD, said during a webinar sponsored by Macular Degeneration Education. The problem is that some of the nutrients in that formulation have negative effects, he said.

“Zinc, for example, increases your risk for developing prostate cancer twofold,” Tolentino said. “We have a fairly good treatment for neovascular AMD, but we still don’t have a very good treatment for prostate cancer. I always put that in front of my patients; I always want to do no harm first, and I have to make sure they’re aware.

“There’s also evidence in the cancer literature that says zinc may have an effect on breast cancer,” he added. “So you’ve got to take that into account, too.”

“The amount of zinc recommended in the AREDS2 formula, 80 mg, is twice the recommended daily intake set by the Institute of Medicine board,” according to Anshel. “While this study did not show the negative effects, it might have been due to the type of zinc used, zinc oxide. Zinc oxide is not well absorbed into the cells. It works better as a sunscreen on your skin. I’ll stick with less than 40 mg.”

According to Adam S. Berger, MD, the secondary randomization suggests no difference in the progression to advanced AMD from lowering the zinc dose, and there appeared to be no differences in terms of adverse side effects between the low zinc and high zinc groups, either.

“As of now, the AREDS team is saying that there is insufficient data to make any recommendations regarding zinc,” Berger said during the webinar.

However, he disagrees with that position.

“I feel the data was fairly convincing,” Berger said. “It didn’t seem to matter whether you had high doses of zinc or low doses of zinc; it certainly would make sense to lower the dose of zinc to prevent some of the other potentially problematic effects that zinc can have on the body.”

Omega-3

“AREDS2 shows that omega-3, certainly by itself and in combination with L/Z, had no effect in either causing harmful or beneficial effects on top of the AREDS supplement,” Chew said. “We were looking for an effect over and above the AREDS1 supplement, and there was none.

“That was surprising for many people because the basic science would suggest that omega-3s should be important,” she continued. “Was there truly a lack of efficacy, or did we get the dose wrong? Are we starting too late [in the progression of the disease]? Are the ratios of the different types of omega-3s incorrect? It’s unlikely we’ll be able to run such a large trial again to test that.”

Dr. Chew said that, “Although we found no statistically significant effect of supplementing with omega-3 fatty acids, it is important to emphasize that our patients eat a diet replete with fish and leafy vegetables.”

“I cannot explain the neutral EPA/DHA omega-3 results,” Stuart Richer, OD, PhD, FAAO, told PCON, “other than to say that the placebo group was highly nourished with these important eye and brain nutrients.”

The AREDS2 findings on fish oil, in the manner it was studied, contradict AREDS1 Report #20 (diets replete with fish) and the preponderance of evidence showing a clear ocular benefit in nearly two dozen other studies, he said.

“In my patient population, I will continue to prescribe omega-3 fats because of their beneficial effects on HDL, triglycerides, blood pressure, heart rate, arrhythmias, violent behavior, mood, cognitive function, cardiovascular disease, cancer and, of course, dry eye and blepharitis,” Richer said.

French NAT 2 Study

Interestingly, recently published results of the Nutritional AMD Treatment (NAT) 2 study, presented at the meeting of the French Society of Ophthalmology, showed a significant reduction of choroidal neovascularization (CNV) incidence in patients who received a supplement of docosahexaenoic acid and had a high biological response to it.

The double-blind randomized study compared the effects of oral DHA supplementation with placebo in the prevention of exudative AMD in 298 patients with early lesions in one eye and wet AMD in the fellow eye. Subjects received either 840 mg/day of DHA and 270 mg/day of EPA from fish oil capsules or received olive oil capsules for 3 years.

According to the study, no significant difference was found in CNV incidence between the two groups at 3 years; however, 25% of the patients in the DHA group were found to have steadily high levels of DHA in the blood. Among these high responders, the DHA level in the red blood cell membrane was doubled. This suggests that CNV incidence was significantly reduced in DHA-supplemented patients showing a steadily high EPA plus DHA index over 3 years, a reduction in risk of developing CNV by almost 70%.

The factors determining the different response in this group of patients require further investigation, the study authors said.

“In the retina,” Anshel said, “most of the fatty acids are DHA, which increases cellular membrane permeability, whereas EPA is the anti-inflammatory agent. The higher amount of DHA could easily account for the differences in results between AREDS2 and NAT 2.

“We also know that the AREDS2 participants were well-nourished and may be taking more fish oil anyway; we can only speculate on why it was not effective in older individuals who were in an inflammatory state,” he said.

In the NAT 2 study, he noted, the subjects’ baseline nutrient levels are unknown.

“The well-nourished AREDS2 population sure could have confounded the study’s statistical analysis,” Richer said. “However, a more likely explanation as to why these subjects didn’t respond to omega-3, as compared to AREDS1, is the fact they were 5 years older, had more advanced AMD and had double the incidence of diabetes. The older and sicker AMD retinas are more difficult to stabilize and likely require higher doses of omega-3 fatty acids — specifically DHA for retinal structure.

“Moreover, it is doubtful that many AREDS2 AMD patients were taking enough DHA to move their red blood cell HS omega index up into the optimal 8% range, as in the NAT 2 study,” he continued. “To my knowledge, the Hollman RBC index was, like macula pigment, not measured within the primary AREDS2 protocol.”

Richer also noted that serum DHA is controlled by dietary folate, and these two nutrients should be evaluated together.

“Up to a third of the U.S. population are ‘folate nonresponders,’ unable to convert food folic acid, such as in bread and pastries, into folate,” he said. “Perhaps a later AREDS2 follow-up analysis will evaluate this important relationship. B vitamins in general and medication-induced depletion may be important as we try to understand why the geographic atrophy subpopulation failed to respond to AREDS2 supplementation.”

Cataracts

“The study results did not reveal a statistically significant impact with regard to the use of L/Z supplementation on the progression of cataracts,” Diana L. Shechtman, OD, FAAO, told PCON in an interview. “However, a subgroup analysis did reveal that among the subjects in the lowest quintile of dietary intake for L/Z, there was a significant reduction in the progression of cataracts when patients took L/Z supplementation. This was evaluated as cataract surgery, lens opacity and progression of severity of cataract.”

According to the study, subjects in that lowest quintile of dietary L/Z intake experienced a 32% reduction in progression to cataract surgery, a 30% reduction in development of any cataract (P = .02) and a 36% reduction in development of any severe cataract (P = .01).

“That’s really interesting,” Chew said, “but it’s really something that needs further evaluation and probably another study before we can make any recommendations.”

“Cataracts are a part of a slowly developing process,” Anshel said, “and other studies have shown that high levels of certain antioxidants can be effective in reducing the incidence and/or progression of cataracts. The levels used in the study were likely not sufficient to achieve such an outcome, however.”

Take-home message

“Studies like AREDS2,” Shechtman said, “and some of the controversies associated with it signify the fact that we are entering an era of personalized medicine, where looking at a case-by-case presentation, as well as genetic profiling, is crucial.”

“While evidence-based medicine (EBM) is a great slogan, and we love these large statistical studies with one-size-fits-all definitive conclusions, populations are not people, and prevention and doctors’ decision-making is always individualized,” Richer said.

“Optometrists should attempt to provide a near-optimal result for each of their patients,” he continued. “That means stepping up to the plate and actually reading level 1 EBM science with a discerning eye, as well as level 2 peer-reviewed scientific studies to obtain a full picture of who will benefit. Be skeptical but not cynical. We will either take abstracts as the gospel truth or start asking each patient what they are eating and measuring macular pigment, visual function or some other measure of retinal pigment epithelium health, such as fundus autofluorescence, parafoveal spectral domain ocular coherence tomography retinal thinning or multispectral imaging melanin migration.

“AREDS2 shows us that it’s tough to treat a disease once it becomes advanced,” Richer said. “So we had better start examining and talking to the children of AMD patients who have a 45% risk of developing AMD when early dietary advice could make a real difference in eventual outcome. AREDS2 shows us that there is no such thing as an average patient, and older and sicker eyes are more difficult to treat.”

Also, Richer noted, many analyses associated with AREDS2 data are still forthcoming.

“Remember, there were more than 30 AREDS1 reports, so there is a treasure trove of data yet to be evaluated,” he said.

For example, Anshel pointed out that genetic evaluations were conducted on only one-fourth of the study population, and those statistics have not yet been released. The same goes for macular pigment density, he said.

Criticisms

“The study is a landmark achievement, and we owe a debt of gratitude to the patients, participating scientists and physicians and the National Eye Institute,” Richer said.

However, the decision to publish in JAMA was perhaps a poor one, he said.

“One has to read the published abstract and conclusions with a discerning eye,” Richer said, “because JAMA both understates and obfuscates what the data is telling us.”

According to Richer, Chew made two separate presentations at the Association for Research in Vision and Ophthalmology (ARVO) meeting that contradicted JAMA’s primary conclusion.

“The ARVO audience is a much narrower audience than the U.S. public,” Richer said. “Other than to say that L/Z is a safer choice than high-dose beta carotene in terms of lung cancer risk, JAMA’s conclusion has generated much unnecessary confusion in the community, and scientific progress has temporarily been reversed.”

As for the study itself, one has to take into account the study design, according to Shechtman.

“AREDS2 had a complex study design with primary and secondary randomization, as well as various analyses,” she said. “This added some confusion when it came to interpreting the results. In addition, the subjects were well-nourished and, hence, the results may not apply to all of our patients.

“In addition, a 5-year follow up may not give us enough time to truly evaluate progression of a chronic degenerative disease like AMD,” she added. “Also of note, AREDS2 never looked at the early stage of the disease or those at risk for AMD. We should keep in mind that as primary care physicians, our role includes patient education and consultation and taking a proactive approach to preventing the disease from ever starting.”

However, Anshel feels the results presented by AREDS2 could have been achieved with a time period shorter than 5 years. “Macular pigments show significant increase in about 6 months,” he explained, “so visual effects might manifest within 1 to 2 years.”

He also cited the fact that the population studied was “well-nourished,” which is unlikely representative of the U.S. population in general.

“The AREDS2 study is significant, sure,” he said. “But will it change the way I address early signs of AMD in my patients? Not according to the study recommendations.”

“Another weakness of the study is that they did not incorporate mesozeaxanthin into the formula,” Berger said, “because they didn’t know about it when the study was designed. Mesozeaxanthin is the third critical carotenoid found in human luteal pigment. Not only is it a stronger antioxidant than either lutein or zeaxanthin, but it is synergistic with those other carotenoids, both in terms of its antioxidant properties and in terms of the speed and magnitude of boosting macular pigment density. It is possible that the results would have been even better had this nutrient been added to the AREDS2 formulation.”

Stratifying patients

While one of the most important findings of AREDS2 was that the formulation reduced progression to advanced stages of AMD by 26% in subjects in the quintile with the lowest dietary intake of L/Z, it has been somewhat impractical to measure carotenoid levels in patients in an everyday clinical environment, Berger said.

“However, there is now actually technology out there that can help us detect low carotenoid intake,” he said. “Measuring macular pigment optical density is a highly accurate way of determining a patient’s intake of carotenoids. The problem is, until now, there hasn’t been a good way of measuring it.

“We can now stratify our patients in terms of whether they are getting enough nutrients in their diet or whether they need to be supplemented,” he added.

“We’re actually very fortunate,” Tolentino said, “because now we understand the disease of macular degeneration so well, we can develop preventive strategies. We know the pathophysiology, we have the tools to identify people at high risk and we also have a good treatment for the development of the blinding condition of the disease, so if patients develop it, we can reverse it.

“This is a great time to be in this field, because we can help our patients,” he added. “We can help prevent them from having visual loss.” – by Daniel R. Morgan