What do Cochrane systematic reviews say about interventions for age-related macular degeneration?

ABSTRACT BACKGROUND: Age-related macular degeneration (AMD) is the third largest cause of blindness worldwide, accounting for 8.7% of all cases. A considerable number of preventive or therapeutic interventions have been used for AMD. OBJECTIVE: This study presents a critical view of the interventions that have been assessed through Cochrane systematic reviews. DESIGN AND SETTING: Review of systematic reviews, conducted in the Discipline of Evidence-Based Medicine, Escola Paulista de Medicina (EPM), Universidade Federal de São Paulo (UNIFESP). METHODS: Review of Cochrane systematic reviews about interventions for AMD. RESULTS: The 18 systematic reviews included assessed the effects of surgical techniques, laser/photo/radiotherapy, intravitreal injections, systemic drugs and phytotherapy/vitamins/supplements. CONCLUSION: The Cochrane systematic reviews found evidence that use of bevacizumab, ranibizumab, pegaptanib, laser photocoagulation, photodynamic therapy and multivitamin compounds may present some benefits for treating AMD. There was insufficient evidence for supporting the use of macular translocation, submacular surgery, steroid implantation, radiotherapy, intravitreal aflibercept, interferon alfa, statins or omega-3 fatty acids for treating AMD; or the use of multivitamin antioxidant vitamins or mineral supplementation for preventing AMD. Future randomized controlled trials are imperative to reduce the uncertainty in several clinical questions regarding AMD.


INTRODUCTION
Age-related macular degeneration (AMD) is a degenerative disease of the macula (central region of the retina) that causes loss of central vision. This type of vision is essential for performing activities of daily living. 1 AMD is the third largest cause of blindness worldwide, accounting for 8.7% of all cases of definitive loss of vision. 1 Currently, 15% to 24% of the population over the age of 65 years are affected by the early stages of AMD. 2 AMD is differentiated into the early (often asymptomatic) or intermediate stages with drusen (amorphous extracellular sediments in the retina) and characteristic pigmentary changes, and the late stages. For clinical purposes, the late stages of AMD have been classified as dry (non-neovascular and atrophic) or wet (neovascular and exudative). In the wet stages, new blood vessels can lead to leakage and tissue lesions. 3 Although the neovascular form represents only 10% of the disease burden, it is responsible for 90% of AMD-related blindness.
A considerable number of preventive or therapeutic interventions are available and have been used for both types of AMD. This study presents a critical view of the interventions that have been assessed through Cochrane systematic reviews (SRs).

OBJECTIVE
To synthetize and present the results from Cochrane SRs assessing interventions for preventing and treating age-related macular degeneration.

Design and setting
We carried out a narrative review of Cochrane SRs in the Discipline of Evidence-Based Medicine of Escola Paulista de Medicina (EPM), Universidade Federal de São Paulo (UNIFESP). This manuscript was elaborated for the section Cochrane Highlights. This initiative is a formal collaboration between the São Paulo Medical Journal and Cochrane, and it is supported by Cochrane Brazil. The aim of this initiative is to disseminate the evidence from Cochrane SRs.

Types of studies
We included only the latest published version of Cochrane SRs.
We did not consider protocols, or any SR marked as "withdrawn" in the Cochrane Database of Systematic Reviews (CDSR).

Types of participants
In relation to reviews examining therapeutic methods, we considered any participant with the diagnosis of AMD, as defined by the review authors' criteria. SRs including cases of AMD and other clinical situations were included only if the subset of data on AMD participants was provided separately. In relation to reviews examining preventive methods, no restrictions on participants were applied.

Types of intervention
We considered any surgical or pharmacological (local or systemic) intervention, compared with placebo, no intervention or any other intervention.

Type of outcomes
We considered all clinical and laboratory outcomes addressed by the SRs.

Search for reviews
We carried out a systematic search in the Cochrane Database of SRs (via Wiley) on January 8, 2019. The search strategy is presented in Table 1.

Selection of systematic reviews
The selection process was performed by two authors (RLP and RR), who independently assessed all titles and abstracts that had initially been obtained through the electronic search for potential reviews. These authors confirmed the eligibility of these SRs by assessing their full texts. Any divergences in the selection process were resolved through reaching a consensus.

Presentation of the results
We summarized and presented the following characteristics from the SRs that were included: PICOs (population, intervention, comparator and outcomes), goals, methods, main findings, certainty of evidence in accordance with the GRADE approach (Grading of Recommendations Assessment, Development and Evaluation) 4 and conclusions.

Implantable mini-telescope for diminishing loss of vision
A mini-telescope is an implantable ophthalmic device for amending visual acuity after impairment of vision due to AMD. It has been used to enlarge objects in the central visual field and focus them onto healthy areas of the retina, thus making it possible to view objects that otherwise could not be seen. This review 5 assessed the effects of an implantable ophthalmic mini-telescope in individuals with late or advanced AMD, but no randomized clinical trials (RCTs) or quasi-RCTs were found.   AMD = age-related macular degeneration; ETDRS = Early Treatment Diabetic Retinopathy Study; NA = not assessed; PDT = photodynamic therapy; RCTs = randomized clinical trials; VEGF = anti-vascular endothelial growth factor. *GRADE (Grading of Recommendations Assessment, Development and Evaluation). This system assesses the certainty of the body of evidence. High certainty of evidence means that there is high confidence that the estimated effect is near to the true effect; moderate certainty means that it is very likely that the estimated effect is close to the real effect, but there is a possibility that it is not; low certainty means that there is only limited confidence in the effect estimate; and very low certainty means that the true effect is likely to be substantially different from the estimated effect. **statistical benefit (clinical benefit is questionable); ***for ranibizumab versus bevacizumab, see the reviews relating to this, in the text. The authors could not draw any conclusions regarding this clinical question. There is one ongoing RCT that is comparing the OriLens intraocular telescope with standard low-vision training for coping with end-stage AMD. The results from this trial are expected in 2020.

Macular translocation
Macular translocation is a surgical procedure that includes displacement of the retina to a less-damaged area, which could improve vision. This review 6  The authors concluded that the current evidence was insufficient for them to be able to recommend macular translocation for AMD, which is also associated with significant harm. This technique is complicated and long surgical training is needed in order to be able to perform it.

Submacular surgery for choroidal neovascularization
Surgical removal of the submacular tissue underlying the macula (within which small new blood vessels grow) might limit the development of AMD. This review 7 aimed to assess the effectiveness of submacular surgery for preserving or improving vision in individuals with AMD and included three RCTs. Two RCTs comparing submacular surgery for AMD with observation were found, including patients with (n = 336) or without (n = 454) blood in the macula. After one year, no difference was found between the intervention arms in relation to:  13.38) were more frequent in the surgical group. Detachment was observed in 5% of the participants without extensive blood under the macula and in 18% of those with this.
In another small pilot RCT, submacular surgery was compared with laser photocoagulation (n = 70) and no difference was found for any of the outcomes measured.
The authors of the SR concluded that submacular surgery for choroidal neovascularization did not provide any benefit for individuals with AMD.

Steroid implantation (intra-and peri-ocular)
Steroids have anti-inflammatory and antiangiogenic properties that can be useful for treating AMD. This review 8  The authors of the SR did not find any evidence that antiangiogenic steroids prevented loss of vision due to AMD.

Photodynamic therapy (PDT)
PDT has been investigated as an option for managing neovascular membranes in cases of AMD without affecting the retina.
This review 11 aimed to assess the effects of PDT for neovascular revascularization in patients with AMD and included four trials (1429 participants) comparing PDT with verteporfin versus PDT with 5% dextrose in water. Verteporfin (Visudyne) is a benzoporphyrin derivative that is used as a photosensitizer during PDT.
The main findings from this review were the following: • Progression of loss of vision at 24 months:

Radiotherapy
This review 12  The frequency of adverse events was low and there were no reports of radiation retinopathy, optic neuropathy or malignancy.
The authors of this SR concluded that there was no convincing evidence that radiotherapy was an effective treatment for neovascular AMD.

Aflibercept
Aflibercept is a biological drug that blocks the biological activity of vascular endothelial growth factor (VEGF) and inhibits abnormal growth of blood vessels. This review 13

Anti-vascular endothelial growth factor
This review 14 assessed the use of anti-vascular endothelial growth factor in patients with AMD. This type of growth factor reduces proliferation of blood vessels, thus preventing AMD.

Complement inhibitors
This review 16

Interferon alpha
Interferon alpha is an antiangiogenic drug that inhibits migration and proliferation of vascular endothelial cells. This review 17 assessed the use of interferon alpha for treating AMD and included one RCT (481 participants). In comparison with placebo, use of interferon alpha was associated with worse results, consisting of loss of three or more lines of vision at 52 weeks (OR 1.60; 95% CI 1.01 to 2.53; one RCT; 391 participants).
This review was published in 2006 and did not assessed the certainty of evidence.
Further RCTs are needed in order to increase confidence in this estimate. The next update of this review will probably assess the certainty of the evidence using the GRADE approach.

Statins
Recent studies have shown that AMD and atherosclerosis present some risk factors in common, and that statins may present benefits for patients with AMD. This review 18  • Drusen score and visual function results were reported to be similar between the groups at 12 months (42 participants), but no effect estimates or confidence intervals were provided.
• Progression of AMD: there was no difference between the groups at three years (OR 0.51; 95% CI 0.23 to 1.09; low certainty of evidence).
• Adverse events: only one RCT reported adverse outcomes, and it was stated that there were no differences between the groups regarding death, muscle aches or acute hepatitis.
The authors of this SR concluded that the current evidence from RCTs was insufficient to confirm that statins had any benefit with regard to preventing or delaying the onset or progression of AMD.

Gingko biloba
Ginkgo biloba extracts are used for treating some health conditions, including peripheral vascular diseases, and may present benefits for treating AMD. This review 21 assessed ginkgo biloba extract for patients with AMD and included two RCTs (119 participants).
In these RCTs, it was reported that ginkgo biloba provided some benefits, but there was insufficient data to pool the results.
The outcomes reported in the RCTs were generally different from those of relevance for the review, and the safety results were very sparse. The certainty of evidence was not assessed. Further RCTs are needed in order to reduce the uncertainty of the evidence and to provide a basis for practical recommendations.
For further details and to access all the analyses, see the orig-  No published RCT was found assessing: (a) use of an implantable ophthalmic mini-telescope device for improving visual acuity after impairment of vision due to AMD (results from an ongoing trial are expected to be published in 2020); and (b) use of complement inhibitors. Among the 18 SRs included, four did not assess the certainty of the body of evidence based on the GRADE approach, since they were developed before this approach became recommended as mandatory in the Cochrane Handbook. It is strongly desirable that SRs should be updated after two years have elapsed, or more frequently if new studies are available. Indeed, the lack of an approach of this nature for supporting SR conclusions is a factor that limits practical applicability.
We observed an issue involving the comparison between ranibizumab and bevacizumab, which was addressed through two different SRs and led to an overlapping of safety assessments. The first SR focused on the overall effects (benefits and harm) of any intravitreal anti-VEGF drug. 14 The second SR focused on safety outcomes for the single comparison of ranibizumab versus bevacizumab. 15 Mainly because of differences between the methodological assumptions used for each SR, the findings regarding serious adverse events were inconsistent between these two reviews. Overlapping of PICOs in Cochrane SRs needs to be avoided, and it is uncommon.
Specifically, in this context, considering the debate around off-label use of bevacizumab for treating AMD, a second Cochrane SR was developed in an attempt to address safety concerns. Additional ongoing Cochrane SRs addressing other interventions for treating AMD will be available over the coming months and may contribute towards expanding the body of evidence available for management of AMD.
Further well-designed and well-conducted randomized controlled trials are still necessary, in order to reduce the uncertainties regarding the clinical questions that surround AMD.

CONCLUSION
This review found 18 Cochrane systematic reviews that evaluated interventions for preventing or treating AMD. Overall, use of bevacizumab, ranibizumab, pegaptanib, laser photocoagulation, photodynamic therapy and multivitamin compounds may present some benefits for treating AMD. Further randomized controlled trials are still necessary, in order to reduce the uncertainties regarding most clinical questions that surround AMD.