900 Diabetes Care Volume 37, April 2014

Ning Cheung,1,2,3 Ian Y. Wong,1 and Ocular Anti-VEGF Therapy for Tien Y. Wong2,3 BENCH TO CLINIC SYMPOSIA Diabetic Retinopathy: Overview of Clinical Efficacy and Evolving Applications Diabetes Care 2014;37:900–905 | DOI: 10.2337/dc13-1990

Ocular anti-vascular endothelial growth factor (VEGF) therapy represents one of the most significant advances in modern medicine. The introduction and widespread use of ocular anti-VEGF therapy for age-related macular degenera- tion heralded a new era in the treatment of vascular and exudative diseases of the retina. Its expanding indications now include diabetic macular edema and proliferative diabetic retinopathy, two vision-threatening forms of diabetic retinopathy. It is widely anticipated that ocular anti-VEGF therapy could spark a dramatic shift in the treatment paradigm for diabetic retinopathy. However, despite its clear efficacy shown in clinical trials, the dynamic landscape of evolving medical, ethical, and economic issues related to this new treatment suggests significant challenges ahead. In this article, we provide a discussion of this topic as part of this two-part Bench to Clinic narrative. Here, our Clinic contribution provides an overview of the current evidence from clinical trials on anti-VEGF therapy for diabetic retinopathy, and highlights the hopes and fears of this new treatment from clinical and public health standpoints. In the Bench narrative that precedes this contribution, Simo´ et al. provide an overview of the role of VEGF in the pathogenesis of diabetic retinopathy.

Ocular anti-vascular endothelial growth factor (VEGF) therapy represents one of the most significant advances in modern medicine. The swift and widespread uptake of this new therapy into clinical practice for treating age-related macular 1 degeneration has saved sight for millions worldwide (1). In fact, national blindness Department of Ophthalmology, The Eye Insti- tute, University of Hong Kong, China Special Ad- registries are already showing declining incidence of blindness related to age- ministrative Region, China related macular degeneration, coinciding with the advent of anti-VEGF therapy 2Singapore Eye Research Institute, Singapore Na- (2). Despite its clear efficacy, however, the safety, cost, and substantial burden tional Eye Centre, National University of Singa- pore, Singapore upon the health care system of this new treatment have generated heated debates 3 in many countries (3). Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital, University of Mel- Now, it is widely anticipated that the use of ocular anti-VEGF therapy will be bourne, Melbourne, Australia extended to treat the vision-threatening forms of diabetic retinopathy (4), which Corresponding author: Ning Cheung, affect an estimated 28 million people around the world (5). In this two-part Bench [email protected]. to Clinic narrative, the Bench article by Simo´ et al. (6) reviews the pathophysiological Received 23 August 2013 and accepted 17 role of VEGF in diabetic retinopathy and the molecular characteristics of antiangio- December 2013. genic agents currently used. Here in the Clinic article, we provide an overview of the © 2014 by the American Diabetes Association. current evidence from clinical trials on anti-VEGF therapy for diabetic retinopathy, See http://creativecommons.org/licenses/by- and highlight the hopes and fears of this new treatment from the clinical and public nc-nd/3.0/ for details. health standpoints. See accompanying article, p. 893. care.diabetesjournals.org Cheung, Wong, and Wong 901

EPIDEMIOLOGY AND NATURAL and PDR among patients with recently steroids (e.g., triamcinolone) have dem- HISTORY OF DIABETIC diagnosed diabetes (10). These encour- onstrated ability to reduce DME and RETINOPATHY aging findings reflect improvement in improve vision, these beneficial effects As a global concern, diabetes affects the systemic management of retinopa- appear to be short-lived, and long-term more than 360 million individuals world- thy risk factors over time for a range of visual outcome was generally not better wide. This number is expected to exceed reasons, such as better devices for self- than conventional laser therapy (4). half a billion by 2030 (7). About one in monitoring of glycemic levels and Furthermore, repeated use of intraocu- three individuals with diabetes has administration of insulin, new and lar steroid injections is associated with signs of retinopathy, and among these, effective hypoglycemic medications, significant ocular side effects (e.g., cat- one-third may have diabetic macular and increased public awareness of the aract, glaucoma). Nevertheless, there edema (DME) or proliferative diabetic ret- need for glycemic and blood pressure are certain advantages in using intraoc- inopathy (PDR), two vision-threatening control through educational and ular steroids (e.g., possibly longer-acting forms of diabetic retinopathy (4). A re- screening programs. Despite these and relatively cheap compared with cent pooled analysis of 35 population- advances in diabetes care, it remains most anti-VEGF agents). Its use might fi based studies in developed countries uncertain whether such a declining thereforebebene cial for selected estimated that more than 90 million trend in the incidence of diabetic patients, such as those who have had individuals have diabetic retinopathy, retinopathy will persist in the context previous cataract surgery, or as an ad- with about 21 million having DME and of expanding diabetes epidemic world- junctive therapy prior to laser (12,13). 17 million having PDR (5). wide, particularly in developing countries OCULAR ANTI-VEGF THERAPY FOR In the Wisconsin Epidemiologic Study where intensive diabetes management DIABETIC RETINOPATHY of Diabetic Retinopathy, about three in and public health resources remain four participants developed retinopa- limited (11). The introduction and widespread use of ocular anti-VEGF therapy for age- thy over a 10-year period, and for CURRENT STRATEGIES FOR related macular degeneration, with participants with retinopathy, about MANAGEMENT OF DIABETIC publication of major clinical trials (1), two-thirds developed more severe RETINOPATHY retinopathy and one in five developed heralded a new era in the treatment of Systemic management of hyperglyce- PDR (4). In terms of progression, dia- vascular and exudative diseases of the mia, hypertension, and dyslipidemia betic retinopathy progresses from retina. The expanding indications for remains the most important and nonproliferative to proliferative reti- ocular anti-VEGF therapy, given via an effective strategy for preventing the nopathy in stages. Nonproliferative di- injection into the vitreal cavity, now development and progression of dia- abetic retinopathy (NPDR) is classified include DME and PDR. betic retinopathy (4). For many decades, as mild, moderate, and severe forms. retinal laser photocoagulation has been Efficacy About 5% mild NPDR, 20% moderate the standard ocular treatment for DME AsshownintheaccompanyingBench NPDR, and 50% severe NPDR may and PDR (4,10). The primary goal for article by Simo´ et al. (6), VEGF has progress to PDR within 1 year (4). most patients receiving laser therapy is long been a therapeutic target for dia- In developed countries, DME has now to preserve any useful vision or to pre- betic retinopathy. In recent years, there overtaken PDR as the more common vent adverse sequalae of PDR. Reversal has been a surge of clinical trials vision-threatening form of diabetic reti- of vision loss is uncommon. In addition, investigating the use of anti-VEGF ther- nopathy, particularly among patients laser therapy is associated with signifi- apy for DME (Table 1) (14–16). These with type 2 diabetes. In the National cant ocular side effects due to its inher- trials provide robust evidence that Health and Nutrition Examination ent destructive nature to the retina. intraocular administration of anti- Survey, DME was shown to be twice as Without timely laser therapy, however, VEGF agents is better than laser therapy commonasPDRintheU.S.(8).The patients may develop blinding neovas- both in preserving and in improving vi- 10-year incidence of DME has been cular complications, such as vitreous sion for patients with DME. Among the reported to be 20% in the Wisconsin hemorrhage and tractional retinal de- four anti-VEGF agents (ranibizumab, Epidemiologic Study of Diabetic Reti- tachment, leading to the need for bevacizumab, pegaptanib, and afliber- nopathy (9). Although DME is usually surgical intervention (vitrectomy). cept), ranibizumab has been the one correlated and accompanied with Over the last decade, intraocular most thoroughly tested. In randomized increasing severity of retinopathy, it administration of pharmacological controlled trials that used ranibizumab may also run an independent course agents (e.g., steroid and anti-VEGF injections, up to 46% of patients and develop even at the early stage of agents) has been evaluated as a new improved vision (vs. 18% with laser diabetic retinopathy. treatment modality for DME and PDR alone; by three lines or more on vision There is evidence to suggest a decline (4,10). Delivery of these agents is chart), and only 4% or less lost more in the incidence and risk of progression achieved by direct injection into the vision (vs. up to 20% with laser alone). for diabetic retinopathy over the last vitreal cavity, a procedure that is usually The studies also suggest that, compared three decades (4,10). The incidence of performed in office setting by with laser therapy alone, ranibizumab visual impairment among people with ophthalmologists using aseptic tech- injections were more effective when diabetic retinopathy has also halved, nique and topical anesthesia. Although used as a monotherapy or in combina- likely as a result of a lower risk of DME intraocular injections of long-acting tion with laser therapy in treating DME 902 Ocular Anti-VEGF Therapy for Retinopathy Diabetes Care Volume 37, April 2014

Table 1—Major recent randomized controlled trials of ocular anti-VEGF therapy for DME Results Number of study Follow-up Trial participants/eyes Anti-VEGF agent Gained vision* Lost vision* (months) RISE and RIDE (42) 377 Ranibizumab 34–46 (12–18) 2–4(9–10) 24 DRCRnet (13) 854 Ranibizumab (1 laser) 28–30 (15) 2 (8) 24 READ-2 (43) 126 Ranibizumab (1/2 laser) 23 (17) 3 (6) 24 RESOLVE (44) 151 Ranibizumab (1/2 laser) 32 (10) 3 (20) 12 RESTORE (45) 345 Ranibizumab (1/2 laser) 23 (8) 1–3 (8) 12 BOLT (46) 80 Bevacizumab 32 (4) 0 (14) 24 Macugen 1013 (47) 207 Pegaptanib (1/2 laser) 23 (15) 3–4(6–9) 24 da Vinci (20) 176 Aflibercept 46 (11) d 12 *, % of patients with $3-line vision gain or loss (vs. with laser therapy alone).

(17). In patients receiving combined demonstrating similar efficacy in re- concerns include cataract formation, in- ranibizumab and laser therapy, best ducing DME based on optical coher- fection (endophthalmitis), vitreous long-term visual outcome could be ence tomography findings (primary hemorrhage, and retinal detachment. achieved with initiation of injections outcome), results on visual outcome The rates of serious sight-threatening followed by deferred laser therapy in this study were considered inconclu- complications are acceptably low, as 6 months later (17). Unlike neovascular sive due to inadequate power (21). shown in studies of not only patients age-related macular degeneration, At present, the role of ocular anti- with diabetic retinopathy, but also of vision gain resulted from ranibizumab VEGF therapy for PDR is less clear, al- patients with age-related macular de- injections in patients with DME could though nationwide studies by groups generation (1,4,16). be maintained with tapering of injec- such as DRCRnet are under way to ad- However, the inherent study design tion frequency over time (17,18). For dress this question. Exploratory analysis of clinical trials hampers the ability to example, the Diabetic Retinopathy from DRCRnet provided the basis for adequately assess systemic safety of Clinical Research Network (DRCRnet) further investigation into the role of in- rare but important events of interest suggests that the average number of traocular anti-VEGF and steroid therapy (e.g., stroke, ischemic heart disease) be- injections in the first, second, and third in reducing risk of retinopathy progres- cause of potential selection bias and lim- year of treatment for DME was 9, 3, and 2, sion (22). Preliminary data from the ited power. The basis of systemic safety respectively, to maintain vision gained DRCRnet did not show significant concern roots from the known risk of (17). Exploratory analysis of trial data short-term benefit of ranibizumab injec- serious adverse events associated with demonstrated that ranibizumab injec- tions in reducing need for surgical in- intravenous anti-VEGF therapy used in tions could reduce risk of progression tervention (vitrectomy) for PDR-related cancer patients, evidence of systemic and increase likelihood of regression of vitreous hemorrhage in the first 4 months. absorption after intraocular anti-VEGF diabetic retinopathy severity among pa- Nonetheless, positive effects were ob- injections (28), and possible safety sig- tients with DME (19). served on secondary outcomes, includ- nals from large epidemiological studies The evidence for the use of the other ing visual acuity improvement, increased of age-related macular degeneration anti-VEGF agents is less robust due laser completion rates, and reduced re- (29–31). Potential adverse effects of sys- to the smaller number of trials with gen- current vitreous hemorrhage rates (23). temic VEGF blockade that are particu- erally shorter follow-up. Nevertheless, Ongoing follow-up of these patients will larly worrisome for diabetic patients all trials reported to date suggest a ben- hopefully offer more clarity in the value include hypertension, proteinuria, im- eficial response to anti-VEGF agents for of anti-VEGF therapy for treating PDR. paired wound healing, and critical vas- DME (Table 1). Being a synthetic fusion While anti-VEGF agents might be useful cular responses to ischemia (32). These protein that has been specifically as a primary treatment, or adjunct to effects may amplify the cardiovascular designed to act like an antibody, afliber- laser or surgical treatments for ad- risk among diabetic patients, particu- cept may require less frequent injec- vanced PDR (24), its use has been re- larly in those with retinopathy, who al- tions and follow-up due to its longer ported to possibly accelerate the ready have two- to threefold higher half-life and durability (20). There is, development or progression of trac- risk of stroke, coronary heart disease, however, a lack of data on comparative tional retinal detachment in a small per- and heart failure than those without efficacy between aflibercept and ranibi- centage of cases (25). retinopathy (33). zumab injections. Despite being an “off-label” therapy, Safety Clinical and Public Health intraocular bevacizumab injections Safety is paramount for any new treat- Implications are commonly used as a much more ment strategy. Although most clinical Despite improvements in diabetes care, affordable alternative to ranibizumab. trials reported a favorable safety pro- the prevalence of diabetic retinopathy A small clinical trial recently compared file, data beyond 2 years of exposure will likely continue to rise, due to popula- the efficacy between ranibizumab and for repeated intraocular anti-VEGF ther- tion growth, aging demographics, and ex- bevacizumab in treating DME. While apy are limited (17,26,27). Ocular safety panding diabetes epidemic worldwide. care.diabetesjournals.org Cheung, Wong, and Wong 903

There are 20 to 30 million individuals with these patients was $1.5 billion between evidence to suggest similar efficacy of DME in developed countries (5). These 2008 and 2009 alone (34). With another ranibizumab and bevacizumab for numbers may double by 2030. Thus, the million adults with vision-threatening treating age-related macular degenera- demand for eye care service will pro- diabetic retinopathy in the U.S. (8), the tion (37), quality data on the compara- foundly increase if anti-VEGF therapy is extent to which this additional financial tive efficacy and safety of these two adopted as the standard treatment for burden will affect the health care sys- agents for DME are still lacking. DME. Such demand is currently un- tem is bound to be significant. Cost- Another major challenge relates to matched by the workforce supply of oph- effectiveness analyses have shown that patient access to ocular anti-VEGF ther- thalmologists, even in many developed substantial cost savings (40–88%) could apy in less developed or developing countries. Alongside the actual procedure be achieved by individualized treatment countries (e.g., India, China, South of the injection itself, the need for regular, strategies for DME (35). Assuming America), where the prevalence of sometimes monthly, follow-up and equivalent effectiveness and similar vision-threatening diabetic retinopathy monitoring of treatment response adds safety profiles between bevacizumab might increase the most in the up- further stress to most health care systems. and ranibizumab injections, the use of coming years. Like any new and expan- The economic burden, for both pa- bevacizumab confers much greater value sive therapies, accessibility is an tients and the society, is a major concern. among different treatment options for inevitable problem due to disparities in Even in the U.S, the economic impact is DME (36). This is due to the substantial health care availability, access, and qual- substantial. For example, over a million cost differential between the two anti- ity entrenched internationally between adults have neovascular age-related mac- VEGF agents. Ranibizumab is up to 40 times developed and developing nations as ular degeneration in the U.S, and the cost more expensive than bevacizumab in well as within countries. There is no sim- of providing ocular anti-VEGF therapy for some countries. Although there is now ple solution, but it should not be

Figure 1—Clinical pathways for ocular treatments of DME. *Clinically significant macular edema is defined by the Early Treatment Diabetic Retinopathy Study (4). 904 Ocular Anti-VEGF Therapy for Retinopathy Diabetes Care Volume 37, April 2014

the reason to overlook such issue. Rather, are important in the pathogenesis of di- contracts of employment, or named positions it should form the basis for more abetic retinopathy (e.g., inflammation, on company boards. No other potential con- fl research in this uncharted area, and to renin-angiotensin) (4). Therapies tar- icts of interest relevant to this article were reported. endorse public health endeavors that geting these pathways (intraocular aim to improve access and cost- steroids, renin-angiotensin blockade) References effectiveness in the delivery of ocular have been shown to have positive ef- 1. Lim LS, Mitchell P, Seddon JM, Holz FG, Wong anti-VEGF therapy to patients with vision- fects in treating diabetic retinopathy TY. Age-related macular degeneration. Lancet threatening diabetic retinopathy in these (4,10). The effects of combining these 2012;379:1728–1738 countries. local and systemic treatments with oc- 2. Cheung N, Wong TY. Changing trends of blindness: the initial harvest from translational ular anti-VEGF therapy remain to be public health and clinical research in ophthal- Unanswered Questions and Future determined. mology. Am J Ophthalmol 2012;153:193–195 Research 3. Cheung CM, Wong TY. Treatment of age- Although evidence supports the use of CONCLUSIONS related macular degeneration. Lancet 2013; anti-VEGF therapy for treating diabetic 382:1230–1232 retinopathy, several key questions remain Visual impairment exerts considerable 4. Cheung N, Mitchell P, Wong TY. Diabetic ret- deleterious impact on quality of life inopathy. Lancet 2010;376:124–136 unanswered. First, it is not a cure. Despite 5. Yau JW, Rogers SL, Kawasaki R, et al.; Meta- and activities of daily living among pa- the possibility of reducing the number of Analysis for Eye Disease (META-EYE) Study anti-VEGF injections over time, repeated tients with diabetic retinopathy. Impor- Group. Global prevalence and major risk factors injections are required to maintain visual tantly, visual impairment may also affect of diabetic retinopathy. Diabetes Care 2012;35: – benefits for many patients. This is due to their ability to manage diabetes and 556 564 other complications. Ocular anti-VEGF 6. Simo´ R, Sundstrom JM, Antonetti DA. Ocular the relatively limited half-lives of the cur- anti-VEGF therapy for diabetic retinopathy: the rently available anti-VEGF agents. While therapy has sparked a dramatic shift in role of VEGF in the pathogenesis of diabetic aflibercept may have longer half-life, the treatment paradigm for diabetic ret- retinopathy. Diabetes Care 2014;37:893–899 fi there is a lack of evidence to date that it inopathy (Fig. 1). Its indisputable ef - 7. International Diabetes Federation. IDF Dia- betes Atlas, 2011. 5th ed. Available from could be used less frequently than other cacy shown in trials has already called for experts to revise clinical and thera- http://www.idf.org/diabetesatlas. Accessed 1 anti-VEGF agents (e.g., ranibizumab) to August 2013 achieve similarly favorable visual out- peutic guidelines, recommending its use 8. Zhang X, Saaddine JB, Chou CF, et al. Preva- fi come. Hence, there is need for studies in some instances as the rst-line pri- lence of diabetic retinopathy in the United – on comparative efficacy between the mary therapy for DME (14). However, States, 2005-2008. JAMA 2010;304:649 656 the dynamic landscape of evolving med- 9. Klein R, Klein BE, Moss SE, Cruickshanks KJ. anti-VEGF agents, and an ongoing effort The Wisconsin Epidemiologic Study of Diabetic to discover new antiangiogenic agents ical, ethical, and economic issues related fi Retinopathy. XV. The long-term incidence of with longer ocular half-lives or novel de- to this new treatment suggests signi - macular edema. Ophthalmology 1995;102:7–16 livery mechanisms (e.g., ocular implants) cant challenges ahead, with legitimate 10. Antonetti DA, Klein R, Gardner TW. Dia- to prolong the effects of anti-VEGF agents concerns regarding systemic safety, betic retinopathy. N Engl J Med 2012;366: 1227–1239 in the eye. cost-effectiveness and sustainability of health care delivery. Furthermore, al- 11. Chan JC, Malik V, Jia W, et al. Diabetes in The use of optical coherence tomog- Asia: epidemiology, risk factors, and pathophys- raphy (OCT) has allowed precise assess- though ocular anti-VEGF therapy could iology. JAMA 2009;301:2129–2140 ment of structural changes in DME in substantially reduce visual impairment 12. Maia OO Jr, Takahashi BS, Costa RA, Scott qualitative and quantitative manners from diabetic retinopathy, ultimately it IU, Takahashi WY. Combined laser and intravi- treal triamcinolone for proliferative diabetic (4). As a component of the diagnostic is not a cure. Only through a continua- tion of the critical ongoing efforts to retinopathy and macular edema: one-year re- algorithms used in major clinical trials, sults of a randomized clinical trial. Am J Oph- OCT has in fact become an indispensible understand pathophysiological mecha- thalmol 2009;147:291–297.e2. tool to manage anti-VEGF therapy for nisms of diabetic retinopathy, and to 13. Elman MJ, Bressler NM, Qin H, et al.; Dia- fi betic Retinopathy Clinical Research Network. patients with DME (38). It also enables nd avenues to prevent diabetes, screen for early retinopathy, and optimize the Expanded 2-year follow-up of ranibizumab objective monitoring of treatment re- plus prompt or deferred laser or triamcinolone management of systemic risk factors sponse. Specificpatternsofmorphologi- plus prompt laser for diabetic macular edema. cal features on OCT have been proposed can we hope to remove diabetic reti- Ophthalmology 2011;118:609–614 “ to predict visual outcome for patients nopathy as the leading cause of pre- 14. Bandello F, Cunha-Vaz J, Chong NV, et al. New approaches for the treatment of diabetic with DME undertaking laser therapy ventable blindness in working-aged people” (4), a finding that has persisted macular oedema: recommendations by an ex- (39,40). Less clear is the potential role pert panel. Eye (Lond) 2012;25:485–493 for more than half a century. of OCT in stratifying risk of progression 15. Zechmeister-Koss I, Huic M. Vascular endo- and predicting therapeutic response to thelial growth factor inhibitors (anti-VEGF) in the anti-VEGF therapy among patients with management of diabetic macular oedema: a sys- Duality of Interest. N.C. has received grants tematic review. Br J Ophthalmol 2012;96:167–178 DME (41). from Bayer and Pfizer. I.Y.W. has received 16. Virgili G, Parravano M, Menchini F, Brunetti Combining anti-VEGF therapy with consultant fees from Bayer; grants from Novartis, M. Antiangiogenic therapy with anti-vascular other existing or novel therapies target- Alcon, Allergan, and Bayer; and lecture hono- endothelial growth factor modalities for dia- ing multiple pathophysiological path- rarium from Global Vision China. T.Y.W. is betic macular oedema. Cochrane Database on advisory boards for Abbott, Allergan, Bayer, Syst Rev 2012;12:CD007419 ways of diabetic retinopathy may Novartis, Pfizer, and Solvay and has received 17. Elman MJ, Qin H, Aiello LP, et al.; Diabetic further optimize visual outcome. Be- travel, honorarium, and research support from Retinopathy Clinical Research Network. Intravi- sides VEGF, several other mechanisms these companies. He has no stocks, equity, treal ranibizumab for diabetic macular edema care.diabetesjournals.org Cheung, Wong, and Wong 905

with prompt versus deferred laser treatment: macular edema: the 36-month results from with diabetic retinopathy. Cochrane Database three-year randomized trial results. Ophthal- two phase III trials: RISE and RIDE. Ophthalmol- Syst Rev 2011;7:CD008081 mology 2012;119:2312–2318 ogy 2013;120:2013–2022 39. Kim NR, Kim YJ, Chin HS, Moon YS. Optical 18. Lang GE, Berta A, Eldem BM, et al.; RESTORE 28. Zehetner C, Kirchmair R, Huber S, Kralinger coherence tomographic patterns in diabetic Extension Study Group. Two-year safety and ef- MT, Kieselbach GF. Plasma levels of vascular macular oedema: prediction of visual outcome ficacy of ranibizumab 0.5 mg in diabetic macular endothelial growth factor before and after after focal laser photocoagulation. Br J Ophthal- edema: interim analysis of the RESTORE exten- intravitreal injection of bevacizumab, ranibizu- mol 2009;93:901–905 sion study. Ophthalmology 2013;120:2004– mab and pegaptanib in patients with age-related 40. Aiello LP, Edwards AR, Beck RW, et al.; Di- 2012 macular degeneration, and in patients with dia- abetic Retinopathy Clinical Research Network. 19. Ip MS, Domalpally A, Hopkins JJ, Wong P, betic macular oedema. Br J Ophthalmol 2013;97: Factors associated with improvement and wors- – Ehrlich JS. Long-term effects of ranibizumab on 454 459 ening of visual acuity 2 years after focal/grid diabetic retinopathy severity and progression. 29. Lim LS, Cheung CM, Mitchell P, Wong TY. photocoagulation for diabetic macular edema. Arch Ophthalmol 2012;130:1145–1152 Emerging evidence concerning systemic safety Ophthalmology 2010;117:946–953 20. Do DV, Nguyen QD, Boyer D, et al.; da Vinci of anti-VEGF agentsdshould ophthalmologists 41. Wu PC, Lai CH, Chen CL, Kuo CN. Optical Study Group. One-year outcomes of the da Vinci be concerned? Am J Ophthalmol 2011;152:329– coherence tomographic patterns in diabetic Study of VEGF Trap-Eye in eyes with diabetic 331 macula edema can predict the effects of intra- macular edema. Ophthalmology 2012;119: 30. Curtis LH, Hammill BG, Schulman KA, 1658–1665 Cousins SW. Risks of mortality, myocardial in- vitreal bevacizumab injection as primary treat- – 21. Nepomuceno AB, Takaki E, Paes De Almeida farction, bleeding, and stroke associated with ment. J Ocul Pharmacol Ther 2012;28:59 64 FP, et al. A prospective randomized trial of intra- therapies for age-related macular degenera- 42. Nguyen QD, Brown DM, Marcus DM, et al.; vitreal bevacizumab versus ranibizumab for the tion. Arch Ophthalmol 2010;128:1273–1279 RISE and RIDE Research Group. Ranibizumab for management of diabetic macular edema. Am J 31. Schmucker C, Loke YK, Ehlken C, et al. Intra- diabetic macular edema: results from 2 phase III Ophthalmol 2013;156:502–510.e2 vitreal bevacizumab (Avastin) versus ranibizumab randomized trials: RISE and RIDE. Ophthalmol- 22. Bressler SB, Qin H, Melia M, et al.; Diabetic (Lucentis) for the treatment of age-related mac- ogy 2012;119:789–801 Retinopathy Clinical Research Network. Exploratory ular degeneration: a safety review. Br J Ophthal- 43. Nguyen QD, Shah SM, Khwaja AA, et al.; analysis of the effect of intravitreal ranibizumab mol 2011;95:308–317 READ-2 Study Group. Two-year outcomes of or triamcinolone on worsening of diabetic reti- 32. Wirostko B, Wong TY, SimoR.Vascularen-´ the ranibizumab for edema of the mAcula in nopathy in a randomized clinical trial. JAMA dothelial growth factor and diabetic complica- diabetes (READ-2) study. Ophthalmology 2010; Ophthalmol 2013;131:1033–1040 tions. Prog Retin Eye Res 2008;27:608–621 117:2146–2151 23. Diabetic Retinopathy Clinical Research Net- 33. Cheung N, Wong TY. Diabetic retinopathy 44. Massin P, Bandello F, Garweg JG, et al. work. Randomized clinical trial evaluating intra- and systemic vascular complications. Prog Retin Safety and efficacy of ranibizumab in diabetic vitreal ranibizumab or saline for vitreous Eye Res 2008;27:161–176 macular edema (RESOLVE Study): a 12-month, hemorrhage from proliferative diabetic retinop- 34. Department of Health and Human Services. randomized, controlled, double-masked, multi- athy. JAMA Ophthalmol 2013;131:283–293 A review of Medicare Part B avastin and lucentis center phase II study. Diabetes Care 2010;33: 24. Smith JM, Steel DH. Anti-vascular endo- treatments for age-related macular degenera- 2399–2405 thelial growth factor for prevention of post- tion [Internet], 2011. Available from http:// 45. Mitchell P, Bandello F, Schmidt-Erfurth U, operative vitreous cavity haemorrhage after oig.hhs.gov/oas/reports/region10/11000514. et al.; RESTORE Study Group. The RESTORE vitrectomy for proliferative diabetic retinopa- pdf. Accessed 1 August 2013 study: ranibizumab monotherapy or combined thy. Cochrane Database Syst Rev 2011;5: 35. Smiddy WE. Clinical applications of cost with laser versus laser monotherapy for diabetic CD008214 analysis of diabetic macular edema treatments. macular edema. Ophthalmology 2011;118:615– 25. Van Geest RJ, Lesnik-Oberstein SY, Tan HS, Ophthalmology 2012;119:2558–2562 625 et al. A shift in the balance of vascular endothe- 36. Stein JD, Newman-Casey PA, Kim DD, 46. Rajendram R, Fraser-Bell S, Kaines A, et al. A lial growth factor and connective tissue growth Nwanyanwu KH, Johnson MW, Hutton DW. 2-year prospective randomized controlled trial factor by bevacizumab causes the angiofibrotic Cost-effectiveness of various interventions for switch in proliferative diabetic retinopathy. Br J newly diagnosed diabetic macular edema. Oph- of intravitreal bevacizumab or laser therapy Ophthalmol 2012;96:587–590 thalmology 2013;120:1835–1842 (BOLT) in the management of diabetic macular 26. Do DV, Nguyen QD, Khwaja AA, et al.; READ-2 37. Martin DF, Maguire MG, Ying GS, edema: 24-month data: report 3. Arch Ophthal- Study Group. Ranibizumab for edema of the mac- Grunwald JE, Fine SL, Jaffe GJ; CATT Research mol 2012;130:972–979 ula in diabetes study: 3-year outcomes and the Group. Ranibizumab and bevacizumab for neo- 47. Sultan MB, Zhou D, Loftus J, Dombi T, Ice KS; need for prolonged frequent treatment. JAMA vascular age-related macular degeneration. Macugen 1013 Study Group. A phase 2/3, mul- Ophthalmol 2013;131:139–145 N Engl J Med 2011;364:1897–1908 ticenter, randomized, double-masked, 2-year 27. Brown DM, Nguyen QD, Marcus DM, et al.; 38. Virgili G, Menchini F, Murro V, Peluso E, Rosa trial of pegaptanib sodium for the treatment RIDE and RISE Research Group. Long-term out- F, Casazza G. Optical coherence tomography of diabetic macular edema. Ophthalmology comes of ranibizumab therapy for diabetic (OCT) for detection of macular oedema in patients 2011;118:1107–1118