PROCEEDINGS
`
`CURRENT CONCEPTS IN THE MANAGEMENT OF
`DIABETIC MACULAR EDEMA*
`—
`
`Ursula Schmidt-Erfurth, MD†
`
`dy
`
`ABSTRACT
`
`Clinically, diabetic macular edema (DME) is
`characterized by retinal thickening and, with
`angiography, by pooled dye in edematous areas.
`For the patient, DME results in vision loss. To
`understand the current approaches to treating
`DME, it is helpful to understand the pathophysio-
`logic changes that precipitate DME and the ther-
`apeutic mechanisms that target those changes.
`This article will focus on 3 pathophysiologic
`changes that are being targeted in treatment:
`vascular leakage into intraretinal layers, extrava-
`sation of plasma lipids, and ischemia-induced
`activation of cytokines (angiogenesis and inflam-
`mation). We now have several options for the
`treatment of DME: focal/grid laser photocoagula-
`tion, anti-vascular endothelial growth factor com-
`pounds, aflibercept, triamcinolone acetonide
`intravitreal injections, and steroid implants. All of
`these compounds show at least some benefit in
`DME. The challenge for clinicians is that none of
`the drugs is currently approved for use in DME,
`thus there are no formal recommendations for
`treatment regimens or changes to the current stan-
`dard of care (focal/grid laser photocoagulation).
`(Adv Stud Ophthalmol. 2010;7(2):52-59)
`
`*Based on a presentation by Dr Schmidt-Erfurth at a
`roundtable symposium held in Fort Lauderdale, Florida, on
`April 30, 2010.
`†Professor and Chair, Department of Ophthalmology,
`Medical University of Vienna, Vienna, Austria.
`Address correspondence to: Ursula Schmidt-Erfurth, MD,
`Professor and Chair, Department of Ophthalmology, Medical
`University of Vienna, Waehringer Guertel 18-20, Vienna,
`Austria. E-mail: ursula.schmidt-erfurth@meduniwien.ac.at.
`
`F or the patient, diabetic macular edema (DME)
`
`results in vision loss due to light scattering,
`impaired cell-to-cell interaction, and distur-
`bances of normal ionic balances in the retina.1 To
`understand the current treatment of DME, it is help-
`ful to understand the pathophysiologic changes that
`underlie DME and the therapeutic mechanisms that
`target those changes.
`
`PATHOPHYSIOLOGIC CHANGES IN DME
`
`A thorough review of DME pathophysiology is
`beyond the scope of this article, but current reviews are
`available.1 This article will focus on the major patho-
`physiologic changes that are being targeted in treatment.
`In general, visual impairment with diabetes mellitus is
`most commonly due to DME, and DME causes vision
`loss through multiple mechanisms, including light scat-
`tering, impaired cell-to-cell interaction, and disturbances
`of normal ionic balances in the retina. Intraretinal fluid
`accumulates due to breakdown of the blood-retinal bar-
`rier (BRB), leading to extravasation of fluid into the
`intraretinal layers, which is rapidly reversible, and
`extravasation of plasma lipids from the intravascular
`lumen, which is less easily reversible. A recent study
`using optical coherence tomography (OCT) showed
`that hyperreflective foci found throughout the retinal
`layers represent extravasated lipoproteins and/or pro-
`teins. Confluent foci are seen as hard exudates clinically.
`Nonconfluent hyperreflective foci were detectable only
`by OCT; in general, their hyperreflectivity did not cor-
`respond with intraretinal hemorrhage nor the typical
`OCT findings seen with intraretinal aneurysms. As such,
`these foci may be a very early subclinical sign of BRB
`breakdown in DME.2 During the progression of the dis-
`ease, lipid foci migrate downwards and form confluent
`plaques.
`
`52
`
`Vol. 7, No. 2 I December 2010
`
`CELLTRION - EXHIBIT 1050
`
`

`

`TI
`
`njections,
`BR
`
`Level of improvement
`
`No change
`>5 letters
`>10 letters
`>15 letters
`>30 letters
`Reduced >5 letters
`Mean improvement
`
`BRVO = branch retinal vein occlusion; CRVO = central retinal vein occlusion; PRN = as needed.
`Data from Campochiaro et al.
`
`PROCEEDINGS
`
`When lipid extravasation reaches the deep retinal
`layers—where photoreceptors reside—functional loss
`occurs. Comparing OCT as anatomical mapping and
`microperimetry as functional mapping, the relevant
`factors for visual function were giant cysts in the outer
`retinal layers at the fovea for serous detachment in the
`fovea and lipid exudates affecting the neurosensory
`layers in the fovea.3
`On a biologic level, DME is considered to be a state
`of low-grade inflammation. As such numerous inflam-
`matory cells, cytokines, factors, and processes are involved
`(Table).4 Inflammatory cytokines are also known to inter-
`act with vascular endothelial growth factor (VEGF),
`which is a vascular permeability factor (with 50 000 times
`the potency of histamine).5 VEGF causes vascular hyper-
`permeability by opening endothelial cell junctions and
`inducing fenestrations and vesiculo-vacuolar organelles.
`VEGF levels correlate with breakdown of the BRB.5,6
`VEGF levels are significantly elevated in eyes with exuda-
`tive retinal disease, including those with DME, retinal
`vein occlusion, and choroidal neovascularization.4
`
`THERAPEUTIC MECHANISMS OF DME TREATMENTS
`
`The primary treatment target is to reduce retinal
`fluid and edema, which is measured by central retinal
`
`thickness as seen on OCT. OCT is the preferred tool to
`measure therapeutic effects in clinical trials, but there is
`much debate regarding whether there is a direct rela-
`tionship between this biomarker and visual acuity. The
`Diabetic Retinopathy Clinical Research (DRCR)
`Network studied this relationship in 1717 patients
`(2058 eyes) with DME.7 The results showed a modest
`general correlation but not an interindividual associa-
`tion (0.52 at baseline and 0.44 3.5 months after laser
`treatment). However, the study did not address whether
`short-term changes on OCT are predictive of long-term
`visual acuity. Also, a wide range of visual acuity may be
`observed for a given degree of retinal edema, and the
`stage of disease and timing of an intervention varies
`widely. At this point, OCT therefore cannot be a surro-
`gate end point for visual acuity in DME.
`However, multiple changes in macular morpholo-
`gy are observed during the treatment of DME. When
`lipid exudates in the retina are examined, there is a
`typical pattern of therapeutic response after laser pho-
`tocoagulation. In one study, during 4 months of fol-
`low-up after laser therapy, in addition to decreased
`retinal thickness, the hyperreflective foci either
`resolved completely or became confluent at the apical
`border of the outer nuclear layer. Upon further follow-
`up, clinically visible hard exudates formed. In areas of
`
`Table. Mean Concentrations (pg/mL) of Cytokines and Inflammatory Factors at Baseline in Patients with DME and
`Controls
`
`CNV
`(n = 28)
`
`Controls
`(n = 28)
`
`53.8
`
`22.4
`
`367.1
`777.6
`38
`85.6
`
`299
`520.9
`58.8
`41.5
`
`IL-1α
`IL-6
`IL-8
`IP-10
`MCP-1
`PDGF-AA
`VEGF
`
`P
`
`NS
`
`NS
`NS
`.04
`.03
`
`RVO
`(n = 13)
`
`Controls
`(n = 13)
`
`40.8
`51.9
`63.9
`581.4
`1224.4
`63.4
`513.3
`
`0
`2.1
`1.8
`197.6
`344.1
`39
`38.2
`
`P
`
`.09
`.01
`.03
`.01
`.003
`.003
`.04
`
`DME
`(n = 10)
`
`Controls
`(n = 10)
`
`P
`
`46.5
`16.2
`
`5.3
`4.9
`
`.003
`<.001
`
`1513.2
`
`493.7
`
`<.001
`
`273.3
`
`61.5
`
`.001
`
`Under detection limits: IL-1β, IL-2, IL-3, IL-4, IL-5, IL-10, IL-12(p40), IL-12(p70), IL-13, IL-15, Eotaxin, ICAM-1, IFNγ, GM-CSF, RANTES, TNF-α, EGF, FGF-2, IL-7,
`MIP-1α, Flt-3, PDGF- AB/BB
`CNV = choroidal neovascularization; DME = diabetic macular edema; EGF = epidermal growth factor; FGF = fibroblast growth factor; Flt-3 = FMS-like tyrosine kinase
`3; GM-CSF = granulocyte/macrophage colony stimulating factor; ICAM-1 = inter-cellular adhesion molecule 1; IFN = interferon; IL = interleukin; IP-10 = inducible pro-
`tein-10; MCP-1 = monocyte chemotactic protein-1; MIP-1α = macrophage inflammatory protein-1α; NS = not significant; PDGF = platelet-derived growth factor;
`RANTES = regulated upon activation, normal T cell expressed and secreted; RVO = retinal vein occlusion; TNF = tumor necrosis factor; VEGF = vascular endothelial
`growth factor.
`Data from Funk et al.4
`
`Johns Hopkins Advanced Studies in Ophthalmology I
`
`53
`
`

`

`PROCEEDINGS
`
`retinal thickening despite laser treatment, the hyper-
`reflective foci maintained their distribution pattern
`throughout all retinal layers, and did not resolve.
`Thus, not only do hyperreflective foci seem to repre-
`sent precursors or components of hard exudates, but
`retinal photocoagulation has a major impact on the
`density and distribution of intraretinal lipid deposits
`to obtain a therapeutic benefit.8
`However, laser therapy also induces destructive
`changes in the retina. With high-resolution OCT dur-
`ing follow-up post laser treatment, marked focal dam-
`age to the outer retinal layers is observed, including the
`outer nuclear outer plexiform, the photoreceptor lay-
`ers, and the retinal pigment epithelium. Thus, with
`laser therapy, there is a permanent destruction of pho-
`toreceptors and adjacent neurosensory elements.
`During the healing process over a few weeks, arcading
`scar formation throughout all retinal layers is
`observed, which leads to a decrease in retinal thickness
`due to atrophy, even if edema was not previously pre-
`sent. Thus, it is important to understand that laser
`therapy reduces retinal thickness, independent of fluid
`resolution. The same is true with laser therapy to the
`central retina in macular edema.8 Due to subsequent
`atrophy of the retinal neurons following focal outer
`retinal thermonecrosis, shrinkage and atrophy show a
`radial pattern.
`With anti-VEGF therapy, the process of macular
`fluid reduction and lipid migration proceeds much
`more rapidly than with laser,9 which may explain why
`vision improvement with anti-VEGF therapy pro-
`ceeds much faster than with laser therapy. VEGF has
`been a major target of DME treatment; however, in
`DME, VEGF therapy does not appear to affect other
`cytokines involved in the disease unlike in neovascu-
`lar age-related macular edema (AMD). In a study
`involving monthly intravitreal injections with beva-
`cizumab (Avastin; Genentech, Inc, South San
`Francisco, CA) in patients with DME, VEGF levels
`decreased to 0 by the first visit at 1 month and
`remained at those levels over 6 visits (ie, 6 months).
`Levels of other cytokines were either unaffected or
`mildly affected by anti-VEGF treatment, which
`explains the relatively slow and limited response of
`DME to anti-VEGF therapy compared to AMD.4
`
`DME TREATMENTS
`
`Although there are several approaches to treating
`
`DME (ie, focal/grid laser photocoagulation, steroids,
`anti-VEGF compounds, or combination therapy),
`laser photocoagulation remains the “gold standard,” as
`long as drugs are not approved. However, such stan-
`dard will change soon as the results of the RESTORE
`(Efficacy and Safety of Ranibizumab [Intravitreal
`Injections; Lucentis; Genentech, Inc, South San
`Francisco, CA] in Patients with Visual Impairment
`Due to DME) phase III study are already submitted to
`the authorities.
`A common question regarding laser therapy is
`which approach to use. The DRCR Network, in a
`study of 263 subjects with DME, compared Early
`Treatment Diabetic Retinopathy Study (ETDRS;
`direct/grid or focal/grid) photocoagulation to mild
`macular grid (MMG) laser photocoagulation. After 12
`months of follow-up, the MMG technique was less
`effective at reducing OCT-measured retinal thickening
`than the ETDRS focal/grid technique, but the visual
`acuity outcomes were not substantially different
`(change in visual acuity at 12 months: 0 letters, mod-
`ified ETDRS group, -2 letters, MMG group, P =
`.10).10 Thus, it does not appear that the particular
`technique of laser therapy is as important as the appli-
`cation of laser therapy in general.
`Steroids have an important role in DME manage-
`ment, particularly with reducing edema, but there are
`risks associated with their use that may not outweigh the
`benefits. The DRCR Network compared 2 doses of
`intravitreal triamcinolone acetonide injectable suspen-
`sion (Trivaris; Allergan, Inc, Irvine, CA) to focal/grid
`laser photocoagulation in 840 eyes with DME. After 1
`year of follow-up, there were no significant differences
`among the treatment groups in visual acuity, but by 2
`years, the mean visual acuity was better in the laser ther-
`apy group compared to the 1-mg triamcinolone group
`and the 4-mg triamcinolone group. OCT results fol-
`lowed the same pattern. In addition, a 10-mm Hg
`increase in intraocular pressure (IOP) occurred much
`more frequently in those receiving triamcinolone (4%
`laser, 16% 1-mg triamcinolone, and 33% 4-mg triamci-
`nolone) as did the need for cataract surgery (13%, 23%,
`and 51%, respectively).11
`There are 2 anti-VEGF agents designed for treating
`DME, although none is yet US Food and Drug
`Administration (FDA) approved for DME. In the
`RESOLVE (Safety and Efficacy of Ranibizumab in
`DME with Center Involvement) study, a phase II
`study comparing 2 doses of ranibizumab to sham
`
`54
`
`Vol. 7, No. 2 I December 2010
`
`

`

`PROCEEDINGS
`
`6 letters. In contrast, laser-treated eyes did not improve
`in best corrected visual acuity (+0.8). Retinal edema
`resolved substantially and rapidly with anti-VEGF
`treatment, but little with laser therapy. Thirty-seven
`percent and 43% of patients in the ranibizumab
`groups improved by at least 2 lines compared to 15%
`in the laser arms and 22% versus 8% improved by at
`least 3 lines with ranibizumab. A mean of 7 treatments
`was required over 12 months. There was an equally
`good response in focal versus diffuse edema and
`patients pretreated or untreated with laser earlier. In
`general, eyes with more intensive edema responded
`better to therapy. Also, it was concluded that the
`results obtained with ranibizumab monotherapy were
`optimal and could not be improved with adjunct laser.
`Finally, the DRCR Network compared intravitreal
`0.5-mg ranibizumab or 4-mg triamcinolone acetonide
`injectable suspension combined with focal/grid laser
`to focal/grid laser alone for treatment of DME in 854
`eyes. As expected, the clinical (Figure 2) and anatomic
`(Figure 3) benefit with both ranibizumab and triamci-
`nolone was significant.18 However, of particular inter-
`est is that the clinical benefit with triamcinolone seen
`
`Figure 1. Ranibizumab vs Laser vs Combination, Mean
`Change in Visual Acuity from Baseline: Results from
`the READ-2 Study
`
`◆
`
`▲
`
`■
`
`■
`◆
`
`◆
`▲
`
`■
`
`◆
`
`▲
`
`■
`
`Baseline
`0.00
`0.00
`0
`00
`
`.
`
`Month 6
`7.24
`-0.43
`3.80
`
`Month 12
`6.61
`2.39
`4.81
`
`Month 24
`7.70
`5.10
`6.80
`
`10
`
`8 6 4 2 0
`
`-2
`
`-4
`
`ETDRS letters read @ 4m
`
`RBZ
`Laser
`RBZ + Laser
`
`■▲◆
`
`At 6 months, the combination therapy group was not statistically different form
`the other 2 groups with regard to mean gain in best corrected visual acuity.
`ETDRS = Early Treatment Diabetic Retinopathy Study; RBZ = ranibizumab;
`READ-2 = Ranibizumab for Edema of the mAcula in Diabetes.
`Adapted with permission from Nguyen et al. Ophthalmology. 2010;117:2146-
`2151.15
`
`injection (www.clinicaltrials.gov: NCT00284050),
`therapeutic benefit with ranibizumab was observed in
`terms of immediate (by month 1) and persistent
`reduced retinal thickening and improved visual acuity
`(mean of 8–10 ETDRS letters after 6 months, com-
`pared to -3 letters in those receiving sham). However,
`the study also showed that the improvement was not
`as rapid as that seen with ranibizumab treatment in
`patients with AMD.12
`The DA VINCI (DME and VEGF Trap-Eye:
`INvestigation of Clinical Impact) study compared 4
`different doses of aflibercept (VEGF Trap-Eye;
`Regeneron Pharmaceuticals, Inc, Tarrytown, NY)—a
`fusion protein of the key binding domains of the
`VEGF receptor-1 and -2 combined with a human
`immunoglobulin G Fc fragment—to laser therapy
`every 16 weeks as needed. Aflibercept blocks all iso-
`forms of VEGF-A as well as placental growth factors-
`1 and -2.13 Preliminary results after 24 weeks showed
`significant improvement in mean change in visual acu-
`ity (gain of 8–12 ETDRS letters) with several different
`dosing regimens, including bimonthly. The OCT
`results paralleled the visual acuity results.14 The study
`will follow these patients for 52 weeks, and it will be
`interesting to see if the results suggest that the dosing
`frequency with anti-VEGF compounds can be
`reduced from monthly injections based on the results.
`The READ-2 (Ranibizumab for Edema of the
`mAcula in Diabetes 2) study compared ranibizumab
`(0.5-mg ranibizumab at baseline and months 1, 3, and
`5) with focal/grid laser (at baseline and month 3 if
`needed) or a combination of both (ranibizumab and
`focal/grid laser at baseline and month 3) in patients
`with DME. After 6 months, patients in both groups
`receiving ranibizumab gained, whereas those receiving
`laser therapy lost vision (Figure 1).15 After 2 years, the
`2 ranibizumab-receiving groups continued to fare bet-
`ter than those in the laser group, but the differences
`among groups were not as great as in the beginning
`(Figure 1).15,16
`The RESTORE trial (phase III) is evaluating the
`efficacy and safety of ranibizumab (0.5 mg) as adjunc-
`tive therapy to laser photocoagulation and monother-
`apy in patients with visual impairment due to DME.
`The study primary outcome completion was January
`2010.17 Patients treated with ranibizumab, either as
`monotherapy or in combination with laser, experi-
`enced a rapid improvement that was sustained over the
`12-month study period with a mean improvement of
`
`Johns Hopkins Advanced Studies in Ophthalmology I
`
`55
`
`

`

`PROCEEDINGS
`
`Figure 2. Ranibizumab + Laser or Triamcinolone +
`Laser, Mean Change in Visual Acuity: Results from the
`DRCR Network Protocol I Study
`
`during the first 6 months of the study was lost during
`the last 6 months, and the anatomic benefit seen dur-
`ing the first year was lost during the second year of fol-
`low-up, suggesting a correlation between functional
`and anatomic changes with triamcinolone plus laser
`treatment, whereas, with ranibizumab + laser prompt
`and deferred there was a persistent decrease in central
`retinal thickness across the 2-year follow-up. In the
`subset of pseudophakic eyes at baseline (n = 273), visu-
`al acuity improvement in the triamcinolone + prompt
`laser group appeared comparable to that in the
`ranibizumab groups. However, there were substantial
`safety issues with triamcinolone. An elevated IOP,
`which required treatment, occurred in approximately
`27% of eyes and cataract surgery was required in
`approximately 50% of the phakic patients.18
`
`CONCLUSIONS
`
`12
`
`11
`
`10
`
`9 8 7 6 5 4 3 2 1 0
`
`Mean change in visual acuity from baseline (letter score)
`
`Sham+Prompt Laser
`Ranibizumab+Prompt Laser
`Ranibizumab+Deferred Laser
`◆◆
`Triamcinolone+Prompt Laser
`N = 799 (52 weeks)
`N = 484 (104 weeks)
`
`■●
`
`▲▲
`
`■
`▲▲
`
`●
`
`■
`
`▲▲
`
`●
`
`◆◆
`
`■
`
`▲▲
`
`●
`◆◆
`
`■
`▲▲
`
`■
`▲▲
`
`▲▲
`
`■
`
`■
`▲▲
`
`▲▲
`■ ■
`
`▲▲
`■ ■
`▲▲
`
`▲▲
`■ ■ ■
`▲▲
`
`◆◆◆◆
`
`◆◆◆◆
`
`◆◆
`
`◆◆
`
`◆◆
`
`●
`
`●
`
`●
`
`●
`
`● ● ●
`
`◆◆
`
`●
`
`◆◆
`●
`
`●
`◆◆
`
`●
`◆◆
`
`■
`▲▲
`
`■
`◆◆◆◆
`
`● ●
`
`●◆◆
`0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92 96 100 104
`
`◆◆
`
`Visit week
`
`The 1-year and 2-year visit completion rates were 94% and 90%,
`respectively.
`DRCR = Diabetic Retinopathy Clinical Research.
`Reprinted with permission from Diabetic Retinopathy Clinical
`Research Network et al. Ophthalmology. 2010;117:1064-1077.18
`
`There is a clear paradigm shift toward intravitreal
`drugs for the treatment of DME: anti-VEGF com-
`pounds (bevacizumab, ranibizumab), aflibercept, tri-
`amcinolone
`intravitreal
`injections, and steroid
`implants (dexamethasone, fluocinolone). All of these
`compounds show at least some benefit in DME. This
`article presents an overview of the current state of
`knowledge regarding treatment mechanisms and
`modalities under investigation for DME. None of the
`pharmacologic agents is currently approved by the
`FDA for the management of DME. Thus, the current
`
`Patient with DME with Poor Response to
`Panretinal Photocoagulation Treatment in the
`Left Eye – Harry W. Flynn, Jr, MD
`
`JM is a 60-year-old man with an 18-year history
`of non–insulin-dependent diabetes mellitus. He pre-
`sents to the ophthalmologist with bilateral decreased
`vision: 20/60 OD (phakic) and 20/200 OS
`(pseudophakic). After laser photocoagulation was
`performed for the DME in the left eye, his vision
`remained poor at 20/200.
`How would you treat his right eye? What are the
`best options when standard-of-care treatment fails?
`When is observation the best course? For further dis-
`cussion of the challenges in managing this real-life situ-
`ation, please go to www.JHASIO.com/retinaldiseases.
`
`Figure 3. Ranibizumab + Laser or Triamcinolone +
`Laser, OCT Central Subfield Thickness <250 µm with
`at Least a 25-µm Decrease in Thickness from
`Baseline: Results from the DRCR Network Protocol I
`Study
`
`Sham+Prompt Laser
`Ranibizumab+Prompt Laser
`Ranibizumab+Deferred Laser
`◆◆
`Triamcinolone+Prompt Laser
`N = 790 (52 weeks)
`N = 444 (104 weeks)
`
`■●
`
`▲▲
`
`▲▲
`
`■
`
`●
`
`◆◆
`
`▲▲
`■
`
`◆◆
`●
`
`■
`▲▲
`
`◆◆
`
`●
`
`▲▲
`
`▲▲▲▲
`▲▲
`◆◆◆◆
`◆◆◆◆◆◆
`■
`■
`■
`
`■
`
`▲▲
`▲▲
`◆◆◆◆
`▲▲
`◆◆
`■■■
`
`●●
`
`●
`
`●
`
`●
`
`●
`
`▲▲
`
`◆◆
`
`■
`
`▲▲
`■■
`◆◆◆◆◆◆
`■
`
`◆◆
`
`■
`
`●●
`
`●
`
`●
`
`●●
`
`●
`
`60%
`
`decrease in thickness from baseline
`
`50%
`
`40%
`
`30%
`
`20%
`
`10%
`
`<250 µm with at least a 25-µm
`OCT central subfield thickness
`
`0%
`
`●◆◆
`0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92 96 100 104
`
`Visit week
`
`DRCR = Diabetic Retinopathy Clinical Research; OCT = optical
`coherence tomography.
`Reprinted with permission from Diabetic Retinopathy Clinical
`Research Network et al. Ophthalmology. 2010;117:1064-1077.18
`
`56
`
`Vol. 7, No. 2 I December 2010
`
`

`

`PROCEEDINGS
`
`standard of care (focal/grid laser photocoagulation) is
`still appropriate, although data from all recent studies
`have clearly demonstrated that pharmacologic agents
`such as VEGF antagonists may lead to further decreas-
`es in retinal thickness and better improvement in visu-
`al acuity for patients with DME.
`
`DISCUSSION
`
`ROLE OF STEROIDS
`Dr Flynn: In any form of therapy, you have to bal-
`ance the risks and the benefits. There is a risk to any
`intravitreal injection. If the patient’s initial visual acu-
`ity is good (ie, 20/40 or better at baseline), in my
`mind, laser therapy is still the preferred initial strategy.
`In the DRCR Protocol I study, the lower limit of
`inclusion was 20/32 with center-involved DME. In
`deciding the initial treatment strategy, you have to
`look at the overall picture. Subgroup analysis of
`pseudophakic eyes at baseline (from Protocol I)
`showed that, at 2 years follow-up, the response to tri-
`amcinolone/laser combination appeared to be similar
`to or better than the 2 ranibizumab treatments, and
`there were fewer injections. Therefore, I would say that
`steroids are still an option because they did work real-
`ly well in pseudophakic eyes at baseline. Elevated IOP
`is still an issue with all intravitreal steroids.
`Dr Kuppermann: But, if you looked at eyes that
`were pseudophakic at 24 months, there was not as
`much benefit. Thus, cataract removal during the trial
`did not seem to have the same results as in those who
`were pseudophakic at baseline, which I thought was
`interesting.
`Dr Flynn: In fact, the DRCR investigators were
`encouraged to proceed with cataract surgery relatively
`early in the course (ie, not to let cataracts be a limiting
`factor in visual acuity improvement). At 1 year in the
`DRCR phakic eyes, the laser group continued to slow-
`ly get better, whereas the triamcinolone group seemed
`to get worse. It became the poorest group at the 2-year
`follow-up point. By 2 years, the number of patients
`undergoing cataract surgery in the triamcinolone
`group was approximately 58%, and there were still
`maybe 33% of the eyes that had a developing cataract
`but had not had surgery by the 2-year time point. We
`all know that some of these cataracts in triamcinolone-
`treated eyes are not your classical posterior subcapsular
`opacity. There can be a subtle milky nuclear sclerotic
`cataract that can blunt vision without a striking
`
`nuclear opacity when you are doing a slit-lamp
`examination.19
`Dr Kuppermann: So, do you think that the resid-
`ual cataract was still an important factor in blunting
`the visual benefit in the triamcinolone arm?
`Dr Flynn: Yes. Steroids are not an outdated treat-
`ment; they still have a role, especially in combination
`therapies.
`
`VISUAL OUTCOMES IN PSEUDOPHAKIA
`Dr Schmidt-Erfurth: Why do we think that
`pseudophakia makes such a difference in the disease,
`which is a retinal disease? Is this a special or different
`population? Do they have different pharmacologic
`dynamics in the eye?
`Dr Flynn: Subgroup analysis papers from the
`DRCR Protocol I will be coming out soon. The mean
`visual acuity at baseline among the groups was around
`20/50, so the groups were well balanced. The details of
`phakic patients at baseline will be discussed.
`Dr Kuppermann: So, it seems that you are saying
`that at 6 months, when the benefits of triamcinolone
`begin to wear off, some of that may be due to subclin-
`ical cataracts that brought down the visual acuity
`result, even though people were encouraged to do
`cataract surgery.
`Dr Do: In the DRCR Protocol B study, which
`compared triamcinolone versus laser, investigators
`evaluated pseudophakic versus phakic status at base-
`line, and there was no difference in visual acuity out-
`comes at 2 or 3 years. The effect of the steroid still was
`inferior to laser.
`Dr Kuppermann: I still have concerns that the tri-
`amcinolone formulation used in the DRCRnet trials
`may have an unexpectedly short duration resulting in
`undertreatment.
`Dr Flynn: The DRCR protocol B reported results
`with monotherapy (laser vs steroid). In my mind, there
`is a 1-2 punch with combination therapy. After you
`treat the patient with intravitreal steroids, you may get
`the macular edema to improve, and then you can more
`accurately apply your treatment in the micro-
`aneurysms and leakage areas when the retina is less
`thick. In DRCR Protocol I, combination therapies
`were evaluated. That is a different strategy from an
`“either/or” monotherapy.
`
`NUMBER OF INJECTIONS AND ENDOPHTHALMITIS
`Dr Flynn: In the ranibizumab groups (in the
`
`Johns Hopkins Advanced Studies in Ophthalmology I
`
`57
`
`

`

`PROCEEDINGS
`
`Protocol I study), the rate of endophthalmitis per
`study patient over a 2-year follow-up was approaching
`1%. And that is a pretty high price to pay, in my opin-
`ion, for the benefit of a few letters gain in visual acu-
`ity. In other words, if you were getting on an airplane
`and there was a 1% chance of crashing over a 2-year
`time interval, taking approximately 8 flights per year, I
`would not get on the airplane.
`Dr Campochiaro: You have to wonder if patients
`with diabetes are more prone to endophthalmitis and
`so these rates were influenced by the disease process
`being studied. It will be interesting to see what the
`rates of endophthalmitis are
`in
`the RISE
`(Ranibizumab Injection in Subjects with Clinically
`Significant Macular Edema with Center Involvement
`Secondary
`to Diabetes Mellitus) and RIDE
`(Ranibizumab Injection in Subjects with Clinically
`Significant Macular Edema with Center Involvement
`Secondary to Diabetes Mellitus) studies. In the MARI-
`NA (Minimally Classic/Occult Trial of the Anti-
`VEGF Antibody Ranibizumab in the Treatment of
`Neovascular AMD) and ANCHOR (Anti-VEGF
`Antibody for the Treatment of Predominantly Classic
`Choroidal Neovascularization in AMD) studies,
`patients with neovascular AMD received monthly
`injections for 2 years (24 injections), and yet the rates
`of endophthalmitis were substantially lower.20,21
`Dr Flynn: The rates of endophthalmitis after
`cataract surgery and other types of surgery are higher
`in patients with diabetes because their immune status
`may be compromised. And there may be other factors
`such as protocol issues. For example, it is very impor-
`tant to use povidone-iodine as a topical antiseptic.
`There is a common misconception that patients with
`a “seafood allergy” or “iodine allergy” are therefore
`allergic to povidone-iodine. There has been hesitancy
`to use the proper povidone-iodine preparation in these
`patient-reported circumstances. I am aware of cases in
`which the physician chose not to use a standard povi-
`done-iodine preparation and instead used a topical
`antibiotic, 4 times, 5 minutes apart. Topical antibiotics
`alone as the preparation are not protective. The causes
`of endophthalmitis are multifactorial.
`The median number of injections in the first year
`were 0 (sham + prompt laser), 8 (ranibizumab +
`prompt laser), 9 (ranibizumab + deferred laser), and 3
`(triamcinolone + prompt laser). There were fewer
`injections in the triamcinolone group, but no cases of
`endophthalmitis.
`
`REFERENCES
`
`5.
`
`1. Singh A, Stewart JM. Pathophysiology of diabetic macular
`edema. Int Ophthalmol Clin. 2009;49:1-11.
`2. Bolz M, Schmidt-Erfurth U, Deak G, et al. Optical coher-
`ence tomographic hyperreflective foci. A morphologic sign
`of lipid extravasation in diabetic macular edema.
`Ophthalmology. 2009;116:914-920.
`3. Deak GG, Bolz M, Ritter M, et al. A systematic correlation
`of morphology and functional alterations in diabetic macu-
`lar edema. Invest Ophthalmol Vis Sci. 2010;Epub ahead
`of print.
`4. Funk M, Schmidinger G, Maar N, et al. Angiogenic and
`inflammatory markers in intraocular fluid of eyes with diabet-
`ic macular edema and influence of therapy with bevacizum-
`ab. Retina. 2010; In press.
`Joussen AM, Smyth N, Niessen C. Pathophysiology of dia-
`betic macular edema. Dev Ophthalmol. 2007;39:1-12.
`6. Amin RH, Frank RN, Kennedy A, et al. Vascular endothelial
`growth factor is present in glial cells of the retina and optic
`nerve of human subjects with nonproliferative diabetic
`retinopathy. Invest Ophthalmol Vis Sci. 1997;38:36-47.
`7. Aiello LP, Edwards AR, Beck RW, et al. Factors associated
`with improvement and worsening of visual acuity 2 years
`after focal/grid photocoagulation for diabetic macular
`edema. Ophthalmology. 2010;117:946-953.
`8. Deak GG, Bolz M, Kriechbaum K, et al. Effect of retinal
`photocoagulation on intraretinal lipid exudates in diabetic
`macular edema documented by optical coherence tomogra-
`phy. Ophthalmology. 2010;117:773-779.
`9. Schmidt-Erfurth U. Impact of antiangiogenic therapy on
`intraretinal exsudates in diabetic macular edema (DME)
`[abstract]. Presented at: Annual Macula Society Meeting,
`Tucson, AZ; February 25, 2010.
`10. Writing Committee for the Diabetic Retinopathy Clinical
`Research Network, Fong DS, Strauber SF, et al.
`Comparison of the modified Early Treatment Diabetic
`Retinopathy Study and mild macular grid laser photocoagu-
`lation strategies for diabetic macular edema. Arch
`Ophthalmol. 2007;125:469-480.
`11. Diabetic Retinopathy Clinical Research Network. A random-
`ized trial comparing intravitreal triamcinolone acetonide
`and focal/grid photocoagulation for diabetic macular
`edema. Ophthalmology. 2008;115:1447-1449.
`12. Schmidt-Erfurth U. Safety and efficacy of ranibizumab in
`diabetic macular edema: 12-month data from the phase II
`RESOLVE trial [abstract]. Presented at: Club Jules Gonin;
`2008; St. Moritz, Switzerland.
`13. Dixon JA, Oliver SC, Olson JL, Mandava N. VEGF Trap-Eye
`for the treatment of neovascular age-related macular degen-
`eration. Expert Opin Ivestig Drugs. 2009;18:1573-1580.
`14. Schmidt-Erfurth U. VEGF Trap-Eye for Diabetic Macular
`Edema: The DA VINCI Study [abstract]. Presented at:
`World Ophthalmology Congress (WOC); June 5-9, 2010;
`Berlin, Germany.
`15. Nguyen QD, Shah SM, Khwaja AA, et al. Two-year out-
`comes of the ranibizumab for edema of the mAcula in dia-
`betes (READ-2) study. Ophthalmology. 2010;117:
`2146-2151.
`16. Do D, Heier JS, Shah SM, et al. Two-year results of the
`READ-2 study: ranibizumab for edema of the macula in dia-
`betes. Presented at: Annual Meeting of the Association for
`Research in Vision and Ophthalmology; May 2-6, 2010;
`Fort Lauderdale, FL. Abstract 5842.
`17. Mitchell P. Efficacy and safety of intravitreal ranibizumab as
`adjunctive therapy to laser photocoagulation or as
`
`58
`
`Vol. 7, No. 2 I December 2010
`
`

`

`PROCEEDINGS
`
`monotherapy in diabetic macular edema: 12-month results
`from the RESTORE study. Presented at: 10th World
`Ophthalmology Congress; June 3-9, 2010; Berlin,
`Germany. Abstract 1629.
`18. Diabetic Retinopathy Clinical Research Network, Elman MJ,
`Aiello LP, et al. Randomized trial evaluating ranibizumab
`plus prompt or deferred laser or triamcinolone plus prompt
`laser for diabetic macular edema. Ophthalmology.
`2010;117:1064-1077.
`
`19. Sarraf D, Vyas N, Jain A, et al. Triamcinolone-associated
`maculopathy. Arch Ophthalmol. 2010;128:685-690.
`20. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for
`neovascular age-related macular degeneration. N Engl J
`Med. 2006;355:1419-1431.
`21. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab ver-
`sus verteporfin for neovascular age-related macular degen-
`eration. N Engl J Med. 2006;355:1432-1444.
`
`Johns Hopkins Advanced Studies in Ophthalmology I
`
`59
`
`