br j ophthalmol 2015 danis bjophthalmol 2015 306823

8202019 Br J Ophthalmol 2015 Danis Bjophthalmol 2015 306823

httpslidepdfcomreaderfullbr-j-ophthalmol-2015-danis-bjophthalmol-2015-306823 17

Anatomical effects of dexamethasone intravitrealimplant in diabetic macular oedema a pooledanalysis of 3-year phase III trials

Ronald P Danis1 Srinivas Sadda2 Xiao-Yan Li3 Harry Cui3 Yehia Hashad3

Scott M Whitcup

3

Additional material ispublished online only To viewplease visit the journal online(httpdxdoiorg101136bjophthalmol-2015-306823)

1Department of Ophthalmologyand Visual Sciences Universityof Wisconsin-MadisonMadison Wisconsin USA2Doheny Image ReadingCenter Doheny Eye InstituteLos Angeles California USA3Allergan Inc Irvine

California USA

Correspondence toDr Ronald P Danis Departmentof Ophthalmology and VisualSciences Fundus PhotographReading Center University of Wisconsin-Madison 2870University Ave Suite 102Madison WI 53705 USArpdaniswiscedu

Received 27 February 2015Revised 27 August 2015Accepted 3 September 2015

To cite Danis RP Sadda SLi X-Y et al Br JOphthalmol PublishedOnline First [ please includeDay Month Year]doi101136bjophthalmol-

2015-306823

ABSTRACTBackgroundaim To assess long-term effects of dexamethasone intravitreal implant (DEX implant)monotherapy on retinal morphology in diabetic macularoedema (DME)Methods Two multicentre masked phase III studieswith identical protocols randomised patients with DMEbest-corrected visual acuity of 34ndash68 Early TreatmentDiabetic Retinopathy Study letters and central sub1047297eldretinal thickness (CSRT) ge300 mm to DEX implant 07035 mg or sham procedure Patients were followed up

for 3 years (39 months if treated at month 36) withretreatment allowed at ge6-month intervals Patientsneeding other macular oedema (ME) therapy exited thestudy Changes from baseline in CSRT macular volumeand ME grade area of retinal thickening macularleakage macular capillary loss and diabetic retinopathyseverity were assessedResults After 3 years more eyes treated with DEXimplant 07 and 035 mg than sham showedimprovement (although small) in ME grade (plt005 vssham) DEX implant 07 mg delayed time to onset of two-step progression in diabetic retinopathy severity bysim12 months DEX implant 07 and 035 mg producedsmall non-sustained reductions in macular leakage but

had no signi1047297cant effect on macular capillary lossConclusions DEX implant 07 or 035 mgadministered at ge6-month intervals over 3 yearsproduced sustained retinal structural improvement inDMETrial registration number NCT00168337 andNCT00168389

INTRODUCTIONOver the last decade new therapeutic approacheshave resulted from improved understanding of thepathophysiological processes responsible for endo-

thelial bloodndash

retinal barrier breakdown in diabeticmacular oedema (DME)1 Several inter-relatedocular in1047298ammatory events are of particular rele-vance the release of vascular permeability factorssuch as vascular endothelial growth factor (VEGF)upregulation of in1047298ammatory mediators increasedexpression of endothelial adhesion molecules andthe in1047298ux and adhesion of leucocytes to the retinalmicrovasculature (leukostasis) resulting in endothe-lial cell injury and apoptosis2ndash5 Several VEGF inhi-bitors including a1047298ibercept bevacizumabpegaptanib and ranibizumab have shown clinicalef 1047297cacy as intravitreal therapies for DME6 7

Current treatment guidelines recognise the role of

ranibizumab in improving visual acuity in patients

with macular centre involvement and vision lossdue to DME8 However per-protocol clinical useof VEGF inhibitors in DME requires monthly intra-vitreal injections8 Moreover many patients exhibitretinal thickening despite anti-VEGF therapy high-lighting the need for additional treatments

Intravitreal corticosteroids downregulate expres-sion of cytokines such as tumour necrosis factor-α9

nuclear factor-κB9 V EGF9 and intercellular adhe-sion molecule-19 10 and inhibit leukostasis andretinal microvasculature leakage9 10 Fluocinolone

acetonide (Retisert Bausch amp Lomb BridgewaterNew Jersey USA Iluvien Alimera Sciences Alpharetta Georgia USA) and dexamethasone(DEX implant Ozurdex Allergan IrvineCalifornia USA) are available commercially asslow-release intravitreal implant systems DEXimplant has recently been approved for the treat-ment of adults with DME In early-phase clinicaltrials in DME DEX implant monotherapy11 andcombination laser therapy12 were effective inimproving central sub1047297eld retinal thickness (CSRT)and best-corrected visual acuity (BCVA) These1047297ndings have recently been con1047297rmed in twopivotal phase III trials of DEX implant in DME

(the MEAD studies) In a pooled analysis of theMEAD study data DEX implant met the primaryef 1047297cacy end point of ge15-letter improvement inBCVA and showed acceptable tolerability13 Theanatomical 1047297ndings from the MEAD studies aredetailed here

METHODSStudy design and participantsData collected from two identically designed3-year multicentre masked phase III trials(NCT00168337 and NCT00168389) of the safetyand ef 1047297cacy of DEX implant in the treatment of

DME (MEAD trials) were pooled The trials wereconducted at 131 sites in 22 countries worldwidebetween February 2005 and June 2012 The proto-col was approved by the institutional review boardethics committee at each study site and the trialswere conducted in accordance with the Declarationof Helsinki All patients provided written informedconsent The protocol is described in detail else-where13 and is summarised below

Adults with diabetic retinopathyndashassociatedmacular oedema involving the fovea previouslytreated with medical or laser therapy and an EarlyTreatment Diabetic Retinopathy Study (ETDRS)BCVA in the study eye of 34ndash68 letters (20200ndash

2050) were enrolled Treatment-naiumlve patients

Danis RP et al Br J Ophthalmol 201501ndash6 doi101136bjophthalmol-2015-306823 1

Clinical science

BJO Online First published on November 27 2015 as 101136bjophthalmol-2015-306823

Copyright Article author (or their employer) 2015 Produced by BMJ Publishing Group Ltd under licence

groupbmjcomon January 26 2016 - Published by httpbjobmjcom Downloaded from



refusing laser treatment or who in the investigatorrsquos opinionwould not bene1047297t from laser treatment also were eligibleRetinal thickness in the central 1 mm macular sub1047297eld measuredby time-domain optical coherence tomography (TD-OCTStratus OCT3 or OCT2 (used for only 1 of study submis-sions) Carl Zeiss Meditec Dublin California USA) wasrequired to be ge300 mm in the study eye Key exclusion criteriawere uncontrolled diabetes (glycosylated haemoglobin gt10)glaucoma ocular hypertension (untreated intraocular pressure

gt23 mm Hg) aphakia active iris or retinal neovascularisationhistory of pars plana vitrectomy or steroid-induced ocularhypertension recent intraocular laser or incisional surgery orintravitreal VEGF inhibitor or triamcinolone treatment andcurrent use of systemic steroids

Patients were randomised (111) to intravitreal DEX implant07 mg DEX implant 035 mg or a sham procedure in the studyeye If both eyes were eligible the eye with the shorter durationof macular oedema was selected Patients were evaluated forretreatment every 3 months during the 3-year study but retreat-ment could not be performed more often than every 6 monthsRetreatment was allowed if retinal thickness in the 1 mm centralmacular sub1047297eld was gt225 mm (revised to gt175 mm or evi-

dence of residual oedema in a protocol amendment in 2010)Patients needing adjunctive or other therapy for macularoedema were required to exit the study prior to administrationof additional treatment Ef 1047297cacy data captured before patientexit were included in the analysis

Study assessments and end pointsThe primary ef 1047297cacy end point in the MEAD trials was the per-centage of patients with ge15-letter improvement in BCVA frombaseline in the study eye at 1047297nal assessment (end of year 3 or39 months for patients treated at month 36) Prespeci1047297ed sec-ondary end points included changes in retinal anatomy mea-sured using OCT fundus photography and 1047298uorescein

angiography Image evaluation (grading) was performed at acentralised reading facility (University of Wisconsin FundusPhotograph Reading Center Madison Wisconsin USA) by cer-ti1047297ed masked technicians

TD-OCT (Stratus OCT3 or if unavailable OCT2) was con-ducted at 3-month intervals Six radial scans each sim6 mm longand centred on the fovea and performed using fast macularthickness map settings (128 A-scansB-scan) were supplementedby high-resolution 6 mm cross-hair scans (512 A-scansB-scan)Stereoscopic 30deg or 35deg colour fundus photographs of the studyeye were taken at baseline every 3 months during the 1047297rst yearand every 6 months during the second and third years Fundusphotographs were assessed for presence and extent of retinal

thickening diabetic retinopathy severity level and presence of clinically signi1047297cant macular oedema (CSME) Diabetic retinop-athy was graded using the ETDRS Final Retinopathy SeverityScale condensed to nine severity categories14 Outcomes of interest included changes from baseline in disc area of centralretinal thickening and macular oedema grade (improvement nochange or worsening)

Fluorescein angiography was performed at baseline months6 12 and 24 and at the end of year 3 (or 39 months for patientstreated at month 36) to assess macular 1047298uorescein leakage andperifoveal capillary integrity Grading protocols were adaptedfrom the ETDRS clinical trials15 The mean change from base-line to study end in total disc area of macular capillary loss andthe proportions of patients with and without ischaemia (de1047297ned

as a total area of macular capillary loss gt05 disc area) at

baseline and the last visit were determined Details of the assess-ments are described in the online supplementary materials

Statistical analysesUnless stated otherwise all ef 1047297cacy analyses were performedwith missing values imputed by last observation carried forward(LOCF) for the intent-to-treat (ITT) population (all randomisedpatients) Treatment comparisons and estimates based on LOCFwere supported by sensitivity analyses using multiple imput-

ation Area under the curve (AUC) analysis of the averagechange from baseline in CSRT during the study used observedvalues in the ITT population missing values were not imputed

Areas of central retinal thickening 1047298uorescein leakage andmacular capillary loss and changes from baseline in CSRT andmacular volume were analysed using analysis of covariance withbaseline value as a covariate Changes from baseline in propor-tions of patients with CSME and central retinal thickening werecompared using Wilcoxon rank-sum test Proportions of patientsin each diabetic retinopathy severity category and proportionsof patients with ge2-step progression from baseline in diabeticretinopathy severity category were analysed using the Cochranndash

MantelndashHaenszel general association test strati1047297ed by study

Time to two-step progression in diabetic retinopathy severitycategory was analysed using the KaplanndashMeier method andcumulative progression rates were compared using the log-ranktest All statistical tests were two-sided and performed atα=005 signi1047297cance level SAS V93 (SAS Institute Cary NorthCarolina USA) was used

RESULTSStudy populationThe pooled ITT population comprised 1048 randomisedpatients (table 1) of whom 607 patients (579) completed allvisits Completion rates were appreciably higher in the DEXimplant 07 mg (641) and 035 mg (663) groups than inthe sham group (434)

The median number of study treatments administered in eachDEX implant group ranged between 4 and 5 compared with 3in the sham group Baseline demographics and study eyecharacteristics did not differ signi1047297cantly among the three treat-ment groups (table 1)

OCT 1047297ndingsDEX implantndashtreated eyes showed marked 1047298uctuation in thereduction in CSRT at consecutive study visits particularlyduring year 1 creating a saw-tooth pattern of treatment effect(1047297gure 1 lt1 of scans were deemed non-gradable) Similarresults were obtained with the observed data analysis Studyeyes treated with DEX implant showed greater reductions from

baseline in CSRT than sham-treated eyes at all time points(1047297gure 1) At the 1047297nal study visit CSRT was reduced by (mean) 1173

and 1278 μm in DEX implant 07ndash and 035 mgndashtreated eyesversus 621 μm in sham-treated eyes (both plt0001 vs shamtable 2) Over the full study period the mean average reductionin CSRT (AUC approach) was greater in DEX implantndashtreatedeyes (table 2) At all study time points from month 3 onwardthe proportion of study eyes with CSRT gt250 μm was lower inthe DEX implant than the sham treatment groups At the 1047297nalstudy visit the decline in the proportion of study eyes in thiscategory was greater with DEX implant 07 mg (from 945(baseline) to 602) and DEX implant 035 mg (from 948 to587) than with sham (from 959 to 716) Likewise at all

study time points DEX implantndashtreated eyes displayed greater

2 Danis RP et al Br J Ophthalmol 201501ndash6 doi101136bjophthalmol-2015-306823

Clinical science




reductions from baseline in macular volume than sham-treatedeyes (both doses plt0001 vs sham at study end 1047297gure 2 and

table 2)

Fundus photography 1047297ndingsThe area of central retinal thickening showed greater reduction

from baseline in DEX implant-treated versus sham-treated eyes

Table 1 Baseline demographic and study eye characteristics of the ITT population

Characteristic DEX implant 07 mg DEX implant 035 mg Sham procedure

Demographic characteristics

Age years N=351 N=347 N=350

Mean (SD) 625 (83) 623 (92) 625 (95)

Range 33ndash85 25ndash84 26ndash88

Gender n () N=351 N=347 N=350

Male 213 (607) 206 (594) 217 (620)Raceethnicity n () N=351 N=347 N=350

Asian 55 (157) 58 (167) 54 (154)

Black 16 (46) 16 (46) 20 (57)

Caucasian 234 (667) 234 (674) 233 (666)

Hispanic 35 (100) 34 (98) 33 (94)

Other 11 (31) 5 (14) 10 (29)

Diabetes type n () N=351 N=347 N=350

Type 1 34 (97) 22 (63) 28 (80)

Type 2 314 (895) 325 (937) 322 (920)

Not available 3 (09) 0 0

Duration of diabetes years N=349 N=347 N=348

Mean (SD) 165 (90) 158 (94) 159 (91)

HbA1c N=347 N=345 N=349

Mean (SD) 76 (12) 75 (11) 75 (11)

Study eye characteristics

ETDRS letter score N=351 N=347 N=350

Mean (SD) 561 (99) 555 (97) 569 (87)

CSRT mm N=348 N=344 N=342

Mean (SD) 4630 (1571) 4668 (1595) 4609 (1326)

Study eyes with CSRT gt250 mm 945 948 959

Duration of DME months N=350 N=347 N=349

Mean (SD) 236 (260) 252 (314) 259 (273)


DME classification n () N=351 N=347 N=350

Focal 127 (362) 136 (392) 142 (406)

Intermediate 134 (382) 124 (357) 122 (349)

Diffuse 69 (197) 60 (173) 72 (206)Not available 16 (46) 25 (72) 10 (29)

None 5 (14) 2 (06) 4 (11)

Previous DME treatment n () N=351 N=347 N=350

Focalgrid laser 231 (658) 224 (646) 243 (694)

Intravitreal steroid 58 (165) 69 (199) 61 (174)

Anti-VEGF 25 (71) 39 (112) 26 (74)

None 104 (296) 98 (282) 89 (254)

Severity of NPDR n () N=351 N=347 N=350

Moderate or better 173 (493) 170 (490) 174 (497)

Severe or worse 151 (430) 151 (435) 149 (426)

Not available 27 (77) 26 (75) 27 (77)

Macular perfusion status n () N=351 N=347 N=350

Ischaemicdagger 43 (123) 31 (89) 27 (77)

Non-ischaemic 257 (732) 260 (749) 284 (811)

Not available 51 (145) 56 (161) 39 (111)

Lens status n () N=351 N=347 N=350

Phakic 265 (755) 259 (749) 249 (711)

Pseudophakic 86 (245) 88 (254) 101 (289)

DME classification based upon clinical assessment by the treating physiciandaggerArea of macular capillary loss on fluorescein angiography gt05 disc areas Analysis based on Fisher exact testIntergroup comparisons performed using Pearsonrsquos χ2 test (categorical variables) and a one-way analysis of variance model (continuous variables)CSRT central subfield retinal thickness DEX implant dexamethasone intravitreal implant DME diabetic macular oedema ETDRS Early Treatment Diabetic Retinopathy Study HbA1cglycosylated haemoglobin ITT intent-to-treat NPDR non-proliferative diabetic retinopathy VEGF vascular endothelial growth factor


Clinical science






Compared with sham-treated patients DEX implant-treatedpatients showed a delay of sim12 months in onset of two-stepprogression in diabetic retinopathy severity (10th percentile of

time to progression was sim

36 months for both the DEX implant07 mg and 035 mg treatment groups vs sim24 months for thesham treatment group p=003 and p=008 respectively) The10th percentile of time to two-step improvement in diabetic ret-inopathy severity was sim24 and sim13 months for the DEXimplant 07 and 035 mg treatment groups versus sim24 monthsfor sham (p=0655 and p=0364 respectively)

The prevalence of CSME declined steadily in both DEXimplant treatment groups and more gradually in the sham group

At study end fewer DEX implant 07ndash and 035 mgndashtreated eyeshad CSME than sham-treated eyes (plt005 and plt001 respect-ively table 2) More DEX implant 07ndash and 035 mgndashtreated eyesshowed an improvement in CSME (shift from a higher to a lowergrade) between baseline and study end than sham-treated eyes

(plt005 and plt001 respectively table 2)

Fluorescein angiography 1047297ndingsDEX implant 07 mgndash and DEX implant 035 mgndashtreated eyesshowed greater reductions (from baseline) in total area of macular 1047298uorescein leakage than sham-treated eyes during the1047297rst year (mean minus035 and minus052 vs +011 disc areas at6 months both plt005 vs sham mean minus058 and minus079 vsminus003 disc areas at 12 months plt001 for DEX implant035 mg vs sham) However these effects were modest andwere not sustained at study end the reductions in total area of 1047298uorescein leakage in the DEX implant treatment groups werenot signi1047297cantly different from those of the sham group

(table 2) Changes (from baseline) in total area of macular capil-lary loss did not differ signi1047297cantly between DEX implantndash andsham-treated eyes at any time (table 2)

DISCUSSIONThis pooled analysis of data from the MEAD trials 13 is the 1047297rstto examine the long-term retinal changes associated with thevisual acuity improvements produced by intravitreal DEXimplant in DME DEX implant administered with a median fre-quency of four to 1047297ve injections over a 3-year period providedsustained improvements in OCT and fundus photographyndashbasedanatomical markers of macular oedema whereas macular 1047298uor-escein leakage showed improvement only over the 1047297rst yearCompared with sham-treated eyes DEX implantndashtreated eyes

displayed signi1047297cant decreases in CSRT macular volume and

central retinal thickening at all study time points with treatmentdifferences emerging as early as month 3 There was no signi1047297-cant effect on macular capillary loss

The present 1047297

ndings extend the limited anatomical informa-tion provided by earlier short-term phase II studies of DEXimplant 07 and 035 mg in patients with DME12 16 17 Inkeeping with previous studies suggesting that the ef 1047297cacy of DEX implant on CSRT peaks at approximately 1ndash3 monthsbefore gradually declining11 12 16 the pro1047297le of mean change inCSRT versus time observed in the present study was charac-terised by a saw-tooth pattern (1047297gure 1) with each cycle of improvement presumably corresponding to retreatment withDEX implant A similar saw-tooth pattern was obtained withobserved data This pattern is most likely explained by thegradual decline over time in DEX release from the implant insitu and suggests that a retreatment interval of less than6 months may be required for a more consistent anatomical

responseThe reduction in macular 1047298uorescein leakage noted during

year 1 is consistent with 1047297ndings from previous short-termphase II studies12 16 17 However the reason for the discrepancybetween the longer-term OCT and 1047298uorescein angiography 1047297nd-ings is unclear Although 1047298uorescein angiography is a reliablemethod for qualitative assessment of 1047298uid leakage a reductionin 1047298uorescein intensity is not always accompanied by a reductionin area of 1047298uorescein leakage Likewise 1047298uorescein angiography1047297ndings do not always mirror OCT 1047297ndings in DME in somecases macular leakage may be evident in the absence of anincrease in retinal thickness in other cases 1047298uorescein angiog-raphy may fail to detect intraretinal or subretinal 1047298uid that is

evident on OCT

18

Additionally the diffuse 1047298

uorescein stainingof non-cystoid oedema in DME may be below the detectionthreshold of the OCT instrument18 19

Strengths of the present study include its large patient popula-tion its extended duration and inclusion of a sham treatmentarm investigator and patient masking to treatment and a rangeof end points The study design ensured that treatment out-comes were not complicated by the effects of adjunctive DMEtherapies However the requirement for subjects requiringadjunctive treatment to exit the study resulted in high attritionparticularly in the control group which may have adverselyaffected ITTLOCF analyses Other potential study limitationsinclude the lack of adjustment for glycaemic and blood pressurecontrol which may affect macular thickness20 and the lower

reproducibility of macular thickness measurements obtained

Figure 2 Mean change from baselinein macular volume versus timeple0002 at all time points fordexamethasone intravitreal (DEX)implants versus sham (analysis of covariance with treatment as a 1047297xedeffect and baseline value as covariate)B baseline F 1047297nal visit


Clinical science




with TD-OCT the standard technology when the study wasinitiated (2004) compared with current spectral-domain OCTinstruments 21 Variability is also likely to have arisen from sub-stitution of centre-point thickness for CSRT owing to scanquality issues and possibly from the natural diurnal variation inmacular thickness in eyes with DME22 Nevertheless manualcentre-point thickness measurements performed by the readingcentre operators showed excellent reproducibility

In summary this pooled data analysis extends the evidence from

previous studies in DME indicating that DEX implant 07 and035 mg provides sustained improvements in anatomical measuresof macular oedema over a 3-year treatment period DEX implantthus has the potential to reduce the need for laser therapy and pro-vides an alternative to anti-VEGF therapy in DME

Acknowledgements Writing and editorial assistance was provided to the authorsby Andrew Fitton PhD of Evidence Scienti1047297c Solutions (Horsham UK) The full listof participating investigators is published in the primary manuscript

Contributors RPD and X-YL are guarantors for this paper All authors participatedin the study design data interpretation and drafting critical review revision andapproval of the manuscript HC conducted the data analysis The study sponsorparticipated in the design of the study data analysis and data interpretation andpreparation review and approval of the manuscript

Funding The study and medical writing support were sponsored by Allergan IncIrvine California USA

Competing interests RPD and SS have received grant support and consulting feesfrom Allergan Inc X-YL and YH are employees of Allergan Inc At the time of thestudy HC and SMW were employees of Allergan Inc

Ethics approval As this was a multicentre study several IECs and IRBs approvedthe study

Provenance and peer review Not commissioned externally peer reviewed

Data sharing statement For additional unpublished data from the study Allerganshould be contacted

Open Access This is an Open Access article distributed in accordance with theCreative Commons Attribution Non Commercial (CC BY-NC 40) license whichpermits others to distribute remix adapt build upon this work non-commerciallyand license their derivative works on different terms provided the original work isproperly cited and the use is non-commercial See httpcreativecommonsorg

licensesby-nc40

REFERENCES1 Klaassen I Van Noorden CJ Schlingemann RO Molecular basis of the inner

blood-retinal barrier and its breakdown in diabetic macular edema and otherpathological conditions Prog Retin Eye Res 20133419ndash48

2 Adamis AP Berman AJ Immunological mechanisms in the pathogenesis of diabeticretinopathy Semin Immunopathol 20083065ndash84

3 Bhagat N Grigorian RA Tutela A et al Diabetic macular edema pathogenesis andtreatment Surv Ophthalmol 2009541ndash32

4 Ehrlich R Harris A Ciulla TA et al Diabetic macular oedema physicalphysiological and molecular factors contribute to this pathological process ActaOphthalmol 201088279ndash91

5 Tang J Kern TS In1047298ammation in diabetic retinopathy Prog Retin Eye Res201130343ndash58

6 Boyer DS Hopkins JJ Sorof J et al Anti-vascular endothelial growth factor therapyfor diabetic macular edema Ther Adv Endocrinol Metab 20134151ndash69

7 Song SJ Wong TY Current concepts in diabetic retinopathy Diabetes Metab J201438416ndash25

8 Bandello F Cunha-Vaz J Chong NV et al New approaches for the treatment of diabetic macular oedema recommendations by an expert panel Eye

201226485ndash

939 Kim YH Choi MY Kim YS et al Triamcinolone acetonide protects the rat retina

from STZ-induced acute in1047298ammation and early vascular leakage Life Sci 2007811167ndash73

10 Tamura H Miyamoto K Kiryu J et al Intravitreal injection of corticosteroidattenuates leukostasis and vascular leakage in experimental diabetic retina Invest Ophthalmol Vis Sci 2005461440ndash4

11 Pacella E Vestri AR Muscella R et al Preliminary results of an intravitrealdexamethasone implant (Ozurdexreg) in patients with persistent diabetic macularedema Clin Ophthalmol 201371423ndash8

12 Callanan DG Gupta S Boyer DS et al for the Ozurdex PLACID Study GroupDexamethasone intravitreal implant in combination with laser photocoagulation forthe treatment of diffuse diabetic macular edema Ophthalmology 20131201843ndash51

13 Boyer DS Yoon YH Belfort R Jr et al for the Ozurdex MEAD Study GroupThree-year randomized sham-controlled trial of dexamethasone intravitreal implant

in patients with diabetic macular edema Ophthalmology 20141211904ndash

1414 [No authors listed] Grading diabetic retinopathy from stereoscopic color fundusphotographsmdashan extension of the modi1047297ed Airlie House classi1047297cation ETDRSreport number 10 Early Treatment Diabetic Retinopathy Study Research GroupOphthalmology 199198(Suppl 5)786ndash806

15 [No authors listed] Classi1047297cation of diabetic retinopathy from 1047298uoresceinangiograms ETDRS report number 11 Early Treatment Diabetic Retinopathy StudyResearch Group Ophthalmology 199198(Suppl 5)807ndash22

16 Boyer DS Faber D Gupta S et al for the Ozurdex CHAMPLAIN Study GroupDexamethasone intravitreal implant for treatment of diabetic macular edema invitrectomized patients Retina 201131915ndash23

17 Haller JA Kuppermann BD Blumenkranz MS et al Randomized controlled trial of an intravitreous dexamethasone drug delivery system in patients with diabeticmacular edema Arch Ophthalmol 2010128289ndash96

18 Kozak I Morrison VL Clark TM et al Discrepancy between 1047298uorescein angiographyand optical coherence tomography in detection of macular disease Retina200828538ndash44

19 Otani T Kishi S Correlation between optical coherence tomography and1047298uorescein angiography 1047297ndings in diabetic macular edema Ophthalmology 2007114104ndash7

20 Stratton IM Kohner EM Aldington SJ et al for the UKPDS Group UKPDS 50 riskfactors for incidence and progression of retinopathy in type II diabetes over 6 yearsfrom diagnosis Diabetologia 200144156ndash63

21 Leung CK Cheung CY Weinreb RN et al Comparison of macular thicknessmeasurements between time domain and spectral domain optical coherencetomography Invest Ophthalmol Vis Sci 2008494893ndash7

22 Frank RN Schulz L Abe K et al Temporal variation in diabetic macular edemameasured by optical coherence tomography Ophthalmology 2004111211ndash17


Clinical science




trialsoedema a pooled analysis of 3-year phase IIIintravitreal implant in diabetic macularAnatomical effects of dexamethasone

and Scott M WhitcupRonald P Danis Srinivas Sadda Xiao-Yan Li Harry Cui Yehia Hashad

published online November 18 2015Br J Ophthalmol

3httpbjobmjcomcontentearly20151127bjophthalmol-2015-30682Updated information and services can be found at

These include

MaterialSupplementary

23DC1htmlhttpbjobmjcomcontentsuppl20151118bjophthalmol-2015-3068Supplementary material can be found at

References

BIBL3httpbjobmjcomcontentearly20151127bjophthalmol-2015-30682This article cites 22 articles 3 of which you can access for free at

Open Access

httpcreativecommonsorglicensesby-nc40 non-commercial Seeprovided the original work is properly cited and the use isnon-commercially and license their derivative works on different termspermits others to distribute remix adapt build upon this workCommons Attribution Non Commercial (CC BY-NC 40) license whichThis is an Open Access article distributed in accordance with the Creative

serviceEmail alerting

box at the top right corner of the online articleReceive free email alerts when new articles cite this article Sign up in the

CollectionsTopic Articles on similar topics can be found in the following collections

(1540)Retina (209)Open access

Notes

httpgroupbmjcomgrouprights-licensingpermissionsTo request permissions go to

httpjournalsbmjcomcgireprintformTo order reprints go to

httpgroupbmjcomsubscribeTo subscribe to BMJ go to




refusing laser treatment or who in the investigatorrsquos opinionwould not bene1047297t from laser treatment also were eligibleRetinal thickness in the central 1 mm macular sub1047297eld measuredby time-domain optical coherence tomography (TD-OCTStratus OCT3 or OCT2 (used for only 1 of study submis-sions) Carl Zeiss Meditec Dublin California USA) wasrequired to be ge300 mm in the study eye Key exclusion criteriawere uncontrolled diabetes (glycosylated haemoglobin gt10)glaucoma ocular hypertension (untreated intraocular pressure

gt23 mm Hg) aphakia active iris or retinal neovascularisationhistory of pars plana vitrectomy or steroid-induced ocularhypertension recent intraocular laser or incisional surgery orintravitreal VEGF inhibitor or triamcinolone treatment andcurrent use of systemic steroids

Patients were randomised (111) to intravitreal DEX implant07 mg DEX implant 035 mg or a sham procedure in the studyeye If both eyes were eligible the eye with the shorter durationof macular oedema was selected Patients were evaluated forretreatment every 3 months during the 3-year study but retreat-ment could not be performed more often than every 6 monthsRetreatment was allowed if retinal thickness in the 1 mm centralmacular sub1047297eld was gt225 mm (revised to gt175 mm or evi-

dence of residual oedema in a protocol amendment in 2010)Patients needing adjunctive or other therapy for macularoedema were required to exit the study prior to administrationof additional treatment Ef 1047297cacy data captured before patientexit were included in the analysis

Study assessments and end pointsThe primary ef 1047297cacy end point in the MEAD trials was the per-centage of patients with ge15-letter improvement in BCVA frombaseline in the study eye at 1047297nal assessment (end of year 3 or39 months for patients treated at month 36) Prespeci1047297ed sec-ondary end points included changes in retinal anatomy mea-sured using OCT fundus photography and 1047298uorescein

angiography Image evaluation (grading) was performed at acentralised reading facility (University of Wisconsin FundusPhotograph Reading Center Madison Wisconsin USA) by cer-ti1047297ed masked technicians

TD-OCT (Stratus OCT3 or if unavailable OCT2) was con-ducted at 3-month intervals Six radial scans each sim6 mm longand centred on the fovea and performed using fast macularthickness map settings (128 A-scansB-scan) were supplementedby high-resolution 6 mm cross-hair scans (512 A-scansB-scan)Stereoscopic 30deg or 35deg colour fundus photographs of the studyeye were taken at baseline every 3 months during the 1047297rst yearand every 6 months during the second and third years Fundusphotographs were assessed for presence and extent of retinal

thickening diabetic retinopathy severity level and presence of clinically signi1047297cant macular oedema (CSME) Diabetic retinop-athy was graded using the ETDRS Final Retinopathy SeverityScale condensed to nine severity categories14 Outcomes of interest included changes from baseline in disc area of centralretinal thickening and macular oedema grade (improvement nochange or worsening)

Fluorescein angiography was performed at baseline months6 12 and 24 and at the end of year 3 (or 39 months for patientstreated at month 36) to assess macular 1047298uorescein leakage andperifoveal capillary integrity Grading protocols were adaptedfrom the ETDRS clinical trials15 The mean change from base-line to study end in total disc area of macular capillary loss andthe proportions of patients with and without ischaemia (de1047297ned

as a total area of macular capillary loss gt05 disc area) at

baseline and the last visit were determined Details of the assess-ments are described in the online supplementary materials

Statistical analysesUnless stated otherwise all ef 1047297cacy analyses were performedwith missing values imputed by last observation carried forward(LOCF) for the intent-to-treat (ITT) population (all randomisedpatients) Treatment comparisons and estimates based on LOCFwere supported by sensitivity analyses using multiple imput-

ation Area under the curve (AUC) analysis of the averagechange from baseline in CSRT during the study used observedvalues in the ITT population missing values were not imputed

Areas of central retinal thickening 1047298uorescein leakage andmacular capillary loss and changes from baseline in CSRT andmacular volume were analysed using analysis of covariance withbaseline value as a covariate Changes from baseline in propor-tions of patients with CSME and central retinal thickening werecompared using Wilcoxon rank-sum test Proportions of patientsin each diabetic retinopathy severity category and proportionsof patients with ge2-step progression from baseline in diabeticretinopathy severity category were analysed using the Cochranndash

MantelndashHaenszel general association test strati1047297ed by study

Time to two-step progression in diabetic retinopathy severitycategory was analysed using the KaplanndashMeier method andcumulative progression rates were compared using the log-ranktest All statistical tests were two-sided and performed atα=005 signi1047297cance level SAS V93 (SAS Institute Cary NorthCarolina USA) was used

RESULTSStudy populationThe pooled ITT population comprised 1048 randomisedpatients (table 1) of whom 607 patients (579) completed allvisits Completion rates were appreciably higher in the DEXimplant 07 mg (641) and 035 mg (663) groups than inthe sham group (434)

The median number of study treatments administered in eachDEX implant group ranged between 4 and 5 compared with 3in the sham group Baseline demographics and study eyecharacteristics did not differ signi1047297cantly among the three treat-ment groups (table 1)

OCT 1047297ndingsDEX implantndashtreated eyes showed marked 1047298uctuation in thereduction in CSRT at consecutive study visits particularlyduring year 1 creating a saw-tooth pattern of treatment effect(1047297gure 1 lt1 of scans were deemed non-gradable) Similarresults were obtained with the observed data analysis Studyeyes treated with DEX implant showed greater reductions from

baseline in CSRT than sham-treated eyes at all time points(1047297gure 1) At the 1047297nal study visit CSRT was reduced by (mean) 1173

and 1278 μm in DEX implant 07ndash and 035 mgndashtreated eyesversus 621 μm in sham-treated eyes (both plt0001 vs shamtable 2) Over the full study period the mean average reductionin CSRT (AUC approach) was greater in DEX implantndashtreatedeyes (table 2) At all study time points from month 3 onwardthe proportion of study eyes with CSRT gt250 μm was lower inthe DEX implant than the sham treatment groups At the 1047297nalstudy visit the decline in the proportion of study eyes in thiscategory was greater with DEX implant 07 mg (from 945(baseline) to 602) and DEX implant 035 mg (from 948 to587) than with sham (from 959 to 716) Likewise at all

study time points DEX implantndashtreated eyes displayed greater


Clinical science





table 2)







Mean (SD) 625 (83) 623 (92) 625 (95)


Gender n () N=351 N=347 N=350


Asian 55 (157) 58 (167) 54 (154)

Black 16 (46) 16 (46) 20 (57)

Caucasian 234 (667) 234 (674) 233 (666)

Hispanic 35 (100) 34 (98) 33 (94)

Other 11 (31) 5 (14) 10 (29)


Type 1 34 (97) 22 (63) 28 (80)

Type 2 314 (895) 325 (937) 322 (920)



Mean (SD) 165 (90) 158 (94) 159 (91)

HbA1c N=347 N=345 N=349

Mean (SD) 76 (12) 75 (11) 75 (11)



Mean (SD) 561 (99) 555 (97) 569 (87)

CSRT mm N=348 N=344 N=342

Mean (SD) 4630 (1571) 4668 (1595) 4609 (1326)



Mean (SD) 236 (260) 252 (314) 259 (273)



Focal 127 (362) 136 (392) 142 (406)

Intermediate 134 (382) 124 (357) 122 (349)


None 5 (14) 2 (06) 4 (11)


Focalgrid laser 231 (658) 224 (646) 243 (694)


Anti-VEGF 25 (71) 39 (112) 26 (74)

None 104 (296) 98 (282) 89 (254)



Severe or worse 151 (430) 151 (435) 149 (426)

Not available 27 (77) 26 (75) 27 (77)



Non-ischaemic 257 (732) 260 (749) 284 (811)

Not available 51 (145) 56 (161) 39 (111)


Phakic 265 (755) 259 (749) 249 (711)

Pseudophakic 86 (245) 88 (254) 101 (289)



Clinical science

















The present 1047297




evident on OCT

18







Clinical science















licensesby-nc40










201226485ndash


















Clinical science








These include



References


Open Access






Notes








table 2)







Mean (SD) 625 (83) 623 (92) 625 (95)


Gender n () N=351 N=347 N=350


Asian 55 (157) 58 (167) 54 (154)

Black 16 (46) 16 (46) 20 (57)

Caucasian 234 (667) 234 (674) 233 (666)

Hispanic 35 (100) 34 (98) 33 (94)

Other 11 (31) 5 (14) 10 (29)


Type 1 34 (97) 22 (63) 28 (80)

Type 2 314 (895) 325 (937) 322 (920)



Mean (SD) 165 (90) 158 (94) 159 (91)

HbA1c N=347 N=345 N=349

Mean (SD) 76 (12) 75 (11) 75 (11)



Mean (SD) 561 (99) 555 (97) 569 (87)

CSRT mm N=348 N=344 N=342

Mean (SD) 4630 (1571) 4668 (1595) 4609 (1326)



Mean (SD) 236 (260) 252 (314) 259 (273)



Focal 127 (362) 136 (392) 142 (406)

Intermediate 134 (382) 124 (357) 122 (349)


None 5 (14) 2 (06) 4 (11)


Focalgrid laser 231 (658) 224 (646) 243 (694)


Anti-VEGF 25 (71) 39 (112) 26 (74)

None 104 (296) 98 (282) 89 (254)



Severe or worse 151 (430) 151 (435) 149 (426)

Not available 27 (77) 26 (75) 27 (77)



Non-ischaemic 257 (732) 260 (749) 284 (811)

Not available 51 (145) 56 (161) 39 (111)


Phakic 265 (755) 259 (749) 249 (711)

Pseudophakic 86 (245) 88 (254) 101 (289)



Clinical science

















The present 1047297




evident on OCT

18







Clinical science















licensesby-nc40










201226485ndash


















Clinical science








These include



References


Open Access






Notes




















The present 1047297




evident on OCT

18







Clinical science















licensesby-nc40










201226485ndash


















Clinical science








These include



References


Open Access






Notes


















licensesby-nc40










201226485ndash


















Clinical science








These include



References


Open Access






Notes











These include



References


Open Access






Notes





br j ophthalmol 2015 danis bjophthalmol 2015 306823

Documents