BMJ Open. For peer review only -

Transcription

1 Comparative efficacy and safety of treatments for macular oedema secondary to branch retinal vein occlusion: a network meta-analysis Journal: Manuscript ID: bmjopen-0-00 Article Type: Research Date Submitted by the Author: -Dec-0 Complete List of Authors: Regnier, Stephane; Novartis Pharma AG, Health Economics and Outcomes Research Larsen, Michael; University of Copenhagen, Department of Ophthalmology, Glostrup Hospital Bezlyak, Vladimir; Novartis Pharma AG, Allen, Felicity; Novartis Pharmaceuticals UK Ltd, <b>primary Subject Heading</b>: Ophthalmology Secondary Subject Heading: Ophthalmology Keywords: Medical retina < OPHTHALMOLOGY, Health economics < HEALTH SERVICES ADMINISTRATION & MANAGEMENT, OPHTHALMOLOGY Note: The following files were submitted by the author for peer review, but cannot be converted to PDF. You must view these files (e.g. movies) online. WinBUGS file for reviewers - Dec 0.odc : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

2 Page of Comparative efficacy and safety of treatments for macular oedema secondary to branch retinal vein occlusion: a network meta-analysis Stephane A Regnier, Michael Larsen, Vladimir Bezlyak, Felicity Allen Author affiliations Novartis Pharma AG, Basel, Switzerland Department of Ophthalmology, Glostrup Hospital, University of Copenhagen, Glostrup, Denmark Novartis Pharmaceuticals UK Ltd, Frimley Business Park, Frimley, Surrey, UK Corresponding author: Stephane Regnier Novartis Campus WSJ CH-0 Basel, Switzerland + (0) 0 Stephane.regnier@novartis.com Suggested keywords/phrases Ranibizumab Macular oedema Branch retinal vein occlusion Anti-vascular endothelial growth factor Bayesian network meta-analysis : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

3 Page of ABSTRACT Objective: To compare the efficacy and safety of treatments for macular oedema secondary to branch retinal vein occlusion (BRVO). Design: Randomized controlled trials (RCTs) evaluating the efficacy and safety of treatments for macular oedema secondary to BRVO were identified from an updated systematic review and a manual search of the ClinicalTrials.gov registry, abstracts from ophthalmology congresses and data held on file. Setting: A Bayesian network meta-analysis of RCTs of treatments for macular oedema secondary to BRVO. Participants: A total of adult patients across the identified RCTs. Interventions: Ranibizumab 0. mg pro re nata (as needed), aflibercept mg monthly (q), dexamethasone 0. mg implant, laser photocoagulation, ranibizumab + laser, or sham intervention. Outcome measures: Efficacy outcomes were mean change in best corrected visual acuity (BCVA; assessed as Early Treatment Diabetic Retinopathy Study letters) and the percentage of patients gaining letters. Safety outcome was the percentage of patients with increased intra-ocular pressure (IOP)/ocular hypertension (OH). Results: Eight RCTs were identified for inclusion. The probability of being the most efficacious treatment at month or month based on letters gained was % for ranibizumab monotherapy, 0% for aflibercept, % for ranibizumab plus laser (adjunctive or prompt), and 0% for dexamethasone implant, laser or sham. Ranibizumab monotherapy was numerically superior to aflibercept (+. letters [% credible interval (CrI):. to.]; odds ratio [OR] for gaining letters:.0 [% CrI: 0. to.]). Dexamethasone implant was associated with significantly higher IOP/OH than anti-vascular endothelial growth factor agents (OR:. [% CrI:. to.]). Conclusions: This Bayesian network meta-analysis confirmed the superiority of ranibizumab over dexamethasone implant or laser photocoagulation for the treatment of macular oedema secondary to BRVO and showed that ranibizumab has the highest likelihood of being the most efficacious treatment. Trial registration numbers: BRAVO (NCT000); BRIGHTER (NCT00); COMRADE-B (NCT00); GENEVA (NCT0); RABAMES (NCT000); VIBRANT (NCT0). - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

4 Page of Strengths and limitations of this study RCTs evaluating the efficacy and safety of approved treatments for macular oedema secondary to BRVO were identified from a published systematic review and database searches. Inclusion/exclusion criteria for the key variables baseline BCVA and duration of disease were matched in the RCTs evaluating anti-vegf therapy. Despite matching for baseline BCVA and duration of disease, substantial heterogeneity existed between anti-vegf RCTs. Two clinical trials included in the meta-analysis were unpublished and therefore their quality could not be assessed. For one study, the standard error for letters gained was graphically estimated. For another study, the standard deviation was assumed. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

5 Page of INTRODUCTION Branch retinal vein occlusion (BRVO) and central retinal vein occlusion (CRVO) are important causes of vision impairment, and are thought to be the result of thrombotic events, external compression or vessel wall pathology. - Occlusion of the major veins of the retinal circulation leads to increased intra-luminal pressure, haemorrhage and oedema. Macular oedema secondary to RVO (BRVO or CRVO) is the second most common retinal vascular disease after diabetic retinopathy. The treatment of choice for patients with macular oedema associated with BRVO has long been considered to be grid laser photocoagulation. - However, the recent introduction of therapies targeted against vascular endothelial growth factor (VEGF), such as ranibizumab and aflibercept, has widened the range of therapeutic options available. The efficacy of ranibizumab, a monoclonal anti-vegf antibody fragment, in the treatment of BRVO was demonstrated in the Branch Retinal Vein Occlusion: Evaluation of Efficacy and Safety (BRAVO) phase trial. Following publication of these results, ranibizumab was approved in the USA 0 and EU for the treatment of macular oedema secondary to RVO. The efficacy of aflibercept, an anti-vegf binding protein, was demonstrated in the VIBRANT phase trial, and aflibercept has been submitted for approval in macular oedema secondary to BRVO in the EU. Intra-vitreal dexamethasone implant is approved for patients with macular oedema secondary to RVO. In addition, triamcinolone, a corticosteroid with a mechanism of action similar to dexamethasone, is used off-label in this treatment setting. - Bevacizumab, an anti-vegf agent, is not licensed in the treatment of VI due to BRVO and was not included in the analysis. Given the number of treatments that have been developed for RVO, there is a need for an evidence-based analysis of the comparative efficacy of the different treatments available. With this in mind, Glanville et al. recently published a systematic review of randomized controlled trials (RCTs) evaluating the efficacy and safety of widely used treatments for macular oedema secondary to RVO. The main findings were that both ranibizumab and dexamethasone implant produced significantly greater improvements in best corrected visual acuity (BCVA) at months in patients with RVO, when compared with individuals receiving sham intervention. The current study built upon the evaluation carried out by Glanville et al., conducting a Bayesian network meta-analysis of all RCTs in order to compare the efficacy and safety of all currently approved treatments or treatments submitted for approval for macular oedema - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

6 Page of secondary to RVO, including ranibizumab 0. mg, aflibercept mg, dexamethasone 0. mg implant and laser photocoagulation. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

7 Page of METHODS Glanville et al. study In brief, this was a systematic review of RCTs reporting the efficacy and safety of available treatments for macular oedema secondary to RVO. The literature search was performed on November 00 using the databases Medline (including Medline In-Process), EMBASE, the Cochrane Library, and Cumulative Index to Nursing and Allied Health Literature. Additional searches were performed in the ClinicalTrials.gov registry and the Association for Research in Vision and Ophthalmology (ARVO) database. The interventions included in the searches were ranibizumab, bevacizumab, dexamethasone and laser photocoagulation. Primary efficacy outcomes were mean change in BCVA from baseline to month (assessed in terms of Early Treatment Diabetic Retinopathy Study [ETDRS] letters) and the number of patients gaining at least 0 letters from baseline to month. To be included, studies had to be RCTs that reported at least one primary efficacy outcome. In addition, RCTs had to comprise at least two comparator arms, including one active comparator of interest. Only RCTs in English were considered. Updated search strategy An updated search of Medline (including Medline In-Process) and the Cochrane Library, using the search terms described in the study by Glanville et al. (see Supplementary materials), was performed on August 0 to identify relevant RCTs that had been published since November 00 (the date of the earlier study). Studies published in English, French and German were considered eligible for inclusion in the review. Inclusion/exclusion criteria for the network meta-analysis For inclusion in the current analysis, RCTs had to meet the following criteria: () report at least one efficacy outcome of interest (mean change in BCVA from baseline or percentage of patients gaining letters from baseline); () the outcome of interest had to be measured at or months from study baseline, with -month data used for the analysis when available; and () have at least two comparators of interest, including sham injections, ranibizumab 0. mg monthly or pro re nata (PRN; as needed), aflibercept mg monthly (q), dexamethasone 0. mg implant (retreatment interval months) and prompt laser photocoagulation therapy. RCTs in which efficacy data for BRVO and CRVO were not presented separately were excluded. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

8 Page of Data sources and extraction RCTs evaluating the efficacy and safety of treatment for macular oedema secondary to RVO were identified from three main sources: () the study by Glanville et al.; () the updated search strategy described above; and () manual searches of the ClinicalTrials.gov registry, proprietary data on file at Novartis Pharma AG, and abstracts from ophthalmology congresses including ARVO, American Academy of Ophthalmology (AAO) and European Society of Retina Specialists (EURETINA) congresses. Study characteristics, including baseline characteristics, number of patients, country and key inclusion/exclusion criteria, and study quality, were captured in Microsoft Excel spreadsheets. Study quality assessment The quality of each RCT was assessed according to the methodology checklist detailed in Appendix C of the National Institute for Health and Care Excellence (NICE) Guidelines Manual 0. 0 In brief, these guidelines allow for assessment of the likelihood of bias in selection (systematic differences between comparator groups), attrition (systematic differences between comparator groups with respect to loss of participants), detection (systematic differences in how outcomes are ascertained, diagnosed or verified), and performance (systematic differences between comparator groups in the care provided, other than in the intervention under investigation). Network meta-analysis A Bayesian network meta-analysis with random treatment effects was used to compare mean change in BCVA from baseline to month, odds ratio (OR) for gaining at least letters and OR for an increase in intra-ocular pressure (IOP)/ocular hypertension (OH) across the RCTs. The rates of increased IOP/OH were small and comparable for ranibizumab (0 %) and aflibercept (%). Therefore, rates of increased IOP/OH for ranibizumab and aflibercept were pooled to give an anti-vegf rate of increased IOP/OH. This assumption is discussed in the Results section. To estimate the posterior distribution for each model, two Markov chain Monte Carlo simulations were initialized using 000 iterations for each simulation. However, results are reported after excluding the first 000 iterations (i.e. results are reported for 000 iterations per chain). A % credible interval (CrI) was created using the. and. percentiles of the posterior distribution. The overall relative treatment effect was calculated using the median value from the posterior distribution. A relative treatment effect was interpreted as - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

9 Page of significant if the % CrI for the OR did not include 0 or if the % CrI for the BCVA gains did not include 0. Analyses were performed using WinBUGS version. (MRC Biostatistics Unit, Cambridge, UK). Sensitivity analyses Several sensitivity analyses were conducted to assess the robustness of the network metaanalysis. The results of a fixed effects and random effects model were compared. The impact of removing RCTs that did not capture outcomes at months was analysed, as was the impact of excluding RCTs with substantially lower baseline BCVAs than the other RCTs. In addition, we evaluated the impact of including the Ranibizumab for Branch Retinal Vein Occlusion Associated Macular Edema Study (RABAMES), in which the ranibizumab regimen was three doses at monthly intervals followed by a -month observation period. Finally, node-splitting analyses were used to compare the direct and indirect evidence of the efficacy of ranibizumab versus laser, ranibizumab versus dexamethasone implant and laser versus sham intervention. In the node-splitting analyses, the two-sided p value for the probability that the direct effect differed from the indirect effect was given by ( pr), if pr was 0% (where pr is the probability that the direct effect was greater that the indirect effect); if pr was < 0%, the p value for the probability that the direct effect differed from the indirect effect was pr. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

10 Page of RESULTS Studies Glanville et al. identified six potentially relevant studies of patients with BRVO. - A further study included both individuals with BRVO and those with CRVO. On full text evaluation, three studies met the inclusion criteria and were included in the analysis (BRAVO, GENEVA, and Battaglia Parodi et al. ). The Branch Vein Occlusion Study Group (BVOS) trial was excluded because the efficacy data were presented at months. The studies by Moradian et al. and Russo et al. were excluded because they did not include two treatments of interest. Finally, the Kuppermann et al. trial was excluded because it did not present separate efficacy outcome results for BRVO and CRVO. The updated systematic review identified an additional eight potentially relevant RCTs; however, full text screening found that only two of these met the inclusion criteria (Tan et al. and RABAMES ). Three of the trials were not relevant because they were extension studies (Brown et al., BRAVO, and Global Evaluation of Implantable Dexamethasone in Retinal Vein Occlusion with Macular Edema [GENEVA] ) and three trials evaluated only one treatment of interest (Moradian et al., Pichi et al. 0 and Azad et al. ). Manual searching identified three further trials (BRIGHTER, COMRADE-B and VIBRANT ). The quality of the published studies is reported in Table. In general, the studies were of good quality, with the exception of the study by Battaglia Parodi et al., which did not report most of the key quality assessment components. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

11 Page 0 of Table Quality appraisal of randomized controlled trials included in the network meta-analysis Was randomization Study carried out appropriately? Was concealment of treatment allocation adequate? Were the groups similar at the outset of the study in terms of prognostic factors? Were the care providers, participants and outcome assessors blind to treatment allocation? Were there any unexpected imbalances in drop-outs between groups? Did the analysis Is there any include an evidence to intention-to-treat suggest that analysis? If so, the authors was this measured appropriate and more were appropriate outcomes methods used to than they account for reported? missing data? Battaglia Parodi et al. Unclear Unclear Yes Unclear Unclear Yes No BRAVO Yes Unclear Yes Yes No No Yes GENEVA Yes Yes Yes Yes No No Yes RABAMES Yes Yes Yes No Yes No Yes Tan et al. Yes Yes No Yes Yes No Yes VIBRANT Yes Yes Yes Yes No No Yes BRIGHTER and COMRADE-B were not assessed for the risk of bias owing to limited study methodology being reported : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

12 Page of A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram showing the screening and selection process is presented in Figure. The network of studies, comprising a total of for the efficacy analysis, is presented in Figure. In total, eight RCTs were identified for the analysis: BRAVO, VIBRANT, Tan, GENEVA, BRIGHTER, COMRADE-B, RABAMES and Battaglia Parodi. Outcomes were reported at month in all studies except the study by Tan et al., for which outcomes were reported at month, and Battaglia Parodi et al., for which outcomes were reported at months and. Treatment regimens Three studies (Tan et al., BRIGHTER, COMRADE-B ) reported results with a ranibizumab PRN regimen (Table ). BRAVO was the only study in which results were reported after patients received monthly doses of ranibizumab. Hence, for the purpose of this analysis, the ranibizumab treatment regimen is referred to as PRN. The mean number of injections in the first months was. for aflibercept in VIBRANT; this compared with. in BRIGHTER,. in COMRADE-B and. in BRAVO for ranibizumab. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

13 Page of Table Baseline patient characteristics and efficacy outcomes Time Mean Study VIBRANT BRAVO Tan et al. GENEVA ** BRIGHTER Treatment (dose) Aflibercept q endpoint Drug number of Baseline Age n measured regimen injections/ BCVA (years) (months) implants x monthly doses Patients Disease BCVA Baseline gaining duration increase CRT (µm) letters (months) (SD) (%) (.)*** Laser (.)*** Ranibizumab 0. mg x monthly doses.... * 0. (.) Sham injection.0.. *. (.) x monthly Ranibizumab doses, then 0. mg PRN.... *. (.) Laser... *. (.) Dexamethasone 0. mg implant Sham procedure Ranibizumab 0. mg implant at month (.)** (.)** x monthly doses, then.... * 0. (0.) PRN Ranibizumab x monthly.... *.0 (.) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

14 Page of Ranibizumab COMRADE-B 0. mg 0. mg + doses, then laser PRN Laser... *. (.) Dexamethasone 0. mg implant x monthly doses, then PRN implant at month (.).... (.) RABAMES Laser (.) Ranibizumab 0. mg Laser + ranibizumab 0. mg Battaglia x monthly doses x monthly doses ~ (.) ~ (.) Parodi et al. Laser and. N/A N/A N/A NA. (.) Control and. N/A N/A N/A NA. (.) *Central foveal thickness. **In GENEVA, baseline characteristics were not split between patients with CRVO and those with BRVO. BCVA letters gained from the NICE assessment file entitled Evidence review: dexamethasone implants (Ozurdex) for macular oedema after retinal vein occlusion (00). Standard error of the mean was graphically estimated. ***SDs were not reported in the VIBRANT publication, but were provided by the authors of the study. BCVA, best-corrected visual acuity (assessed in terms of Early Treatment Diabetic Retinopathy Study letters); BRVO, branch retinal vein occlusion; CRT, central retinal thickness; CRVO, central retinal vein occlusion; N/A, not applicable; NICE, National Institute for Health and Care Excellence; PRN, pro re nata (as needed); RCT, randomized controlled trial; SD, standard deviation; q, mg monthly. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

15 Page of In RABAMES, patients received only three injections of ranibizumab 0. mg at monthly intervals, which were then followed by months of observation. The number of injections was substantially lower than in the other studies evaluating ranibizumab, so the RABAMES results were included only in one sensitivity analysis. In Battaglia Parodi et al., the mean BCVA gains from baseline at month and month were used in the analysis as a proxy for the efficacy at month. We could not use the standard deviations (SDs) from Battaglia Parodi et al. because they were reported on a non-linear transformation of the ETDRS scale. Instead, we assumed that the SD was the same as for BRAVO (.), which is in the mid-range of SD values in Table (minimum:.; maximum:.). Network meta-analysis Eligibility criteria differed among RCTs (see Supplementary Table ). Therefore, patient-level data for three of the ranibizumab trials (BRAVO, BRIGHTER, COMRADE-B ) were reanalysed to match the key eligibility criteria from the VIBRANT aflibercept study (i.e. criteria relating to BCVA and duration of disease at baseline). Specifically, patients were excluded from the analysis if they had a baseline BCVA of less than letters or a duration of disease of more than months. Using these criteria, only three patients were excluded from BRAVO and three patients were excluded from COMRADE-B, but of (0.%) patients were excluded from BRIGHTER. The exclusion of these patients did not, however, create a major imbalance between comparator arms (Supplementary Tables and ). On the one hand, the gender distribution became unbalanced following adjustment (p = 0.0) but, on the other hand, the imbalance in BCVA distribution became non-significant (p = 0.). The baseline characteristics and efficacy outcomes of interest after this step are shown in Table. Rates of increased IOP/OH are presented in Table. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

16 Page of Table Study The proportion of patients reporting IOP and OH adverse events from baseline to month by study and treatment group Ranibizumab 0. mg PRN n/n (%) Aflibercept q n/n (%) Laser n/n (%) Sham n/n (%) BRAVO /0 () / () VIBRANT / () 0/ (0) Tan et al. Not reported Not reported Dexamethasone 0. mg implant n/n (%) GENEVA / () / () COMRADE-B * / () / (0) Ranibizumab 0. mg PRN + laser photocoagulation n/n (%) BRIGHTER /0 () / (%) / () Battaglia Parodi Not reported Not reported et al. RABAMES 0/0 (0) 0/0 (0) 0/0 (0) Total / () / () /0 () /0 () 0/0 () / () Reported numbers are the sum of IOP and OH events. BRAVO adverse events from the clinical study report. BRIGHTER and COMRADE-B adverse events from data on file. In GENEVA, the adverse event split between BRVO and CRVO was done using the NICE assessment file entitled Evidence review: dexamethasone implants (Ozurdex) for macular oedema after retinal vein occlusion (00). *Treatment-emergent adverse events in the study eye. BRVO, branch retinal vein occlusion; CRVO, central retinal vein occlusion; IOP, intra-ocular pressure; n, number of patients in treatment arm with adverse events; N, total number of patients in treatment arm; NICE, National Institute for Health and Care Excellence; OH, ocular hypertension; PRN, pro re nata (as needed); q, mg monthly. on 0 October 0 by guest. Protected by copyright. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from

17 Page of Statistically significant mean (% CrI) changes in BCVA from baseline were found for ranibizumab monotherapy (+. [. to.]), ranibizumab plus laser combination therapy (+0. [. to.]) and aflibercept (+0. [. to.]) when compared with laser therapy alone (Table ), and for ranibizumab monotherapy versus dexamethasone implant (+.0 [.0 to.]). Pairwise ORs for gaining at least letters from baseline in the random treatment effects model are presented in Table. Only ranibizumab monotherapy was statistically superior to laser therapy alone. ORs (% CrI) compared with laser therapy were. (.0 to.) for ranibizumab monotherapy and.0 (0. to.) for aflibercept. Mean (% CrI) BCVA letter gains were numerically higher for ranibizumab monotherapy versus aflibercept (+. letters [. to.]) and the OR (% CrI) for gaining at least letters was.0 [0. to.]). Patients treated with dexamethasone implant had statistically significant higher rates of IOP/OH than those receiving the anti-vegf monotherapies (OR:. [. to.]). Table Incremental BCVA letters gained at months versus laser therapy alone (random treatment effects model) Incremental letters Treatment Ranibizumab 0. mg PRN Ranibizumab 0. mg PRN + laser photocoagulation Dexamethasone 0. mg implant Aflibercept q Sham *p < 0.0. gained vs laser photocoagulation alone. (..)* 0. (..)*. (..) 0. (..)*.0 (..) Studies included in the base case analysis are VIBRANT, BRAVO, BRIGHTER, COMRADE-B, Tan, Battaglia Parodi, GENEVA. BCVA, best-corrected visual acuity; PRN, pro re nata (as needed); q, mg monthly. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

18 Page of Table Pairwise odds ratios (% CrI) for gaining letters from baseline (random treatment effects model) Ranibizumab Comparator Ranibizumab 0. Laser Dexamethasone 0. mg PRN + Sham Aflibercept q mg PRN photocoagulation 0. mg implant laser photocoagulation Sham Ranibizumab 0. mg PRN. (.0.)*.0 (0..). (.0.)* Aflibercept q. (0..) 0. (0..).0 (0..) Laser photocoagulation. (0..) 0. (0.0 0.)* 0. (0.0.) Dexamethasone 0. mg implant. (0..) 0. (0.0.) 0. (0.0.). (0..) Ranibizumab 0. mg PRN + laser. (0. 0.) 0. (0..) 0. (0..). (0..). (0..) photocoagulation *p < 0.0. Pairwise odds ratios indicate the relative treatment effect for the treatments compared in the network meta-analysis. A statistically significant odds ratio greater than indicates that the treatment in the corresponding row is superior to the treatment in the corresponding column. Studies included in the base-case analysis are VIBRANT, BRAVO, BRIGHTER, COMRADE-B, Tan et al., Battaglia Parodi et al., GENEVA. CrI, credible interval; PRN, pro re nata (as needed); q, mg monthly. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

19 Page of The probabilities of being the most efficacious treatment in the study network are presented in Figure. The probability of being the most efficacious treatment based on BCVA letters gained was % for ranibizumab monotherapy, 0% for aflibercept and % for ranibizumab plus laser photocoagulation and 0% for dexamethasone 0. mg implant. The probability that ranibizumab monotherapy is a better treatment than aflibercept is %, which can be interpreted as follows: if 00 patients receive ranibizumab monotherapy in the right eye and aflibercept in the left eye, assuming the right and left eyes have the same baseline characteristics, on average patients would have a bigger gain observed in the right eye than in the left eye. The probability of being the most efficacious treatment based on the percentage of patients gaining at least letters was % for aflibercept, % for ranibizumab monotherapy, % for ranibizumab plus laser photocoagulation, % for dexamethasone, % for sham intervention and < % for laser monotherapy. Node-splitting analysis The node-splitting results are presented in Supplementary Table. The model showed an incremental gain (% CrI) of. letters (. to.) for ranibizumab monotherapy versus dexamethasone implant from the direct evidence (COMRADE-B ), whereas the indirect evidence (all trials except COMRADE-B ) provided an incremental gain of. letters (0. to.). This difference between direct and indirect evidence was non-significant (p = 0.). Similarly, there was no statistically significant difference between the direct (Battaglia Parodi et al. ) and indirect evidence of the relative efficacy of laser monotherapy versus sham intervention. The difference between direct and indirect evidence was higher when considering the incremental efficacy of ranibizumab monotherapy over laser monotherapy (.-letter difference between the indirect and the direct evidence), although this difference remained non-significant (p = 0.). The choice of a random effects model (rather than a fixed effects model) was justified because of the heterogeneity among trials (between-study SD:.). However, the fixed effects model provided similar results to the random effects model, with ranibizumab monotherapy having a % probability of being the best treatment (in terms of letters gained) compared with % for aflibercept, 0% for ranibizumab plus laser and 0% for dexamethasone implant, laser and sham. Sensitivity analyses Including RABAMES in the network had limited impact on the results. Mean (% CrI) BCVA letters gained from baseline for ranibizumab monotherapy over aflibercept remained numerically superior (+. [. to.]), and the probability of ranibizumab monotherapy - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

20 Page of being the best treatment based on letters gained increased to %, compared with % for aflibercept and % for ranibizumab plus laser. The inclusion of RABAMES slightly increased the ranibizumab numerical advantage in terms of percentage of patients gaining at least letters (OR vs aflibercept:. [0. to.]). The probability that ranibizumab monotherapy was the best treatment increased from % to 0% (vs 0% for aflibercept, % for ranibizumab plus laser and % for dexamethasone implant). When data from the study by Battaglia Parodi et al. were excluded from the network, patients receiving ranibizumab monotherapy showed the same non-significant mean gain (% CrI) of. letters (. to.) over aflibercept. After excluding data from Tan et al. from the network, patients receiving ranibizumab monotherapy showed a non-significant mean gain (% CrI) of. letters (. to.) compared with those receiving aflibercept. Removing the results of the study by Tan et al. also made the baseline BCVA more homogeneous across studies. When removing Tan et al. or Battaglia Parodi et al., ranibizumab monotherapy remained the agent with the highest probability of being the best treatment (%, both). We attempted to analyse the rates of increased IOP/OH when results for ranibizumab and aflibercept were not combined. However, it was not possible to compare the relative rates of events between the two anti-vegf therapies because the model did not converge. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

21 Page 0 of DISCUSSION Glanville et al. concluded that it was not possible to conduct a network meta-analysis in BRVO owing to inter-study differences in design and baseline characteristics. However, the BRVO analysis in their study included only three studies, BRAVO, Battaglia Parodi et al. and GENEVA. Since November 00, when Glanville et al. conducted the searches for their review, the preliminary results of new clinical trials have been made available, allowing the network meta-analysis described here to be conducted. Specifically, BRIGHTER enables comparison between laser therapy and ranibizumab monotherapy, and also between laser therapy and ranibizumab + laser combination therapy; COMRADE-B allows direct comparison between ranibizumab monotherapy and dexamethasone implant, and VIBRANT provides information on aflibercept versus laser therapy. In general, the results from this network meta-analysis confirm the results from head-to-head clinical trials. In particular, our analysis confirms that anti-vegf monotherapies are more efficacious than laser therapy, as shown in VIBRANT, BRIGHTER, RABAMES and by Tan et al.. Our results also confirm the superiority, in terms of letters gained in BCVA, of ranibizumab monotherapy over dexamethasone implant, as shown in COMRADE-B. A key finding was that the efficacy of laser was similar to sham intervention, suggesting that the role of laser in the treatment of BRVO should be reappraised. The efficacy of ranibizumab 0. mg PRN was numerically, but not statistically, superior to aflibercept q. The results were shown to be robust in a number of sensitivity analyses, including an analysis from which trials that did not report outcomes at months were excluded, and an analysis in which a trial with only three injections in the first months was included. In addition, there was no evidence of inconsistency between direct and indirect evidence. It was not possible to compare adverse event rates between ranibizumab monotherapy and aflibercept monotherapy because the model did not converge owing to the low incidence of increased IOP/OH. This meant that we had to merge ranibizumab and aflibercept treatments into an anti-vegf therapy group. Similarly, the incidence of systemic adverse events was not analysed. The lack of convergence does not mean that there are no differences in increased IOP/OH rates between treatments: it could mean that the included studies did not have sufficient power to detect any such differences. A preliminary analysis indicated that two samples of individuals would be required to reach a power of 0% to detect significant differences between treatments if the event rates were %. Hence, analyses of real-world data could shed some light on the relative safety and tolerability of anti-vegf and dexamethasone implant therapy and help to refine the full benefit risk ratio of approved treatments for macular oedema secondary to BRVO : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

22 Page of The retreatment period for dexamethasone implant in the clinical trials was months. Therefore, we used a -month perspective to evaluate the relative efficacy of dexamethasone implant. However, based on the GENEVA and COMRADE-B trials, the efficacy (and rate of increased IOP/OH) of dexamethasone implant peaks at month before decreasing to month. While we cannot formally determine how dexamethasone implant would have performed with a -month retreatment regimen, the BCVA was. letters higher at month than month in GENEVA and approximately. letters higher at month than month in COMRADE-B. Therefore, a bi-monthly retreatment regimen for dexamethasone implants may reduce the efficacy advantage of ranibizumab 0. mg PRN found in this network meta-analysis (+.0 [.0 to.] letters vs dexamethasone implant). Additional trials assessing the efficacy and safety of bi-monthly or tri-monthly dexamethasone would be useful, allowing us to investigate this issue. The results presented here indicate that the value of adjunctive laser photocoagulation therapy for macular oedema secondary to BRVO is uncertain. Our analysis was not able to demonstrate that the combination of laser and ranibizumab therapy provided higher efficacy gains than ranibizumab monotherapy. In addition, in BRIGHTER the mean (SD) number of ranibizumab injections was similar for the combination (. [.] injections) and ranibizumab monotherapy arms (. [.0] injections). Therefore, it is not clear whether adjunctive laser therapy would decrease the ranibizumab injection costs. The main challenge in any network meta-analysis is to compare studies that may have different populations (e.g. owing to different inclusion/exclusion criteria). The key strength of the present analysis is that inclusion/exclusion criteria for baseline BCVA and duration of disease were matched in three trials containing anti-vegf therapies. Those two variables have been shown to influence the outcome in RVO and need to be accounted for in order to reduce the risk of heterogeneity. Indeed, when the ranibizumab clinical trials,, were analysed without matching key inclusion/exclusion criteria to those of VIBRANT, there were no differences in letters gained between ranibizumab 0. mg PRN and aflibercept q (0.0 letters [. to.]). This demonstrates the importance of adjusting for baseline characteristics that influence clinical outcomes. In theory, a meta-regression with baseline BCVA and duration of disease as covariates could be run to account for heterogeneity. However, the coefficients for baseline BCVA or duration of disease did not converge (potentially because of a lack of degrees of freedom in the network). - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

23 Page of The main limitation of the study is that, at the time of completion of this manuscript, two clinical trials were not yet published in the peer-reviewed literature and therefore their quality could not be assessed. A second limitation was that the SEs of the BCVA gains in GENEVA were only graphically estimated because they were not reported in the study. However, this assumption has a limited impact. If we increase the SDs to. for the sham arm and to. for the dexamethasone implant arm (corresponding to a % confidence interval of ± letter for the estimated mean in GENEVA), the results remain similar: the ranibizumab 0. mg PRN advantage over dexamethasone implant remains at.0 (.0 to.) letters. Finally, the definition of increased IOP was not usually reported in the publications and it was not possible to assess whether there were substantial differences and, if so, whether the differences were relevant. Conclusions This Bayesian network meta-analysis confirmed the superiority of ranibizumab monotherapy over dexamethasone implant or laser for the treatment of macular oedema secondary to BRVO and showed that ranibizumab monotherapy has the highest likelihood of the agents investigated of being the most efficacious treatment for visual impairment due to BRVO based on absolute ETDRS letters gained. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

24 Page of Funding and competing interests We have read and understood BMJ policy on declaration of interests and declare the following interests: This study was funded by Novartis Pharma AG. The funder provided support in the form of salaries for authors SAR, FA and VB, but did not have any additional role in the study design, data collection and analysis, decision to publish or preparation of the manuscript. However, Novartis Pharma AG and Genentech, Inc. reviewed the manuscript before submission. Third-party medical writing assistance was funded by Novartis Pharma AG. ML has received financial support and or consulting fees from Novartis AG, Bayer, Allergan, Roche, GlaxoSmithKline, AstraZeneca and Novo Nordisk. However, ML did not receive any financial support for this manuscript. Acknowledgement Medical writing support was provided by Fernando Gibson, PhD, and Robert Gillies, PhD, of PharmaGenesis London. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

25 Page of REFERENCES. Rehak J, Rehak M. Branch retinal vein occlusion: pathogenesis, visual prognosis, and treatment modalities. Curr Eye Res 00;():.. Green WR, Chan CC, Hutchins GM, et al. Central retinal vein occlusion: a prospective histopathologic study of eyes in cases. Trans Am Ophthalmol Soc ;:.. Hayreh SS, Zimmerman MB, Podhajsky P. Incidence of various types of retinal vein occlusion and their recurrence and demographic characteristics. Am J Ophthalmol ;():.. McIntosh RL, Rogers SL, Lim L, et al. Natural history of central retinal vein occlusion: an evidence-based systematic review. Ophthalmology 00;():.. Rogers S, McIntosh RL, Cheung N, et al. The prevalence of retinal vein occlusion: pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology 00;():.. Finkelstein D. Argon laser photocoagulation for macular edema in branch vein occlusion. Ophthalmology ;():.. The Branch Vein Occlusion Study Group. Argon laser photocoagulation for macular edema in branch vein occlusion. Am J Ophthalmol ;():.. Battaglia Parodi M, Saviano S, Ravalico G. Grid laser treatment in macular branch retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol ;():0.. Brown DM, Campochiaro PA, Singh RP, et al. Ranibizumab for macular edema following central retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology 00;():. 0. Genentech Press Release 0. FDA Approves Lucentis (Ranibizumab Injection) for the Treatment of Macular Edema Following Retinal Vein Occlusion. Available at: Accessed November, 0.. Novartis Europharm Limited 0. Lucentis (ranibizumab) Summary of Product Characteristics. Available at: - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

26 Page of Regeneron. EYLEA (aflibercept) Injection Submitted for EU Marketing Authorization for the Treatment of Patients with Macular Edema Secondary to Branch Retinal Vein _Product_Information/human/000/WC000.pdf. Accessed November, 0.. Campochiaro PA, Clark WL, Boyer DS, et al. Intravitreal Aflibercept for Macular Edema Following Branch Retinal Vein Occlusion: The -Week Results of the VIBRANT Study. Ophthalmology 0. Occlusion (BRVO). Available at: abc-c-ba-abaabeb/regn_news_0 General_Releases.pdf. 0.. Haller JA, Bandello F, Belfort R, Jr., et al. Randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with macular edema due to retinal vein occlusion. Ophthalmology 00;():.. Haller JA, Bandello F, Belfort R, Jr., et al. Dexamethasone intravitreal implant in patients with macular edema related to branch or central retinal vein occlusion twelve-month study results. Ophthalmology 0;(): 0.. McAllister IL, Vijayasekaran S, Chen SD, et al. Effect of triamcinolone acetonide on vascular endothelial growth factor and occludin levels in branch retinal vein occlusion. Am J Ophthalmol 00;():.. Scott IU, Ip MS, VanVeldhuisen PC, et al. A randomized trial comparing the efficacy and safety of intravitreal triamcinolone with standard care to treat vision loss associated with macular Edema secondary to branch retinal vein occlusion: the Standard Care vs Corticosteroid for Retinal Vein Occlusion (SCORE) study report. Arch Ophthalmol 00;():.. Ip MS, Scott IU, VanVeldhuisen PC, et al. A randomized trial comparing the efficacy and safety of intravitreal triamcinolone with observation to treat vision loss associated with macular edema secondary to central retinal vein occlusion: the Standard Care vs Corticosteroid for Retinal Vein Occlusion (SCORE) study report. Arch Ophthalmol 00;(): : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

27 Page of Glanville J, Patterson J, McCool R, et al. Efficacy and safety of widely used treatments for macular oedema secondary to retinal vein occlusion: a systematic review. BMC Ophthalmol 0;:. 0. National Institute for Health and Care Excellence 0. Appendix C: Methodology checklist: randomised controlled trials. Available at: Accessed November, 0.. Ford JA, Elders A, Shyangdan D, et al. The relative clinical effectiveness of ranibizumab and bevacizumab in diabetic macular oedema: an indirect comparison in a systematic review. Bmj 0;:e.. Pielen A, Mirshahi A, Feltgen N, et al. Ranibizumab for Branch Retinal Vein Occlusion Associated Macular Edema Study (RABAMES): six-month results of a prospective randomized clinical trial. Acta Ophthalmol 0.. Campochiaro PA, Heier JS, Feiner L, et al. Ranibizumab for macular edema following branch retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology 00;():0.. Moradian S, Faghihi H, Sadeghi B, et al. Intravitreal bevacizumab vs. sham treatment in acute branch retinal vein occlusion with macular edema: results at months (Report ). Graefes Arch Clin Exp Ophthalmol 0;(): 00.. Russo V, Barone A, Conte E, et al. Bevacizumab compared with macular laser grid photocoagulation for cystoid macular edema in branch retinal vein occlusion. Retina 00;():.. Kuppermann BD, Blumenkranz MS, Haller JA, et al. Randomized controlled study of an intravitreous dexamethasone drug delivery system in patients with persistent macular edema. Arch Ophthalmol 00;():0.. Tan MH, McAllister IL, Gillies ME, et al. Randomized controlled trial of intravitreal ranibizumab versus standard grid laser for macular edema following branch retinal vein occlusion. Am J Ophthalmol 0;():.. Brown DM, Campochiaro PA, Bhisitkul RB, et al. Sustained benefits from ranibizumab for macular edema following branch retinal vein occlusion: -month outcomes of a phase III study. Ophthalmology 0;(): : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

28 Page of Campochiaro PA, Hafiz G, Channa R, et al. Antagonism of vascular endothelial growth factor for macular edema caused by retinal vein occlusions: two-year outcomes. Ophthalmology 00;():. 0. Pichi F, Specchia C, Vitale L, et al. Combination therapy with dexamethasone intravitreal implant and macular grid laser in patients with branch retinal vein occlusion. Am J Ophthalmol 0;():0.. Azad R, Vivek K, Sharma Y, et al. Ranibizumab as an adjunct to laser for macular edema secondary to branch retinal vein occlusion. Indian J Ophthalmol 0;0():.. Novartis Pharmaceuticals 0. Efficacy and Safety of Ranibizumab With or Without Laser in Comparison to Laser in Branch Retinal Vein Occlusion (BRIGHTER). Available at: Accessed November, 0.. Hattenbach L-O. Efficacy and Safety of 0. mg Ranibizumab compared with 0. mg dexamethasone intravitreal implant in patients with branch retinal vein occlusion over months: The COMRADE-B study. Invest Ophthalmol Vis Sci 0;:E Abstract 0.. Allergan 00. A Study of the Safety and Efficacy of a New Treatment for Macular Edema Resulting From Retinal Vein Occlusion. Available from: Accessed November, 0.. Ford JA, Lois N, Royle P, et al. Current treatments in diabetic macular oedema: systematic review and meta-analysis. 0;(). - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

29 Page of Figure legends Figure Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram BRVO, branch retinal vein occlusion; RCT, randomized controlled trial; SR, systematic review. Figure Meta-analysis study network Figure Probability that each treatment is the most efficacious in the study network BCVA assessed in terms of ETDRS letters. BCVA, best corrected visual acuity; ETDRS, Early Treatment Diabetic Retinopathy Study; PRN, pro re nata (as needed). - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

30 Page of xmm (00 x 00 DPI) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

31 Page 0 of x0mm ( x DPI) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

32 Page of xmm (00 x 00 DPI) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

33 Page of Supplementary Table Inclusion criteria Study Baseline BCVA Disease duration Battaglia Parodi et al. letters* Not available BRAVO letters (0/00 0/0) weeks BRIGHTER letters (0/00 0/0) months COMRADE-B letters (0/00 0/0) months GENEVA letters (0/00 0/0) Duration of macular oedema weeks RABAMES letters (0/0 0/0) > to < months Tan et al. letters (0/00 0/0) weeks to months VIBRANT letters (0/0 0/0) weeks *Decimal score converted into ETDRS letters. BCVA, best-corrected visual acuity; ETDRS, Early Treatment Diabetic Retinopathy Study. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

34 Page of Supplementary Table Impact of adjusting inclusion criteria for BRIGHTER on baseline characteristics. Baseline characteristic Before adjustment After adjustment Ranibizumab 0. mg PRN Ranibizumab 0. mg PRN + laser Laser p value Ranibizumab 0. mg PRN Ranibizumab 0. mg PRN Laser p value + laser Patients (n) 0 0 Female (%) Duration of disease (months) Age (years) BCVA (letters) p values calculated with an analysis of variance test for continuous variables (duration of disease, age, BCVA) and with a Χ-squared test for the categorical variable (gender). BCVA, best-corrected visual acuity; PRN, pro re nata (as needed). Before adjustment, the number of patients is based on the full analysis set and the other demographics are based on the randomized set. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

35 Page of Supplementary Table Outcomes before adjustment of baseline characteristics for BRAVO, BRIGHTER and COMRADE-B Study Outcome Treatment group BRAVO n/n (%) n/n (%) COMRADE-B LG (SD) n/n (%) BRIGHTER LG (SD) Ranibizumab 0. mg PRN 0/ () Laser Sham / () LG (SD). (.). (.0) / (). (.) /0 (). (0.) /0 ().0 (.) Dexamethasone 0. mg implant / (). (.) Ranibizumab 0. mg PRN + laser / (). (.0) BCVA, best-corrected visual acuity; LG, BCVA letters gained from baseline to month or month ; n, number of patients in treatment arm with adverse events; N, total number of patients in treatment arm; PRN, pro re nata (as needed); SD, standard deviation. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

36 Page of Supplementary Table Node-splitting (consistency between direct and indirect evidence): BCVA letters gained Comparison Difference between Direct evidence Indirect evidence direct and indirect estimates (% CrI) estimates (% CrI) estimates (% CrI) Ranibizumab 0. mg PRN vs dexamethasone. letters (..). letters (0..) 0. letters (..) 0. mg implant Laser vs sham. letters (.0.0) 0. letters (..) 0. letters (.0.) Ranibizumab 0. mg PRN vs laser. letters (..). letters (..). letters (..) BCVA, best-corrected visual acuity; CrI, credible interval; PRN, pro re nata (as needed); - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

37 Page of Supplementary materials. Medline search strategy for RCTs published since 00 Database: Ovid MEDLINE(R) < to July Week 0>, Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations <August 0, 0> exp Randomized Controlled Trials as Topic/ () exp Clinical Trials as Topic/ () randomised controlled trial.mp. [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] () exp Random Allocation/ () exp Double-Blind Method/ (0) exp Single-Blind Method/ () exp Clinical Trial/ () or or or or or or (0) ((singl$ or doubl$ or treb$ or tripl$) adj (blind$ or mask$)).tw. () 0 (clinic$ adj trial$).tw. (0) exp Placebos/ (0) placebo$.tw. (0) randomly allocated.tw. () (allocated adj random).tw. () or 0 or or or or () or () case report.tw. (0) exp Letter/ (0) exp Historical Article/ () 0 review of reported cases.pt. (0) or or or 0 (0) not () exp Macular Edema/ (0) exp Edema/ () (macular adj (edema or oedema or odema)).ti,ab. () exp Retinal Vein/ () exp Retinal Vein Occlusion/ () ((vein or veins or veinous) adj (occlusion$ or occluded or obstruction$ or obstructed or closed or closure$ or stricture$ or stenosis or stenosed or block or blocks or blockage$ or blocking or embolism$ or emboli) adj retina$).ti,ab. () - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

38 Page of (crvo or cvo or rvo or brvo or bvo or crvome).ti,ab. () 0 (branch vein adj occlu$).ti,ab. () (central vein adj occlu$).ti,ab. () or or or or or or or 0 or () exp Antibodies, Monoclonal/ () antibodies/tu () exp Vascular Endothelial Growth Factors/ () (vascular endothelial growth or vegf$ or antivef$).ti,ab,rn. () (ranibizumab or lucentis or rhufab v or --).ti,ab,rn. (0) exp Angiogenesis Inhibitors/ (0) triamcinolone acetonide.mp. or exp Triamcinolone Acetonide/ () 0 ivta.ti,ab,rn. () exp Dexamethasone/ () ozurdex.ti,ab,rn. (0) exp Light Coagulation/ (0) (photocoagulation or laser coagulation).ti,ab. () aflibercept.mp. or Vascular Endothelial Growth Factor A/ () Eylea.mp. () or/- () Animals/ () (editorial or letter or news).pt. (00) 0 and () 0 not ( or ) (0) limit to english language () limit to ed= () and () - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

39 Comparative efficacy and safety of approved treatments for macular oedema secondary to branch retinal vein occlusion: a network meta-analysis Journal: Manuscript ID: bmjopen-0-00.r Article Type: Research Date Submitted by the Author: -Mar-0 Complete List of Authors: Regnier, Stephane; Novartis Pharma AG, Health Economics and Outcomes Research Larsen, Michael; University of Copenhagen, Department of Ophthalmology, Glostrup Hospital Bezlyak, Vladimir; Novartis Pharma AG, Allen, Felicity; Novartis Pharmaceuticals UK Ltd, <b>primary Subject Heading</b>: Ophthalmology Secondary Subject Heading: Ophthalmology Keywords: Medical retina < OPHTHALMOLOGY, Health economics < HEALTH SERVICES ADMINISTRATION & MANAGEMENT, OPHTHALMOLOGY Note: The following files were submitted by the author for peer review, but cannot be converted to PDF. You must view these files (e.g. movies) online. WinBUGS file for reviewers - Dec 0.odc : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

40 Page of Comparative efficacy and safety of approved treatments for macular oedema secondary to branch retinal vein occlusion: a network meta-analysis Stephane A Regnier, Michael Larsen, Vladimir Bezlyak, Felicity Allen Author affiliations Novartis Pharma AG, Basel, Switzerland Department of Ophthalmology, Glostrup Hospital, University of Copenhagen, Glostrup, Denmark Novartis Pharmaceuticals UK Ltd, Frimley Business Park, Frimley, Surrey, UK Corresponding author: Stephane Régnier Novartis Campus WSJ CH-0 Basel, Switzerland Tel: + (0) 0 stephane.regnier@novartis.com Suggested keywords/phrases Ranibizumab Macular oedema Branch retinal vein occlusion Anti-vascular endothelial growth factor Bayesian network meta-analysis : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

41 Page of ABSTRACT Objective: To compare the efficacy and safety of approved treatments for macular oedema secondary to branch retinal vein occlusion (BRVO). Design: Randomized controlled trials (RCTs) evaluating the efficacy and safety of approved treatments for macular oedema secondary to BRVO were identified from an updated systematic review. Setting: A Bayesian network meta-analysis of RCTs of treatments for macular oedema secondary to BRVO. Interventions: Ranibizumab 0. mg pro re nata (as needed), aflibercept mg monthly (q), dexamethasone 0. mg implant, laser photocoagulation, ranibizumab + laser, or sham intervention. Outcome measures: Efficacy outcomes were mean change in best corrected visual acuity (BCVA; assessed as Early Treatment Diabetic Retinopathy Study letters) and the percentage of patients gaining letters. Safety outcome was the percentage of patients with increased intra-ocular pressure (IOP)/ocular hypertension (OH). Results: Eight RCTs were identified for inclusion with adult patients. The probability of being the most efficacious treatment at month or month based on letters gained was % for ranibizumab monotherapy, 0% for aflibercept, % for ranibizumab plus laser (adjunctive or prompt), and 0% for dexamethasone implant, laser or sham. The probability of being the most efficacious treatment for patients gaining letters was % for ranibizumab monotherapy, % for aflibercept, % for ranibizumab plus laser, % for dexamethasone implant, and less than % laser or sham. There was no statistical difference between ranibizumab monotherapy and aflibercept for letters gained (+. letters for ranibizumab vs. aflibercept with % credible interval (CrI) of. to +. letters) or the odds ratio [OR] for gaining letters:.0 [% CrI: 0. to.]). Dexamethasone implant was associated with significantly higher IOP/OH than anti-vascular endothelial growth factor agents (OR:. [% CrI:. to.]). Conclusions: There was no statistically significant difference between ranibizumab and aflibercept. : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

42 Page of Strengths and limitations of this study RCTs evaluating the efficacy and safety of approved treatments for macular oedema secondary to BRVO were identified from a published systematic review and database searches. Inclusion/exclusion criteria for the key variables baseline BCVA and duration of disease were matched in the RCTs evaluating anti-vegf therapy. Despite matching for baseline BCVA and duration of disease, substantial heterogeneity existed between anti-vegf RCTs. Two clinical trials included in the meta-analysis were unpublished and therefore their quality could not be assessed. For one study, the standard error for letters gained was graphically estimated. For another study, the standard deviation was assumed. : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

43 Page of INTRODUCTION Branch retinal vein occlusion (BRVO) and central retinal vein occlusion (CRVO) are important causes of visual impairment, and are thought to be the result of thrombotic events, external compression or vessel wall pathology. - Occlusion of the major veins of the retinal circulation leads to increased intra-luminal pressure, haemorrhage and oedema. Macular oedema secondary to RVO (BRVO or CRVO) is the second most common retinal vascular disease after diabetic retinopathy. The treatment of choice for patients with macular oedema associated with BRVO has long been considered to be grid laser photocoagulation. - However, the recent introduction of pharmacotherapies specifically targeting vascular endothelial growth factor (VEGF), such as ranibizumab and aflibercept, has widened the range of therapeutic options. The efficacy of ranibizumab, a monoclonal anti-vegf antibody fragment, in the treatment of BRVO was demonstrated in the Branch Retinal Vein Occlusion: Evaluation of Efficacy and Safety (BRAVO) phase trial. Following publication of these results, ranibizumab was approved in the USA 0 and EU for the treatment of macular oedema secondary to RVO. The efficacy of aflibercept, an anti-vegf binding protein, was demonstrated in the VIBRANT phase trial, and aflibercept has been submitted for approval in macular oedema secondary to BRVO in the EU. An intravitreal dexamethasone implant is approved for patients with macular oedema secondary to RVO. Other therapies include triamcinolone, a corticosteroid with a mechanism of action similar to dexamethasone, which is used off-label in this treatment setting; - and bevacizumab, an anti-vegf agent, which is not licensed for the treatment of visual impairment of any etiology. Thus, triamcinolone and bevacizumab were both excluded from the current analysis. Given the number of treatments that have been developed for RVO, there is a need for an evidence-based analysis of the comparative efficacy of the different treatments available. With this in mind, Glanville et al. recently published a systematic review of randomized controlled trials (RCTs) evaluating the efficacy and safety of widely used treatments for macular oedema secondary to RVO. The main findings were that both ranibizumab and dexamethasone implant produced significantly greater improvements in best corrected visual acuity (BCVA) at months in patients with RVO, when compared with individuals receiving sham intervention. The current analysis adds to the Glanville et al. study by conducting a Bayesian network meta-analysis of all relevant RCTs, with the aim of comparing the efficacy and safety of all currently approved treatments or treatments submitted for approval for macular oedema : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

44 Page of secondary to RVO, comprising ranibizumab 0. mg, aflibercept mg, dexamethasone 0. mg implant and laser photocoagulation. : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

45 Page of METHODS Glanville et al. study In brief, this was a systematic review of RCTs reporting the efficacy and safety of available treatments for macular oedema secondary to RVO. The literature search was performed on November 00 using the databases Medline (including Medline In-Process), EMBASE, the Cochrane Library, and Cumulative Index to Nursing and Allied Health Literature. Additional searches were performed in the ClinicalTrials.gov registry and the Association for Research in Vision and Ophthalmology (ARVO) database. The interventions included in the searches were ranibizumab, bevacizumab, dexamethasone and laser photocoagulation. Primary efficacy outcomes were mean change in BCVA from baseline to month (assessed in terms of Early Treatment Diabetic Retinopathy Study [ETDRS] letters) and the number of patients gaining at least 0 letters from baseline to month. To be included, studies had to be RCTs that reported at least one primary efficacy outcome. In addition, RCTs had to comprise at least two treatment arms, including one active comparator of interest. Only RCTs in English were considered. Updated search strategy An updated search of Medline (including Medline In-Process) and the Cochrane Library, using the search terms described in the study by Glanville et al. (see Supplementary materials), was performed on August 0 to identify relevant RCTs that had been published since November 00 (the date of the earlier study). Studies published in English, French and German were considered eligible for inclusion in the review. Inclusion/exclusion criteria for the network meta-analysis For inclusion in the current analysis, RCTs had to meet the following criteria: () report at least one efficacy outcome of interest (mean change in BCVA from baseline or percentage of patients gaining letters from baseline); () the outcome of interest had to be measured at or months from study baseline, with -month data used for the analysis when available; and () have at least two interventions of interest, including sham injections, ranibizumab 0. mg monthly or pro re nata (PRN; as needed), aflibercept mg monthly (q), dexamethasone 0. mg implant (retreatment interval months) and prompt laser photocoagulation therapy () patients receiving anti-vegf had a PRN or monthly regimen. RCTs in which efficacy data for BRVO and CRVO were not presented separately were excluded. : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

46 Page of Data sources and extraction RCTs evaluating the efficacy and safety of treatment for macular oedema secondary to RVO were identified from three main sources: () the study by Glanville et al.; () the updated search strategy described above; and () manual searches of the ClinicalTrials.gov registry, proprietary data on file at Novartis Pharma AG, and abstracts from ophthalmology congresses including ARVO, American Academy of Ophthalmology (AAO) and European Society of Retina Specialists (EURETINA) congresses. Study characteristics, including baseline characteristics, number of patients, country and key inclusion/exclusion criteria, and study quality, were captured in Microsoft Excel spreadsheets. Study quality assessment The quality of each RCT was assessed according to the methodology checklist detailed in Appendix C of the National Institute for Health and Care Excellence (NICE) Guidelines Manual 0. 0 In brief, these guidelines allow for assessment of the likelihood of bias in selection (systematic differences between comparator groups), attrition (systematic differences between comparator groups with respect to loss of participants), detection (systematic differences in how outcomes are ascertained, diagnosed or verified), and performance (systematic differences between comparator groups in the care provided, other than in the intervention under investigation). Network meta-analysis A Bayesian network meta-analysis with random treatment effects was used to compare mean change in BCVA from baseline to month, odds ratio (OR) for gaining at least letters and OR for an increase in intra-ocular pressure (IOP)/ocular hypertension (OH) across the RCTs. The rates of increased IOP/OH were small and comparable for ranibizumab (0 %) and aflibercept (%). Therefore, rates of increased IOP/OH for ranibizumab and aflibercept were pooled to give an anti-vegf rate of increased IOP/OH. This assumption is discussed in the Results section. To estimate the posterior distribution for each model, two Markov chain Monte Carlo simulations were initialized using 000 iterations for each simulation. However, results are reported after excluding the first 000 iterations (i.e. results are reported for 000 iterations per chain). A % credible interval (CrI) was created using the. and. percentiles of the posterior distribution. The overall relative treatment effect was calculated using the : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

47 Page of median value from the posterior distribution. A relative treatment effect was interpreted as significant if the % CrI for the OR did not include 0 or if the % CrI for the BCVA gains did not include 0. Analyses were performed using WinBUGS version. (MRC Biostatistics Unit, Cambridge, UK). Sensitivity analyses Several sensitivity analyses were conducted to assess the robustness of the network meta- analysis. The results of a fixed effects and random effects model were compared. The impact of removing RCTs that did not capture outcomes at months was analysed, as was the impact of excluding RCTs with substantially lower baseline BCVAs than the other RCTs. In addition, we evaluated the impact of including the Ranibizumab for Branch Retinal Vein Occlusion Associated Macular Edema Study (RABAMES), in which the ranibizumab regimen was three doses at monthly intervals followed by a -month observation period. Finally, node-splitting analyses were used to compare the direct and indirect evidence of the efficacy of ranibizumab versus laser, ranibizumab versus dexamethasone implant and laser versus sham intervention. In the node-splitting analyses, the two-sided p value for the probability that the direct effect differed from the indirect effect was given by ( pr), if pr was 0% (where pr is the probability that the direct effect was greater that the indirect effect); if pr was < 0%, the p value for the probability that the direct effect differed from the indirect effect was pr. : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

48 Page of RESULTS Studies Glanville et al. identified six potentially relevant studies of patients with BRVO: the Branch Vein Occlusion Study Group (BVOS) trial, Battaglia Parodi et al., GENEVA, BRAVO, Moradian et al., and Russo et al. An additional study (Kuppermann et al. ) included both individuals with BRVO and those with CRVO. On full text evaluation, three studies met the inclusion criteria and were included in the analysis (BRAVO, GENEVA, and Battaglia Parodi et al. ). BVOS was excluded because the efficacy data were presented at months. The studies by Moradian et al. and Russo et al. were also excluded because they did not include two treatments of interest. Finally, the Kuppermann et al. trial was excluded because it did not present separate efficacy outcome results for BRVO and CRVO. The updated systematic review identified an eight potentially relevant RCTs; however, full text screening found that only one of these, Tan et al., met the inclusion criteria. In RABAMES, patients received only three injections of ranibizumab 0. mg at monthly intervals, which were then followed by months of observation. The number of injections was substantially lower than in the other studies evaluating ranibizumab, so the RABAMES results were included only for sensitivity analysis. Three trials were excluded because they were extension studies (Brown et al., Campochiaro et al., and Haller et al. ) and three trials were excluded because they evaluated only one treatment of interest (Moradian et al., Pichi et al. 0 and Azad et al. ). Manual searching identified three further trials (BRIGHTER, COMRADE-B and VIBRANT ). The quality of the published studies is reported in Table. In general, the studies were of good quality, with the exception of the study by Battaglia Parodi et al., which did not report most of the key quality assessment components. : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

49 Page 0 of Table Quality appraisal of randomized controlled trials included in the network meta-analysis Did the analysis Study Battaglia Parodi et Was randomization carried out appropriately? Was concealment of treatment allocation adequate? Were the groups similar at the outset of the study in terms of prognostic factors? Were the care providers, participants and outcome assessors blind to treatment allocation? Were there any unexpected imbalances in drop-outs between groups? Is there any evidence to suggest that the authors measured more outcomes than they reported? include an intention-totreat analysis? If so, was this appropriate and were appropriate methods used to account for missing data? al. Unclear Unclear Yes Unclear Unclear Yes No BRAVO Yes Unclear Yes Yes No No Yes GENEVA Yes Yes Yes Yes No No Yes RABAMES Yes Yes Yes No Yes No Yes Tan et al. Yes Yes No Yes Yes No Yes VIBRANT Yes Yes Yes Yes No No Yes BRIGHTER and COMRADE-B were not assessed for the risk of bias owing to limited study methodology being reported : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

50 Page of A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram showing the screening and selection process is presented in Figure. The network of studies, comprising a total of adult patients for the efficacy analysis, is presented in Figure. In total, seven RCTs were identified for the analysis: BRAVO, VIBRANT, Tan, GENEVA, BRIGHTER, COMRADE-B, and Battaglia Parodi. RABAMES was used in sensitivity analysis. Outcomes were reported at month in all studies except the study by Tan et al., for which outcomes were reported at month, and Battaglia Parodi et al., for which outcomes were reported at months and. Treatment regimens Three studies (Tan et al., BRIGHTER, COMRADE-B ) reported results with a ranibizumab PRN regimen (Table ). BRAVO was the only ranibizumab study in which results were reported after patients received monthly doses of ranibizumab. Hence, for the purpose of this analysis, the ranibizumab treatment regimen is referred to as PRN. The mean number of injections in the first months was. for aflibercept in VIBRANT; this compared with. in BRIGHTER,. in COMRADE-B and. in BRAVO for ranibizumab. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

51 Page of Table Baseline patient characteristics and efficacy outcomes Time Mean Study Treatment (dose) endpoint measured (months) Drug number of n regimen injections/ implants Disease Baseline Age Baseline duration BCVA (years) CRT (µm) (months) Patients gaining letters (%) x VIBRANT Aflibercept q monthly...0. doses BRAVO**** Tan et al. GENEVA ** BRIGHTER**** BCVA increase (SD).0 (.)*** Laser (.)*** x Ranibizumab monthly 0. mg doses.... * 0. (.) Sham injection.0.. *. (.) x Ranibizumab 0. mg monthly doses, then PRN.... *. (.) Laser... *. (.) Dexamethasone 0. mg implant implant at month (.)** Sham procedure (.)** Ranibizumab x 0. mg monthly.... * 0. (0.) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

52 Page of Ranibizumab 0. mg + laser COMRADE- Ranibizumab B**** 0. mg Dexamethasone 0. mg implant doses, then PRN x monthly doses,.... *.0 (.) then PRN Laser... *. (.) x monthly doses, then PRN implant at month (.).... (.) RABAMES Laser (.) Ranibizumab 0. mg Laser + ranibizumab 0. mg Battaglia x monthly ~ (.) doses x monthly doses ~ (.) Parodi et al. Laser and. N/A N/A N/A NA. (.) Control and. N/A N/A N/A NA. (.) *Central foveal thickness. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

53 Page of **In GENEVA, baseline characteristics were not split between patients with CRVO and those with BRVO. BCVA letters gained from the NICE assessment file entitled Evidence review: dexamethasone implants (Ozurdex) for macular oedema after retinal vein occlusion (00). Standard error of the mean was graphically estimated. ***SDs were not reported in the VIBRANT publication, but were provided by the authors of the study. BCVA, best-corrected visual acuity (assessed in terms of Early Treatment Diabetic Retinopathy Study letters); BRVO, branch retinal vein occlusion; CRT, central retinal thickness; CRVO, central retinal vein occlusion; N/A, not applicable; NICE, National Institute for Health and Care Excellence; PRN, pro re nata (as needed); RCT, randomized controlled trial; SD, standard deviation; q, mg monthly. ****Data for BRAVO, BRIGHTER, COMRADE-B is reported after patient-level data analysis. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

54 Page of In Battaglia Parodi et al., the mean BCVA gains from baseline at month and month were used in the analysis as a proxy for the efficacy at month. We could not use the standard deviations (SDs) from Battaglia Parodi et al. because they were reported on a non-linear transformation of the ETDRS scale. Instead, we assumed that the SD was the same as for BRAVO (.), which is in the mid-range of SD values in Table (minimum:.; maximum:.). Network meta-analysis Eligibility criteria differed among RCTs (see Supplementary Table ). Therefore, patient-level data for three of the ranibizumab trials (BRAVO, BRIGHTER, COMRADE-B ) were reanalysed to match the key eligibility criteria from the VIBRANT aflibercept study (i.e. criteria relating to BCVA and duration of disease at baseline). Specifically, patients were excluded from the analysis if they had a baseline BCVA of less than letters or a duration of disease of more than months. Using these criteria, only three patients were excluded from BRAVO and three patients were excluded from COMRADE-B, but of (0.%) patients were excluded from BRIGHTER. The exclusion of these patients did not, however, create a major imbalance between treatment arms in BRIGHTER (Supplementary Tables and ). The baseline characteristics and efficacy outcomes of interest after this step are shown in Table. Rates of increased IOP/OH are presented in Table. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

55 Page of Table The proportion of patients reporting IOP and OH adverse events from baseline to month by study and treatment group Study Ranibizumab Aflibercept Laser Sham Dexamethasone Ranibizumab 0. mg PRN n/n (%) q n/n (%) n/n (%) 0. mg implant 0. mg PRN + n/n (%) n/n (%) laser photocoagulation n/n (%) BRAVO /0 () / () VIBRANT / () 0/ (0) Tan et al. Not reported Not reported GENEVA / () / () COMRADE-B * / () / (0) BRIGHTER /0 () / (%) / () Battaglia Parodi et al. Not reported Not reported RABAMES 0/0 (0) 0/0 (0) 0/0 (0) Total / () / () /0 () /0 () 0/0 () / () Reported numbers are the sum of IOP and OH events. BRAVO adverse events from the clinical study report. BRIGHTER and COMRADE-B adverse events from data on file. In GENEVA, the adverse event split between BRVO and CRVO was done using the NICE assessment file entitled Evidence review: dexamethasone implants (Ozurdex) for macular oedema after retinal vein occlusion (00). - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

56 Page of *Treatment-emergent adverse events in the study eye. BRVO, branch retinal vein occlusion; CRVO, central retinal vein occlusion; IOP, intra-ocular pressure; n, number of patients in treatment arm with adverse events; N, total number of patients in treatment arm; NICE, National Institute for Health and Care Excellence; OH, ocular hypertension; PRN, pro re nata (as needed); q, mg monthly. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

57 Page of Statistically significant mean (% CrI) changes in BCVA from baseline were found for ranibizumab monotherapy (+. [. to.]), ranibizumab plus laser combination therapy (+0. [. to.]) and aflibercept (+0. [. to.]) when compared with laser therapy alone (Table ), and for ranibizumab monotherapy versus dexamethasone implant (+.0 [.0 to.]). Pairwise ORs for gaining at least letters and absolute letters gains from baseline in the random treatment effects model are presented in Table. Only ranibizumab monotherapy was statistically superior to laser therapy alone. ORs (% CrI) compared with laser therapy were. (.0 to.) for ranibizumab monotherapy and.0 (0. to.) for aflibercept. Mean (% CrI) BCVA letter gains were not statistically significant for ranibizumab monotherapy versus aflibercept (+. letters [. to.]) and the OR (% CrI) for gaining at least letters was.0 [0. to.]). Patients treated with dexamethasone implant had statistically significant higher rates of IOP/OH than those receiving the anti-vegf monotherapies (OR:. [. to.]). Table Incremental BCVA letters gained at months versus laser therapy alone (random treatment effects model) Incremental letters Treatment Ranibizumab 0. mg PRN Ranibizumab 0. mg PRN + laser photocoagulation Dexamethasone 0. mg implant Aflibercept q Sham *p < 0.0. gained vs laser photocoagulation alone. (..)* 0. (..)*. (..) 0. (..)*.0 (..) Studies included in the base case analysis are VIBRANT, BRAVO, BRIGHTER, COMRADE-B, Tan, Battaglia Parodi, GENEVA. BCVA, best-corrected visual acuity; PRN, pro re nata (as needed); q, mg monthly. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

58 Page of Table a Pairwise odds ratios (% CrI) for gaining letters from baseline (random treatment effects model) Comparator Sham Ranibizumab Ranibizumab 0. Laser Dexamethasone 0. mg PRN + Aflibercept q mg PRN photocoagulation 0. mg implant laser photocoagulation Sham Ranibizumab 0. mg PRN. (.0.)*.0 (0..). (.0.)* Aflibercept q. (0..) 0. (0..).0 (0..) Laser photocoagulation. (0..) 0. (0.0 0.)* 0. (0.0.) Dexamethasone 0. mg implant. (0..) 0. (0.0.) 0. (0.0.). (0..) Ranibizumab 0. mg PRN + laser. (0. 0.) 0. (0..) 0. (0..). (0..). (0..) photocoagulation *p < 0.0. Pairwise odds ratios indicate the relative treatment effect for the treatments compared in the network meta-analysis. A statistically significant odds ratio greater than indicates that the treatment in the corresponding row is superior to the treatment in the corresponding column. Studies included in the base-case analysis are VIBRANT, BRAVO, BRIGHTER, COMRADE-B, Tan et al., Battaglia Parodi et al., GENEVA. CrI, credible interval; PRN, pro re nata (as needed); q, mg monthly. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

59 Page 0 of Table b Pairwise difference (% CrI) for letters gained from baseline (random treatment effects model) Comparator Sham Ranibizumab Ranibizumab 0. Laser Dexamethasone 0. mg PRN + Aflibercept q mg PRN photocoagulation 0. mg implant laser photocoagulation Sham - Ranibizumab 0. mg PRN 0. (.,.)* - Aflibercept q. (.,.)* -. (-.,.) - Laser -.0 (-.,.) -. (-.,-.)* photocoagulation -0. (-.,.)* Dexamethasone 0. mg implant. (-.,.) -.0 (-.,-.0)* -. (-.0,.). (-.,.) - Ranibizumab 0. mg PRN + laser. (.,.)* -. (-.,.) -0.0 (-.,.) 0. (.,.)*. (0.-.) - photocoagulation *p < : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -

60 Page of The probabilities of being the most efficacious treatment in the study network are presented in Figure. The probability of being the most efficacious treatment based on BCVA letters gained was % for ranibizumab monotherapy, 0% for aflibercept and % for ranibizumab plus laser photocoagulation and 0% for dexamethasone 0. mg implant. The probability that ranibizumab monotherapy is a better treatment than aflibercept is %, which can be interpreted as follows: if 00 patients receive ranibizumab monotherapy in the right eye and aflibercept in the left eye, assuming the right and left eyes have the same baseline characteristics, on average patients would have a bigger gain observed in the right eye than in the left eye. The probability of being the most efficacious treatment based on the percentage of patients gaining at least letters was % for aflibercept, % for ranibizumab monotherapy, % for ranibizumab plus laser photocoagulation, % for dexamethasone, % for sham intervention and < % for laser monotherapy. Node-splitting analysis The node-splitting results are presented in Supplementary Table. The model showed an incremental gain (% CrI) of. letters (. to.) for ranibizumab monotherapy versus dexamethasone implant from the direct evidence (COMRADE-B ), whereas the indirect evidence (all trials except COMRADE-B ) provided an incremental gain of. letters (0. to.). This difference between direct and indirect evidence was non-significant (p = 0.). Similarly, there was no statistically significant difference between the direct (Battaglia Parodi et al. ) and indirect evidence of the relative efficacy of laser monotherapy versus sham intervention. The difference between direct and indirect evidence was higher when considering the incremental efficacy of ranibizumab monotherapy over laser monotherapy (.-letter difference between the indirect and the direct evidence), although this difference remained non-significant (p = 0.). The choice of a random effects model (rather than a fixed effects model) was justified because of the heterogeneity among trials (between-study SD:.). However, the fixed effects model provided similar results to the random effects model, with ranibizumab monotherapy having a % probability of being the best treatment (in terms of letters gained) compared with % for aflibercept, 0% for ranibizumab plus laser and 0% for dexamethasone implant, laser and sham. Sensitivity analyses The sensitivity analyses results are summarized in Figures and (Supplementary materials). Including RABAMES in the network had limited impact on the results. Mean (% CrI) BCVA letters gained from baseline for ranibizumab monotherapy over aflibercept - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

61 Page of remained non-statistically significant (+. [. to.]), and the probability of ranibizumab monotherapy being the best treatment based on letters gained increased to %, compared with % for aflibercept and % for ranibizumab plus laser. The inclusion of RABAMES slightly increased the ranibizumab numerical advantage in terms of percentage of patients gaining at least letters (OR vs aflibercept:. [0. to.]). The probability that ranibizumab monotherapy was the best treatment increased from % to 0% (vs 0% for aflibercept, % for ranibizumab plus laser and % for dexamethasone implant). When data from the study by Battaglia Parodi et al. were excluded from the network, patients receiving ranibizumab monotherapy showed a non-significant mean (% CrI) gain of. letters (. to.) over aflibercept. After excluding data from Tan et al. from the network, patients receiving ranibizumab monotherapy showed a non-significant mean (% CrI) gain of. letters (. to.) compared with those receiving aflibercept. When removing Tan et al. or Battaglia Parodi et al., ranibizumab monotherapy remained the agent with the highest probability of being the best treatment (%, both). We attempted to analyse the rates of increased IOP/OH when results for ranibizumab and aflibercept were not combined. However, it was not possible to compare the relative rates of events between the two anti- VEGF therapies because the model did not converge. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

62 Page of DISCUSSION Glanville et al. concluded that it was not possible to conduct a network meta-analysis in BRVO owing to inter-study differences in design and baseline characteristics. However, the BRVO analysis in their study included only three studies, BRAVO, Battaglia Parodi et al. and GENEVA. Since November 00, when Glanville et al. conducted the searches for their review, the preliminary results of new clinical trials have been made available, allowing the network meta-analysis described here to be conducted. Specifically, BRIGHTER enables comparison between laser therapy and ranibizumab monotherapy, and also between laser therapy and ranibizumab + laser combination therapy; COMRADE-B allows direct comparison between ranibizumab monotherapy and dexamethasone implant, and VIBRANT provides information on aflibercept versus laser therapy. In general, the results from this network meta-analysis confirm the results from head-to-head clinical trials. In particular, our analysis confirms that anti-vegf monotherapies are more efficacious than laser therapy, as shown in VIBRANT, BRIGHTER, RABAMES and by Tan et al.. Our results also confirm the superiority, in terms of letters gained in BCVA, of ranibizumab monotherapy over dexamethasone implant, as shown in COMRADE-B. A key finding was that the efficacy of laser was similar to sham intervention, suggesting that the role of laser in the treatment of BRVO should be reappraised. The credible intervals were broad and there was no statistically significant difference between ranibizumab 0. mg PRN and aflibercept q in either letters gained or proportion of patients gaining more than letters. The results were shown to be robust in a number of sensitivity analyses, including an analysis from which trials that did not report outcomes at months were excluded, and an analysis in which a trial with only three injections in the first months was included. In addition, there was no evidence of inconsistency between direct and indirect evidence. It was not possible to compare adverse event rates between ranibizumab monotherapy and aflibercept monotherapy because the model did not converge owing to the low incidence of increased IOP/OH. This meant that we had to merge ranibizumab and aflibercept treatments into an anti-vegf therapy group. Similarly, the incidence of systemic adverse events was not analysed. The lack of convergence does not mean that there are no differences in increased IOP/OH rates between treatments: it could mean that the included studies did not have sufficient power to detect any such differences. A preliminary analysis indicated that two samples of individuals would be required to reach a power of 0% to detect significant differences between treatments if the event rates were %. Hence, analyses of real-world data could shed some light on the relative safety and tolerability of anti-vegf and - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

63 Page of dexamethasone implant therapy and help to refine the full benefit risk ratio of approved treatments for macular oedema secondary to BRVO. The retreatment period for dexamethasone implant in the clinical trials was months. Therefore, we used a -month perspective to evaluate the relative efficacy of dexamethasone implant. However, based on the GENEVA and COMRADE-B trials, the efficacy (and rate of increased IOP/OH) of dexamethasone implant peaks at month before decreasing to month. While we cannot formally determine how dexamethasone implant would have performed with a -month retreatment regimen, the BCVA was. letters higher at month than month in GENEVA and approximately. letters higher at month than month in COMRADE-B. Therefore, a bi-monthly retreatment regimen for dexamethasone implants may reduce the efficacy advantage of ranibizumab 0. mg PRN found in this network meta-analysis (+.0 [.0 to.] letters vs dexamethasone implant). Additional trials assessing the efficacy and safety of bi-monthly or tri-monthly dexamethasone would be useful, allowing us to investigate this issue. The results presented here indicate that the value of adjunctive laser photocoagulation therapy for macular oedema secondary to BRVO is uncertain. Our analysis was not able to demonstrate that the combination of laser and ranibizumab therapy provided higher efficacy gains than ranibizumab monotherapy. In addition, in BRIGHTER the mean (SD) number of ranibizumab injections was similar for the combination (. [.] injections) and ranibizumab monotherapy arms (. [.0] injections). Therefore, it is not clear whether adjunctive laser therapy would decrease the ranibizumab injection costs. The main challenge in any network meta-analysis is to compare studies that may have different populations (e.g. owing to different inclusion/exclusion criteria). The key strength of the present analysis is that inclusion/exclusion criteria for baseline BCVA and duration of disease were matched in three trials containing anti-vegf therapies. Those two variables have been shown to influence the outcome in RVO and need to be accounted for in order to reduce the risk of heterogeneity. Indeed, when the ranibizumab clinical trials,, were analysed without matching key inclusion/exclusion criteria to those of VIBRANT, there were no differences in letters gained between ranibizumab 0. mg PRN and aflibercept q (0.0 letters [. to.]). This demonstrates the importance of adjusting for baseline characteristics that influence clinical outcomes. In theory, a meta-regression with baseline BCVA and duration of disease as covariates could be run to account for heterogeneity. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

64 Page of However, the coefficients for baseline BCVA or duration of disease did not converge (potentially because of a lack of degrees of freedom in the network). The main limitation of the study is that, at the time of completion of this manuscript, two clinical trials were not yet published in the peer-reviewed literature and therefore their quality could not be assessed. A second limitation was that the SEs of the BCVA gains in GENEVA were only graphically estimated because they were not reported in the study. However, this assumption had a limited impact on the results. If we increase the SDs to. for the sham arm and to. for the dexamethasone implant arm (corresponding to a % confidence interval of ± letter for the estimated mean in GENEVA), the results remain similar: the ranibizumab 0. mg PRN advantage over dexamethasone implant remains at.0 (.0 to.) letters. A further limitation was that the definition of increased IOP was not usually reported in the publications and it was not possible to assess whether there were substantial differences and, if so, whether the differences were relevant. Finally, data from only a limited number of trials were included in this analysis and future analyses will be strengthened once additional clinical trial data becomes available. Conclusions This Bayesian network meta-analysis confirmed the superiority of ranibizumab monotherapy over dexamethasone implant or laser for the treatment of macular oedema secondary to BRVO and showed that there was no statistical difference between ranibizumab monotherapy and aflibercept. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

65 Page of Funding and competing interests We have read and understood BMJ policy on declaration of interests and declare the following interests: This study was funded by Novartis Pharma AG. The funder provided support in the form of salaries for authors SAR, FA and VB, but did not have any additional role in the study design, data collection and analysis, decision to publish or preparation of the manuscript. However, Novartis Pharma AG and Genentech, Inc. reviewed the manuscript before submission. Third-party medical writing assistance was funded by Novartis Pharma AG. ML has received financial support and or consulting fees from Novartis AG, Bayer, Allergan, Roche, GlaxoSmithKline, AstraZeneca and Novo Nordisk. However, ML did not receive any financial support for this manuscript. Novartis has exclusive rights to market ranibizumab outside the USA and Genentech has exclusive rights to market ranibizumab in the USA. Acknowledgement Medical writing support was provided by Fernando Gibson, PhD, and Robert Gillies, PhD, of PharmaGenesis London. Competing Interest: We have read and understood BMJ policy on declaration of interests and declare the following interests: This study was funded by Novartis Pharma AG. The funder provided support in the form of salaries for authors SAR, FA and VB, but did not have any additional role in the study design, data collection and analysis, decision to publish or preparation of the manuscript. However, Novartis Pharma AG and Genentech, Inc. reviewed the manuscript before submission. Third-party medical writing assistance was funded by Novartis Pharma AG. ML has received financial support and or consulting fees from Novartis AG, Bayer, Allergan, Roche, GlaxoSmithKline, AstraZeneca and Novo Nordisk. However, ML did not receive any financial support for this manuscript. Novartis has exclusive rights to market ranibizumab outside the USA and Genentech has exclusive rights to market ranibizumab in the USA. Contributorship Statement: SR designed the study, collected the data, run the network meta-analysis, wrote and approved the manuscript ML: identified the relevant endpoints, discussed the validity of the trials, revised and approved the - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

66 Page of manuscript VB re-analyzed BRAVO and BRIGHTER data to match VIBRANT inclusion/exclusion criteria and approved the manuscript FA conducted the systematic review, evaluated the quality of the studies, participated in the elaboration of the manuscript and approved the manuscript. Data Sharing Statement: Two studies (BRIGHTER and COMRADE-B) used in the network meta-analysis are not yet publsihed. In addition, for those two studies (plus BRAVO), the authors reanalyzed clinical trials results to match VIBRANT inclusion criteria. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

67 Page of REFERENCES. Rehak J, Rehak M. Branch retinal vein occlusion: pathogenesis, visual prognosis, and treatment modalities. Curr Eye Res 00;():.. Green WR, Chan CC, Hutchins GM, et al. Central retinal vein occlusion: a prospective histopathologic study of eyes in cases. Trans Am Ophthalmol Soc ;:.. Hayreh SS, Zimmerman MB, Podhajsky P. Incidence of various types of retinal vein occlusion and their recurrence and demographic characteristics. Am J Ophthalmol ;():.. McIntosh RL, Rogers SL, Lim L, et al. Natural history of central retinal vein occlusion: an evidence-based systematic review. Ophthalmology 00;():.. Rogers S, McIntosh RL, Cheung N, et al. The prevalence of retinal vein occlusion: pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology 00;():.. Finkelstein D. Argon laser photocoagulation for macular edema in branch vein occlusion. Ophthalmology ;():.. The Branch Vein Occlusion Study Group. Argon laser photocoagulation for macular edema in branch vein occlusion. Am J Ophthalmol ;():.. Battaglia Parodi M, Saviano S, Ravalico G. Grid laser treatment in macular branch retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol ;():0.. Brown DM, Campochiaro PA, Singh RP, et al. Ranibizumab for macular edema following central retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology 00;():. 0. Genentech Press Release 0. FDA Approves Lucentis (Ranibizumab Injection) for the Treatment of Macular Edema Following Retinal Vein Occlusion. Available at: Accessed November, 0.. Novartis Europharm Limited 0. Lucentis (ranibizumab) Summary of Product Characteristics. Available at: _Product_Information/human/000/WC000.pdf. Accessed November, 0.. Campochiaro PA, Clark WL, Boyer DS, et al. Intravitreal Aflibercept for Macular Edema Following Branch Retinal Vein Occlusion: The -Week Results of the VIBRANT Study. Ophthalmology : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

68 Page of dexamethasone intravitreal implant in patients with macular edema due to retinal vein occlusion. Ophthalmology 00;():.. Haller JA, Bandello F, Belfort R, Jr., et al. Dexamethasone intravitreal implant in patients with macular edema related to branch or central retinal vein occlusion twelve-month study results. Ophthalmology 0;(): 0.. McAllister IL, Vijayasekaran S, Chen SD, et al. Effect of triamcinolone acetonide on vascular endothelial growth factor and occludin levels in branch retinal vein occlusion. Am J Ophthalmol 00;():.. Scott IU, Ip MS, VanVeldhuisen PC, et al. A randomized trial comparing the efficacy and safety of intravitreal triamcinolone with standard care to treat vision loss associated with macular Edema secondary to branch retinal vein occlusion: the Standard Care vs Corticosteroid for Retinal Vein Occlusion (SCORE) study report. Arch Ophthalmol 00;():.. Ip MS, Scott IU, VanVeldhuisen PC, et al. A randomized trial comparing the efficacy and safety of intravitreal triamcinolone with observation to treat vision loss associated with macular edema secondary to central retinal vein occlusion: the Standard Care vs Corticosteroid for Retinal Vein Occlusion (SCORE) study report. Arch Ophthalmol 00;():0.. Glanville J, Patterson J, McCool R, et al. Efficacy and safety of widely used treatments for macular oedema secondary to retinal vein occlusion: a systematic review. BMC Ophthalmol 0;:. 0. National Institute for Health and Care Excellence 0. Appendix C: Methodology checklist: randomised controlled trials. Available at:. Regeneron. EYLEA (aflibercept) Injection Submitted for EU Marketing Authorization for the Treatment of Patients with Macular Edema Secondary to Branch Retinal Vein Occlusion (BRVO). Available at: abc-c-ba-abaabeb/regn_news_0 General_Releases.pdf. 0.. Haller JA, Bandello F, Belfort R, Jr., et al. Randomized, sham-controlled trial of Accessed November, 0.. Ford JA, Elders A, Shyangdan D, et al. The relative clinical effectiveness of ranibizumab and bevacizumab in diabetic macular oedema: an indirect comparison in a systematic review. Bmj 0;:e. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

69 Page 0 of Pielen A, Mirshahi A, Feltgen N, et al. Ranibizumab for Branch Retinal Vein Occlusion Associated Macular Edema Study (RABAMES): six-month results of a prospective randomized clinical trial. Acta Ophthalmol 0.. Campochiaro PA, Heier JS, Feiner L, et al. Ranibizumab for macular edema following branch retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology 00;():0.. Moradian S, Faghihi H, Sadeghi B, et al. Intravitreal bevacizumab vs. sham treatment in acute branch retinal vein occlusion with macular edema: results at months (Report ). Graefes Arch Clin Exp Ophthalmol 0;(): 00.. Russo V, Barone A, Conte E, et al. Bevacizumab compared with macular laser grid photocoagulation for cystoid macular edema in branch retinal vein occlusion. Retina 00;():.. Kuppermann BD, Blumenkranz MS, Haller JA, et al. Randomized controlled study of an intravitreous dexamethasone drug delivery system in patients with persistent macular edema. Arch Ophthalmol 00;():0.. Tan MH, McAllister IL, Gillies ME, et al. Randomized controlled trial of intravitreal ranibizumab versus standard grid laser for macular edema following branch retinal vein occlusion. Am J Ophthalmol 0;():.. Brown DM, Campochiaro PA, Bhisitkul RB, et al. Sustained benefits from ranibizumab for macular edema following branch retinal vein occlusion: -month outcomes of a phase III study. Ophthalmology 0;(): 0.. Campochiaro PA, Hafiz G, Channa R, et al. Antagonism of vascular endothelial growth factor for macular edema caused by retinal vein occlusions: two-year outcomes. Ophthalmology 00;():. 0. Pichi F, Specchia C, Vitale L, et al. Combination therapy with dexamethasone intravitreal implant and macular grid laser in patients with branch retinal vein occlusion. Am J Ophthalmol 0;():0.. Azad R, Vivek K, Sharma Y, et al. Ranibizumab as an adjunct to laser for macular edema secondary to branch retinal vein occlusion. Indian J Ophthalmol 0;0():.. Novartis Pharmaceuticals 0. Efficacy and Safety of Ranibizumab With or Without Laser in Comparison to Laser in Branch Retinal Vein Occlusion (BRIGHTER). Available at: Accessed November, 0.. Hattenbach L-O. Efficacy and Safety of 0. mg Ranibizumab compared with 0. mg dexamethasone intravitreal implant in patients with branch retinal vein occlusion over : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

70 Page of months: The COMRADE-B study. Invest Ophthalmol Vis Sci 0;:E Abstract 0.. Allergan 00. A Study of the Safety and Efficacy of a New Treatment for Macular Edema Resulting From Retinal Vein Occlusion. Available from: Accessed November, 0.. Ford JA, Lois N, Royle P, et al. Current treatments in diabetic macular oedema: systematic review and meta-analysis. 0;(). - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

71 Page of Figure legends Figure Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram. BRVO, branch retinal vein occlusion; RCT, randomized controlled trial; SR, systematic review. Figure Figure Meta-analysis study network Probability that each treatment is the most efficacious in the study network BCVA assessed in terms of ETDRS letters. BCVA, best corrected visual acuity; ETDRS, Early Treatment Diabetic Retinopathy Study; PRN, pro re nata (as needed). Figure Sensitivity analysis: mean BCVA letters gained from baseline for ranibizumab monotherapy over aflibercept Figure Sensitivity analysis: probability that ranibizumab monotherapy is the most efficacious treatment based on BCVA letters gained from baseline - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

72 Page of xmm (00 x 00 DPI) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

73 Page of x0mm ( x DPI) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

74 Page of xmm (00 x 00 DPI) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

75 Page of Sensitivity analysis: mean BCVA letters gained from baseline for ranibizumab monotherapy over aflibercept x0mm ( x DPI) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

76 Page of Sensitivity analysis: probability that ranibizumab monotherapy is the most efficacious treatment based on BCVA letters gained from baseline xmm (0 x 0 DPI) - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

77 Page of Supplementary Table Inclusion criteria Study Baseline BCVA Disease duration Battaglia Parodi et al. letters* Not available BRAVO letters (0/00 0/0) weeks BRIGHTER letters (0/00 0/0) months COMRADE-B letters (0/00 0/0) months GENEVA letters (0/00 0/0) Duration of macular oedema weeks RABAMES letters (0/0 0/0) > to < months Tan et al. letters (0/00 0/0) weeks to months VIBRANT letters (0/0 0/0) weeks *Decimal score converted into ETDRS letters. BCVA, best-corrected visual acuity; ETDRS, Early Treatment Diabetic Retinopathy Study. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

78 Page of Supplementary Table Impact of adjusting inclusion criteria for BRIGHTER on baseline characteristics. Baseline characteristic Before adjustment After adjustment Ranibizumab 0. mg PRN Ranibizumab 0. mg PRN + laser Laser p value Ranibizumab 0. mg PRN Ranibizumab 0. mg PRN Laser p value + laser Patients (n) 0 0 Female (%) Duration of disease (months) Age (years) BCVA (letters) p values calculated with an analysis of variance test for continuous variables (duration of disease, age, BCVA) and with a Χ-squared test for the categorical variable (gender). BCVA, best-corrected visual acuity; PRN, pro re nata (as needed). Before adjustment, the number of patients is based on the full analysis set and the other demographics are based on the randomized set. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

79 Page 0 of Supplementary Table Outcomes before adjustment of baseline characteristics for BRAVO, BRIGHTER and COMRADE-B Study Outcome Treatment group BRAVO n/n (%) n/n (%) COMRADE-B LG (SD) n/n (%) BRIGHTER LG (SD) Ranibizumab 0. mg PRN 0/ () Laser Sham / () LG (SD). (.). (.0) / (). (.) /0 (). (0.) /0 ().0 (.) Dexamethasone 0. mg implant / (). (.) Ranibizumab 0. mg PRN + laser / (). (.0) BCVA, best-corrected visual acuity; LG, BCVA letters gained from baseline to month or month ; n, number of patients in treatment arm with adverse events; N, total number of patients in treatment arm; PRN, pro re nata (as needed); SD, standard deviation. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

80 Page of Supplementary Table Node-splitting (consistency between direct and indirect evidence): BCVA letters gained Comparison Difference between Direct evidence Indirect evidence direct and indirect estimates (% CrI) estimates (% CrI) estimates (% CrI) Ranibizumab 0. mg PRN vs dexamethasone. letters (..). letters (0..) 0. letters (..) 0. mg implant Laser vs sham. letters (.0.0) 0. letters (..) 0. letters (.0.) Ranibizumab 0. mg PRN vs laser. letters (..). letters (..). letters (..) BCVA, best-corrected visual acuity; CrI, credible interval; PRN, pro re nata (as needed); - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

81 Page of Supplementary materials. Medline search strategy for RCTs published since 00 Database: Ovid MEDLINE(R) < to July Week 0>, Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations <August 0, 0> exp Randomized Controlled Trials as Topic/ () exp Clinical Trials as Topic/ () randomised controlled trial.mp. [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] () exp Random Allocation/ () exp Double-Blind Method/ (0) exp Single-Blind Method/ () exp Clinical Trial/ () or or or or or or (0) ((singl$ or doubl$ or treb$ or tripl$) adj (blind$ or mask$)).tw. () 0 (clinic$ adj trial$).tw. (0) exp Placebos/ (0) placebo$.tw. (0) randomly allocated.tw. () (allocated adj random).tw. () or 0 or or or or () or () case report.tw. (0) exp Letter/ (0) exp Historical Article/ () 0 review of reported cases.pt. (0) or or or 0 (0) not () exp Macular Edema/ (0) exp Edema/ () (macular adj (edema or oedema or odema)).ti,ab. () exp Retinal Vein/ () exp Retinal Vein Occlusion/ () ((vein or veins or veinous) adj (occlusion$ or occluded or obstruction$ or obstructed or closed or closure$ or stricture$ or stenosis or stenosed or block or blocks or blockage$ or blocking or embolism$ or emboli) adj retina$).ti,ab. () - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

82 Page of (crvo or cvo or rvo or brvo or bvo or crvome).ti,ab. () 0 (branch vein adj occlu$).ti,ab. () (central vein adj occlu$).ti,ab. () or or or or or or or 0 or () exp Antibodies, Monoclonal/ () antibodies/tu () exp Vascular Endothelial Growth Factors/ () (vascular endothelial growth or vegf$ or antivef$).ti,ab,rn. () (ranibizumab or lucentis or rhufab v or --).ti,ab,rn. (0) exp Angiogenesis Inhibitors/ (0) triamcinolone acetonide.mp. or exp Triamcinolone Acetonide/ () 0 ivta.ti,ab,rn. () exp Dexamethasone/ () ozurdex.ti,ab,rn. (0) exp Light Coagulation/ (0) (photocoagulation or laser coagulation).ti,ab. () aflibercept.mp. or Vascular Endothelial Growth Factor A/ () Eylea.mp. () or/- () Animals/ () (editorial or letter or news).pt. (00) 0 and () 0 not ( or ) (0) limit to english language () limit to ed= () and () - : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright.

83 Page of Section/topic TITLE PRISMA 00 Checklist # Checklist item Reported on page # Title Identify the report as a systematic review, meta-analysis, or both. Page 0 ABSTRACT Structured summary Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, Page participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number. INTRODUCTION Rationale Describe the rationale for the review in the context of what is already known. Page 0 Objectives Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, Page outcomes, and study design (PICOS). METHODS Protocol and registration Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including registration number. Protocol not registered Eligibility criteria Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, Page language, publication status) used as criteria for eligibility, giving rationale. 0 Information sources Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched. Page Search Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated. Study selection State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, Page included in the meta-analysis). Data collection process 0 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes Page for obtaining and confirming data from investigators. 0 Data items List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and Page simplifications made. Risk of bias in individual Describe methods used for assessing risk of bias of individual studies (including specification of whether this was Page studies done at the study or outcome level), and how this information is to be used in any data synthesis. Summary measures State the principal summary measures (e.g., risk ratio, difference in means). Page - on 0 October 0 by guest. Protected by copyright. Page of supplement : first published as 0./bmjopen-0-00 on June 0. Downloaded from

84 Page of Synthesis of results PRISMA 00 Checklist Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I ) for each meta-analysis. Page of Page Reported on page # Section/topic # Checklist item 0 Risk of bias across studies Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective Page reporting within studies). Additional analyses Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating Page which were pre-specified. RESULTS Study selection Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at Figure each stage, ideally with a flow diagram. 0 Study characteristics For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and Table provide the citations. Risk of bias within studies Present data on risk of bias of each study and, if available, any outcome level assessment (see item ). Table Results of individual studies 0 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each Table intervention group (b) effect estimates and confidence intervals, ideally with a forest plot. Synthesis of results Present results of each meta-analysis done, including confidence intervals and measures of consistency. Pages and.tables 0 and a Risk of bias across studies Present results of any assessment of risk of bias across studies (see Item ). Page 0, Table Additional analysis Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item ]). Page and DISCUSSION Summary of evidence Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers). Page 0 Limitations Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of Page identified research, reporting bias). Conclusions Provide a general interpretation of the results in the context of other evidence, and implications for future research. Page FUNDING on 0 October 0 by guest. Protected by copyright. - : first published as 0./bmjopen-0-00 on June 0. Downloaded from

85 Page of Funding PRISMA 00 Checklist Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review. From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (00). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med (): e0000. doi:0./journal.pmed0000 For more information, visit: on 0 October 0 by guest. Protected by copyright. Page of - Page : first published as 0./bmjopen-0-00 on June 0. Downloaded from

86 Comparative efficacy and safety of approved treatments for macular oedema secondary to branch retinal vein occlusion: a network meta-analysis Journal: Manuscript ID: bmjopen-0-00.r Article Type: Research Date Submitted by the Author: -Apr-0 Complete List of Authors: Regnier, Stephane; Novartis Pharma AG, Health Economics and Outcomes Research Larsen, Michael; University of Copenhagen, Department of Ophthalmology, Glostrup Hospital Bezlyak, Vladimir; Novartis Pharma AG, Allen, Felicity; Novartis Pharmaceuticals UK Ltd, <b>primary Subject Heading</b>: Ophthalmology Secondary Subject Heading: Ophthalmology Keywords: Medical retina < OPHTHALMOLOGY, Health economics < HEALTH SERVICES ADMINISTRATION & MANAGEMENT, OPHTHALMOLOGY : first published as 0./bmjopen-0-00 on June 0. Downloaded from on 0 October 0 by guest. Protected by copyright. -