640 Full

Clinical science
Br J Ophthalmol: first published as 10.1136/bjophthalmol-2020-317262 on 4 January 2021. Downloaded from http://bjo.bmj.com/ on October 24, 2024 by guest. Protected by copyright.
Short-term efficacy of latanoprostene bunod for the
treatment of open-angle glaucoma and ocular
hypertension: a systematic literature review and a
network meta-analysis
Paul Harasymowycz,1 Catherine Royer,2 Amy Xianying Cui,3 Martin Barbeau,3
Katherine Jobin-Gervais,3 Karine Mathurin,2,4 Jean Lachaine ‍ ‍,2,4
Catherine Beauchemin2,4
► Additional material is ABSTRACT vision loss and the second leading cause of blind-
published online only. To view, Background/aims To assess the comparative efficacy ness worldwide.2 3 Primary open- angle glaucoma
please visit the journal online
(http://d x.doi.o rg/10.1136/ of latanoprostene bunod (LBN), a novel prostaglandin (POAG) is the most common form of the disease
bjophthalmol-2020-317262). analogue (PGA), to other medications for open- in North America with a prevalence of 3.3% (2.7
angle glaucoma and ocular hypertension on lowering million people) in adults aged between 40 and 80
1
Ophthalmology, University of intraocular pressure (IOP). years in 2013.4
Montreal, Montreal, Québec, The goal of treatment is to reduce intraocular
Methods A systematic literature review adapted
Canada
2
Pharmacoeconomics and from the Li et al (Ophthalmology, 2016) study was pressure (IOP), which is the only modifiable risk
Outcomes Research, PeriPharm conducted. Medline, Embase and PubMed were searched factor at this time.3 5 Initial treatment consists
Inc, Montreal, Québec, Canada for randomised controlled trials published between of topical therapies with several classes avail-
3
Market Access, Bausch Health 1 January 2014 and 19 March 2020. Studies had to able, including prostaglandin analogues (PGAs),
Canada Inc, Laval, Quebec,
Canada report IOP reduction after 3 months for at least two α-adrenergic agonist, beta- blockers and carbonic
4
Pharmacy, Université de different treatments among placebo, PGAs (bimatoprost anhydrase inhibitors and parasympathomimetic
Montréal, Montreal, Quebec, 0.01%, bimatoprost 0.03%, latanoprost, LBN, tafluprost, agents.3 6 7 Among these, PGAs are the most effec-
Canada unoprostone) or apraclonidine, betaxolol, brimonidine, tive medication because of their unmatched safety
brinzolamide, carteolol, dorzolamide, levobunolol, profile, IOP-lowering capabilities and their once-
Correspondence to timolol, travoprost. A Bayesian network meta-analysis daily administration, with latanoprost, bimatoprost
Professor Jean Lachaine,
Pharmacy, Université de was performed to provide the relative effect in terms of and travoprost being the most frequently used.8 Of
Montréal, Montreal, H3T 1J4, mean difference (95% credible interval) of IOP reduction note, in 2012, Lumigan (bimatoprost 0.03%) was
Canada; and ranking probabilities. Surface under the cumulative discontinued and replaced by Lumigan RC (bimato-
jean.lachaine@umontreal.ca ranking curve (SUCRA) was generated. prost 0.01%) due to its favourable tolerability
Results A total of 106 trials were included with data profile.9
Received 18 June 2020
Revised 4 November 2020 for 18 523 participants. LBN was significantly more In order to compare the different treatments a
Accepted 4 December 2020 effective than unoprostone (−3.45 (−4.77 to −2.12)). comprehensive assessment of their relative efficacy
Published Online First Although relative effect was not significative, compared is crucial for clinicians and healthcare decision-
4 January 2021 with other PGAs, LBN numerically outperformed makers,8 10 however, no head- to-
head trials
latanoprost (−0.70 (−1.83 to 0.43)) and tafluoprost comparing all relevant competing therapies have
(−0.41 (−1.87 to 1.07)), was similar to bimatoprost been published. In the absence of direct evidence,
0.01% (-0.02(−1.59 to 1.55)) and was slightly the use of a network meta- analysis (NMA) may
disadvantaged by bimatoprost 0.03% (−0.17 (−1.42 provide useful evidence.10 In 2016, Li et al published
to 1.07)). LBN was significantly more efficient than the the results of a systematic review and an NMA
beta-blockers apraclonidine, betaxolol, brimonidine, which aimed to compare the effectiveness of first-
brinzolamide, carteolol, dorzolamide and timolol. line medications for patients with POAG or ocular
According to SUCRA, LBN was ranked second after hypertension (OH) and to provide relative ranking
bimatoprost 0.03%, followed by bimatoprost 0.01%. of these treatments. The authors conducted a
Conclusion LBN was significantly more effective than systematic review in March 2014 in order to identify
the PGA unoprostone and most of the beta-blockers. all randomised controlled trials (RCTs) comparing
Compared with the most widely used PGAs, LBN single active topical medication with no treatment/
numerically outperformed latanoprost and travoprost and placebo or with another single topical treatment.
© Author(s) (or their Following a systematic review of 114 eligible trials,
was similar to bimatoprost 0.01%.
employer(s)) 2022. Re-use
permitted under CC BY-NC. No results of the NMA indicated that, compared with
commercial re-use. See rights beta-blockers, α-adrenergic agonists and carbonic
and permissions. Published anhydrase inhibitors, PGAs were more efficacious in
by BMJ. INTRODUCTION reducing IOP at 3 months. Authors also concluded
To cite: Harasymowycz P, Glaucoma is a group of progressive optic neurop- that drugs within the PGA class, namely bimato-
Royer C, Cui AX, et al. athies characterised by degeneration of retinal prost, latanoprost and travoprost were among the
Br J Ophthalmol ganglion cells which may lead to vision loss and most efficacious, with intraclass difference found to
2022;106:640–647. blindness.1 It is the number one cause of irreversible be small and not clinically meaningful.11
640 Harasymowycz P, et al. Br J Ophthalmol 2022;106:640–647. doi:10.1136/bjophthalmol-2020-317262
Clinical science
Vyzulta (latanoprostene bunod (LBN) ophthalmic solution, trial design, location of trial, sample size, patients’ baseline char-
0.024% w/v), a novel nitric oxidedonating prostaglandin F2α acteristics, intervention characteristics and quantitative results
analogue has received approbation for commercialisation in six with regard to treatment effect. For studies presenting multiple
different countries.12 The safety and efficacy of LBN has been treatment durations, the duration closest to 3 months was used.
well established through clinical studies (APOLLO and LUNAR If many IOP measures were available, the selection was made
studies),13 14 where LBN demonstrated enhanced efficacy in this order: mean diurnal IOP, 24-hour mean IOP, peak IOP
compared with latanoprost and timolol.13–15 However, the effec- reduction and morning IOP. Any discrepancies were resolved by
tiveness of LBN in comparison to other topical therapies other consensus or with the help of a third reviewer.
than latanoprost and timolol has not yet been evaluated. The
objective of this study was to assess, through a systematic review
and an NMA, the relative efficacy, as well as provide a relative Quality assessment
ranking, of LBN compared with other topical medications, with As part of their systematic review, Li et al assessed the quality of
a focus on PGAs, for the treatment of POAG and OH. included trials using the Cochrane Risk of Bias Tool, where the
following seven methodological domains were graded as ‘low’,
‘high’ or ‘unclear’ risk of bias: sequence generation, allocation
MATERIAL AND METHODS concealment, blinding of participants and personnel, blinding of
The pool of studies included in Li et al11 previously described was
outcome assessors, funding of the trial and financial relation-
considered and an exhaustive literature review was performed for
ship reported by the authors.11 17 Based on their work, new trials
studies published after 2013. An NMA was conducted according
identified by the current systematic review were assessed using
to a predefined protocol and was conformed to the Preferred
the same method.
Reporting Items for Systematic Reviews and Meta-Analyses16
extension for NMA. The review question was established using
the population, intervention, comparators, outcomes (PICO) Outcome definition
framework. This systematic review was adapted from the work The primary outcome was defined as the mean reduction (MR)
by Li et al previously described.11 of IOP in continuous mmHg units after 3 months of treatment.
The mean difference (MD) of the MR of IOP between two
Search strategy treatments with a 95% CI or credible interval (CrI) was calcu-
MEDLINE, EMBASE and PubMed databases were searched on lated. An MD under 0 indicated that the treatment of reference
19 March 2020 to identify RCTs published in English or French performed a higher IOP reduction relative to its comparator and
between 1 January 2014 and 19 March 2020. A manual search was therefore more effective.
of reference lists was also performed to identify potentially rele-
vant papers and systematic reviews. The same search strategy
Data synthesis and analysis
elaborated by Li et al was used but ‘latanoprostene bunod’ was
Using the ‘meta’ package in R, a pairwise meta- analysis (ie,
added as a keyword.11 Detailed search strategies are presented in
direct comparisons) with a random-effect model was conducted
online supplemental appendix A.
for every treatment comparison with at least two trials. Statis-
tical heterogeneity between studies was assessed using the I2
Eligibility statistic, which describes the percentage of variability in effect
Studies were selected if they reported relative efficacy between estimates that is due to heterogeneity rather than sampling error
at least two different treatments (placebo, bimatoprost 0.01%, (chance).18 Cochrane Handbook developed a rough guide for
bimatoprost 0.03%, latanoprost, LBN, tafluprost, unoprostone, interpretation of I2: less than 40% might not be important,
apraclonidine, betaxolol, brimonidine, brinzolamide, carteolol, 30%–60% may represent moderate heterogeneity; 50%–90%
dorzolamide, levobunolol, timolol or travoprost) in terms of IOP may represent substantial heterogeneity and 75%–100% repre-
reduction after 3 months of usage. All eligibility criteria were sented considerable heterogeneity.18 Pairwise comparisons with
defined a priori and were rigorously considered assuming the an I2 value greater than 65% were investigated to identify studies
similarity assumption. Inclusion criteria included the following: possibly causing heterogeneity.
RCTs with a parallel-group design (cross-over trials excluded); An NMA, which combined direct and indirect compari-
at least 60% of patients with a diagnostic of POAG and/or sons, was conducted using a Bayesian random- effect model
OH; trials that assess a monotherapy regimen (combination of with Markov Chain Monte Carlo simulations executed with
medical treatments excluded); studies published in English and the ‘gemtc’ package in R.19 Using four parallel chains, 50 000
French between 1 January 2014 and 19 March 2020.Trials were samples after 20 000- sample burn-in were obtained in each
excluded if they enrolled fewer than 10 participants in each chain. Convergence of the model was assessed using the Brooks-
group or if they evaluated a combination of medical treatments. Gelman-Rubin diagnostic in the ‘coda’ package in R. Consis-
Although no maximum or minimum duration of treatment was tency of the NMA, defined as a statistical discrepancy between
required, participants had to be followed for at least 28 days direct and indirect comparison results, was evaluated using a
after randomisation. node-splitting approach with the ‘gemtc’ package in R.19
The model ranked each treatment by their relative effect
Study selection and data extraction (probabilities of being more effective). Cumulative probability
Two reviewers independently screened the titles and abstracts of of being the most effective treatment was calculated. With that,
publications for potential eligibility. Using a predefined eligibility the surface under the cumulative ranking curve (SUCRA) of
form (online supplemental appendix B), both reviewers screened each treatment is obtained.20 Specifically, SUCRA is a numeric
the full text of all potentially eligible trials. Any disagreements presentation of the overall ranking and presents a single number,
were resolved by consensus or with the help of a third reviewer. ranging from 0% to 100%, associated with each treatment,
Data extraction was performed by two independent reviewers. where 0% represents the least effective treatment and 100%
Data extracted included: first author’s name, year of publication, represents the most effective treatment.21
Harasymowycz P, et al. Br J Ophthalmol 2022;106:640–647. doi:10.1136/bjophthalmol-2020-317262 641
Clinical science
Figure 1 Organisational chart of the literature review. *Among the
128 excluded studies, seven were included in the Li et al publication.
NMA, network meta-analysis; POAG, primary open-angle glaucoma.
Figure 2 Network Graph. The nodes are weighted according to

the number of participants randomised to that drug. The edges are
Sensitivity analyses were conducted to assess the impact of weighted according to the number of direct comparison studies
heterogeneity between studies and inconsistency results by between drugs.
removing studies identified as possibly causing heterogeneity
and including inconsistent combinations, respectively.
Supplementary analyses were conducted to evaluate the
heterogeneity between baseline characteristics among trials compared with timolol in two studies and latanoprost in one
included. study. In both cases, LBN significantly lowers IOP more than
All concentrations of the same medication were combined in the other treatments after 3 months (LBN vs timolol: MD (95%
the same group except for bimatoprost 0.01% and bimatoprost CI)=−1.42 (−1.84 to −1.01) and LBN vs latanoprost: −1.23
0.03%. (−1.76 to −0.70)).
Results of the NMA indicate that, when compared with
RESULTS placebo, all active drugs demonstrate an improved reduction
Of the 2642 publications identified by the systematic review and of IOP at 3 months (table 2 and online supplemental appendix
the 114 studies used by Li et al, 106 RCTs met the a priori eligi- F). More specifically, the MDs in IOP reduction at 3 months for
bility criteria and were included (figure 1; references of these active drug in comparison to placebo range from −1.97 mm
RCTs are listed in online supplemental appendix C). Of these, Hg for unoprostone to −5.59 mm Hg for bimatoprost 0.03%
11 (10%) were published between 2014 and 2020. The total and are all statistically significative. Importantly, LBN shows
number of participants contributing to this network is 18 523 the second greatest reduction in IOP vs placebo with an MD
(complete characteristics of included studies are listed in online (95% CrI) of −5.42 mm Hg (−6.68 to –4.16). Furthermore,
supplemental appendix D. these results highlight the statistically significant superiority in
Of the 106 trials, risk of selection bias (online supplemental efficacy of LBN compared with the PGA unoprostone (−3.45
appendix E) was rated as low for 54 (51%) and 33 (31%) studies (−4.77 to −2.12)) and the beta-blockers apraclonidine (−2.55
when assessing sequence generation or allocation concealment, (−4.52 to −0.55)), betaxolol (−2.89 (−4.17 to −1.60)),
respectively, whereas the remaining trials were rated as having an brimonidine (−1.75 (−3.02 to −0.49)), brinzolamide (−2.88
‘unclear risk’ except for one study with a ‘high risk’ in allocation (−-4.29 to −1.47)), carteolol (−2.17 (−3.65 to −0.69)),
concealment. Risk of performance bias, associated with blinding dorzolamide (−2.87 (−4.17 to −1.55)) and timolol (−1.69
of participants, was rated as low (ie, reported blinding), high (ie, (−2.80 to −0.58)). Although the relative effect was not signif-
reported not blinding) or unclear (ie, not reported or unclear), icative, compared with other PGAs, LBN numerically outper-
for 42%, 37% and 21% of studies, respectively. Risk of detection formed latanoprost (−0.70 (−1.83 to 0.43)) and tafluoprost
bias, associated with blinding of the outcome assessor, was rated (−0.41 (−1.87 to 1.07)), was similar to bimatoprost 0.01%
as low (ie, reported blinding), high (ie, reported not blinding) (−0.02 (−1.59 to 1.55)) and bimatoprost 0.03% demonstrated
or unclear risk (ie, not reported) for 24%, 63% and 13% of a slightly advantage over LBN (−0.17 (−1.42 to 1.07)) (table 2
studies, respectively. Of the 69 articles who reported funding for and online supplemental appendix F). The model ranked each
their research, 64 (93%) were funded by the industry. Of the 55 treatment by their relative effect (probabilities of being more
articles that reported financial relationship, 15 (27%) declared effective) (table 3). According to these results, treatment with
having no financial conflict of interest. the higher probability of being ranked first is bimatoprost
The 106 studies included compared 16 interventions (figure 2). 0.03% with a probability of 37%, followed by LBN with a
A total of 138 direct comparisons were performed based on 93 probability of 29%. LBN has a probability of 51% to be under
two-arm trials, 11 three-arm trials and 2 four-arm trials. Results the two best treatments and 70% to be under the three best
of the pairwise meta-analysis are presented in table 1. LBN was treatments. Cumulative probability of being the most effective
Clinical science
Table 1 Summary estimates for intraocular pressure at 3 months derived from the pairwise meta-analysis
95%CI
Mean
Control Experimental Total no of studies difference* Low Up τ2† I2‡
Placebo Bimatoprost 0.01% 1 −4.60 −5.60 −3.60 NA NA
Latanoprost 1 −3.10 −3.98 −2.22 NA NA
Unoprostone 1 −0.30 −1.50 0.90 NA NA
Betaxolol 2 −3.16 −4.17 −2.15 0.3 52%
Brimonidine 1 −2.30 −3.99 −0.61 NA NA
Brinzolamide 1 −2.22 −3.48 −0.96 NA NA
Dorzolamide 4 −2.48 −3.84 −1.12 1.3 76%
Levobunolol 2 −7.90 −8.94 −6.85 0.0 0%
Timolol 4 −3.75 −4.75 −2.76 0.6 58%
Bimatoprost 0.01% Latanoprost 2 1.02 0.68 1.37 0.0 0%
Tafluprost 1 2.30 −0.91 5.51 NA NA
Travoprost 2 1.50 −1.98 4.97 5.2 80%
Bimatoprost 0.03% Latanoprost 7 0.99 0.46 1.53 0.3 61%
Travoprost 8 0.44 −0.52 1.40 1.4 86%
Latanoprost Latanoprostene bunod 1 −1.23 −1.76 −0.70 NA NA
Tafluprost 3 −0.99 −1.92 −0.07 0.0 0%
Unoprostone 6 2.90 2.16 3.63 0.3 37%
Travoprost 7 −0.15 −1.30 1.00 1.9 87%
Apraclonidine Timolol 2 −0.44 −3.91 3.03 5.6 89%
Betaxolol Latanoprost 2 −1.84 −3.22 −0.47 0.0 0%
Unoprostone 1 0.60 0.09 1.11 NA NA
Dorzolamide 2 −0.21 −0.82 0.40 0.0 0%
Levobunolol 2 −4.65 −10.13 0.84 13.3 84%
Timolol 6 −1.30 −2.46 −0.13 1.2 67%
Brimonidine Latanoprost 5 −1.22 −2.13 −0.31 0.8 78%
Betaxolol 1 2.00 0.90 3.10 NA NA
Brinzolamide 2 0.90 0.39 1.42 0.0 0%
Timolol 4 0.42 0.04 0.81 0.0 0%
Travoprost 1 −1.20 −3.77 1.37 NA NA
Brinzolamide Dorzolamide 2 −0.34 −0.84 0.16 0.0 0%
Carteolol Levobunolol 1 −2.90 −4.59 −1.21 NA NA
Timolol 4 −0.27 −1.11 0.57 0.4 60%
Dorzolamide Latanoprost 1 −2.90 −3.70 −2.10 0.0 NA
Levobunolol Timolol 9 0.11 −0.40 0.62 0.1 15%
Timolol Bimatoprost 0.03% 6 −2.06 −2.36 −1.75 0.0 0%
Latanoprost 15 −1.18 −1.65 −0.70 0.6 76%
Latanoprostene bunod 2 −1.42 −1.84 −1.01 0.0 0%
Tafluprost 2 −0.50 −1.12 0.12 0.1 38%
Unoprostone 2 0.94 −0.43 2.31 0.9 87%
Brinzolamide 3 1.10 0.52 1.69 0.0 0%
Dorzolamide 4 0.99 0.34 1.64 0.1 26%
Travoprost 4 −0.89 −1.26 −0.52 0.0 0%
Travoprost Tafluprost 1 −1.30 −2.93 0.33 NA NA
Total 16 drugs 138§
‍ ‍, PGA
*Difference between the reduction in IOP during the study of the experimental drug and the control drug (mean difference under 0 favours the experimental drug). Results
presented in bold are significant.
†τ2 describes the underlying between-study variability.
‡I2 is the percentage of variability in the treatment estimates which is attributable to heterogeneity.
§106 trials considered: 93 two-arm trials, 11 three-arm trials and 2 four-arm trials.
IOP, intraocular pressure; NA, not available; PGA, prostaglandin analogue.
treatment was calculated and the cumulative ranking curve of bimatoprost 0.01% (87%), tafluprost (78%), travoprost (73%),
each treatment (presented in online supplemental appendix G) levobunolol (72%), latanoprost (68%), timolol (48%), brimoni-
was obtained to calculate the SUCRA. According to SUCRA dine (47%), carteolol (38%), apraclonidine (30%), dorzolamide
results, LBN (SUCRA=88%) emerges as the second best treat- (23%), brinzolamide (22%), betaxolol (22%), unoprostone
ment after bimatoprost 0.03% (94%) and followed in order by (11%) and placebo (0%).
644
Table 2 Mean difference in intraocular pressure at 3 months (95% credible interval) derived from the NMA
−5.39 −5.59 −4.72 −5.42 −5.00 −1.97 −2.86 −2.53 −3.66 −2.53 −3.24 −2.55 −4.79 −3.73 −4.84
(−5.69
Placebo (−6.60 to −4.21) (−6.43 to−4.75) (−5.39 to−4.04) (−6.68 to−4.16) (−6.15 to−3.88) (−2.88 to−1.06) (−4.65 to−1.12) (−3.32 to−1.75) (−4.48 to−2.85) (−3.52 to−1.56) (−4.40 to−2.09) (−3.33 to−1.78) (−5.64 to−3.95) (−4.36 to−3.10) to−3.98)
Clinical science
5.39 Bimatoprost −0.20 0.67 −0.02 0.39 3.42 2.53 2.86 1.73 2.85 2.15 2.84 0.60 1.66 0.55
0.01%
(4.21 to 6.60) (−1.41 to1.03) (−0.44 to1.80) (−1.59 to1.55) (−1.06 to1.82) (2.11 to 4.75) (0.51 to 4.51) (1.59 to 4.15) (0.47 to 2.99) (1.45 to 4.27) (0.65 to 3.65) (1.55 to 4.13) (−0.70 to1.92) (0.52 to 2.81) (−0.60 to1.72)
5.59 0.20 Bimatoprost 0.88 0.17 0.59 3.62 2.73 3.06 1.93 3.06 2.35 3.04 0.80 1.86 0.75
0.03%
(4.75 to 6.43) (−1.03 to1.41) (0.26 to 1.48) (−1.07 to1.42) (−0.54 to1.69) (2.69 to 4.55) (0.94 to 4.47) (2.18 to 3.95) (1.09 to 2.76) (2.01 to 4.10) (1.18 to 3.50) (2.12 to 3.94) (−0.12 to1.71) (1.24 to 2.48) (0.08 to 1.43)
4.72 −0.67 −0.88 Latanoprost −0.70 −0.29 2.75 1.85 2.18 1.05 2.18 1.47 2.17 −0.08 0.99 −0.12
(4.04 to 5.39) (−1.80 to0.44) (−1.48 to−0.26) (−1.83 to0.43) (−1.28 to0.69) (2.01 to 3.49) (0.13 to 3.53) (1.46 to 2.92) (0.41 to 1.69) (1.27 to 3.09) (0.41 to 2.53) (1.41 to 2.92) (−0.85 to0.70) (0.58 to 1.39) (−0.75 to0.52)
5.42 0.02 −0.17 0.70 Latanoprostene 0.41 3.45 2.55 2.89 1.75 2.88 2.17 2.87 0.62 1.69 0.58
Bunod
5.00 −0.39 −0.59 0.29 −0.41 Tafluprost 3.04 2.14 2.47 1.34 2.47 1.76 2.46 0.21 1.28 0.17
1.97 −3.42 −3.62 −2.75 −3.45 −3.04 Unoprostone −0.90 −0.56 −1.70 −0.57 −1.28 −0.58 −2.82 −1.76 −2.87
(1.06 to 2.88) (−4.75 to−2.11) (−4.55 to−2.69) (−3.49 to−2.01) (−4.77 to−2.12) (−4.25 to−1.83) (−2.73 to0.90) (−1.50 to0.38) (−2.63 to−0.75) (−1.69 to0.56) (−2.52 to−0.03) (−1.57 to0.41) (−3.83 to−1.81) (−2.53 to−0.99) (−3.82
to−1.92)
2.86 −2.53 −2.73 −1.85 −2.55 −2.14 0.90 Apraclonidine 0.33 −0.80 0.33 −0.38 0.31 −1.93 −0.86 −1.98
(1.12 to 4.65) (−4.51 to−0.51) (−4.47 to−0.94) (−3.53 to−0.13) (−4.52 to−0.55) (−4.05 to−0.20) (−0.90 to2.73) (−1.43 to2.14) (−2.54 to0.99) (−1.51 to2.21) (−2.28 to1.54) (−1.46 to2.13) (−3.70 to−0.13) (−2.49 to0.81) (−3.73
to−0.18)
2.53 −2.86 −3.06 −2.18 −2.89 −2.47 0.56 −0.33 Betaxolol −1.13 0.00 −0.71 −0.02 −2.26 −1.20 −2.31
(1.75 to 3.32) (−4.15 to−1.59) (−3.95 to−2.18) (−2.92 to−1.46) (−4.17 to−1.60) (−3.66 to−1.31) (−0.38 to1.50) (−2.14 to1.43) (−1.99 to−0.29) (−1.06 to1.03) (−1.90 to0.46) (−0.87 to0.81) (−3.17 to−1.37) (−1.88 to−0.52) (−3.21
to−1.40)
3.66 −1.73 −1.93 −1.05 −1.75 −1.34 1.70 0.80 1.13 Brimonidine 1.13 0.42 1.11 −1.13 −0.06 −1.17
(2.85 to 4.48) (−2.99 to−0.47) (−2.76 to−1.09) (−1.69 to−0.41) (−3.02 to−0.49) (−2.49 to−0.21) (0.75 to 2.63) (−0.99 to2.54) (0.29 to 1.99) (0.21 to 2.04) (−0.75 to1.59) (0.23 to 1.99) (−2.05 to−0.21) (−0.70 to0.57) (−2.02
to−0.32)
2.53 −2.85 −3.06 −2.18 −2.88 −2.47 0.57 −0.33 0.00 −1.13 Brinzolamide −0.71 −0.01 −2.26 −1.19 −2.30
(1.56 to 3.52) (−4.27 to−1.45) (−4.10 to−2.01) (−3.09 to−1.27) (−4.29 to−1.47) (−3.77 to−1.18) (−0.56 to1.69) (−2.21 to1.51) (−1.03 to1.06) (−2.04 to−0.21) (−2.01 to0.60) (−0.99 to0.95) (−3.34 to−1.17) (−2.06 to−0.32) (−3.36
to−1.24)
3.24 −2.15 −2.35 −1.47 −2.17 −1.76 1.28 0.38 0.71 −0.42 0.71 Carteolol 0.69 −1.55 −0.48 −1.59
(2.09 to 4.40) (−3.65 to−0.65) (−3.50 to−1.18) (−2.53 to−0.41) (−3.65 to−0.69) (−3.15 to−0.38) (0.03 to 2.52) (−1.54 to2.28) (−0.46 to1.90) (−1.59 to0.75) (−0.60 to2.01) (−0.51 to1.89) (−2.68 to−0.42) (−1.46 to0.50) (−2.76
to−0.42)
2.55 −2.84 −3.04 −2.17 −2.87 −2.46 0.58 −0.31 0.02 −1.11 0.01 −0.69 Dorzolamide −2.24 −1.18 −2.29
(1.78 to 3.33) (−4.13 to−1.55) (−3.94 to−2.12) (−2.92 to−1.41) (−4.17 to−1.55) (−3.65 to−1.27) (−0.41 to1.57) (−2.13 to1.46) (−0.81 to0.87) (−1.99 to−0.23) (−0.95 to0.99) (−1.89 to0.51) (−3.18 to−1.29) (−1.88 to−0.47) (−3.21
to−1.35)
4.79 −0.60 −0.80 0.08 −0.62 −0.21 2.82 1.93 2.26 1.13 2.26 1.55 2.24 Levobunolol 1.06 −0.05
(3.95 to 5.64) (−1.92 to0.70) (−1.71 to0.12) (−0.70 to0.85) (−1.92 to0.68) (−1.41 to0.97) (1.81 to 3.83) (0.13 to 3.70) (1.37 to 3.17) (0.21 to 2.05) (1.17 to 3.34) (0.42 to 2.68) (1.29 to 3.18) (0.38 to 1.75) (−0.98 to0.89)
3.73 −1.66 −1.86 −0.99 −1.69 −1.28 1.76 0.86 1.20 0.06 1.19 0.48 1.18 −1.06 Timolol −1.11
(3.10 to 4.36) (−2.81 to−0.52) (−2.48 to−1.24) (−1.39 to−0.58) (−2.80 to−0.58) (−2.27 to−0.30) (0.99 to 2.53) (−0.81 to2.49) (0.52 to 1.88) (−0.57 to0.70) (0.32 to 2.06) (−0.50 to1.46) (0.47 to 1.88) (−1.75 to−0.38) (−1.76
to−0.46)
4.84 −0.55 −0.75 0.12 −0.58 −0.17 2.87 1.98 2.31 1.17 2.30 1.59 2.29 0.05 1.11 Travoprost
(3.98 to 5.69) (−1.72 to0.60) (−1.43 to−0.08) (−0.52 to0.75) (−1.85 to0.69) (−1.28 to0.92) (1.92 to 3.82) (0.18 to 3.73) (1.40 to 3.21) (0.32 to 2.02) (1.24 to 3.36) (0.42 to 2.76) (1.35 to 3.21) (−0.89 to0.98) (0.46 to1.76)
‍ ‍, PGA
‍ ‍, others
Results presented in bold are significant.
Mean difference under 0 favours the drug in the column.
NAM, network meta-analysis; PGA, prostaglandin analogue.
Harasymowycz P, et al. Br J Ophthalmol 2022;106:640–647. doi:10.1136/bjophthalmol-2020-317262

Table 3 Ranking probabilities and SUCRA
A B C D E F G H I J K L M N O P
Bimatoprost Bimatoprost Latanoprostene
Ranks Placebo 0.01% 0.03% Latanoprost bunod Tafluprost Unoprostone Apraclonidine Betaxolol Brimonidine Brinzolamide Carteolol Dorzolamide Levobunolol Timolol Travoprost
1 0.000 0.266 0.367 0.000 0.288 0.067 0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.009 0.000 0.002
2 0.000 0.221 0.374 0.002 0.220 0.117 0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.039 0.000 0.025
3 0.000 0.204 0.190 0.023 0.193 0.180 0.000 0.002 0.000 0.000 0.000 0.000 0.000 0.104 0.000 0.104
4 0.000 0.126 0.055 0.103 0.118 0.198 0.000 0.004 0.000 0.000 0.000 0.000 0.000 0.176 0.000 0.220
5 0.000 0.072 0.011 0.251 0.071 0.144 0.000 0.005 0.000 0.000 0.000 0.001 0.000 0.195 0.000 0.250
6 0.000 0.052 0.002 0.358 0.050 0.120 0.000 0.007 0.000 0.002 0.000 0.003 0.000 0.193 0.000 0.214
7 0.000 0.052 0.000 0.256 0.052 0.153 0.000 0.016 0.000 0.020 0.000 0.010 0.000 0.263 0.005 0.174
8 0.000 0.006 0.000 0.008 0.006 0.016 0.000 0.089 0.000 0.328 0.001 0.107 0.000 0.018 0.409 0.011
9 0.000 0.001 0.000 0.000 0.001 0.004 0.000 0.063 0.001 0.353 0.005 0.120 0.002 0.001 0.446 0.001
10 0.000 0.001 0.000 0.000 0.001 0.001 0.003 0.150 0.036 0.230 0.058 0.352 0.041 0.000 0.128 0.000
11 0.000 0.000 0.000 0.000 0.000 0.000 0.015 0.201 0.152 0.062 0.170 0.232 0.157 0.000 0.011 0.000
12 0.000 0.000 0.000 0.000 0.000 0.000 0.041 0.107 0.259 0.004 0.231 0.088 0.271 0.000 0.000 0.000
Harasymowycz P, et al. Br J Ophthalmol 2022;106:640–647. doi:10.1136/bjophthalmol-2020-317262

13 0.000 0.000 0.000 0.000 0.000 0.000 0.087 0.094 0.271 0.001 0.220 0.049 0.278 0.000 0.000 0.000
14 0.000 0.000 0.000 0.000 0.000 0.000 0.220 0.123 0.223 0.000 0.214 0.029 0.192 0.000 0.000 0.000
15 0.001 0.000 0.000 0.000 0.000 0.000 0.635 0.136 0.059 0.000 0.102 0.009 0.058 0.000 0.000 0.000
16 0.999 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
SUCRA* 0.0% 87.2% 93.5% 68.4% 87.6% 77.9% 10.6% 30.1% 22.2% 46.7% 22.3% 37.8% 22.7% 71.8% 48.5% 72.7%
Ranking 16 3 1 7 2 4 15 11 14 9 13 10 12 6 8 5
according
SUCRA*
‍ ‍, PGA
*Higher is the SUCRA, better is the rank of this treatment.
SUCRA, surface under the cumulative ranking curve.
Clinical science
645
Clinical science
Sensitivity analyses the databases exploited for this study. Nonetheless, results
A total of 10 direct comparisons were identified as possibly presented herein are consistent with the findings of Li et al.
causing heterogeneity (online supplemental appendix H). Indeed, when comparing PGAs in terms of IOP reduction at 3
When excluding these studies, the sensitivity analysis revealed months, the intraclass differences are relatively small and not
no significant change in the NMA results (online supplemental significantly meaningful. In addition, this systematic review
appendix I). LBN was still significantly better than unoprostone and NMA, which include the most recent PGA, namely LBN,
and non-PGAs treatments, except for levobunolol and travoprost provides new findings relevant to clinicians and decision-
that was numerically superior. Although the relative effect was makers as it allows for the comparison of drugs that had not
not significative, compared with other PGAs, LBN numerically yet been evaluated in head-to-head trials.
outperformed latanoprost (−0.72 (−1.60 to 0.16)), tafluoprost It should be noted that there are some limitations associated
(−0.60 (−1.80 to 0.61)) and bimatoprost 0.01% (-0.40 (−1.70 with this NMA. First, although an NMA represents a powerful
to 0.83)) and bimatoprost 0.03% demonstrated a slight advan- tool and may provide crucial information, an inherent limitation
tage over LBN (0.13 (−0.88 to 1.10)). associated with NMA resides in the variability and the risk of
The node-splitting approach allowed for the identification of biases of studies included. Due to possible variability between
two inconsistent nodes (levobunolol vs placebo and timolol vs studies and between the comparisons made, a critical step when
levobunolol) (online supplemental appendix J). When excluding performing an NMA consists of validating the homogeneity and
these nodes, the sensitivity analysis revealed no significant consistency assumptions. The sensitivity analyses conducted did
change in the NMA results. Compared with unoprostone (PGA) not significantly alter the results, suggesting that the assumptions
and other non-PGAs, results indicated that LBN was significantly and conclusions made based on the statistical analysis are reliable
better, excluding travoprost but including levobunolol (online and robust.
supplemental appendix I). Compared with other PGAs, although Second, this NMA focused on IOP reduction and did not
the relative effect was not significative, LBN was still numeri- include visual field outcomes. We acknowledge that the
cally superior to latanoprost (−0.66 (−1.60 to 0.31)), similar to ultimate goal in the management of glaucoma consists of
bimatoprost 0.01% (0.09 (−1.30 to 1.50)) and disadvantaged by slowing or stopping structural damages leading to vision loss
bimatoprost 0.03% (0.20 (−0.87 to 1.30)). and that, consequently, visual field outcomes would be more
Four supplementary analyses were also conducted to evaluate clinically meaningful than IOP when comparing treatment
the heterogeneity between baseline characteristics among trials response. However, due to the lag time between onset of
by considering: (1) only studies published from 2000 onward, optic neuropathy and clinically detectable visual field defects,
(2) studies with a washout period before randomisation, (3) the use of visual field outcomes to assess relative effective-
studies that excluded prior glaucoma and cataract surgery, and ness of different interventions requires an extended time
(4) studies that excluded prior glaucoma laser. These analyses frame which poses challenges to the conduct of RCTs. Thus,
revealed that heterogeneity between baseline characteristics had although IOP does not measure structural of functional glau-
no significant impact on the NMA results (online supplemental comatous optic neuropathy, it remains the most commonly
appendix K). used surrogate endpoint of RCTs.10 20 22 Li et al reported that
Also, the Brooks- Gelman- Rubin plot (online supplemental only 11% of trials included in their NMA reported any anal-
appendix L) illustrates that the NMA model converges. ysable visual field data. Moreover, the authors mentioned
that since visual field data were reported in many different
ways, the conduct of a pairwise meta-analysis or NMA would
DISCUSSION have been impossible.11 Finally, although our study provided
The objective of this study was to assess the relative efficacy a relative ranking of topical treatments for glaucoma based on
of a new IOP- lowering medication, LBN, compared with IOP reduction at 3 months, the choice of treatment remains
other topical medications for the treatment of POAG and OH a multifactorial decision to take into consideration different
and to provide a relative ranking of these treatments. Find- factors, such as patient’s medical history and preference, risk
ings from this NMA confirm that all drugs are more effective factors and likelihood of compliance.
when compared with the placebo. Importantly, results also
indicate that LBN is significantly more effective than unopro-
stone (PGA) and other non-PGAs drugs except levobunolol CONCLUSION
and travoprost for which LBN is numerically better although Results from the NMA showed that, LBN was significantly
not significant. This demonstrates that LBN is more effective more effective than the PGA unoprostone and most of the
than timolol, which aligns with the conclusion drawn from beta-blockers. Although there was no significant relative effect,
the individual studies (APOLLO and LUNAR).13 14 More- compared with the most widely used PGAs, LBN was numeri-
over, compared with other PGAs, LBN was numerically more cally more efficient than latanoprost and tafluoprost, was similar
effective than tafluoprost, similar to bimatoprost 0.01% and to bimatoprost 0.01% and was slightly disadvantaged by bimato-
slightly disadvantaged by bimatoprost 0.03%. prost 0.03%. LBN could potentially become a promising option
This systematic review was adapted from the one conducted for glaucoma patients.
by Li et al that was previously published in a peer-reviewed
journal.11 A clear research question was formed using the Contributors PH analysed the data and drafted and revised the manuscript for
important intellectual content. CR designed the study, acquired and analysed the
PICO framework and the analysis was conducted based on data, drafted the initial manuscript and reviewed the manuscript. AXC, MB, KJ-G,
the predefined protocol. It should be noted that of the 114 KM, JL and CB designed the study and reviewed the manuscript.
trials eligible in the NMA published by Li et al, 19 were not Funding This research was funded by Bausch Health, Canada Inc.
included in this NMA. This is explained by the fact that our
Competing interests PH has received consultant honoraria from Bausch Health,
systematic review was limited to English or French publica- Canada. CR is an employee of PeriPharm Inc. AXC, MB and KJ-G are employees of
tions, whereas Li et al did not impose any language restric- Bausch Health, Canada. KM is an employee of PeriPharm and Université de Montréal. JL
tion. Moreover, some full-text articles were not accessible via and CB have received research funds from Bausch Health, Canada to conduct this study.

Clinical science
Patient consent for publication Not required. 8 Jansen JP, Fleurence R, Devine B, et al. Interpreting indirect treatment comparisons
and network meta-analysis for health-care decision making: report of the ISPOR Task
Provenance and peer review Not commissioned; externally peer reviewed.
force on indirect treatment comparisons good research practices: Part 1. Value Health
Data availability statement Data are available on reasonable request. 2011;14:417–28.
Supplemental material This content has been supplied by the author(s). It 9 Higher formulation of bimatoprost to be discontinued [press release] 2012.
has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have 10 Rouse B, Cipriani A, Shi Q, et al. Network meta-analysis for clinical practice guidelines:
been peer-reviewed. Any opinions or recommendations discussed are solely those a case study on first-line medical therapies for primary open-angle glaucoma. Ann
of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and Intern Med 2016;164:674–82.
responsibility arising from any reliance placed on the content. Where the content 11 Li T, Lindsley K, Rouse B, et al. Comparative effectiveness of first-line medications
includes any translated material, BMJ does not warrant the accuracy and reliability for primary open-angle glaucoma: a systematic review and network meta-analysis.
of the translations (including but not limited to local regulations, clinical guidelines, Ophthalmology 2016;123:129–40.
terminology, drug names and drug dosages), and is not responsible for any error 12 Nicox’s Partner Secures Additional Approval of VYZULTA® (latanoprostene bunod
and/or omissions arising from translation and adaptation or otherwise. ophthalmic solution), 0.024% in Taiwan [press release] 2020.
Open access This is an open access article distributed in accordance with the 13 Medeiros FA, Martin KR, Peace J, et al. Comparison of Latanoprostene Bunod 0.024%
Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which and Timolol Maleate 0.5% in Open-Angle Glaucoma or Ocular Hypertension: The
permits others to distribute, remix, adapt, build upon this work non-commercially, LUNAR Study. Am J Ophthalmol 2016;168:250–9.
and license their derivative works on different terms, provided the original work is 14 Weinreb RN, Liebmann JM, Martin KR, et al. Latanoprostene Bunod 0.024% in
properly cited, appropriate credit is given, any changes made indicated, and the use Subjects With Open-angle Glaucoma or Ocular Hypertension: Pooled Phase 3 Study
is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/. Findings. J Glaucoma 2018;27:7–15.
15 Weinreb RN, Ong T, Scassellati Sforzolini B, et al. A randomised, controlled comparison
ORCID iD of latanoprostene bunod and latanoprost 0.005% in the treatment of ocular
Jean Lachaine http://orcid.org/0000-0002-0791-2518 hypertension and open angle glaucoma: the VOYAGER study. Br J Ophthalmol
2015;99:738–45.
16 Hutton B, Salanti G, Caldwell DM, et al. The PRISMA extension statement for
REFERENCES reporting of systematic reviews incorporating network meta-analyses of health care
1 Weinreb RN, Aung T, Medeiros FA. The pathophysiology and treatment of glaucoma: a
interventions: checklist and explanations. Ann Intern Med 2015;162:777–84.
review. JAMA 2014;311:1901–11.
17 The Cochrane collaboration. Assessing risk of bias in included studies. In: Cochrane
2 Pascolini D, Mariotti SP. Global estimates of visual impairment: 2010. Br J Ophthalmol
Handbook for Systematic Reviews of Interventions, 2011. http://handbook-5-1.
2012;96:614–8.
3 Harasymowycz P, Birt C, Gooi P, et al. Medical management of glaucoma in the 21st cochrane.org/
century from a Canadian perspective. J Ophthalmol 2016;2016:1–22. 18 Higgins JPT, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses.
4 Tham Y-C, Li X, Wong TY, et al. Global prevalence of glaucoma and projections BMJ 2003;327:557–60.
of glaucoma burden through 2040: a systematic review and meta-analysis. 19 Gert van Valkenhoef JK. Package ’gemtc’, 2016. Available: https://cran.r-project.org/
Ophthalmology 2014;121:2081–90. web/packages/gemtc/gemtc.pdf [Accessed 1 Aug 2018].
5 Chang EE, Goldberg JL. Glaucoma 2.0: neuroprotection, neuroregeneration, 20 Salanti G, Ades AE, Ioannidis JPA. Graphical methods and numerical summaries for
neuroenhancement. Ophthalmology 2012;119:979–86. presenting results from multiple-treatment meta-analysis: an overview and tutorial. J
6 Canadian Ophthalmological Society. Canadian Ophthalmological Society evidence-based Clin Epidemiol 2011;64:163–71.
clinical practice guidelines for the management of glaucoma in the adult eye, 2009. 21 Mbuagbaw L, Rochwerg B, Jaeschke R, et al. Approaches to interpreting and choosing
7 MacIver S, MacDonald D, Prokopich CL. Screening, diagnosis, and management of the best treatments in network meta-analyses. Syst Rev 2017;6:79.
open angle glaucoma: an evidence-based guideline for Canadian optometrists. Can J 22 Medeiros FA. Biomarkers and surrogate endpoints in glaucoma clinical trials. Br J
Optometry 2017;79. Ophthalmol 2015;99:599–603.

640 Full

Uploaded by

Copyright:

Available Formats

640 Full

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

640 Full

Uploaded by

Copyright:

Available Formats

Clinical science

Figure 2 Network Graph. The nodes are weighted according to

Harasymowycz P, et al. Br J Ophthalmol 2022;106:640–647. doi:10.1136/bjophthalmol-2020-317262

Harasymowycz P, et al. Br J Ophthalmol 2022;106:640–647. doi:10.1136/bjophthalmol-2020-317262

646 Harasymowycz P, et al. Br J Ophthalmol 2022;106:640–647. doi:10.1136/bjophthalmol-2020-317262

Harasymowycz P, et al. Br J Ophthalmol 2022;106:640–647. doi:10.1136/bjophthalmol-2020-317262 647

You might also like