Article Text
PDF
Abstract
Background/aims We characterised the relationships between monitoring frequency, ranibizumab injection need and vision in patients receiving as-needed (pro re nata; PRN) ranibizumab for macular oedema due to branch retinal vein occlusion (BRVO) or central retinal vein occlusion (CRVO) in this post-hoc analysis of SHORE and HORIZON.
Methods Patients aged 18 years and older with macular oedema due to BRVO/CRVO were included in this analysis. Injection frequency and best-corrected visual acuity (BCVA) were evaluated by PRN injection frequency in the PRN dosing phase (months (M) 7–15) of SHORE and through 12 months of HORIZON. Prespecified PRN re-treatment criteria for each trial were based on protocol-prespecified BCVA and optical coherence tomography outcomes.
Results After the initial 7 monthly ranibizumab injections, patients in SHORE gained a mean of 18.3 letters from baseline. Patients randomised to PRN, on average, maintained these gains. However, some patients experienced additional mean gains, whereas others suffered losses (range 4.0 (95% CI 0.7 to 7.3) to −4.6 (95% CI −11.8 to 2.6) letters in patients who received 0 and 6–7 PRN injections, respectively). In BRAVO and CRUISE (lead-in trials), patients experienced mean gains from baseline to M6 (monthly dosing) of 19.3 and 15.0 letters, respectively, with gains maintained with PRN from M6 to M12. However, mean BCVA changes from baseline to M12 varied in HORIZON (range −0.4 (95% CI −2.5 to 1.6) to −3.6 (95% CI −6.2 to −1.0) letters in patients who received zero and six injections, respectively, during the preceding PRN phase of BRAVO and CRUISE).
Conclusion The BRVO/CRVO population is heterogenous with a varied response to ranibizumab treatment.
- Retina
- Angiogenesis
- Clinical Trial
- Macula
- Treatment Medical
Data availability statement
Data are available upon reasonable request. For eligible studies, qualified researchers may request access to individual patient-level clinical data through a data request platform. At the time of writing, this request platform is Vivli. https://vivli.org/ourmember/roche/. For up-to-date details on Roche’s Global Policy on the Sharing of Clinical Information and how to request access to related clinical study documents, see here: https://go.roche.com/data_sharing. Anonymised records for individual patients across more than one data source external to Roche cannot, and should not, be linked due to a potential increase in risk of patient re-identification.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Despite the efficacy achieved with monthly office visits and optimally timed intravitreal anti-vascular endothelial growth factor injections in clinical trials, retinal vein occlusion (RVO) data from real-world studies suggest that this treatment paradigm might not be feasible outside of a trial setting as the need for frequent monthly monitoring is burdensome for patients, caregivers, physicians and the healthcare system.
WHAT THIS STUDY ADDS
Findings in this study suggest prognoses may be better in patients who require fewer pro re nata injections and worse in those who require more injections.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Our findings suggest that there are unmet needs in management of patients with RVO; novel agents with improved durability, sustained drug delivery approach or new mechanism of action may be needed to enable reduction of monitoring frequency and potentially attain optimal vision outcomes across all patients.
Introduction
American Academy of Ophthalmology Preferred Practice Pattern guidelines recommend intravitreal anti–vascular endothelial growth factor (VEGF) monotherapy as the first-line strategy to manage macular oedema due to retinal vein occlusion (RVO).1 Specifically, monthly injections of ranibizumab 0.5 mg and aflibercept 2.0 mg are US Food and Drug Administration approved for treatment of patients with macular oedema due to RVO.2 3 In the European Union and many other countries and regions, as-needed (pro re nata; PRN) and/or treat-and-extend (TAE) regimens have been approved, with monitoring and treatment intervals dependent on visual/anatomical signs of disease activity.4–9
Safety and efficacy of anti-VEGF therapy for macular oedema due to RVO were first demonstrated in two pivotal clinical trials of ranibizumab in patients with branch (BRVO; BRAVO10 11) and central (CRVO; CRUISE) retinal vein occlusion.12 13 Both trials reported superior best-corrected visual acuity (BCVA) improvements with monthly ranibizumab versus sham at month 6 from baseline,10 12 with vision gains maintained through 1 year with monthly monitoring visits and PRN ranibizumab.11 13 Patients receiving sham injections for the first 6 months crossed over to PRN ranibizumab at month 6. Patients who completed 12 months of BRAVO/CRUISE were eligible to enter the HORIZON open-label extension (OLE) trial, during which they were monitored at least every 3 months and received PRN ranibizumab.14 Vision outcomes remained stable among patients with BRVO; however, in patients with CRVO, less-frequent follow-up, and, subsequently, fewer PRN injections, were associated with vision decline. The phase IV open-label SHORE trial compared monthly and PRN ranibizumab in patients who achieved stable disease after ≥7 monthly injections.15 Patients who achieved disease stability were randomised to monthly or PRN ranibizumab. Patients not achieving disease stability continued monthly ranibizumab. Among randomised patients, mean vision gains achieved at month 7 were maintained through month 15, with statistical non-inferiority achieved between monthly and PRN ranibizumab.
Despite the efficacy achieved with monthly office visits and optimally timed intravitreal anti-VEGF injections in clinical trials, RVO data from real-world studies16 suggest that this treatment paradigm might not be feasible outside of a trial setting. The need for frequent monthly monitoring is burdensome for patients, caregivers, physicians and the healthcare system. In a questionnaire of patients receiving intravitreal injection therapy for RVO or diabetic macular oedema, a higher proportion attributed their reduced quality of life to their retinal disease than other comorbidities.17 Moreover, when asked to recommend improvements to their current treatment, patients desired fewer injections (42% of those surveyed) and fewer appointments (22%).
TAE regimens may lessen treatment burden while optimising visual and anatomical outcomes; however, this strategy has not been extensively studied in patients with RVO. The SCORE2 trial investigated efficacy of monthly aflibercept versus monthly bevacizumab for macular oedema due to CRVO/hemiretinal RVO at month 6,18 followed by re-randomisation of patients who achieved disease stability at month 6 to compare efficacy of aflibercept and bevacizumab dosed either monthly or TAE from month 6 to month 12.19 The second year of SCORE2 assessed efficacy outcomes in patients treated with anti-VEGF agents per investigator discretion.20 On average, vision gains achieved at month 6 were maintained through month 12, with no significant difference between the TAE and monthly treatment groups or between the two anti-VEGF agents19; however, patients consequently lost vision with less-frequent monitoring and fewer injections during the per-investigator-discretion follow-up in the second year of the study.20 Our post-hoc analyses aimed to characterise relationships between monitoring frequency, intravitreal anti-VEGF injection need and vision outcomes during the PRN treatment phases of SHORE, BRAVO, CRUISE and HORIZON.
Materials and methods
Data source
This was a post-hoc analysis of the SHORE (ClinicalTrials.gov identifier, NCT01277302), BRAVO (NCT00486018), CRUISE (NCT00485836) and HORIZON (NCT00379795) clinical trials of intravitreal ranibizumab therapy in patients with macular oedema due to BRVO and/or CRVO. Full methods and primary outcomes for each trial have been described previously.10 12 14 15 Each trial was conducted in accordance with the International Conference on Harmonisation E6 Guideline for Good Clinical Practice (SHORE, BRAVO and CRUISE), Declaration of Helsinki (HORIZON) and applicable national requirements (all studies). Study protocols were approved by institutional review boards or ethics committees at each study site as applicable, and complied with the Health Insurance Portability and Accountability Act as applicable. All patients provided written informed consent to participate.
Online supplemental table 1 summarises key aspects of the SHORE, BRAVO, CRUISE and HORIZON trial designs. Briefly, SHORE was a 15-month study where patients could be randomised to monthly or PRN ranibizumab 0.5 mg only after achieving stable disease starting at the month 7 visit onward, which followed a period of monthly (every 4 weeks) ranibizumab injections from day 0 until month 6 (seven initial injections) in all patients. BRAVO and CRUISE were 12-month trials with patients randomised to monthly ranibizumab 0.3 mg, monthly ranibizumab 0.5 mg or monthly sham. After the initial 6 monthly injections from day 0 to month 5, patients receiving ranibizumab switched to PRN ranibizumab at their assigned dose, whereas sham-treated patients crossed over to PRN ranibizumab 0.5 mg starting at month 6. HORIZON was an OLE study of patients who completed BRAVO or CRUISE. Patients in HORIZON received PRN ranibizumab 0.5 mg. Patients in SHORE, BRAVO and CRUISE were monitored monthly whereas patients in HORIZON were monitored at least every 3 months. Prespecified PRN re-treatment criteria for each trial were based on protocol-prespecified BCVA and optical coherence tomography (OCT) outcomes.
Supplemental material
Outcomes
The primary outcome for this post-hoc analysis was the relationship between PRN injection frequency and vision outcomes in the PRN-treated subset in SHORE and HORIZON. In SHORE, we evaluated the mean number of PRN injections received during the second part of the study in patients randomised to PRN ranibizumab. Most patients were randomised to PRN ranibizumab at months 7 and 8 (n=69, 80%). Therefore, to attain at least 7 months of follow-up and reduce variability in length of follow-up, only these patients were included in the current analysis of SHORE (figure 1). The outcome was the proportion of patients who received zero, one to three, four to five or six to seven PRN injections during the PRN period, and 15-month BCVA outcomes for each PRN injection subgroup. In post-hoc analyses of HORIZON, BCVA outcomes over 12 months (mean BCVA change from OLE baseline at months 3, 6, 9 and 12) were evaluated in patients stratified by PRN injection frequency during the 6-month PRN phase of the original core studies, BRAVO or CRUISE (ie, zero, one to three, four to five or six PRN injections from months 6–12 in BRAVO/CRUISE). In patients with sufficient follow-up, mean number of PRN injections received over 12 months in HORIZON were estimated (figure 1), stratified by PRN injection subgroup (zero, one to three, four to five or six to seven PRN injections) in BRAVO or CRUISE.
Treatment periods in the BRAVO and CRUISE phase III trials, the HORIZON OLE and the SHORE phase IV trial. BRVO, branch retinal vein occlusion; CRVO, central retinal vein occlusion; HRVO, hemiretinal vein occlusion; OLE, open-label extension; PRN, as-needed (pro re nata); q4w, every 4 weeks; q12w, every 12 weeks.
Statistical analysis
Analyses were completed using observed data, with no imputation for missing values. PRN injection frequencies and BCVA outcomes are summarised using descriptive statistics. Data from HORIZON are reported overall and by BRVO and CRVO subgroups.
Results
Baseline demographics and clinical characteristics
In SHORE, 171 patients with BRVO or CRVO were randomised to monthly ranibizumab 0.5 mg (n=85) or PRN ranibizumab 0.5 mg (n=86) after achieving disease stability on or after month 7, whereas 31 patients who never reached disease stability were never randomised and continued monthly ranibizumab treatment until study end. Among patients who received PRN ranibizumab in SHORE, 69 were randomised at months 7 and 8 and included in the present analysis.
HORIZON included 304 patients with BRVO from BRAVO and 304 patients with CRVO from CRUISE. Baseline patient characteristics were similar between HORIZON and SHORE. On average, baseline BCVA was higher for patients in HORIZON versus SHORE (59.8–73.6 Early Treatment Diabetic Retinopathy Study (ETDRS) letters vs 47.7–54.9 ETDRS letters across trial subgroups). SHORE used spectral-domain OCT to measure central subfield thickness, whereas HORIZON used time-domain OCT; therefore, central subfield thickness data are not directly comparable between the two studies.
PRN injection frequency and vision outcomes in SHORE
Increasing injection need during the variable-length PRN phase of SHORE was associated with worsening vision outcomes over 15 months (figure 2). Patients who achieved disease stability and were randomised to PRN ranibizumab 0.5 mg at months 7 or 8 received an average of 3.2 PRN injections through month 15 (figure 2A).
(A) Number of ranibizumab injections required during the PRN phase of SHORE, and mean (95% CI) BCVA change (B) from baseline and (C) from randomisation over 15 months, stratified by PRN injection frequency. Observed data from patients randomised to PRN ranibizumab at (A) M7 or M8 of SHORE (n=69), and (B, C) those with evaluable BCVA data at M15 (n=65). Error bars represent 95% CIs. BCVA, best-corrected visual acuity; ETDRS, Early Treatment Diabetic Retinopathy Study; M, month; PRN, as-needed (pro re nata).
Baseline mean (95% CI) BCVA was 52.5 (35.9 to 69.1), 54.0 (48.1 to 59.9), 52.5 (47.7 to 57.4) and 60.6 (47.6 to 73.6) letters in patients who required zero, one to three, four to five or six to seven injections during the PRN phase of SHORE, respectively. Initial mean (95% CI) BCVA gains with monthly ranibizumab treatment from day 0 until randomisation at months 7 or 8 are shown in figure 2B. Patients who required zero injections during the PRN phase of the studies had the best vision outcomes during the initial monthly dosing period, but patients in the other groups still demonstrated substantial vision gains.
After switching to PRN treatment, patients who received zero PRN injections (6/69 patients) experienced further mean BCVA gain, whereas those who required six to seven PRN injections through study end (5/69 patients) experienced average BCVA loss. The remaining majority of patients maintained their initial BCVA gains (figure 2C).
PRN injection frequency and vision outcomes in HORIZON
Post-hoc analyses of HORIZON also revealed heterogeneity of patients’ BCVA outcomes when assessed by injection frequency. There was an association between worse vision outcomes and increased injection frequency in analyses of BCVA change from HORIZON baseline to month 12 based on injection frequency during the preceding 6-month PRN phase of BRAVO/CRUISE (figures 3–5). In the pooled RVO cohort, initial mean (95% CI) BCVA gains from day 0 to month 6 of BRAVO/CRUISE are shown in figure 3A. Patients who required zero injections during the PRN phase of BRAVO/CRUISE had the best vision outcomes during the initial monthly dosing period, but patients in the other groups still demonstrated vision gains throughout BRAVO/CRUISE. Patients on average received fewer injections with less-frequent monitoring during the 12-month HORIZON trial than during the preceding 6-month PRN phase of BRAVO/CRUISE (figure 3B). There was considerable heterogeneity among patients with 12-month follow-up during HORIZON. Patients (n=28) who received no PRN injections during the 6-month PRN phase of BRAVO/CRUISE received, on average, 0.3 injections (95% CI −0.03 to 0.6) during 12 months of HORIZON; 23 of 28 required no injections, 4 of 28 required one injection and 1 of 28 required four injections. This subset of patients also had the highest BCVA at HORIZON baseline (mean BCVA 79.8 (95% CI 75.8 to 83.8) letters). In contrast, patients who required monthly PRN injections throughout the core trials received, on average, 3.7 injections (95% CI 3.1 to 4.3) in HORIZON and had the lowest BCVA at HORIZON baseline (mean BCVA 58.0 (95% CI 55.5 to 60.6) letters). At the start of HORIZON, patients who required one to three PRN injections during the 6-month PRN phase of BRAVO/CRUISE had a mean BCVA of 75.5 letters (95% CI 73.6 to 77.4). At the start of HORIZON, patients who needed four to five PRN injections during BRAVO/CRUISE had a mean BCVA of 63.1 letters (95% CI 60.4 to 65.7).
(A) Mean (95% CI) BCVA change from baseline to month 6 and end of BRAVO/CRUISE and (B) mean PRN injection frequencies and (C) BCVA outcomes over 12 months in HORIZON, stratified by PRN injection frequency during BRAVO/CRUISE (pooled retinal vein occlusion cohort). Observed data from patients who completed BRAVO/CRUISE (all treatment arms pooled) and had sufficient follow-up at each HORIZON time point. Error bars represent 95% CIs. BCVA, best-corrected visual acuity; ETDRS, Early Treatment Diabetic Retinopathy Study; PRN, as-needed (pro re nata).
(A) Mean (95% CI) BCVA change from baseline to month 6 and end of BRAVO and (B) mean PRN injection frequencies and (C) BCVA outcomes over 12 months in HORIZON, stratified by PRN injection frequency during BRAVO (branch retinal vein occlusion cohort). Observed data from patients who completed BRAVO (all treatment arms pooled) and had sufficient follow-up at each HORIZON time point. Error bars represent 95% CIs. BCVA, best-corrected visual acuity; ETDRS, Early Treatment Diabetic Retinopathy Study; PRN, as-needed (pro re nata).
(A) Mean (95% CI) BCVA change from baseline to month 6 and end of CRUISE and (B) mean PRN injection frequencies and (C) BCVA outcomes over 12 months in HORIZON, stratified by PRN injection frequency during CRUISE (central retinal vein occlusion cohort). Observed data from patients who completed CRUISE (all treatment arms pooled) and had sufficient follow-up at each HORIZON time point. Error bars represent 95% CIs. BCVA, best-corrected visual acuity; ETDRS, Early Treatment Diabetic Retinopathy Study; PRN, as-needed (pro re nata).
On average, patients who required fewer PRN injections during BRAVO/CRUISE maintained vision over 12 months in HORIZON, whereas those who required frequent PRN injections in BRAVO/CRUISE lost vision with less-frequent monitoring during HORIZON (figure 3C).
Patients with CRVO tended to require more injections and had worse vision outcomes than patients with BRVO (figures 4 and 5). On average, patients with BRVO who required zero and one to three PRN injections during BRAVO maintained vision over this period (figure 4). Conversely, patients who required four to five or six (ie, monthly) PRN injections during BRAVO rapidly lost vision on entering HORIZON.
In patients with CRVO, mean number of PRN injections received over 12 months in HORIZON varied from zero injections among those who previously required no PRN injections during CRUISE to 4.3 injections (95% CI 3.5 to 5.2) in patients who required 6 monthly injections (figure 5). At month 12 of HORIZON, mean (95% CI) change in BCVA from HORIZON baseline was −4.3 (−18.0 to 9.3), −2.7 (−5.5 to 0.0), −4.5 (−7.7 to −1.4) and −6.6 (−10.7 to −2.4) ETDRS letters in patients who required zero, one to three, four to five or six PRN injections, respectively, during CRUISE.
Discussion
In these post-hoc analyses of PRN ranibizumab in SHORE and HORIZON, a high degree of heterogeneity was found among patients with macular oedema due to RVO in terms of disease activity, frequency of treatments and visual outcomes. Comparison of management courses over 15–24 months showed that there was a group of patients who had no, or few, episodes of disease activity requiring PRN re-treatment. In contrast, another group of patients did not achieve macular oedema resolution over the study periods, or had frequent recurrences, and continued to require monthly or near-monthly PRN injections. Most patients fell between these two groups, receiving a moderate number of re-treatments once they entered the PRN phase of management.
In SHORE,15 15.3% (31/202) of patients with BRVO/CRVO failed to achieve full resolution of macular oedema at any point and required monthly treatments over the entire 15-month study period or until early discontinuation. Most patients did achieve resolution of macular oedema; of these, few patients required no further PRN re-treatment over the ensuing 7–8 months, and a few patients displayed continuing disease recurrences requiring resumption of monthly PRN injections until the end of the 15-month study. However, the largest group of patients in SHORE included those who required individualised dosing at intermediate frequency, with 40.6% of randomised patients requiring one to three injections over the 7-month to 8-month phase of the PRN protocol and 43.5% requiring four to five injections. These data suggest that most patients with macular oedema due to RVO may be appropriately managed with a less-than-monthly PRN treatment, providing they undergo monthly disease activity assessments at the clinic.
HORIZON study findings suggest that patients with BRVO and CRVO who required frequent re-treatments in the initial PRN phase of BRAVO/CRUISE subsequently also required frequent re-treatments over the long term. Specifically, patients with high injection need in BRAVO/CRUISE also received a higher frequency of PRN injections over the 12-month extension in HORIZON, although at a reduced rate versus BRAVO/CRUISE, which may have been a consequence of less-than-monthly monitoring in HORIZON, rather than a true reduction in intravitreal anti-VEGF injection need. In contrast, patients who required no PRN re-treatments during BRAVO/CRUISE continued, on average, to receive few, if any, injections over the ensuing 12 months in HORIZON, and generally tended to maintain vision gains achieved at the end of BRAVO/CRUISE.
In HORIZON, patients with a higher need for treatment tended to have higher mean vision losses, potentially implicating undertreatment due to less-than-monthly disease activity assessment visits (only quarterly monitoring visits were required in HORIZON). Similarly, in SHORE, despite monthly monitoring throughout the 15-month study period, patients with highest need (requiring monthly or near-monthly PRN injections) trended, on average, toward losing vision. These findings were not unexpected because receiving more frequent PRN injections tended to reflect persistent or frequently recurrent disease activity, leading to greater vision declines despite treatment versus patients who achieved stable disease control requiring few or no PRN injections. Furthermore, these data suggest that some patients with RVO were unable to achieve disease resolution or maintain clinically significant vision gains with maximal intravitreal anti-VEGF monotherapy, and therefore may require additional or novel therapies with mode of action beyond VEGF inhibition alone.
Injection needs and vision outcomes tended to be more pronounced for patients with macular oedema due to CRVO versus BRVO, probably because patients with CRVO in HORIZON suffer more extensive disease with risk of serious complications, such as retinal ischaemia or retinal neovascularisation, and therefore trended towards greater vision loss and higher need for PRN injections versus patients with BRVO.14 This finding suggests that patients with CRVO may be even more susceptible to ongoing disease activity, may require more frequent monitoring and re-treatment and may be at risk for greater vision declines under PRN management. Further research is needed to evaluate these possibilities.
Findings in this study suggest prognoses may be better in patients who require fewer PRN injections and worse in those who require more injections. However, other factors may also impact prognosis. For example, nature (subretinal vs intraretinal fluid), location (fovea involving vs extrafoveal), ischaemia presence and recurrent exudation severity may have additional prognostic value regarding both injection frequency and vision outcomes. Some initial work on this has been completed, but these analyses only focused on potential predictive OCT biomarkers of visual outcomes in SHORE patients.21 It would also be useful to determine whether baseline aqueous VEGF levels are predictive as suggested by findings of a previously published study.22 However, this would require a new study to enable collection of aqueous humour samples.
These post-hoc analyses are subject to inherent limitations. SHORE, BRAVO, CRUISE and HORIZON were not powered to evaluate subgroups by PRN injection frequency; therefore, our analyses are descriptive only. The four studies incorporated into the post-hoc analyses were not uniform in study design, patient population, ranibizumab dose concentrations and visit frequencies for monitoring patients. Particularly, the HORIZON trial design required only quarterly monitoring visits; by the time SHORE was conducted, the study design required monthly monitoring visits throughout the PRN phase of the study, reflecting further clinical experience and study results.23 Furthermore, the monthly monitoring visits throughout the PRN phase of SHORE may explain the smaller proportion of patients with vision loss compared with HORIZON.
Collectively, results from SHORE and HORIZON suggest that the RVO patient population is heterogenous, with wide differences in ranibizumab treatment response and macular oedema disease activity. Our data support PRN monthly disease assessment criteria, which can lead to correct identification of patients who require higher frequency of injections. As expected, we showed that reduced need for PRN injections was associated with better vision outcomes and that there are patients who do not reach disease quiescence despite continued monthly treatments. Our results also show that some patients experienced vision loss when PRN dosing was supported with less-than-monthly disease assessment visits in the OLE study. Currently, no reliable predictive biomarker tools exist in clinical practice to inform at the start of treatment whether the patient will need a high or low number of injections. To ensure ideal outcomes with intravitreal anti-VEGF monotherapy, patients with RVO need to be monitored frequently, which in real-world practice is burdensome and detrimental to patient quality of life.17 Together, the current analyses suggest that there are unmet needs in management of patients with RVO; novel agents with improved durability, sustained drug delivery approach or new mechanism of action may be needed to enable reduction of monitoring frequency and potentially attain optimal vision outcomes across all patients.
Data availability statement
Data are available upon reasonable request. For eligible studies, qualified researchers may request access to individual patient-level clinical data through a data request platform. At the time of writing, this request platform is Vivli. https://vivli.org/ourmember/roche/. For up-to-date details on Roche’s Global Policy on the Sharing of Clinical Information and how to request access to related clinical study documents, see here: https://go.roche.com/data_sharing. Anonymised records for individual patients across more than one data source external to Roche cannot, and should not, be linked due to a potential increase in risk of patient re-identification.
Ethics statements
Patient consent for publication
Ethics approval
Each trial was conducted in accordance with the International Conference on Harmonisation E6 Guideline for Good Clinical Practice (SHORE, BRAVO and CRUISE), Declaration of Helsinki (HORIZON) and applicable national requirements (all studies). Study protocols were approved by institutional review boards or ethics committees at each study site as applicable, complied with the Health Insurance Portability and Accountability Act as applicable. All patients provided written informed consent to participate. SHORE study institutional review boards (IRBs), chairs and addresses were as follows: University of Chicago; Chair, Christopher Daugherty, MD; 5841 S. Maryland Avenue, MR 5030, Chicago, IL 60637, USA; Sterling IRB; Chair, Steven L. Saltzman, MD; 6300 Powers Ferry Road Suite 600-351, Atlanta, GA 30339, USA. BRAVO study IRBs, chairs and addresses were as follows: Biological Science Division/University of Chicago Hospitals Institutional Review Board; Chair; Christopher Daugherty; 5841 S. Maryland Avenue, AMB S-152 MC1108, Chicago, IL 60637, USA; Copernicus Group Institutional Review Board; Chair, Glenn C. Veit; One Triangle Drive, Suite 100, PO Box 110605, Research Triangle Park, NC 27709, USA; Emory University Institutional Review Board; 1599 Clifton Road NE, Atlanta, GA 30322, USA; Johns Hopkins Institutional Review Board; Chair, Richard Moore; Reed Hall B-130, 1620 McElderry Street, Baltimore, MD 21205, USA; Kaiser Permanente Southern California Institutional Review Board; 393 E. Walnut Street, 2nd Floor, Pasadena, CA 91183. USA; Lahey Clinic Institutional Review Board; Chair, Sarkis Soukiasian; 41 Mall Road, Burlington, MA 01805-0002, USA; Massachusetts Eye and Ear Infirmary Human Studies Committee; 243 Charles Street, 12th floor, Boston, MA 02114, USA; Mayo Clinic Institutional Review Board; Chair, William J. Tremaine; 200 First Street SW, 201 Building, Room 4-60, Rochester, MN 55905, USA; MedStar Research Institute Institutional Review Board; Chair, Charles MacKay; 6495 New Hampshire Avenue, Suite 201, Hyattsville, MD 20783, USA; Porter and Littleton Adventist Hospitals Joint Institutional Review Board; Chair, Ben Vernon; 2525 S. Downing Street, Denver, CO 80210, USA; Scott and White Memorial Hospital, and Scott, Sherwood, and Brindley Foundation Institutional Review Board; 2401 South 31st Street Temple, TX 76508, USA; University of Missouri-Kansas City Adult Health Sciences Adult Health Sciences Institutional Review Board; Chair, Julie Wright; 5319 Rockhill Road, Kansas City, MO 64108, USA; University of California, San Francisco Committee on Human Research; Chair, Susan Sniderman; Office of Research, 3333 California Street, Suite 315, San Francisco, CA 94118, USA; University of Texas Southwestern Medical Center Institutional Review Board; Chair, Jody L. Jensen; 5323 Harry Hines Boulevard, C1 206, Dallas, TX 75390-8843, USA; Washington University School of Medicine, Human Research Protection Office; Chair, Philip Ludbrook; 660 S. Euclid Avenue, Campus Box 8089, St. Louis, MO 63110, USA; Western Institutional Review Board; Chair, Theodore D. Schultz; 3535 Seventh Avenue SW, Olympia, WA 98502-5010, USA; Wills Eye Hospital Institutional Review Board; Chair, Ralph C. Eagle; 840 Walnut Street, 15th Floor, Philadelphia, PA 19107, USA. CRUISE study IRBs, chairs and addresses were as follows: Biological Science Division/University of Chicago Hospitals Institutional Review Board; Chair, Christopher Daugherty; 5841 S. Maryland Avenue, AMB S-152 MC1108, Chicago, IL 60637, USA; Cleveland Clinic Foundation Institutional Review Board; Chair, Alan Lichtin; 9500 Euclid Avenue, A16, Cleveland, OH 44195, USA; Copernicus Group Institutional Review Board; Chair, Glenn C. Veit; One Triangle Drive Suite 100, PO Box 110605, Research Triangle Park, NC 27709, USA; Emory University Institutional Review Board; 1599 Clifton Road NE, Atlanta, GA 30322, USA; Henry Ford Health System Institutional Review Board; 2799 West Grand Boulevard K10, Detroit, MI 48202, USA; Johns Hopkins Institutional Review Board; Chair, Richard Moore; Reed Hall B-130. 1620 McElderry Street, Baltimore, MD 21205, USA; Kaiser Permanente Southern California Institutional Review Board; 393 E. Walnut Street, 2nd Floor, Pasadena, CA 91183, USA; Lahey Clinic Institutional Review Board; Chair, Sarkis Soukiasian; 41 Mall Road, Burlington, MA 01805-0002, USA; Loma Linda University Adventist Health Institutional Review Board; Chair, Rhodes L. Rigsby; 11188 Anderson Street, Loma Linda, CA 92354, USA; Massachusetts Eye and Ear Infirmary Human Studies Committee; 243 Charles Street, 12th floor, Boston, MA 02114, USA; Mayo Clinic Institutional Review Board; Chair, William J. Tremaine; 200 First Street SW, 201 Building, Room 4-60, Rochester, MN 55905, USA; MedStar Research Institute Institutional Review Board; Chair, Charles MacKay; 6495 New Hampshire Avenue, Suite 201, Hyattsville, MD 20783, USA; Porter and Littleton Adventist Hospitals Joint Institutional Review Board; Chair, Ben Vernon; 2525 S. Downing Street, Denver, CO 80210, USA. HORIZON study IRBs, chairs and addresses were as follows: Western Institutional Review Board; Chair, Theodore D. Schultz; 3535 Seventh Avenue SW, Olympia, WA 98502-5010, USA; Spectrum Health Research and Human Rights Committee; Chair, Jeffrey Jones, MD; 100 Michigan NE MC 38, Grand Rapids, MI 49503, USA; University of Louisville, Human Subjects Protection Program Office; Chair, Laura Clark, MD; MedCenter One 501 E. Broadway, Suite 200, Louisville, KY 40202, USA; University of California, San Francisco, Committee on Human Research; Chair; Reese T. Jones, MD; Office of Research 3333 California Street, Suite 315 University of California, San Francisco, CA 94118, USA; University of Iowa, Institutional Review Board; Chair, J. Andrew Bertolatus, MD; Human Subjects Office, Office of the Vice President for Research, Hardin Library for the Health Sciences, Office 105, 600 Newton Road Iowa City, IA 52242, USA; Medical College of Wisconsin, Institutional Review Board Office; Chair, Ryan Spellecy, Phd; Medical College of Wisconsin, Milwaukee, WI 53226, USA; Duke University Health System, Institutional Review Board; Chair, John Harrelson, MD; Hock Plaza, Suite 405, 2424 Erwin Road, Campus Box 2712, Durham, NC 27705, USA; California Pacific Medical Center Institutional Review Board; Chair, David Busch, MD; 2200 Webster Street, 5th Floor, Pacific Campus, San Francisco, CA 94115, USA; New York Eye and Ear Infirmary Institutional Review Board, Chair, Joseph Walsh, MD; 310 East. 14th Street, South Building, 6th Floor, New York, NY 10003, USA; University of Michigan Medical School IRBMED; Chair, John Weg, MD; 2800 Plymouth Road, Building 200, Room 2086, Ann Arbor, MI 48109, USA; University of San Diego, Human Research Protections Program; Chair, Michael Caligiuri, PhD; 9500 Gilman Drive, Mail Code 0052, La Jolla, CA 92093, USA; Cleveland Clinic Foundation Institutional Review Board; Chair, Alan Lichtin, MD; 9500 Euclid Avenue, HSB-103, Cleveland, OH 44195, USA; University of Oklahoma Health Sciences Center Office for Human Research Participant Protection, Chair, Karen Beckman, MD; Evans Hall, Room 316, 660 Parrington Oval, Norman, OK 73019, USA; Hawaii Pacific Health Institutional Review Board; Chair, David T. Horio, MD; 1100 Ward Avenue, Suite 1045, Honolulu, HI 96814, USA; University of California, Irvine Human Research Protections Program, Chair, Robert A. Burger, MD; Office or Research Administration, University of California, Irvine 300 University Tower, Irvine, CA 92697, USA; University of California, Davis, Institutional Review Board; Chair, David Asmuth, MD; 2921 Stockton Boulevard Suite 1400, Room 1429, Sacramento, CA 95817, USA; University of Arizona, Human Subjects Protection Program; Chair, David G. Johnson, MD; 1618 E. Helen Street, The University of Arizona, P.O. Box 245137, Tucson, AZ 85724, USA; University of South Florida; Chair, Barry Bercu, MD; 3702 Spectrum Boulevard, Suite 155, Tampa, FL 33612, USA; Tufts Medical Center and Tufts University Health Sciences; Chair, David Chelmow, MD; Tufts Medical Center/Tufts University Institutional Review Board Office, 800 Washington Street, Box 817, Boston, MA 02111, USA; University of Southern California Institutional Review Board; Chair, Darcy Spicer, MD; Health Sciences Campus General Hospital, Suite 4700, 1200 North State Street, Los Angeles, CA 90033, USA; Stanford University Administrative Panel on Human Subjects in Medical Research; Chair, Ronald Ariagno, MD; Research Compliance Office, 1501 S. California Avenue, Mail Code: 5579, Palo Alto, CA 94304, USA; William Beaumont Hospital Human Investigation Committee; Chair, Phillip Bendick, PhD; 3811 West Thirteen Mile Road, Royal Oak, MI 48073, USA; University of Utah Institutional Review Board; Chair, Dennis O’Rourke, PhD; Research Administration Building, 75 South 2000 East, Salt Lake City, UT 84112, USA; Wills Eye Hospital Institutional Review Board; Chair, Ralph C. Eagle, MD; Administrative Offices, 840 Walnut Street, 15th Floor, Philadelphia, PA 19107, USA; COAST Institutional Review Board, LLC; Chair, Melissa Cortes, MEd, ET/P; 5475 Mark Dabling Boulevard, Suite 351, Colorado Springs, CO 80918, USA. Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors All authors made substantial contributions to the conception or design of the manuscript; or the acquisition, analysis or interpretation of data for the manuscript; AND drafting of the manuscript/critically reviewing for important intellectual content; AND gave final approval of the version to be published; AND agree to be accountable for all aspects of the manuscript in ensuring that questions related to the accuracy or integrity of any part of the manuscript are appropriately investigated and resolved.
Funding Funding was provided by Genentech, Inc., a member of the Roche Group, for the study and third-party writing assistance, provided by Karina D Hamilton-Peel, PhD, CMPP, and Trishan Gajanand, PhD, of Envision Pharma Group. The sponsor participated in the design of the studies; collection, management, analysis and interpretation of the data; and preparation, review, approval and decision to submit this manuscript.
Competing interests RBB is a consultant for Rezolute, Ribomic, Unity Bio and Visgenx; receives financial support from Genentech, Inc. and NGM Bio; and holds a patent or personal financial interest in Oculinea. PAC is an advisory board member for Aerpio, Allegro, Applied Genetic Technologies Corporation, Exonate, Merck and Roche/Genentech, Inc.; co-founder of Graybug Vision; consultant for AsclepiX, Bausch + Lomb, CureVac, Graybug Vision, Novartis, Perfuse Therapeutics and Wave Life Sciences; and investigator for Aerpio, Oxford Biomedica, Regeneron, Regenxbio, Roche/Genentech, Inc. and Sanofi Genzyme. SB, CQ-R and ZH are employees of Genentech, Inc. ML indicates no conflicts of interest.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.