Port Delivery System With Ranibizumab vs Monitoring in Nonproliferative Diabetic Retinopathy Without Macular Edema: The Pavilion Randomized Clinical Trial

Dante J Pieramici; Carl C Awh; Margaret Chang; Andres Emanuelli; Nancy M Holekamp; Allen Y Hu; Ivan J Suñer; Charles C Wykoff; Christopher Brittain; Dena Howard; Carlos Quezada-Ruiz; Anjana Santhanakrishnan; Paul Latkany

doi:10.1001/jamaophthalmol.2025.0001

. 2025 Mar 6;143(4):317–325. doi: 10.1001/jamaophthalmol.2025.0001

Port Delivery System With Ranibizumab vs Monitoring in Nonproliferative Diabetic Retinopathy Without Macular Edema

The Pavilion Randomized Clinical Trial

Dante J Pieramici ^1,^✉, Carl C Awh ², Margaret Chang ³, Andres Emanuelli ^4,⁵, Nancy M Holekamp ⁶, Allen Y Hu ⁷, Ivan J Suñer ⁸, Charles C Wykoff ⁹, Christopher Brittain ¹⁰, Dena Howard ¹¹, Carlos Quezada-Ruiz ^10,^12,¹³, Anjana Santhanakrishnan ¹⁰, Paul Latkany ¹⁰

¹California Retina Consultants, Santa Barbara

²Tennessee Retina, Nashville

³Retinal Consultants Medical Group, Sacramento, California

⁴Emanuelli Research and Development, Retina Care, Arecibo, Puerto Rico

⁵University of Puerto Rico, School of Medicine, Department of Ophthalmology, San Juan, Puerto Rico

⁶F. Hoffmann-La Roche Ltd, Basel, Switzerland

⁷Cumberland Valley Retina Consultants, Hagerstown, Maryland

⁸Retina Associates of Florida, Tampa

⁹Retina Consultants of Texas, Retina Consultants of America, Blanton Eye Institute, Houston Methodist Hospital, Houston

¹⁰Genentech, Inc, South San Francisco, California

¹¹Roche Products Ltd, Welwyn Garden City, United Kingdom

¹²Clínica de Ojos Garza Viejo, San Pedro Garza García, Nuevo León, Mexico

¹³Retina Department, Instituto de Oftalmologia Fundación Conde de Valenciana, Mexico City, Mexico

Accepted for Publication: December 6, 2024.

Published Online: March 6, 2025. doi:10.1001/jamaophthalmol.2025.0001

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License, which does not permit alteration or commercial use, including those for text and data mining, AI training, and similar technologies. © 2025 Pieramici DJ et al. JAMA Ophthalmology.

^✉

Corresponding Author: Dante J. Pieramici, MD, California Retina Consultants, 525 E Micheltorena St, Santa Barbara, CA 93103 (dpieramici@yahoo.com).

Author Contributions: Dr Pieramici had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Pieramici, Hu, Brittain, Howard, Quezada-Ruiz, Latkany.

Acquisition, analysis, or interpretation of data: Pieramici, Awh, Chang, Emanuelli, Holekamp, Hu, Suñer, Wykoff, Howard, Quezada-Ruiz, Santhanakrishnan, Latkany.

Drafting of the manuscript: Pieramici, Chang, Emanuelli, Hu, Suñer, Brittain, Howard, Quezada-Ruiz, Latkany.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Suñer, Howard, Quezada-Ruiz.

Administrative, technical, or material support: Emanuelli, Hu, Wykoff, Quezada-Ruiz.

Supervision: Pieramici, Awh, Emanuelli, Holekamp, Hu, Wykoff, Brittain, Quezada-Ruiz, Latkany.

Conflict of Interest Disclosures: Dr Pieramici reported receiving grants and personal fees from Genentech/Roche during the conduct of the study; and serving as a consultant for Adverum, Gemini, Genentech, Iveric Bio, NGM Bio, Regeneron, and Regenxbio, receiving research support from Adverum, Apellis, Chengdu Kanghong, F. Hoffmann-La Roche, Genentech, Iveric Bio, Kodiak Sciences, Novartis, Ocular Therapeutix, Regeneron, and Unity, serving as a steering committee member for Genentech, and having equity in Gemini outside the submitted work. Dr Awh reported receiving grants, personal fees, and nonfinancial support from Genentech during the conduct of the study; and receiving grants from Regeneron, serving as a consultant for Bausch and Lomb and Genentech, serving as an investigator for 4D Molecular Therapeutics, Adverum, Allergan, Apellis, Astellas/AIRM, Aura, Aviceda, Chengdu Kanghong, Cognition, Eluminex, Genentech, Ionis, Kodiak Sciences, Mylan, NGM Bio, Ocular Therapeutix, Oxurion, Regeneron, Regenxbio, and Stealth, receiving royalties from Bausch and Lomb and FCI Ophthalmics/Katalyst Surgical, serving as a speaker for Genentech, and having stock in ArcticDX outside the submitted work. Dr Chang reported receiving personal fees and nonfinancial support from Genentech/Roche, AbbVie/Regenxbio, Ocular Therapeutix, and Opthea, personal fees from EyePoint, and nonfinancial support from Outlook and Kodiak Sciences during the conduct of the study; and receiving personal fees and nonfinancial support from Astellas and Oculis, receiving personal fees from Zeiss, receiving nonfinancial support from Novartis, NGM Bio, OcuTerra, Alexion, Mylan, and Boehringer Ingelheim, and receiving grants or research support from Alexion, EyeBio, NGM Bio, Novartis, OcuTerra, and Opthea outside the submitted work. Dr Emanuelli reported receiving grants from Roche/Genentech during the conduct of the study; and serving as a consultant for Genentech and Novartis and receiving grants or research support from AbbVie, Adverum, Apellis, Genentech, F. Hoffmann-La Roche, Nanoscope, Novartis, Novartis Institute of Biomedical Research, OcuTerra, Opthea, Regeneron, 4D Molecular Therapeutics, Kodiak Sciences, EyeBio, Regenxbio, and Adverum Biotechnologies outside the submitted work. Dr Holekamp reported receiving personal fees from Genentech, Biogen, Apellis, EyePoint, Bausch and Lomb, Regeneron, and Stealth, serving as a consultant for AbbVie, Apellis, Bausch and Lomb, Biogen, Boehringer Ingelheim, EyePoint, F. Hoffman-La Roche and Roche Products Ltd, Genentech, Genentech USA, Notal Vision, and Regeneron Pharmaceuticals and Regeneron Healthcare Solutions, serving as a speaker for Apellis, Bausch and Lomb, Genentech, and Regeneron, receiving research support from Genentech and Notal Vision, serving on the data monitoring and safety committee for Editas, Ocuphire Pharma, and Roche, and having stock or stock options in Nacuity, Notal Vision, and Roche outside the submitted work. Dr Hu reported receiving grants from Genentech/Roche during the conduct of the study; and receiving personal fees from Genentech/Roche, receiving grants or contracts from 4DMT, AbbVie, Acelyrin, Alexion, Alimera, Annexin, Annexon, Apellis, Ashvattha, Aviceda, Bayer, Boehringer Ingelheim, Chengdu Kanghong, Clearside, Cognition, Curacle, EyeBioTech, EyePoint, Genentech, Graybug, Ionis, Iveric Bio, Janssen, Kodiak Sciences, LMRI, Lumithera, Lupin, Neurotech, Novartis, Ocular Therapeutix, Oculis, Ocuphire, Ophthotech, Opthea, Outlook Therapeutics, Oxurion, Regeneron, Regenxbio, Rezolute, Roche, Sandoz, Shanghai Henlius, Smilebiotek Zhuhai, Stealth, and Unity, serving as a consultant for Annexon, Apellis, Bayer, Clearside, Genentech, Janssen, Oculis, Outlook Therapeutics, Regeneron, Roche, and Stealth, serving as a speaker for Apellis, Genentech, and Iveric Bio, and participating on the advisory or data safety monitoring board for Clearside outside the submitted work. Dr Suñer reported receiving personal fees from Genentech, serving as a consultant for Alimera, Allergan, Apellis, Aura Biosciences, Bausch and Lomb, Eclipse Biosciences, Genentech, Iveric Bio, Ocular Therapeutix, ONL, and Regeneron, and receiving research support from Alimera, Apellis, Eclipse Biosciences, EyeBio, EyePoint, Genentech, Kodiak Sciences, Ocular Therapeutix, and Regeneron outside the submitted work. Dr Wykoff reported receiving consulting fees from 4DMT, AbbVie, Adverum, Aerie, AGTC, Alcon, Alimera, Alkeus, Allergan, Allgenesis, AMC Sciences, Annexon, Apellis, Arrowhead, Ascidian, Aviceda, Bausch and Lomb, Bayer, Biocryst, Bionic Vision, Boehringer Ingelheim, Chengdu Kanghong, Cholgene, Clearside, Curacle, Emmecell, EyeBiotech, EyePoint, Foresite, Frontera, Genentech, Gyroscope, IACTA, InGel, Iveric Bio, Janssen, Kato, Kiora, Kodiak Sciences, Kriya, Merck, Merit, Nanoscope, Neurotech, NGM Bio, Notal Vision, Novartis, OccuRx, Ocular Therapeutix, Ocuphire, OcuTerra, OliX, ONL, Opthea, Osanni, Oxular, Palatin, Perceive Bio, Perfuse, PolyPhotonix, Ray, RecensMedical, Regeneron, Regenxbio, Resonance, RetinAI, Roche, Sandoz, Sanofi, Santen, SciNeuro, Stealth, Surrozen, Suzhou Raymon, Sylentis, Takeda, Thea, Therini, TissueGen, Valo, Verana, Visgenx, Vitranu, and Zeiss, receiving grants or research support from 4DMT, Adverum, Aerie, AffaMed, Aldeyra, Alexion, Alimera, Alkahest, Allergan, Allgenesis, Amgen, Annexin, Annexon, Apellis, Ascidian, Asclepix, Aviceda, Bayer, Boehringer Ingelheim, Chengdu Kanghong, Chengdu Origen, Clearside, Curacle, Eluminex, EyeBiotech, EyePoint, Gemini, Genentech, GlaxoSmithKline, Graybug, Gyroscope, Ionis, Iveric Bio, Janssen, Kalaris, Kodiak Sciences, Kyoto DDD, Kyowa Kirin, Nanoscope, Neurotech, NGM Bio, Novartis, Ocugen, Ocular Therapeutix, Oculis, Ocuphire, OcuTerra, OliX, Opthea, Outlook Therapeutics, Oxular, Oxurion, Oyster Point, Perceive Bio, Pykus, RecensMedical, Regeneron, Regenxbio, Rezolute, Roche, SamChunDang Pharm, Sandoz, Shanghai Henlius, Skyline, Stealth, Unity, Valo, Verily, and Xbrane, and having stock options in InGel, ONL, Osanni, Panther, PolyPhotonix, RecensMedical, TissueGen, Visgenx, and Vitranu outside the submitted work. Dr Santhanakrishnan reported having stock options in Roche/Genentech outside the submitted work. No other disclosures were reported.

Funding/Support: This study was funded by F. Hoffmann-La Roche Ltd.

Role of the Funder/Sponsor: F. Hoffmann-La Roche Ltd provided support for the study and participated in the design and conduct of the study; collection, analysis, and interpretation of the data; preparation of the manuscript; and decision to submit the manuscript for publication. Funding was provided by F. Hoffmann-La Roche Ltd for third-party writing assistance, which was provided by Rachel Verdon, PhD, Envision Pharma Group. Data were collected by investigators and analyzed by statisticians employed by the sponsor.

Data Sharing Statement: See Supplement 4.

Additional Contributions: We thank the patients who participated in these trials and their families, the investigators, and staff at all the Pavilion clinical sites, and the members of the independent data and safety monitoring committee.

^✉

Corresponding author.

PMCID: PMC11886866 PMID: 40048178

Key Points

Question

Does the Port Delivery System (PDS) with ranibizumab, 100 mg/mL, refilled every 36 weeks, improve Diabetic Retinopathy Severity Scale (DRSS) levels in patients with nonproliferative diabetic retinopathy without center-involved diabetic macular edema (CI-DME) compared with monitoring?

Findings

In this randomized clinical trial, substantially more participants receiving PDS vs monitoring achieved at least 2-step DRSS level improvement and avoided CI-DME, proliferative diabetic retinopathy, or anterior segment neovascularization through 52 weeks. With PDS, vision decreased transiently 4 weeks after implantation, resolving 8 weeks after that; no vision change from baseline occurred in either group at week 52.

Meaning

PDS improved DRSS levels in nonproliferative diabetic retinopathy without CI-DME.

Abstract

Importance

Frequent prophylactic intravitreal anti–vascular endothelial growth factor injections can reduce risk of progression to vision-threatening complications in nonproliferative diabetic retinopathy (NPDR). A refillable drug delivery system for continuous intraocular ranibizumab release could offer less frequent treatment regimens.

Objective

To evaluate the Port Delivery System (PDS) with ranibizumab, 100 mg/mL, with refill-exchange procedures every 36 weeks (PDS Q36W), vs no PDS (control) in moderately severe to severe NPDR without center-involved diabetic macular edema (CI-DME), monitoring both groups every 4 weeks.

Design, Setting, and Participants

This was a randomized clinical trial at 50 US investigational sites. Participants aged 18 years or older with moderately severe or severe NPDR (Diabetic Retinopathy Severity Scale [DRSS] level 47 or 53) secondary to type 1 or 2 diabetes were eligible. Data analysis was performed from August 10, 2020, to October 3, 2022.

Intervention

Participants were randomized (unmasked) 5:3 to PDS Q36W vs control. Both groups could receive intravitreal ranibizumab injections if CI-DME, proliferative diabetic retinopathy (PDR), or anterior segment neovascularization (ASNV) developed.

Main Outcomes and Measures

Proportion of participants with an improvement of at least 2 levels in Early Treatment Diabetic Retinopathy Study DRSS from baseline at week 52.

Results

A total of 174 participants (mean [SD] age, 53.9 [11.7] years; 74 [42.5%] female) were randomized to PDS Q36W (n = 106) or control (n = 68). At week 52, 80.1% of those receiving PDS Q36W vs 9.0% of control participants had at least a 2-step DRSS improvement from baseline (difference, 71.1% [95% CI, 61.0% to 81.2%]; P < .001). Secondary outcomes included rate of development of CI-DME, PDR, or ASNV through week 52 (PDS Q36W, 7.1%; control, 47.0%; hazard ratio, 0.12 [95% CI, 0.05 to 0.28]; P < .001) and best-corrected visual acuity (BCVA) change from baseline to week 52 (+1.4 letters [95% CI, –0.5 to 3.3 letters] for those receiving PDS Q36W vs –2.6 letters [95% CI, –5.0 to –0.1 letters] for control participants; difference, 4.0 letters [95% CI, 0.9 to 7.1 letters]; P = .01). The PDS Q36W group had a transient BCVA decrease of 7.4 letters (95% CI, –10.3 to –4.5 letters) at 4 weeks after implantation, resolving 8 weeks later. Ocular adverse events of special interest occurred in 17 of 105 participants (16.2%) receiving PDS Q36W (cataract, 7 participants [6.7%]; vitreous hemorrhage, 6 participants [5.7%]; conjunctival bleb, conjunctival retraction, and hyphema, each 2 participants [1.9%]; conjunctival erosion and retinal detachment, each 1 participant [1.0%]), with no endophthalmitis reported through week 52.

Conclusions and Relevance

At 1 year, PDS Q36W resulted in substantially more participants achieving at least a 2-step DRSS improvement and a reduced risk of developing CI-DME, PDR, or ASNV compared with control participants, with safety outcomes consistent with previous reports. These findings should be balanced with the transient, postoperative decrease in BCVA 4 through 12 weeks after implantation and the need for longer-term BCVA and safety outcomes.

Trial Registration

ClinicalTrials.gov Identifier: NCT04503551

This randomized clinical trial evaluates whether the Port Delivery System with ranibizumab, compared with monitoring, improves Diabetic Retinopathy Severity Scale levels among patients with nonproliferative diabetic retinopathy without center-involved diabetic macular edema.

Introduction

Diabetic retinopathy (DR) is a sight-threatening disease affecting approximately one-third of individuals with diabetes. The global prevalence of diabetes is expected to reach 783 million people by 2045, and DR will likely remain a leading cause of vision loss and irreversible blindness.^1,2,3,4,5,6 Patients with moderately severe to severe nonproliferative DR (NPDR) are at high risk for progression to proliferative DR (PDR) and vision loss associated with development of diabetic macular edema (DME) and complications including nonclearing vitreous hemorrhage or traction retinal detachment.^7,8 The risk of progression to PDR within 1 year approximately doubles from 26% in moderately severe NPDR to 52% in severe NPDR.⁷ A retrospective review of a large cohort from the Intelligent Research in Sight (IRIS) Registry showed that patients with moderately severe or severe NPDR could develop visual acuity (VA) of 20/200 or worse for at least 3 months in 1 eye after 2 years.⁸

The PANORAMA⁹ randomized clinical trial in participants with moderately severe to severe NPDR and the DRCR Retina Network Protocol W¹⁰ randomized clinical trial in participants with moderate to severe NPDR showed that prophylactic, periodic anti–vascular endothelial growth factor (VEGF) injections significantly improved DR severity and reduced the risk of sight-threatening complications vs sham injections through approximately 2 years. In both trials, no difference in vision outcomes was observed between groups. In Protocol W, additional follow-up in years 2 to 4 (participant retention, approximately 75%), with treatment administered only if vision-threatening complications developed, showed that while there was still no difference in vision outcomes between groups, the anatomic benefits were maintained. These benefits of anti-VEGF treatment come with burden of frequent treatment; drug-device combinations have the potential to reduce this treatment burden.

The Port Delivery Platform is a drug delivery system comprising a surgically implanted, refillable ocular implant and associated drug product to provide continuous intraocular drug delivery, including the Port Delivery System (PDS) with ranibizumab, which has US Food and Drug Administration (FDA) approval for treatment of neovascular age-related macular degeneration and DME and is under investigation for DR and DME.^11,12,13 During an investigation into septum dislodgement cases in the PDS phase 3 clinical trial program, a need for additional testing of the commercial implant supply was identified, leading to the voluntary recall of the PDS implant by Roche/Genentech, Inc in October 2022. The FDA classified this voluntary recall as class III, ie, use of the product is not likely to cause adverse health consequences. Component level changes and manufacturing process improvements were implemented, with no septum failures subsequently observed using the updated implant and updated refill needle combination through more than 30 years of simulated use.¹² The voluntary recall was lifted in April 2024. The updated implant and updated refill needle received FDA approval in July 2024 and are currently available in the US. Herein, we present primary analyses from the Pavilion trial, in which both groups had monthly clinical monitoring and the PDS was assessed with refill-exchange procedures every 36 weeks (Q36W; after 2 intravitreal ranibizumab loading doses) vs no PDS (control) in participants with moderately severe to severe NPDR without center-involved DME (CI-DME), wherein participants in either group could receive intravitreal ranibizumab injections if CI-DME, PDR, or anterior segment neovascularization (ASNV) developed.

Methods

Trial Design and Oversight

The Pavilion trial is a phase 3, multicenter, visual assessor–masked, superiority, randomized clinical trial of the PDS for NPDR without CI-DME, conducted across 50 US sites (trial protocol available in Supplement 1, statistical analysis plan in Supplement 2, and additional details in eAppendix 1 in Supplement 3). Pavilion adhered to the tenets of the Declaration of Helsinki¹⁴ and was conducted in accordance with the International Conference on Harmonisation E6 for Good Clinical Practice.¹⁵ Trial sites received institutional review board approval before trial initiation, and all participants provided written informed consent before enrollment. Participants were compensated for their travel expenses. Study conduct was monitored and unmasked safety data were reviewed routinely by an independent data monitoring committee. The primary analysis was conducted when all participants completed the week 52 visit. The Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines were followed. Data analysis was performed from August 10, 2020, to October 3, 2022.

Trial Population

Participants aged 18 years or older with moderately severe or severe NPDR (Diabetic Retinopathy Severity Scale [DRSS] level 47 or 53) secondary to type 1 or 2 diabetes were eligible.^16,17 Study eye inclusion criteria included a best-corrected VA (BCVA) letter score of at least 69 (20/40 approximate Snellen equivalent). Study eye exclusion criteria included presence of CI-DME (central subfield thickness [CST] ≥325 μm) and prior treatment for DR or DME. Full eligibility criteria are provided in eAppendix 2 in Supplement 3.

Trial Procedures

Participants were randomized 5:3 to PDS Q36W or control (no PDS) (eFigure 1 in Supplement 3), with all participants monitored monthly. Study participants and study personnel were not masked to study eye or treatment assignment due to the surgical nature of the trial. VA examiners were masked to study eye, treatment assignment, and study visit type to minimize bias. Reading center graders were masked in a similar fashion for the primary outcome. Participants randomized to PDS Q36W received 2 loading doses with intravitreal ranibizumab 0.5-mg injections (Genentech, Inc) at day 1 and week 4. The PDS implant (with a customized formulation of ranibizumab, 100 mg/mL) was surgically inserted 1 to 14 days after the second loading dose, and PDS refill-exchange procedures with ranibizumab, 100 mg/mL, occurred on a fixed Q36W interval thereafter, as previously described.^11,18,19 Participants randomized to the PDS group attended safety assessments visits 1 day and 7 (±2) days after implantation, at week 8 (±7 days), and every 4 weeks thereafter through week 52 (total of 17 scheduled study visits through week 52). Participants randomized to the control underwent study visits every 4 weeks for observation and comprehensive clinical monitoring through week 52 (total of 14 scheduled study visits through week 52). Participants in either group could receive intravitreal ranibizumab 0.5-mg injections per unmasked investigator discretion if CI-DME (CST ≥325 μm), PDR, or ASNV developed.

Outcomes

The prespecified primary efficacy end point was the proportion of participants with at least a 2-step Early Treatment Diabetic Retinopathy Study (ETDRS)–DRSS improvement from baseline at week 52 based on color fundus photographs. Key secondary end points included rates of developing CI-DME or developing PDR and/or ASNV, rate of developing at least a 2-step or 3-step ETDRS-DRSS worsening from baseline, and proportion of participants with at least a 3-step ETDRS-DRSS improvement from baseline. Additional secondary end points included BCVA (ETDRS letter score) change from baseline, CST change from baseline, and receipt of supplemental treatment.

The safety objective was to evaluate the safety and tolerability of ranibizumab, 100 mg/mL, delivered via the PDS. Safety outcomes included the incidence and severity of ocular and nonocular adverse events (AEs) and prespecified ocular AEs of special interest (AESIs; defined as cataract, conjunctival retraction, conjunctival bleb or filtering bleb leak, conjunctival erosion, endophthalmitis, hyphema, implant dislocation, retinal detachment, and vitreous hemorrhage). A full list of objectives is provided in eAppendix 3 in Supplement 3.

Statistical Analysis

A sample size of 160 participants was determined to provide more than 99% power to demonstrate at least a 35% absolute improvement in the proportion of participants with at least a 2-step ETDRS-DRSS improvement at week 52, assuming an achievement of at least a 2-step ETDRS-DRSS improvement at week 52 in 15% of control participants²⁰ by Fisher exact test, a 4.96% 2-sided type I error rate (adjusted for interim safety analyses), and a 15% dropout rate by week 52. Efficacy was analyzed in the intent-to-treat population, and safety was assessed in the safety-evaluable population (eAppendix 4 in Supplement 3). The primary end point was analyzed using Cochran-Mantel-Haenszel (CMH) test stratified by baseline ETDRS-DRSS level (47 vs 53) and intraretinal or subretinal fluid status (present vs absent). Participants who received supplemental treatment, prohibited therapy, or panretinal photocoagulation were considered nonresponders, and their ETDRS-DRSS level at week 52 was imputed using the assumption that they did not achieve at least a 2-step improvement from baseline. Participants with missing baseline outcomes were excluded.

Overall type I error was set at a 2-sided significance level of .0496 to account for 4 independent data monitoring committee interim safety monitoring meetings (penalty of 0.0001 per meeting). A fixed-sequence testing procedure controlled overall type I error for primary and key secondary end points, with the testing order specified in the statistical analysis plan.²¹

The rate of participants developing CI-DME or developing PDR and/or ASNV and the rate of participants developing at least a 2-step or 3-step ETDRS-DRSS worsening were estimated using the Kaplan-Meier method based on time to the first event; receipt of supplemental treatment, prohibited therapy, or panretinal photocoagulation was considered an event. The proportion of participants with at least a 3-step ETDRS-DRSS improvement was analyzed as described for the primary end point. Continuous secondary end points were analyzed using a mixed-effects model with repeated measures. SAS version 9.4 (SAS Institute Inc) was used for statistical analysis. See eAppendix 4 in Supplement 3 for additional details on statistical analysis.

Results

Trial Population

From August 10, 2020, to September 30, 2021, 174 participants were randomized to PDS Q36W (n = 106) or control (n = 68) (Figure 1). Baseline demographic characteristics were generally well balanced (eTable 1 in Supplement 3). Mean (SD) age was 53.9 (11.7) years, with 74 female participants (42.5%). Self-reported race data were as follows: 18 (10.3%) African American or Black, 4 (2.3%) Alaska Native or American Indian, 17 (9.8%) Asian, 2 (1.1%) Native Hawaiian or Other Pacific Islander, 125 (71.8%) White, and 8 (4.6%) not available; 72 participants (41.4%) self-reported as Hispanic or Latino. Mean (SD) time since DR diagnosis was 12.8 (20.2) months. Mean (SD) BCVA letter score at baseline was 82.1 (5.9) (approximate Snellen equivalent, 20/25) in the PDS Q36W group and 82.9 (4.7) (approximate Snellen equivalent, 20/25) in the control group.

Seven PDS Q36W participants discontinued before receiving the PDS implant; none discontinued after implantation. Six control participants discontinued the study. At week 52, 99 participants receiving PDS Q36W (93.4%) and 62 control participants (91.2%) were ongoing in the study. Mean (SD) time receiving treatment was 49.6 (11.0) weeks for PDS Q36W participants and 50.8 (5.7) weeks for control participants (eTable 2 in Supplement 3). All participants who received the PDS implant received a refill exchange after approximately 9 months. The percentage of participants who completed their week 52 visit was 89.6% (95 of 106) in the PDS Q36W group and 82.4% (56 of 68) in the control group. Further treatment details are available in eTable 2 in Supplement 3.

Through week 52, major protocol deviations occurred in 42 PDS Q36W participants (39.6%) and 22 control participants (32.4%) (eTable 3 in Supplement 3), deemed unlikely to impact study conclusions (procedural deviations were balanced across groups, ambient temperature excursions were assessed and considered to have low impact on efficacy and safety, and all participants reconsented). One participant in each group had a major protocol deviation of a missed visit due to COVID-19. As these visits were procedural in nature and 1 participant in each group was affected, these deviations were judged to have minor impact.

Efficacy

For the primary outcome, PDS Q36W was superior to control, with significantly more participants achieving at least a 2-step ETDRS-DRSS improvement from baseline in the study eye at week 52. The CMH-weighted estimates were 80.1% for PDS Q36W and 9.0% for control (difference in CMH-weighted estimates, 71.1%; 95% CI, 61.0%-81.2%; P < .001) (Figure 2). PDS Q36W was superior to control for rates of participants developing CI-DME, PDR, or ASNV in the study eye through week 52 (7.1% vs 47.0%, respectively; hazard ratio [HR], 0.12 [95% CI, 0.05-0.28]; P < .001) (Figure 3). The rate of developing CI-DME through week 52 was significantly lower for PDS Q36W participants (7.1%) than for control participants (47.0%) (HR, 0.12 [95% CI, 0.05-0.28]; P < .001). Similarly, the rate of developing PDR or ASNV was significantly lower for PDS Q36W participants (1.0%) than for control participants (42.4%) (HR, 0.02 [95% CI, <0.01-0.14]; P < .001). The Kaplan-Meier plot for time to participants first developing CI-DME, PDR, or ASNV through week 52 is presented in eFigure 2 in Supplement 3.

Figure 2. — This analysis shows data for the intent-to-treat population randomized either to the Port Delivery System (PDS) with ranibizumab every 36 weeks or to the control (monitoring). The main estimand strategy was used; participants who received supplemental treatments, prohibited therapy, or panretinal photocoagulation were considered nonresponders regardless of their observed outcome after the corresponding visit. The weighted estimate was based on the Cochran-Mantel-Haenszel (CMH) method stratified by baseline ETDRS-DRSS level (47 vs 53) and baseline intraretinal or subretinal fluid status (present vs absent). 95% CI is a rounding of 95.04% CI; type I error was adjusted for interim safety monitoring. Missing ETDRS-DRSS level was imputed using the last observed outcome before week 52. Participants with missing baseline outcomes were excluded.

^aP value was calculated using the CMH test.

Figure 3. — This analysis shows data for the intent-to-treat population randomized either to the Port Delivery System (PDS) with ranibizumab every 36 weeks or to the control (monitoring). The main estimand strategy was used; receipt of supplemental treatments, prohibited therapy, or panretinal photocoagulation was considered to be an event in the analysis. Participants who did not experience CI-DME and/or PDR or ASNV or these intercurrent events were censored at the end of their time at risk. Analysis was stratified by baseline Early Treatment Diabetic Retinopathy Study Diabetic Retinopathy Severity Scale level (47 vs 53) and baseline intraretinal or subretinal fluid status (present vs absent). CI-DME was defined as central subfield thickness of 325 μm or greater on spectral-domain optical coherence tomography. The hazard ratios (HRs) and 95% CIs were estimated using a Cox proportional hazards regression model. 95% CI is a rounding of 95.04% CI; type I error was adjusted for interim safety monitoring. KM indicates Kaplan-Meier.

^aP value was calculated using the stratified log-rank test.

At week 52, PDS Q36W participants demonstrated significantly lower rates of at least a 2-step DRSS worsening from baseline in the study eye compared with control participants (3.0% vs 46.4%, respectively; HR, 0.05 [95% CI, 0.02-0.16]; P < .001), and results were similar for at least a 3-step DRSS worsening (3.0% vs 45.1%, respectively; HR, 0.05 [95% CI, 0.02-0.17]; P < .001) (eTable 4 in Supplement 3). The proportion of participants who achieved at least a 3-step DRSS improvement from baseline was significantly greater for PDS Q36W participants (16 of 106 [15.1%]) than for control participants (0 of 68) (difference, 15.1% [95% CI, 8.3%-21.9%]; P < .001).

Following PDS implantation at week 4, there was a transient, surgery-related decrease in BCVA recorded in the study eye with PDS Q36W at 4 weeks after implantation (adjusted mean [SD] BCVA change from baseline, –7.4 [14.5] letters [95% CI, –10.3 to –4.5 letters; IQR, –10.0 to 0.0 letters]; mean [SD] absolute BCVA letter score, 75.9 [16.0]; approximate Snellen equivalent, 20/32; IQR, 72.0 to 84.0 letters [approximate Snellen equivalent, 20/40 to 20/20]). This transient decrease in mean BCVA resolved approximately 12 weeks after implantation, and mean BCVA was maintained thereafter (Figure 4A). At week 52, the adjusted mean BCVA change from baseline was +1.4 (95% CI, –0.5 to 3.3) ETDRS letters in the PDS Q36W group and –2.6 (95% CI, –5.0 to –0.1) ETDRS letters in the control group (difference, 4.0 letters [95% CI, 0.9-7.1 letters]; P = .01). The proportion of participants who lost less than 15, less than 10, and less than 5 letters in BCVA from baseline over time at week 52 are shown in eTable 5 in Supplement 3.

Figure 4. — Adjusted mean changes from baseline in best-corrected visual acuity (BCVA) (A) and central subfield thickness (CST) (B) are shown for the intent-to-treat population randomized either to the Port Delivery System (PDS) with ranibizumab every 36 weeks or to the control (monitoring). The mixed-effects model with repeated measures was used; the model was adjusted for treatment group, visit, treatment-by-visit interaction, baseline Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale level (47 vs 53), and baseline intraretinal or subretinal fluid status (present vs absent). The BCVA model was adjusted for baseline BCVA score, and the CST model was adjusted for baseline CST. An unstructured covariance structure was used. 95% CI is a rounding of 95.04% CI; type I error was adjusted for interim safety monitoring. All observed data were included in the analysis, regardless of whether a participant had experienced an intercurrent event. Missing data were implicitly imputed by the mixed-effects model with repeated measures, assuming a missing-at-random mechanism.

In the PDS Q36W group, CST decreased slightly from baseline to week 4, followed by a transient increase between weeks 4 and 8 (after the second loading dose injection and PDS implantation surgery) that remained below baseline levels. Mean CST then declined through week 52, with CST improvement in the PDS Q36W group compared with the control group (adjusted mean change from baseline at week 52, –18.8 μm vs +2.5 μm, respectively; difference, –21.3 μm [95% CI, –31.7 μm to –10.9 μm]; P < .001) (Figure 4B).

Of 99 participants assessed in the PDS Q36W group, none received supplemental treatment through week 52. Of 68 participants assessed in the control group, 27 (39.7%) received supplemental treatment, among whom the mean (SD) number of injections was 2.9 (2.0) (eTable 2 in Supplement 3). See eAppendix 5 in Supplement 3 for additional efficacy outcomes.

Safety

Safety outcomes reported herein are for the study eye of the PDS Q36W safety-evaluable population (n = 105) through week 52. AEs are not summarized for the control group because only key safety events were collected through week 52 until study treatment initiation (eAppendix 4 in Supplement 3). In total, 271 ocular AEs were reported in 84 PDS Q36W participants (80.0%), of which 3 (1.1%) were serious (eTables 6 and 7 in Supplement 3). Ocular AESIs were reported in 17 of 105 PDS Q36W participants (16.2%) (Table) (cataract in 7 participants [6.7%]; vitreous hemorrhage in 6 participants [5.7%]; conjunctival bleb, conjunctival retraction, and hyphema, each in 2 participants [1.9%]; conjunctival erosion and retinal detachment, each in 1 participant [1.0%]). Of these ocular AESIs, 2 (1.9%) were serious (1 event each of retinal detachment and vitreous hemorrhage) and 1 (1.0%) was severe (vitreous hemorrhage). Six PDS Q36W participants (5.7%) experienced vitreous hemorrhage; most events were nonserious, with no grade 4 reported; 1 participant required vitrectomy. All the vitreous hemorrhage events occurred within 28 days of PDS implantation. No events of endophthalmitis, septum dislodgement, retinal occlusive vasculitis, or implant dislocation were reported with PDS Q36W through week 52. See eAppendix 6 in Supplement 3 for further details on AESIs reported with PDS Q36W.

Table. Ocular Adverse Events of Special Interest (AESIs) in the Safety-Evaluable Population Through Week 52 in the Study Eyes Receiving the Port Delivery System With Ranibizumab, 100 mg/mL, Every 36 Weeks.

Analysis	All AESIs^a				Serious AESIs^a
Analysis	Preimplant	Onset within 37 d	Onset after 37 d	Overall	Preimplant	Onset within 37 d	Onset after 37 d	Overall
AESIs, total No.^b	0	13	9	22	0	2	0	2
Participants with ≥1 AESI, No. (%)	0	10 (9.5)	9 (8.6)	17 (16.2)	0	2 (1.9)	0	2 (1.9)
AESI, No. (%)
Cataract^c	0	2 (1.9)	6 (5.7)	7 (6.7)	0	0	0	0
Conjunctival bleb^d	0	1 (1.0)	1 (1.0)	2 (1.9)	0	0	0	0
Conjunctival erosion	0	0	1 (1.0)	1 (1.0)	0	0	0	0
Conjunctival retraction	0	1 (1.0)	1 (1.0)	2 (1.9)	0	0	0	0
Implant dislocation^e	0	0	0	0 (0.0)	0	0	0	0
Endophthalmitis	0	0	0	0 (0.0)	0	0	0	0
Hyphema	0	2 (1.9)	0	2 (1.9)	0	0	0	0
Retinal detachment^f	0	1 (1.0)	0	1 (1.0)	0	1 (1.0)	0	1 (1.0)
Vitreous hemorrhage	0	6 (5.7)	0	6 (5.7)	0	1 (1.0)	0	1 (1.0)

Open in a new tab

^{^a}

Periods were defined as follows: preimplant, the period from the day of the first loading dose through the day before implantation; onset within 37 days, the period from the day of implantation through 37 days after implantation; onset after 37 days, the period from 38 days after implantation through week 52; and overall, the period from the day of the first loading dose through week 52.

^{^b}

Frequency counts are by the Medical Dictionary for Regulatory Activities Preferred Term. Multiple occurrences of the same adverse event in the same individual are counted only once per eye.

^{^c}

Includes cataract, cataract nuclear, cataract cortical, cataract subcapsular, and traumatic cataract.

^{^d}

Includes conjunctival bleb (defined as an elevation of the conjunctiva above the implant flange, which may be secondary to subconjunctival tissue thickening or transient fluid accumulation), conjunctival filtering bleb leak, conjunctival cyst, subconjunctival cyst, and implant site cyst.

^{^e}

Reported as device dislocation.

^{^f}

Includes retinal detachment, rhegmatogenous retinal detachment, and tractional retinal detachment.

Discussion

Pavilion demonstrated effects of continuous anti-VEGF blockade and monthly clinical monitoring vs monthly clinical monitoring with no PDS (control) in participants with moderately severe to severe NPDR without CI-DME. Significantly more participants achieved at least a 2-step ETDRS-DRSS improvement at week 52 with PDS Q36W vs control (80.1% vs 9.0%, respectively; difference, 71.1% [95% CI, 61.0%-81.2%]; P < .001). In PANORAMA, mirroring Pavilion participant demographics, participants received intravitreal aflibercept 2-mg injections either every 8 weeks (Q8W) after 5 initial monthly doses or every 16 weeks (Q16W) after 3 initial monthly doses.⁹ With aflibercept Q8W, 79.9% of eyes achieved at least a 2-step DRSS improvement by week 52, vs 65.2% in the aflibercept Q16W group, suggesting that frequent injections are required for optimal outcomes. In the DRCR Retina Network Protocol W study of aflibercept vs sham treatment in participants with moderate to severe NPDR without CI-DME (with intravitreal injections at baseline, 1, 2, and 4 months, and every 4 months through 2 years), aflibercept treatment resulted in 44.8% of eyes achieving at least a 2-step DRSS improvement at year 2 vs 13.7% of eyes with sham.¹⁰ However, differences in both the study design and participant baseline characteristics limit comparison between Protocol W and Pavilion. Notably, participants in the aflibercept group of Protocol W had milder baseline disease (16.5%, 59.5%, and 24.0% with baseline DRSS levels 43, 47, and 53, respectively) compared with PDS Q36W participants in Pavilion (82.1% and 17.9% with DRSS levels 47 and 53, respectively) and better baseline VA vs Pavilion (median BCVA letter score, 88 [approximate Snellen equivalent 20/20] vs 82 [approximate Snellen equivalent 20/25]).

CI-DME, PDR, and ASNV are complications of DR that threaten ocular well-being and can lead to irreversible vision loss if not treated promptly. In Pavilion, PDS Q36W participants (7.1%) showed a significantly lower rate of developing CI-DME, PDR, or ASNV vs control participants (47.0%) at week 52. These findings align with PANORAMA, where event rates for aflibercept-treated eyes developing CI-DME, PDR, or ASNV by week 52 were 10.1% (Q16W) and 10.8% (Q8W) vs 41.8% (sham), and similarly with Protocol W, wherein the 2-year probability of developing CI-DME with vision loss or PDR was 16.3% with aflibercept vs 43.5% with sham.^9,10

PDS Q36W significantly reduced the rate of at least a 2-step or 3-step DRSS worsening vs control, suggesting potential to slow disease progression. The PDS also showed a trend toward improved CST vs control (adjusted mean baseline change, –18.8 μm vs +2.5 μm, respectively, at week 52). In the PDS Q36W group, there was a transient, postsurgical VA decrease of 7.4 letters at 4 weeks after implantation that resolved 8 weeks later. This vision decrease, also observed in neovascular age-related macular degeneration trials, may be driven by media changes in the anterior vitreous or associated with the conjunctival sutures. At week 52, vision outcomes were comparable between the PDS Q36W and control groups (adjusted mean BCVA change from baseline of +1.4 vs –2.6 letters, respectively); this is likely due to the monthly follow-up schedule and initiation of treatment based on anatomic thresholds before BCVA decline, as previously demonstrated in PANORAMA and Protocol W where no difference in VA outcomes was reported between groups in the study follow-up periods (PANORAMA: BCVA change from baseline, +1.7, +1.3, and +0.5 letters at week 52 and +1.5, +0.8, and +0.6 letters at week 100 for aflibercept Q16W, Q8W, and sham, respectively; Protocol W: BCVA change from baseline, –0.9 vs –2.0 letters at year 2 and –2.7 vs –2.4 letters at year 4 for aflibercept vs sham, respectively).^9,10,22

In Pavilion, 93.4% of PDS Q36W participants were ongoing at week 52, with all discontinuations occurring before PDS implantation (99 of 99 implanted participants remaining at week 52). In PANORAMA,⁹ 92.5% and 90.4% of participants completed week 52 in the aflibercept Q8W and Q16W groups, respectively. The retention rate of Protocol W^10,22 was 80% through the 2 years of follow-up in the aflibercept group.

Ocular AESIs occurred in 17 PDS Q36W participants (16.2%) (cataract, 6.7%; vitreous hemorrhage, 5.7%; conjunctival bleb, conjunctival retraction, and hyphema, each 1.9%; conjunctival erosion and retinal detachment; each 1.0%). The 6 PDS Q36W participants (5.7%) who experienced vitreous hemorrhage had an onset within 28 days of PDS implantation. There were no events of endophthalmitis, retinal occlusive vasculitis, or implant dislocation through week 52 with the PDS. Additional strengths of Pavilion include the diverse participant population enrolled. Of note, 41.4% of participants were of Hispanic or Latino origin, for whom real-world prevalence is reportedly higher than for other ethnic groups.^5,23

In clinical practice, patients with diabetic eye disease receiving intravitreal anti-VEGF injections often deviate from the approved prescribing regimen due to the treatment burden, leading to suboptimal outcomes and progression to advanced stages of DR with risk of eventual vision loss.^{24,25,26,27,28} In Pavilion, PDS participants were monitored every 4 weeks and did not receive any supplemental treatment through week 52 (per investigator discretion), yet they achieved improvement in ETDRS-DRSS level and other clinically relevant end points, with only 1 refill-exchange procedure at approximately 9 months after PDS implantation. As a continuous-release anti-VEGF therapy for DR, the PDS offers reduced treatment burden compared with conventional intravitreal anti-VEGF therapies, which often require monthly injections to achieve comparable outcomes. Pavilion was designed to assess treatment burden with scheduled treatment Q36W; however, it was not specifically designed to investigate the impact on visit burden, as all participants were monitored every 4 weeks per protocol. Future studies may help define optimal management strategies for the PDS as a treatment option in the clinical practice setting.

Limitations

Limitations include frequency of monitoring visits in the control group, which were done more than would be expected in clinical practice. Given the interventional nature of the PDS implant, participants and study personnel were not masked to treatment allocation, which may introduce bias in some assessments. Image analysis for the primary outcome, however, was performed at independent reading centers by experienced, masked graders. This primary analysis was limited to 1 year of follow-up, with longer-term follow-up ongoing to provide further insight into the efficacy and safety of the PDS in this patient population.

Conclusions

For patients with diabetic eye disease, the PDS offers a unique strategy for continuous VEGF inhibition that can maintain vision and reduce disease progression with infrequent treatment administration. This approach may help prevent vision-threatening complications by providing sustained anti-VEGF therapy, particularly in cases where lapses in care or delayed monitoring might otherwise lead to substantial vision loss before treatment is initiated. These potential benefits should be balanced with the transient, postoperative decrease in BCVA from 4 to 12 weeks after implantation and AEs through 52 weeks. Longer-term studies are required to evaluate the safety and efficacy of the PDS as a treatment option vs intravitreal injections for patients with NPDR without CI-DME, which may help inform optimal management strategies for the PDS in clinical practice.

Supplement 1.

Trial Protocol

jamaophthalmol-e250001-s001.pdf^{(2.6MB, pdf)}

Supplement 2.

Statistical Analysis Plan

jamaophthalmol-e250001-s002.pdf^{(1.7MB, pdf)}

Supplement 3.

eAppendix 1. Pavilion Investigators and Study Sites

eAppendix 2. Pavilion Eligibility Criteria

eAppendix 3. Study Objectives and Prespecified Endpoints in Pavilion

eAppendix 4. Additional Methods

eAppendix 5. Additional Results

eAppendix 6. Additional Safety

eFigure 1. Design of the Pavilion Trial

eFigure 2. Time to Participants First Developing CI-DME (CST ≥325 µm on SD-OCT) and/or PDR or ASNV in the Study Eye Through Week 52

eTable 1. Demographic and Baseline Characteristics of Pavilion Participants

eTable 2. Treatment Exposure Through Week 52

eTable 3. Major Protocol Deviations Through Week 52

eTable 4. Rate of Participants Developing an ≥2-Step or ≥3-Step Worsening on ETDRS-DRSS From Baseline Through Week 52

eTable 5. Proportion of Participants Who Lost <15, <10, and <5 Letters in BCVA From Baseline in Study Eye Over Time at Week 52

eTable 6. Safety Summary of PDS Q36W Through Week 52

eTable 7. Ocular and Nonocular SAEs in Study Eye With PDS Q36W Through Week 52

jamaophthalmol-e250001-s003.pdf^{(585.2KB, pdf)}

Supplement 4.

Data Sharing Statement

jamaophthalmol-e250001-s004.pdf^{(21.5KB, pdf)}

References

1.International Diabetes Federation . IDF Diabetes Atlas. 10th ed. International Diabetes Federation; 2021. Accessed April 16, 2024. https://diabetesatlas.org/atlas/tenth-edition/
2.Flaxel CJ, Adelman RA, Bailey ST, et al. Diabetic Retinopathy Preferred Practice Pattern®. Ophthalmology. 2020;127(1):66-P145. doi: 10.1016/j.ophtha.2019.09.025 [DOI] [PubMed] [Google Scholar]
3.Flaxman SR, Bourne RRA, Resnikoff S, et al. ; Vision Loss Expert Group of the Global Burden of Disease Study . Global causes of blindness and distance vision impairment 1990-2020: a systematic review and meta-analysis. Lancet Glob Health. 2017;5(12):e1221-e1234. doi: 10.1016/S2214-109X(17)30393-5 [DOI] [PubMed] [Google Scholar]
4.Klein R, Klein BE. Vision disorders in diabetes. In: Harris MI, Cowie CC, Stern MP, Boyko EJ, Reiber GE, Bennett PH, eds. Diabetes in America. 2nd ed. National Institute of Diabetes and Digestive and Kidney Diseases; 1995:293-338. [Google Scholar]
5.Teo ZL, Tham YC, Yu M, et al. Global prevalence of diabetic retinopathy and projection of burden through 2045: systematic review and meta-analysis. Ophthalmology. 2021;128(11):1580-1591. doi: 10.1016/j.ophtha.2021.04.027 [DOI] [PubMed] [Google Scholar]
6.Yau JW, Rogers SL, Kawasaki R, et al. ; Meta-Analysis for Eye Disease (META-EYE) Study Group . Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care. 2012;35(3):556-564. doi: 10.2337/dc11-1909 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Early Treatment Diabetic Retinopathy Study Research Group . Fundus photographic risk factors for progression of diabetic retinopathy: ETDRS report number 12. Ophthalmology. 1991;98(5)(suppl):823-833. doi: 10.1016/S0161-6420(13)38014-2 [DOI] [PubMed] [Google Scholar]
8.Wykoff CC, Khurana RN, Nguyen QD, et al. Risk of blindness among patients with diabetes and newly diagnosed diabetic retinopathy. Diabetes Care. 2021;44(3):748-756. doi: 10.2337/dc20-0413 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Brown DM, Wykoff CC, Boyer D, et al. Evaluation of intravitreal aflibercept for the treatment of severe nonproliferative diabetic retinopathy: results from the PANORAMA randomized clinical trial. JAMA Ophthalmol. 2021;139(9):946-955. doi: 10.1001/jamaophthalmol.2021.2809 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Maturi RK, Glassman AR, Josic K, et al. ; DRCR Retina Network . Effect of intravitreous anti-vascular endothelial growth factor vs sham treatment for prevention of vision-threatening complications of diabetic retinopathy: the Protocol W randomized clinical trial. JAMA Ophthalmol. 2021;139(7):701-712. doi: 10.1001/jamaophthalmol.2021.0606 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Campochiaro PA, Marcus DM, Awh CC, et al. The Port Delivery System with ranibizumab for neovascular age-related macular degeneration: results from the randomized phase 2 ladder clinical trial. Ophthalmology. 2019;126(8):1141-1154. doi: 10.1016/j.ophtha.2019.03.036 [DOI] [PubMed] [Google Scholar]
12.Pieramici DJ, Pearce I, Campochiaro PA, et al. Technical updates to Port Delivery System with ranibizumab (PDS) and clinical experience with updated refill needle. Presented at: American Society of Retina Specialists 42nd Annual Meeting; July 18, 2024; Stockholm, Sweden. [Google Scholar]
13.Susvimo. Package insert. Genentech, Inc; 2025. [Google Scholar]
14.World Medical Association . WMA Declaration of Helsinki—ethical principles for medical research involving human subjects. Accessed April 16, 2024. https://www.wma.net/policies-post/wma-declaration-of-helsinki/
15.European Medicines Agency . ICH harmonised guideline integrated addendum to ICH E6(R1): guideline for good clinical practice ICH E6(R2). Accessed April 16, 2024. https://ichgcp.net/
16.American Diabetes Association . Standards of care in diabetes—2023 abridged for primary care providers. Clin Diabetes. 2022;41(1):4-31. doi: 10.2337/cd23-as01 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.World Health Organization . Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia: report of a WHO/IDF consultation. Updated January 31, 2024. Accessed April 16, 2024. https://www.who.int/publications/i/item/definition-and-diagnosis-of-diabetes-mellitus-and-intermediate-hyperglycaemia
18.Khanani AM, Graff JM, Marcus DM, et al. Refill-exchange procedure of the Port Delivery System with ranibizumab: overview and clinical trial experience. Ophthalmic Surg Lasers Imaging Retina. 2022;53(5):257-265. doi: 10.3928/23258160-20220412-01 [DOI] [PubMed] [Google Scholar]
19.Pieramici DJ, Wieland MR, Stewart JM, et al. Implant insertion procedure of the Port Delivery System with ranibizumab: overview and clinical pearls. Ophthalmic Surg Lasers Imaging Retina. 2022;53(5):249-256. doi: 10.3928/23258160-20220408-01 [DOI] [PubMed] [Google Scholar]
20.Eylea (aflibercept) injection. Prescribing information. Regeneron Pharmaceuticals Inc; 2023.
21.Westfall PH, Krishen A. Optimally weighted, fixed sequence and gatekeeper multiple testing procedures. J Stat Plan Inference. 2001;99(1):25-40. doi: 10.1016/S0378-3758(01)00077-5 [DOI] [Google Scholar]
22.Maturi RK, Glassman AR, Josic K, et al. ; DRCR Retina Network . Four-year visual outcomes in the Protocol W randomized trial of intravitreous aflibercept for prevention of vision-threatening complications of diabetic retinopathy. JAMA. 2023;329(5):376-385. doi: 10.1001/jama.2022.25029 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Kempen JH, O’Colmain BJ, Leske MC, et al. ; Eye Diseases Prevalence Research Group . The prevalence of diabetic retinopathy among adults in the United States. Arch Ophthalmol. 2004;122(4):552-563. doi: 10.1001/archopht.122.4.552 [DOI] [PubMed] [Google Scholar]
24.Best AL, Fajnkuchen F, Nghiem-Buffet S, et al. Treatment efficacy and compliance in patients with diabetic macular edema treated with ranibizumab in a real-life setting. J Ophthalmol. 2018;2018:4610129. doi: 10.1155/2018/4610129 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Ehlken C, Helms M, Böhringer D, Agostini HT, Stahl A. Association of treatment adherence with real-life VA outcomes in AMD, DME, and BRVO patients. Clin Ophthalmol. 2017;12:13-20. doi: 10.2147/OPTH.S151611 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Holekamp NM, Campbell J, Almony A, et al. Vision outcomes following anti–vascular endothelial growth factor treatment of diabetic macular edema in clinical practice. Am J Ophthalmol. 2018;191:83-91. doi: 10.1016/j.ajo.2018.04.010 [DOI] [PubMed] [Google Scholar]
27.Weiss M, Sim DA, Herold T, et al. Compliance and adherence of patients with diabetic macular edema to intravitreal anti-vascular endothelial growth factor therapy in daily practice. Retina. 2018;38(12):2293-2300. doi: 10.1097/IAE.0000000000001892 [DOI] [PubMed] [Google Scholar]
28.Wubben TJ, Johnson MW; Anti-VEGF Treatment Interruption Study Group . Anti-vascular endothelial growth factor therapy for diabetic retinopathy: consequences of inadvertent treatment interruptions. Am J Ophthalmol. 2019;204:13-18. doi: 10.1016/j.ajo.2019.03.005 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

Trial Protocol

jamaophthalmol-e250001-s001.pdf^{(2.6MB, pdf)}

Supplement 2.

Statistical Analysis Plan

jamaophthalmol-e250001-s002.pdf^{(1.7MB, pdf)}

Supplement 3.

eAppendix 1. Pavilion Investigators and Study Sites

eAppendix 2. Pavilion Eligibility Criteria

eAppendix 3. Study Objectives and Prespecified Endpoints in Pavilion

eAppendix 4. Additional Methods

eAppendix 5. Additional Results

eAppendix 6. Additional Safety

eFigure 1. Design of the Pavilion Trial

eFigure 2. Time to Participants First Developing CI-DME (CST ≥325 µm on SD-OCT) and/or PDR or ASNV in the Study Eye Through Week 52

eTable 1. Demographic and Baseline Characteristics of Pavilion Participants

eTable 2. Treatment Exposure Through Week 52

eTable 3. Major Protocol Deviations Through Week 52

eTable 4. Rate of Participants Developing an ≥2-Step or ≥3-Step Worsening on ETDRS-DRSS From Baseline Through Week 52

eTable 5. Proportion of Participants Who Lost <15, <10, and <5 Letters in BCVA From Baseline in Study Eye Over Time at Week 52

eTable 6. Safety Summary of PDS Q36W Through Week 52

eTable 7. Ocular and Nonocular SAEs in Study Eye With PDS Q36W Through Week 52

jamaophthalmol-e250001-s003.pdf^{(585.2KB, pdf)}

Supplement 4.

Data Sharing Statement

jamaophthalmol-e250001-s004.pdf^{(21.5KB, pdf)}

[eoi250001r1] 1.International Diabetes Federation . IDF Diabetes Atlas. 10th ed. International Diabetes Federation; 2021. Accessed April 16, 2024. https://diabetesatlas.org/atlas/tenth-edition/

[eoi250001r2] 2.Flaxel CJ, Adelman RA, Bailey ST, et al. Diabetic Retinopathy Preferred Practice Pattern®. Ophthalmology. 2020;127(1):66-P145. doi: 10.1016/j.ophtha.2019.09.025 [DOI] [PubMed] [Google Scholar]

[eoi250001r3] 3.Flaxman SR, Bourne RRA, Resnikoff S, et al. ; Vision Loss Expert Group of the Global Burden of Disease Study . Global causes of blindness and distance vision impairment 1990-2020: a systematic review and meta-analysis. Lancet Glob Health. 2017;5(12):e1221-e1234. doi: 10.1016/S2214-109X(17)30393-5 [DOI] [PubMed] [Google Scholar]

[eoi250001r4] 4.Klein R, Klein BE. Vision disorders in diabetes. In: Harris MI, Cowie CC, Stern MP, Boyko EJ, Reiber GE, Bennett PH, eds. Diabetes in America. 2nd ed. National Institute of Diabetes and Digestive and Kidney Diseases; 1995:293-338. [Google Scholar]

[eoi250001r5] 5.Teo ZL, Tham YC, Yu M, et al. Global prevalence of diabetic retinopathy and projection of burden through 2045: systematic review and meta-analysis. Ophthalmology. 2021;128(11):1580-1591. doi: 10.1016/j.ophtha.2021.04.027 [DOI] [PubMed] [Google Scholar]

[eoi250001r6] 6.Yau JW, Rogers SL, Kawasaki R, et al. ; Meta-Analysis for Eye Disease (META-EYE) Study Group . Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care. 2012;35(3):556-564. doi: 10.2337/dc11-1909 [DOI] [PMC free article] [PubMed] [Google Scholar]

[eoi250001r7] 7.Early Treatment Diabetic Retinopathy Study Research Group . Fundus photographic risk factors for progression of diabetic retinopathy: ETDRS report number 12. Ophthalmology. 1991;98(5)(suppl):823-833. doi: 10.1016/S0161-6420(13)38014-2 [DOI] [PubMed] [Google Scholar]

[eoi250001r8] 8.Wykoff CC, Khurana RN, Nguyen QD, et al. Risk of blindness among patients with diabetes and newly diagnosed diabetic retinopathy. Diabetes Care. 2021;44(3):748-756. doi: 10.2337/dc20-0413 [DOI] [PMC free article] [PubMed] [Google Scholar]

[eoi250001r9] 9.Brown DM, Wykoff CC, Boyer D, et al. Evaluation of intravitreal aflibercept for the treatment of severe nonproliferative diabetic retinopathy: results from the PANORAMA randomized clinical trial. JAMA Ophthalmol. 2021;139(9):946-955. doi: 10.1001/jamaophthalmol.2021.2809 [DOI] [PMC free article] [PubMed] [Google Scholar]

[eoi250001r10] 10.Maturi RK, Glassman AR, Josic K, et al. ; DRCR Retina Network . Effect of intravitreous anti-vascular endothelial growth factor vs sham treatment for prevention of vision-threatening complications of diabetic retinopathy: the Protocol W randomized clinical trial. JAMA Ophthalmol. 2021;139(7):701-712. doi: 10.1001/jamaophthalmol.2021.0606 [DOI] [PMC free article] [PubMed] [Google Scholar]

[eoi250001r11] 11.Campochiaro PA, Marcus DM, Awh CC, et al. The Port Delivery System with ranibizumab for neovascular age-related macular degeneration: results from the randomized phase 2 ladder clinical trial. Ophthalmology. 2019;126(8):1141-1154. doi: 10.1016/j.ophtha.2019.03.036 [DOI] [PubMed] [Google Scholar]

[eoi250001r12] 12.Pieramici DJ, Pearce I, Campochiaro PA, et al. Technical updates to Port Delivery System with ranibizumab (PDS) and clinical experience with updated refill needle. Presented at: American Society of Retina Specialists 42nd Annual Meeting; July 18, 2024; Stockholm, Sweden. [Google Scholar]

[eoi250001r13] 13.Susvimo. Package insert. Genentech, Inc; 2025. [Google Scholar]

[eoi250001r14] 14.World Medical Association . WMA Declaration of Helsinki—ethical principles for medical research involving human subjects. Accessed April 16, 2024. https://www.wma.net/policies-post/wma-declaration-of-helsinki/

[eoi250001r15] 15.European Medicines Agency . ICH harmonised guideline integrated addendum to ICH E6(R1): guideline for good clinical practice ICH E6(R2). Accessed April 16, 2024. https://ichgcp.net/

[eoi250001r16] 16.American Diabetes Association . Standards of care in diabetes—2023 abridged for primary care providers. Clin Diabetes. 2022;41(1):4-31. doi: 10.2337/cd23-as01 [DOI] [PMC free article] [PubMed] [Google Scholar]

[eoi250001r17] 17.World Health Organization . Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia: report of a WHO/IDF consultation. Updated January 31, 2024. Accessed April 16, 2024. https://www.who.int/publications/i/item/definition-and-diagnosis-of-diabetes-mellitus-and-intermediate-hyperglycaemia

[eoi250001r18] 18.Khanani AM, Graff JM, Marcus DM, et al. Refill-exchange procedure of the Port Delivery System with ranibizumab: overview and clinical trial experience. Ophthalmic Surg Lasers Imaging Retina. 2022;53(5):257-265. doi: 10.3928/23258160-20220412-01 [DOI] [PubMed] [Google Scholar]

[eoi250001r19] 19.Pieramici DJ, Wieland MR, Stewart JM, et al. Implant insertion procedure of the Port Delivery System with ranibizumab: overview and clinical pearls. Ophthalmic Surg Lasers Imaging Retina. 2022;53(5):249-256. doi: 10.3928/23258160-20220408-01 [DOI] [PubMed] [Google Scholar]

[eoi250001r20] 20.Eylea (aflibercept) injection. Prescribing information. Regeneron Pharmaceuticals Inc; 2023.

[eoi250001r21] 21.Westfall PH, Krishen A. Optimally weighted, fixed sequence and gatekeeper multiple testing procedures. J Stat Plan Inference. 2001;99(1):25-40. doi: 10.1016/S0378-3758(01)00077-5 [DOI] [Google Scholar]

[eoi250001r22] 22.Maturi RK, Glassman AR, Josic K, et al. ; DRCR Retina Network . Four-year visual outcomes in the Protocol W randomized trial of intravitreous aflibercept for prevention of vision-threatening complications of diabetic retinopathy. JAMA. 2023;329(5):376-385. doi: 10.1001/jama.2022.25029 [DOI] [PMC free article] [PubMed] [Google Scholar]

[eoi250001r23] 23.Kempen JH, O’Colmain BJ, Leske MC, et al. ; Eye Diseases Prevalence Research Group . The prevalence of diabetic retinopathy among adults in the United States. Arch Ophthalmol. 2004;122(4):552-563. doi: 10.1001/archopht.122.4.552 [DOI] [PubMed] [Google Scholar]

[eoi250001r24] 24.Best AL, Fajnkuchen F, Nghiem-Buffet S, et al. Treatment efficacy and compliance in patients with diabetic macular edema treated with ranibizumab in a real-life setting. J Ophthalmol. 2018;2018:4610129. doi: 10.1155/2018/4610129 [DOI] [PMC free article] [PubMed] [Google Scholar]

[eoi250001r25] 25.Ehlken C, Helms M, Böhringer D, Agostini HT, Stahl A. Association of treatment adherence with real-life VA outcomes in AMD, DME, and BRVO patients. Clin Ophthalmol. 2017;12:13-20. doi: 10.2147/OPTH.S151611 [DOI] [PMC free article] [PubMed] [Google Scholar]

[eoi250001r26] 26.Holekamp NM, Campbell J, Almony A, et al. Vision outcomes following anti–vascular endothelial growth factor treatment of diabetic macular edema in clinical practice. Am J Ophthalmol. 2018;191:83-91. doi: 10.1016/j.ajo.2018.04.010 [DOI] [PubMed] [Google Scholar]

[eoi250001r27] 27.Weiss M, Sim DA, Herold T, et al. Compliance and adherence of patients with diabetic macular edema to intravitreal anti-vascular endothelial growth factor therapy in daily practice. Retina. 2018;38(12):2293-2300. doi: 10.1097/IAE.0000000000001892 [DOI] [PubMed] [Google Scholar]

[eoi250001r28] 28.Wubben TJ, Johnson MW; Anti-VEGF Treatment Interruption Study Group . Anti-vascular endothelial growth factor therapy for diabetic retinopathy: consequences of inadvertent treatment interruptions. Am J Ophthalmol. 2019;204:13-18. doi: 10.1016/j.ajo.2019.03.005 [DOI] [PubMed] [Google Scholar]

PERMALINK

Port Delivery System With Ranibizumab vs Monitoring in Nonproliferative Diabetic Retinopathy Without Macular Edema

Dante J Pieramici, MD

Carl C Awh, MD

Margaret Chang, MD

Andres Emanuelli, MD

Nancy M Holekamp, MD

Allen Y Hu, MD

Ivan J Suñer, MD, MBA

Charles C Wykoff, MD, PhD

Christopher Brittain, MD

Dena Howard, PhD

Carlos Quezada-Ruiz, MD

Anjana Santhanakrishnan, BPharm (Hons), RPh

Paul Latkany, MD

Key Points

Question

Findings

Meaning

Abstract

Importance

Objective

Design, Setting, and Participants

Intervention

Main Outcomes and Measures

Results

Conclusions and Relevance

Trial Registration

Introduction

Methods

Trial Design and Oversight

Trial Population

Trial Procedures

Outcomes

Statistical Analysis

Results

Trial Population

Figure 1. Pavilion Randomized Clinical Trial Participant Allocation and Disposition.

Efficacy

Figure 2. Proportion of Participants With at Least a 2-Step Improvement From Baseline to Week 52 on the Early Treatment Diabetic Retinopathy Study Diabetic Retinopathy Severity Scale (ETDRS-DRSS).

Figure 3. Rate of Participants Developing Center-Involved Diabetic Macular Edema (CI-DME) and/or Proliferative Diabetic Retinopathy (PDR) or Anterior Segment Neovascularization (ASNV) in the Study Eye Through Week 52.

Figure 4. Vision and Anatomic Outcomes in the Study Eye.

Safety

Table. Ocular Adverse Events of Special Interest (AESIs) in the Safety-Evaluable Population Through Week 52 in the Study Eyes Receiving the Port Delivery System With Ranibizumab, 100 mg/mL, Every 36 Weeks.

Discussion

Limitations

Conclusions

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases