[Skip to Navigation]
Sign In
Visual Abstract. Effect of Intravitreal Aflibercept vs Laser Photocoagulation on Treatment Success of Retinopathy of Prematurity
Effect of Intravitreal Aflibercept vs Laser Photocoagulation on Treatment Success of Retinopathy of Prematurity
Figure.  Screening, Randomization, and Follow-up in the FIREFLEYE Trial of Intravitreal Aflibercept for the Treatment of Retinopathy of Prematurity
Screening, Randomization, and Follow-up in the FIREFLEYE Trial of Intravitreal Aflibercept for the Treatment of Retinopathy of Prematurity

FIREFLEYE indicates Aflibercept for ROP—IVT Injection versus Laser Therapy.

aSee eMethods in Supplement 3 for additional details of inclusion and exclusion criteria.

bOne infant with retinopathy of prematurity only in zone III was screened but not randomized.

cRandomization was stratified by retinopathy of prematurity category (zones) and country of enrollment. Randomization and evaluation was by infant with each infant demonstrating retinopathy in 1 eye or both eyes.

dOne infant was discontinued from the trial after an adverse event of sinus tachycardia.

eOne infant was discontinued from the trial after an adverse event of retinal detachment.

Table 1.  Baseline Demographics and Characteristics
Baseline Demographics and Characteristics
Table 2.  Primary Outcome of Treatment Success Rate Based on Bayesian Analyses Overall and in Prespecified Subgroups
Primary Outcome of Treatment Success Rate Based on Bayesian Analyses Overall and in Prespecified Subgroups
Table 3.  Secondary Outcomesa
Secondary Outcomesa
Table 4.  Additional Secondary Outcomes
Additional Secondary Outcomes
1.
Fierson  WM; American Academy of Pediatrics Section on Ophthalmology; American Academy of Ophthalmology; American Association for Pediatric Ophthalmology and Strabismus; American Association of Certified Orthoptists.  Screening examination of premature infants for retinopathy of prematurity.   Pediatrics. 2018;142(6):e20183061. doi:10.1542/peds.2018-3061PubMedGoogle ScholarCrossref
2.
Blencowe  H, Lawn  JE, Vazquez  T, Fielder  A, Gilbert  C.  Preterm-associated visual impairment and estimates of retinopathy of prematurity at regional and global levels for 2010.   Pediatr Res. 2013;74(suppl 1):35-49. doi:10.1038/pr.2013.205PubMedGoogle ScholarCrossref
3.
Early Treatment for Retinopathy of Prematurity Cooperative Group.  Revised indications for the treatment of retinopathy of prematurity: results of the Early Treatment for Retinopathy of Prematurity randomized trial.   Arch Ophthalmol. 2003;121(12):1684-1694. doi:10.1001/archopht.121.12.1684PubMedGoogle ScholarCrossref
4.
Good  WV; Early Treatment for Retinopathy of Prematurity Cooperative Group.  Final results of the Early Treatment for Retinopathy of Prematurity (ETROP) randomized trial.   Trans Am Ophthalmol Soc. 2004;102:233-248.PubMedGoogle Scholar
5.
Tan  QQ, Christiansen  SP, Wang  J.  Development of refractive error in children treated for retinopathy of prematurity with anti-vascular endothelial growth factor (anti-VEGF) agents: a meta-analysis and systematic review.   PLoS One. 2019;14(12):e0225643. doi:10.1371/journal.pone.0225643PubMedGoogle ScholarCrossref
6.
Lashkari  K, Hirose  T, Yazdany  J, McMeel  JW, Kazlauskas  A, Rahimi  N.  Vascular endothelial growth factor and hepatocyte growth factor levels are differentially elevated in patients with advanced retinopathy of prematurity.   Am J Pathol. 2000;156(4):1337-1344. doi:10.1016/S0002-9440(10)65004-3PubMedGoogle ScholarCrossref
7.
Sukgen  EA, Koçluk  Y.  Comparison of clinical outcomes of intravitreal ranibizumab and aflibercept treatment for retinopathy of prematurity.   Graefes Arch Clin Exp Ophthalmol. 2019;257(1):49-55. doi:10.1007/s00417-018-4168-5PubMedGoogle ScholarCrossref
8.
Salman  AG, Said  AM.  Structural, visual and refractive outcomes of intravitreal aflibercept injection in high-risk prethreshold type 1 retinopathy of prematurity.   Ophthalmic Res. 2015;53(1):15-20. doi:10.1159/000364809PubMedGoogle ScholarCrossref
9.
Huang  CY, Lien  R, Wang  NK,  et al.  Changes in systemic vascular endothelial growth factor levels after intravitreal injection of aflibercept in infants with retinopathy of prematurity.   Graefes Arch Clin Exp Ophthalmol. 2018;256(3):479-487. doi:10.1007/s00417-017-3878-4PubMedGoogle ScholarCrossref
10.
Mintz-Hittner  HA, Kennedy  KA, Chuang  AZ; BEAT-ROP Cooperative Group.  Efficacy of intravitreal bevacizumab for stage 3+ retinopathy of prematurity.   N Engl J Med. 2011;364(7):603-615. doi:10.1056/NEJMoa1007374PubMedGoogle ScholarCrossref
11.
Stahl  A, Lepore  D, Fielder  A,  et al.  Ranibizumab versus laser therapy for the treatment of very low birthweight infants with retinopathy of prematurity (RAINBOW): an open-label randomised controlled trial.   Lancet. 2019;394(10208):1551-1559. doi:10.1016/S0140-6736(19)31344-3PubMedGoogle ScholarCrossref
12.
International Committee for the Classification of Retinopathy of Prematurity.  The International Classification of Retinopathy of Prematurity revisited.   Arch Ophthalmol. 2005;123(7):991-999. doi:10.1001/archopht.123.7.991PubMedGoogle ScholarCrossref
13.
Smith  LEH, Hellström  A, Stahl  A,  et al; Retinopathy of Prematurity Workgroup of the International Neonatal Consortium.  Development of a Retinopathy of Prematurity Activity Scale and clinical outcome measures for use in clinical trials.   JAMA Ophthalmol. 2019;137(3):305-311. doi:10.1001/jamaophthalmol.2018.5984PubMedGoogle ScholarCrossref
14.
Hu  J, Blair  MP, Shapiro  MJ, Lichtenstein  SJ, Galasso  JM, Kapur  R.  Reactivation of retinopathy of prematurity after bevacizumab injection.   Arch Ophthalmol. 2012;130(8):1000-1006. doi:10.1001/archophthalmol.2012.592PubMedGoogle ScholarCrossref
15.
Snyder  LL, Garcia-Gonzalez  JM, Shapiro  MJ, Blair  MP.  Very late reactivation of retinopathy of prematurity after monotherapy with intravitreal bevacizumab.   Ophthalmic Surg Lasers Imaging Retina. 2016;47(3):280-283. doi:10.3928/23258160-20160229-12PubMedGoogle ScholarCrossref
16.
Wong  RK, Hubschman  S, Tsui  I.  Reactivation of retinopathy of prematurity after ranibizumab treatment.   Retina. 2015;35(4):675-680. doi:10.1097/IAE.0000000000000578PubMedGoogle ScholarCrossref
17.
Chiang  MF, Quinn  GE, Fielder  AR,  et al.  International Classification of Retinopathy of Prematurity, Third Edition.   Ophthalmology. 2021;128(10):e51-e68. doi:10.1016/j.ophtha.2021.05.031Google ScholarCrossref
Original Investigation
July 26, 2022

Effect of Intravitreal Aflibercept vs Laser Photocoagulation on Treatment Success of Retinopathy of Prematurity: The FIREFLEYE Randomized Clinical Trial

Author Affiliations
  • 1Department of Ophthalmology, University Medicine Greifswald, Greifswald, Germany
  • 2Department of Ophthalmology, Adana City Training and Research Hospital, Adana, Turkey
  • 3Department of Ophthalmology, Linkou Chang Gung Memorial Hospital, and College of Medicine, Chang Gung University, Taoyuan, Taiwan
  • 4Department of Geriatrics and Neuroscience, Catholic University of the Sacred Heart, A. Gemelli Foundation IRCSS, Rome, Italy
  • 5Research and Development Center for New Medical Frontiers, Department of Advanced Medicine, Division of Neonatal Intensive Care Medicine, Kitasato University School of Medicine, Sagamihara, Japan
  • 6Department of Neonatology, Poznan University of Medical Sciences, Poznan, Poland
  • 7Byers Eye Institute, Stanford University School of Medicine, Stanford, California
  • 8Regeneron Pharmaceuticals, Tarrytown, New York
  • 9Bayer AG, Wuppertal, Germany
  • 10Bayer Inc, Toronto, Ontario, Canada
  • 11Bayer AG, Berlin, Germany
  • 12Bayer Consumer Care AG, Basel, Switzerland
  • 13Department of Ophthalmology, National Center for Child Health and Development, Tokyo, Japan
JAMA. 2022;328(4):348-359. doi:10.1001/jama.2022.10564
Key Points

Question  Among preterm infants with retinopathy of prematurity requiring treatment, is treatment with a 0.4-mg dose of intravitreal aflibercept noninferior to laser photocoagulation?

Findings  In this noninferiority randomized clinical trial that included 113 treated infants, treatment success (measured as the proportion of infants without active retinopathy of prematurity and unfavorable structural outcomes at week 24) was 85.5% with intravitreal aflibercept compared with 82.1% with laser photocoagulation, a difference that did not meet the noninferiority margin of 5%.

Meaning  Among infants with retinopathy of prematurity, intravitreal aflibercept vs laser photocoagulation did not meet criteria for noninferiority with respect to treatment success at week 24, and further data would be required for more definitive conclusions regarding the comparative effects of these treatment approaches.

Abstract

Importance  Laser photocoagulation, which is the standard treatment for retinopathy of prematurity (ROP), can have adverse events. Studies of anti–vascular endothelial growth factor injections have suggested efficacy in the treatment of ROP, but few studies have directly compared them with laser treatments.

Objective  To compare intravitreal aflibercept vs laser photocoagulation in infants with ROP requiring treatment.

Design, Setting, and Participants  This noninferiority, phase 3, 24-week, randomized clinical trial was conducted in 27 countries (64 hospital sites) throughout Asia, Europe, and South America. Overall, 118 infants (gestational age ≤32 weeks at birth or birth weight ≤1500 g) with ROP severity (zone I stage 1+ [stage 1 plus increased disease activity], zone I stage 2+, zone I stage 3, zone I stage 3+, zone II stage 2+, or zone II stage 3+) requiring treatment or with aggressive posterior ROP in at least 1 eye were enrolled between September 25, 2019, and August 28, 2020 (the last visit occurred on February 12, 2021).

Interventions  Infants were randomized 2:1 to receive a 0.4-mg dose of intravitreal aflibercept (n = 75) or laser photocoagulation (n = 43) at baseline. Additional treatment was allowed as prespecified.

Main Outcomes and Measures  The primary outcome was the proportion of infants without active ROP and unfavorable structural outcomes 24 weeks after starting treatment (assessed by investigators). The requirement for rescue treatment was considered treatment failure. Intravitreal aflibercept was deemed noninferior if the lower limit of the 1-sided 95% bayesian credible interval for the treatment difference was greater than −5%.

Results  Among 118 infants randomized, 113 were treated (mean gestational age, 26.3 [SD, 1.9] weeks; 53 [46.9%] were female; 16.8% had aggressive posterior ROP, 19.5% had zone I ROP, and 63.7% had zone II ROP) and 104 completed the study. Treatment (intravitreal aflibercept: n = 75; laser photocoagulation: n = 38) was mostly bilateral (92.9%), and 82.2% of eyes in the intravitreal aflibercept group received 1 injection per eye. Treatment success was 85.5% with intravitreal aflibercept vs 82.1% with laser photocoagulation (between-group difference, 3.4% [1-sided 95% credible interval, −8.0% to ∞]). Rescue treatment was required in 4.8% (95% CI, 1.9% to 9.6%) of eyes in the intravitreal aflibercept group vs 11.1% (95% CI, 4.9% to 20.7%) of eyes in the laser photocoagulation group. The serious adverse event rates were 13.3% (ocular) and 24.0% (systemic) in the intravitreal aflibercept group compared with 7.9% and 36.8%, respectively, in the laser photocoagulation group. Three deaths, which occurred 4 to 9 weeks after intravitreal aflibercept treatment, were considered unrelated to aflibercept by the investigators.

Conclusions and Relevance  Among infants with ROP, intravitreal aflibercept compared with laser photocoagulation did not meet criteria for noninferiority with respect to the primary outcome of the proportion of infants achieving treatment success at week 24. Further data would be required for more definitive conclusions regarding the comparative effects of intravitreal aflibercept and laser photocoagulation in this population.

Trial Registration  ClinicalTrials.gov Identifier: NCT04004208

Introduction

Retinopathy of prematurity (ROP) is a disorder of the developing retinal blood vessels in preterm infants.1 There are 5 stages of ROP indicating disease severity from stage 1 (demarcation line) to stage 5 (total retinal detachment). The addition of the plus sign describes increased disease activity in which retinal blood vessels at the ocular posterior pole are particularly dilated and tortuous. Most cases of ROP are mild and regress spontaneously and the cases warranting treatment are rare. However, once the need for treatment is identified, timely action is required to prevent retinal detachment with consequent vision impairment and loss.2

Laser photocoagulation of the avascular retina has been the standard treatment of ROP,3,4 with the aim of removing the source of excessive angiogenic stimulus originating from the avascular peripheral retina. However, with its tissue destructive nature, laser photocoagulation has been associated with negative sequelae, including irreversible visual field loss and a high level of myopia.5 Upregulation of vascular endothelial growth factor (VEGF), and the role of VEGF in promoting abnormal vascular proliferation during the vasoproliferative phase of ROP, has been well established.6 This has led to the investigation of anti-VEGF agents for the treatment of ROP.

Preliminary evidence from retrospective case series and small-scale exploratory studies with anti-VEGF agents administered intravitreally, including aflibercept, have suggested a favorable benefit-risk profile overall.7-9 Two larger-scale randomized clinical trials have investigated the treatment of ROP with anti-VEGF agents compared with laser photocoagulation.10,11 The BEAT-ROP (Bevacizumab Eliminates the Angiogenic Threat of Retinopathy of Prematurity) trial10 was conducted in the US and assessed only stage 3+ ROP in zone I and posterior zone II. The multinational RAINBOW (Ranibizumab Compared With Laser Therapy for the Treatment of Infants Born Prematurely With Retinopathy of Prematurity) trial11 assessed patients bilaterally treated for aggressive posterior ROP; zone I stage 1+, zone I stage 2+, zone I stage 3, or zone I stage 3+; and zone II stage 3+.

The FIREFLEYE (Aflibercept for ROP—IVT Injection versus Laser Therapy) trial aimed to evaluate whether a 0.4-mg dose of intravitreal aflibercept was noninferior to laser photocoagulation in preterm infants with severe ROP.

Methods
Trial Design and Oversight

This was a noninferiority, phase 3, 24-week, open-label randomized clinical trial assessing the efficacy and adverse event profile of intravitreal aflibercept vs laser photocoagulation in infants with ROP requiring treatment. Patients were enrolled between September 25, 2019, and August 28, 2020 (the last treatment visit occurred on February 12, 2021), at 64 hospital sites in 27 countries across Asia, Europe, and South America. The trial protocol (appears in Supplement 1) was reviewed and approved by each site’s independent ethics committee or institutional review board before study initiation. Written informed consent by the parents or legally authorized representatives was obtained before enrollment. An independent data monitoring committee assessed study progress and patient safety. The statistical analysis plan appears in Supplement 2.

Study Population

Eligible infants (1) were born at a gestational age of 32 weeks or younger or had a birth weight of 1500 g or less, (2) weighed 800 g or greater at the time of treatment, and (3) had ROP within the spectrum of severity requiring treatment (zone I stage 1+, zone I stage 2+, zone I stage 3, zone I stage 3+, zone II stage 2+, zone II stage 3+, or aggressive posterior ROP [according to the International Classification of Retinopathy of Prematurity12] in at least 1 eye). The most posterior region, zone I, is a circle with a radius twice the estimated distance from the optic disc center to the foveal center.12 Zone II is a ring-shaped region extending nasally from the outer limit of zone I to the nasal ora serrata, with a similar distance temporally, superiorly, and inferiorly.12

Retinopathy of prematurity in zone I is more likely to progress and become more severe than ROP in zone II.12 Staging is defined by the appearance of a structure at the vascular-avascular juncture: demarcation line (stage 1), ridge (stage 2), and extraretinal neovascular proliferation or flat neovascularization (stage 3).12 If more than 1 ROP stage was present, the eye was classified by the most severe stage.12 Plus disease is defined by the appearance of dilation and tortuosity of the retinal vessels.12 Aggressive posterior ROP was used to describe a severe, rapidly progressive form of ROP in posterior zones I or II.12

Infants were not eligible if they had (1) ROP stages 4 or 5 or ROP involving only zone III, (2) active ocular infection within 5 days before investigational treatment, or (3) neurological comorbidities potentially confounding visual function.

Information on race was collected as part of standard demographic parameters, and classified by the investigator in fixed predefined categories (African American or Black, American Indian or Alaska Native, Asian, and White), considering information from medical records and parents. The full inclusion and exclusion criteria appear in the eMethods in Supplement 3.

Randomization

Eligible infants were randomized 2:1 to a 0.4-mg dose of intravitreal aflibercept or laser photocoagulation (Figure), stratified by ROP category (zone I [excluding aggressive posterior ROP], zone II [excluding aggressive posterior ROP], and aggressive posterior ROP) and country of enrollment (sites in and outside Japan to meet regulatory expectations). Allocation was conducted via a computer-generated randomization list. Additional details appear in the eMethods in Supplement 3.

Procedures

Infants received either a single 0.4-mg dose of intravitreal aflibercept or transpupillary conventional laser photocoagulation applied to the entire avascular peripheral retina for each eye requiring treatment at baseline. Infants were seen on the day after treatment, at least weekly during the first month, every 2 weeks until week 12, and at least monthly thereafter. Investigators examined the retina and evaluated ROP using indirect ophthalmoscopy and digital wide-field retinal imaging. In addition, images were transmitted to an external reading center. The criteria for retreatment (defined as receiving the same modality as the one administered at baseline) and rescue treatment (defined as receiving laser photocoagulation in the intravitreal aflibercept group or receiving intravitreal aflibercept in the laser photocoagulation group) are further described in the Box.

Box Section Ref ID
Box.

Descriptions of End Points and Definitions for the Terms Used in the FIREFLEYE Trial

Descriptions of End Points
Primary Outcome
  • The primary outcome was the proportion of infants without active retinopathy of prematurity (ROP) and unfavorable structural outcomes 24 weeks after starting treatment (assessed by the investigators).

  • One eye or both eligible eyes of an infant were included to determine the primary outcome (primarily in the full analysis set) by randomized treatment group. If both eyes were treated, both needed to respond for the primary outcome to be met. Infants who required rescue treatment did not meet the primary outcome.

Secondary Efficacy Outcomes
  • The proportion of infants requiring an intervention with a second treatment modality (either rescue treatment as defined below and in the trial protocol in Supplement 1 or any other surgical or nonsurgical treatment for ROP) from baseline to week 24.

  • The proportion of infants with recurrence of ROP until week 24 (ie, the need for retreatment or rescue therapy after prior improvement to a disease stage not requiring treatment).a

  • The number of intravitreal aflibercept doses (0.4-mg dose injections) and the number of laser photocoagulation treatments received from baseline to week 24. In cases in which multiple sessions of laser photocoagulation were necessary within 1 week from baseline, these sessions were counted as a single treatment.

  • The distribution and changes from baseline over time in the Retinopathy of Prematurity Activity Scale scores and the 3 subcategories (mild, moderate, or severe) were determined based on the ROP grading by the central reading center and the number and percentage of eyes with at least a 2-point decrease by the visit. Retinopathy of Prematurity Activity Scale scores range from 0 to 22 (0-7 is considered mild; 8-12, moderate; and 13-22, severe).

Exploratory Efficacy Outcomes
  • The duration of the treatment procedure included the time required to administer intravitreal aflibercept or perform laser photocoagulation per eye and per infant and was calculated by a stop time minus a start time.

  • The type of anesthesia included general, sedation, local, or other required during administration of the study interventions.

  • The time between the 2 treatments was measured for infants who required retreatment and who received the same treatment (intravitreal aflibercept or laser photocoagulation both times).

Definitions for the Terms Used
Active ROP
  • Defined as ROP requiring treatment.

Unfavorable Structural Outcomes
  • Defined as retinal detachment, macular dragging, macular fold, or retrolental opacity.

Second Treatment Modality
  • Defined as either rescue treatment or treatment with any other surgical or nonsurgical treatment for ROP (eg, intravitreal anti–vascular endothelial growth factor injection, ablative laser therapy, cryotherapy, or vitrectomy) captured as concomitant medication or surgeries after baseline treatment.

Rescue Treatment
  • Defined as additional treatment with laser photocoagulation for the intravitreal aflibercept group and additional treatment with intravitreal aflibercept for the laser photocoagulation group.

  • Rescue treatment with a 0.4-mg dose of intravitreal aflibercept was permitted if the fundus examination revealed the original treatment with laser photocoagulation was complete (as judged by the investigator) and if there was either (1) a worsening of ROP compared with the examination prior to the infant receiving laser photocoagulation or (2) ROP persisted and required treatment 28 days or longer after the infant received treatment with laser photocoagulation.

  • Rescue treatment with laser photocoagulation was permitted if there was either (1) worsening of ROP compared with the examination before the previous 0.4-mg dose during the 27 days after receiving intravitreal aflibercept or (2) ROP was present and required treatment after a total of 3 doses of intravitreal aflibercept and the interval since the last dose was 28 days or longer.

  • Rescue treatment was considered treatment failure.

Retreatment
  • For the infants randomized to receive a 0.4-mg dose of intravitreal aflibercept, up to 2 additional 0.4-mg doses could be administered in each eye if (1) there was presence of ROP requiring treatment and (2) the interval since the last dose of intravitreal aflibercept was 28 days or longer.

  • For the infants randomized to receive laser photocoagulation, retreatment with laser photocoagulation could be administered if (1) there was presence of ROP requiring treatment and (2) the fundus examination revealed the prior treatment was incomplete (as judged by the investigator). When multiple sessions were necessary within 1 week from baseline, they were counted as a single treatment. Thereafter, supplementary treatments with laser photocoagulation were considered as retreatment.

  • Retreatment did not constitute treatment failure.

FIREFLEYE indicates Aflibercept for ROP—IVT Injection versus Laser Therapy.

a The term recurrence of ROP activity was used in the current trial as well as in the BEAT-ROP (Bevacizumab Eliminates the Angiogenic Threat of Retinopathy of Prematurity) trial10 and in the RAINBOW (Ranibizumab Compared With Laser Therapy for the Treatment of Infants Born Prematurely With Retinopathy of Prematurity) trial.11 The third edition of the International Classification of Retinopathy of Prematurity17 recommends the use of the term reactivation referring to the recurrence of acute phase features.

Study Outcomes

The primary outcome was the proportion of infants without active ROP and unfavorable structural outcomes (retinal detachment, macular dragging, macular fold, or retrolental opacity) 24 weeks after starting treatment (assessed by investigators; Box). If both eyes were treated, both needed to respond for the primary outcome to be met. Eyes were considered nonresponders if rescue treatment was given.

The secondary outcomes addressing the primary efficacy objective (Box) were the requirement for intervention with a second treatment modality from baseline to week 24, any recurrence of ROP from baseline to week 24, and disease severity based on Retinopathy of Prematurity Activity Scale13 score (measured on a scale of 0-22; scores from 0-7 are considered mild; 8-12, moderate; and 13-22, severe) as proposed by the International Neonatal Consortium. Further secondary outcomes were the number of doses of intravitreal aflibercept administered and the number of laser photocoagulation treatments from baseline to week 24. In addition, exploratory outcomes included the duration of the treatment procedure and the type of anesthesia used. A post hoc outcome was the interval between treatments.

Safety assessments included ophthalmic examinations, physical examinations (including growth parameters), vital signs, laboratory evaluations, and central nervous system imaging. The proportion of infants with ocular and systemic treatment-emergent adverse events and serious adverse events by week 24 was evaluated. The treatment-emergent adverse events were defined as those adverse events observed or reported after the first administration of treatment and no later than 30 days after the last administration of treatment. Adverse events were reported by the treating physician, the caregiver or surrogate, or the infants’ legally authorized representative. Adverse events were assessed for causality by investigators.

The secondary outcome of systemic exposure to free (ie, pharmacologically active) aflibercept in plasma (sparse sampling) on day 1 and during weeks 2, 4, 8, 12, and 24 after the first administration of treatment was detected using a validated enzyme-linked immunosorbent assay with a lower level of quantitation of 15.6 ng/mL. The trial protocol was amended to collect sampling beyond week 4 to week 24 (Supplement 1). The presence of antidrug antibodies before and 12 weeks after receiving intravitreal aflibercept was an additional secondary outcome.

Sample Size Calculation

The sample size rationale was based on the primary efficacy outcome and the success criterion that the response probability for intravitreal aflibercept was greater than that for laser photocoagulation minus 5%, with at least 95% posterior probability. The sample size was determined via simulation, assuming the response probability for laser photocoagulation was similar to historical data10,11 (expressed by a beta distribution with the parameters 34.7 and 13.8, which is centered on 72%), and the response probability for intravitreal aflibercept was 15% higher than that for laser photocoagulation (but not >95%). A similar difference also has been reported for ranibizumab.11 With at least 102 infants evaluable for the primary analysis and a 2:1 randomization ratio, the defined success criterion would be achieved with a power of 81%.

Statistical Analysis

The primary outcome was analyzed using bayesian statistical methods and a bivariate binomial model accounting for the correlation between the infant’s 2 eyes. Noninformative prior distributions were used for the response probability for a single eye for both treatment groups, and an informative prior distribution allowing only positive values for the correlation coefficient. The parameter of interest was the probability for a response in both eyes. The primary analysis was performed on the full analysis set (infants who received study treatment and had assessments of efficacy at baseline and at least once after baseline) according to randomized treatment assignment. Missing data for week 24 for a treated eye was imputed by the last available response status at week 16 or later, if available, with a multiple imputation approach otherwise (eMethods in Supplement 3). Intravitreal aflibercept was deemed to be noninferior to laser photocoagulation, with a noninferiority margin of 5%, if the lower limit of the 1-sided 95% bayesian credible interval (CrI) for the treatment difference was above −5%.

A noninferiority success criterion was considered appropriate because differentiation between laser photocoagulation and anti-VEGF approaches in unfavorable structural outcomes has not been consistently demonstrated when treating patients with ROP in zone II. Laser photocoagulation requires burning and destruction of the retina where vessels have not yet developed and is associated with development of a high level of myopia, whereas anti-VEGF agents allow further development of retinal vascularization and potential future functional benefits. The proposed noninferiority margin of 5% is smaller than the smallest difference between the laser photocoagulation and ranibizumab groups in the RAINBOW trial,11 and not greater than the difference between the 2 ranibizumab doses. Treatment differences of 5% or less were assumed to not relevantly change a treatment decision.

The estimated treatment success rates are expressed as medians of the posterior distributions (with 90% CrIs) as prespecified in the statistical analysis plan (Supplement 2).

The sensitivity analyses were prespecified to evaluate the effect of missing data and to analyze the primary outcome according to traditional frequentist methods. In addition, a post hoc analysis evaluated a potential investigator site effect by including a random site effect into the model.

The secondary outcomes of (1) requirement for an intervention with a second treatment modality from baseline to week 24 and (2) recurrence of ROP from baseline to week 24 were analyzed descriptively and with a similar bayesian statistical model as the primary outcome. The remaining secondary efficacy outcomes (Retinopathy of Prematurity Activity Scale score, the number of doses of intravitreal aflibercept administered, and the number of laser photocoagulation treatments administered) were analyzed descriptively only.

For primary and selected secondary efficacy outcomes, subgroup analyses were performed by ROP zones (zone I and zone II; each excluding aggressive posterior ROP) and by presence of aggressive posterior ROP. In addition, the exploratory and post hoc outcomes and the secondary outcomes for pharmacokinetics and antidrug antibodies were analyzed descriptively.

The statistical evaluations were performed using SAS version 9.4 (SAS Institute Inc).

Results
Study Population

Overall, 121 infants were screened, 118 were randomized, and 113 were treated. Enrollment by country appears in eTable 1 in Supplement 3. The reasons for study discontinuation appear in the Figure; the most common reason was withdrawal by a parent or guardian, which occurred in 6 randomized infants. Five infants in the laser photocoagulation group did not receive any study treatment because their parents became concerned after randomization for 4 of the infants and decided they preferred anti-VEGF injections and because the clinician decided laser photocoagulation treatment was not appropriate for 1 infant, given the infant’s medical instability. These 5 infants were excluded from the full analysis set, which comprised 113 treated infants (intravitreal aflibercept: n = 75; laser photocoagulation: n = 38). Of the treated infants, 68/75 (90.7%) in the intravitreal aflibercept group and 36/38 (94.7%) in the laser photocoagulation group completed the study.

Baseline demographics and characteristics appear in Table 1. The mean gestational age was 26.3 weeks (SD, 1.9 weeks) and the mean birth weight was 862.1 g (SD, 282.9 g). Aggressive posterior ROP (mostly zone I) was present in 16.8% of infants at baseline, ROP in zone I (without aggressive posterior ROP) was present in 19.5% of infants, and ROP in zone II (without aggressive posterior ROP) was present in 63.7% of infants. Treatment was mostly bilateral (92.9%) and 82.2% of eyes in the intravitreal aflibercept group received 1 injection per eye. Bilateral treatment was performed in a higher proportion of infants in the intravitreal aflibercept group (71/75 [94.7%]) vs the laser photocoagulation group (34/38 [89.5%]).

Efficacy
Primary Outcome

The bayesian-estimated treatment success rate with intravitreal aflibercept was 85.5% (90% CrI, 78.0% to 91.3%) compared with 82.1% (90% CrI, 70.5% to 90.8%) with laser photocoagulation (Table 2). The between-group difference was 3.4% (1-sided 95% CrI, −8.0% to ∞) in favor of intravitreal aflibercept. However, because the lower limit of the 95% CrI for the treatment difference was −8.0%, and not greater than the prespecified value of −5.0%, noninferiority could not be concluded.

Active ROP was found in 9 eyes (6.2%) in the intravitreal aflibercept group and in 2 eyes (2.8%) in the laser photocoagulation group. Unfavorable structural outcomes were found in 10 eyes (6.8%) in the intravitreal aflibercept group and in 4 eyes (5.6%) in the laser photocoagulation group; the most common outcome was retinal detachment (eTable 2 in Supplement 3).

The bayesian-estimated treatment success rate in the intravitreal aflibercept group was 92.0% (90% CrI, 84.1%-96.9%) in the 46 infants with zone II ROP excluding aggressive posterior ROP compared with 84.9% (90% CrI, 71.5%-94.0%) in the 26 infants in the laser photocoagulation group. The bayesian-estimated treatment success rate in the intravitreal aflibercept group was 70.8% (90% CrI, 50.7%-86.6%) in the 15 infants with zone I ROP excluding aggressive posterior ROP compared with 64.4% (90% CrI, 36.7%-87.1%) in the 7 infants in the laser photocoagulation group. The bayesian-estimated treatment success rate in the intravitreal aflibercept group was 73.3% (90% CrI, 52.1%-89.1%) in the 14 infants with aggressive posterior ROP compared with 72.2% (90% CrI, 40.0%-93.0%) in the 5 infants in the laser photocoagulation group. The treatment differences and the 1-sided 95% CrIs appear in Table 2.

The results of the sensitivity analyses and the post hoc analyses to assess the potential effect of missing data or adjusting for a potential investigator site effect were consistent with the primary analysis as was the frequentist analysis (eTables 3-5 in Supplement 3).

Secondary Outcomes

In the intravitreal aflibercept group, 17.8% of eyes received 1 retreatment (Figure). No infant received more than 2 doses (1 retreatment) per eye. In total, 8.2% of eyes in the intravitreal aflibercept group (10.7% of infants) received any second treatment modality, which was rescue treatment with laser photocoagulation in 4.8% (95% CI, 1.9%-9.6%) of eyes (6.7% of infants) (eTable 2 in Supplement 3). In the laser photocoagulation group, 9.7% of eyes were retreated more than 1 week after the baseline session (Figure). In total, 12.5% of eyes (13.2% of infants) in the laser photocoagulation group received any second treatment modality, which was rescue treatment with intravitreal aflibercept in 11.1% (95% CI, 4.9%-20.7%) of eyes (10.5% of infants) (eTable 2 in Supplement 3). Of the eyes receiving rescue treatment, the outcomes at 24 weeks appear in eTable 6 in Supplement 3.

Retinopathy of prematurity reactivation or recurrence up until week 24 occurred in 32/146 eyes (21.9%) in the intravitreal aflibercept group and in 6/72 eyes (8.3%) in the laser photocoagulation group. Results of the bayesian analysis of recurrence and requirement for a second treatment modality appears in Table 3. For recurrence, the between-group difference was 9.6% (90% CrI, 1.9%-17.5%).

The Retinopathy of Prematurity Activity Scale mean scores decreased from baseline in both treatment groups at each visit after the baseline visit (Table 4). At week 24, the mean change from baseline was –15.4 (from a score of 16.2 to 0.9) in the intravitreal aflibercept group and was –14.8 (from a score of 15.6 to 0.5) in the laser photocoagulation group (Table 4 and eFigure 1 in Supplement 3).

Exploratory Outcomes

The mean duration of administration for the injections was 3.7 (SD, 5.5) minutes per infant (median, 2.0 [IQR, 0-4.0] minutes per infant) or 1.9 minutes per eye in the intravitreal aflibercept group. The mean duration of the laser photocoagulation treatments was 121.5 (SD, 85.4) minutes per infant (median, 106.5 [IQR, 55.0-156.0] minutes per infant) or 64.1 minutes per eye in the laser photocoagulation group. The treatments were conducted under general anesthesia in 54.1% of infants in the intravitreal aflibercept group and in 84.7% of infants in the laser photocoagulation group (eTable 7 in Supplement 3).

Post Hoc Outcomes

The mean interval between the first and second dose of intravitreal aflibercept was 79.3 days (range, 29-121 days). In the laser photocoagulation group, 3 infants (7.9%) received their initial laser photocoagulation treatment performed in more than 1 session within the first week of baseline. As predefined in the trial protocol, these early retreatments were considered as a single treatment; therefore, 18.1% of eyes received more than 1 laser photocoagulation treatment overall. After the first laser photocoagulation treatment, subsequent laser photocoagulation treatments were given after a mean of 31.6 days (range, 15-48 days). For infants requiring rescue treatment, intravitreal aflibercept was given between day 4 and day 75 after receiving the initial laser photocoagulation treatment (mean, 34.3 [SD, 27.9] days; median, 29 [IQR, 13.0-55.5] days) (eTable 7 in Supplement 3).

Adverse Events

The serious adverse event rates were 13.3% (ocular) and 24.0% (systemic) in the intravitreal aflibercept group compared with 7.9% and 36.8%, respectively, in the laser photocoagulation group. Ocular treatment-emergent adverse events in the treated eyes were reported among 38.7% of infants in the intravitreal aflibercept group and 36.8% in the laser photocoagulation group (eTable 8 in Supplement 3). No endophthalmitis or any other intraocular inflammatory events were reported. Systemic treatment-emergent adverse events were reported in 52.0% of infants in the intravitreal aflibercept group and in 63.2% of infants in the laser photocoagulation group. The rate of any treatment-emergent serious adverse event was lower in the intravitreal aflibercept group (12.0%) than in the laser photocoagulation group (26.3%), primarily because of a lower rate of systemic serious adverse events in the intravitreal aflibercept group (6.7% vs 18.4% in the laser photocoagulation group). Treatment-emergent ocular serious adverse events were reported in 8.0% of infants in the intravitreal aflibercept group and in 7.9% of infants in the laser photocoagulation group. Causes of treatment-emergent serious adverse events appear in eTable 8 in Supplement 3.

Three deaths were reported overall (2.7%); all occurred in infants born before 27 weeks’ gestational age in the intravitreal aflibercept group. Among the reported causes of death, there was 1 each for bronchiolitis and bronchopulmonary dysplasia and 1 infant had both bronchopulmonary dysplasia and pneumothorax. Deaths occurred 4 to 9 weeks after intravitreal aflibercept treatment and were considered by the investigators and the sponsor to be unrelated to the study drug, but were instead considered related to complications of preterm birth (eTable 9 in Supplement 3). There were no clinically relevant between-group differences in the change from baseline to week 24 for body length, head circumference, weight, or blood pressure (eTable 10 in Supplement 3).

Pharmacokinetics

Plasma concentrations of free (ie, pharmacologically active) aflibercept declined to values below or close to the lower limit of quantification within approximately 8 weeks (eFigure 2 in Supplement 3). One infant (1.3%) who received a single 0.4-mg dose of intravitreal aflibercept in 1 eye had non–neutralizing antidrug antibodies (low titer, 1:30) 12 weeks after receiving the baseline treatment. Treatment was successful in this infant.

Discussion

In this noninferiority randomized clinical trial of infants with ROP, intravitreal aflibercept compared with laser photocoagulation did not meet the noninferiority margin of 5% with regard to treatment success. Overall, the study showed a response rate to intravitreal aflibercept within the expected range, considering randomized clinical trials of other anti-VEGF agents for the treatment of ROP.10,11

The fact that noninferiority of intravitreal aflibercept could not be demonstrated may be attributable to an observed laser photocoagulation response rate that was markedly higher than the treatment response to laser photocoagulation in the RAINBOW and BEAT-ROP trials.10,11 Power calculations using the laser photocoagulation response rate observed in the current trial would have necessitated a larger study population to demonstrate noninferiority given the smaller than expected difference between the laser photocoagulation and intravitreal aflibercept treatments. It is unlikely that the differences in the study population enrolled and the definition of the primary outcome or its analysis between the current trial, the BEAT-ROP trial,10 and particularly the RAINBOW trial,11 fully explain the discrepancies in laser photocoagulation performance across the 3 trials.

No major changes in laser photocoagulation technology were noted between the completion of the RAINBOW trial (conducted from 2015-2017) and the conduct of the current trial. In the RAINBOW trial,11 any laser photocoagulation treatment received 11 days after baseline was considered treatment failure. In the current trial, approximately 10% of both eyes and infants in the laser photocoagulation group received additional laser photocoagulation after 11 days, which was not considered treatment failure to meet the primary outcome.

Reactivation of ROP (termed recurrence in many studies) after anti-VEGF treatment has been commonly described.14-16 However, different end point definitions were used in different studies. In the RAINBOW trial,11 recurrence was defined as any ROP requiring treatment after baseline treatment. In contrast, in the current trial, recurrence was defined as the presence of initial disease improvement to a stage not requiring treatment, and subsequent worsening to require treatment again. The higher rate of recurrence in the intravitreal aflibercept group, primarily managed with a second dose as in clinical practice, compared with laser photocoagulation, may not necessarily imply that intravitreal aflibercept treatment is less efficacious than laser photocoagulation treatment.

The adverse event profile of a 0.4-mg dose of intravitreal aflibercept in infants with ROP was consistent with the established profile of a 2.0-mg dose of intravitreal aflibercept in adults, and no new concerns were identified in this vulnerable pediatric population. No cases of endophthalmitis or any other intraocular inflammatory events were reported. The 3 deaths in the study were considered to be causally unrelated to the trial drug or injection procedure, but rather attributable to complications of premature birth.

The current trial incorporated a robust design with an active control in its use of laser photocoagulation. The broad trial population (including all forms of ROP requiring treatment according to international classifications)17 and geographic reach (Asia, Europe, and South America) highlights the generalizability of the results to infants with ROP in clinical practice. Also, with the inclusion of ROP of stage 2+ in zone II and the option to include unilateral treatment, this was the first study, to our knowledge, to truly reflect all treatment scenarios that are recommended according to the criteria by the Early Treatment for Retinopathy of Prematurity Cooperative Group.3,4 Long-term ocular and overall clinical outcomes up to 5 years of chronological age are being evaluated in the ongoing extension study (final results are expected in 2026; NCT04015180).

Limitations

This study has several limitations. First, the trial had a relatively small sample size, although the size of the intravitreal aflibercept group is consistent with the anti-VEGF treatment groups in the RAINBOW trial.11 In particular, the study was powered for an effect in the overall population without accounting for potential differences in efficacy between the categories of ROP. Second, as with other anti-VEGF ROP studies,10,11 treatment masking was not possible due to visible laser photocoagulation effects and sham procedures would not have been considered ethical.

Conclusions

Among infants with ROP, intravitreal aflibercept compared with laser photocoagulation did not meet criteria for noninferiority with respect to the primary outcome of the proportion of infants achieving treatment success at week 24. Further data would be required for more definitive conclusions regarding the comparative effects of intravitreal aflibercept and laser photocoagulation in this population.

Back to top
Article Information

Corresponding Author: Andreas Stahl, MD, Department of Ophthalmology, University Medicine Greifswald, Ferdinand Sauerbruch Straße, 17475 Greifswald, Germany (andreas.stahl@med.uni-greifswald.de).

Correction: This article was corrected August 22, 2022, to fix the spelling of a coauthor’s name from "Evra Koefuencue" to "Evra Köfüncü" in the byline, author affiliations, author contributions, and conflict of interest disclosures.

Accepted for Publication: June 6, 2022.

Author Contributions: Dr Stahl 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: Stahl, Sukgen, Wu, Lepore, Nakanishi, Moshfeghi, Vitti, Athanikar, Chu, Iveli, Zhao, Schmelter, Leal, Köfüncü.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Stahl, Lepore, Moshfeghi, Vitti, Iveli, Schmelter, Köfüncü.

Critical revision of the manuscript for important intellectual content: Stahl, Sukgen, Wu, Lepore, Nakanishi, Mazela, Moshfeghi, Vitti, Athanikar, Chu, Zhao, Schmelter, Leal, Köfüncü, Azuma.

Statistical analysis: Schmelter.

Obtained funding: Vitti, Leal.

Administrative, technical, or material support: Stahl, Mazela, Vitti, Athanikar, Chu, Iveli, Zhao, Leal.

Supervision: Stahl, Sukgen, Wu, Lepore, Moshfeghi, Vitti, Chu, Iveli, Zhao, Leal, Köfüncü.

Conflict of Interest Disclosures: Dr Stahl reported serving on scientific advisory boards for Alcon, Apellis, Bayer, Novartis, and Roche; receiving speaker fees from Allergan, Bayer, and Novartis; and receiving grants and personal fees from Bayer and Novartis. Dr Sukgen reported receiving speaker fees from Allergan and receiving grants from Bayer, Novartis, and TRPharm. Dr Wu reported receiving personal fees from Allergan, Bayer, and Novartis. Dr Lepore reported receiving grants from Bayer and Novartis and serving as a paid member of the steering committee for the current trial. Dr Mazela reported receiving speaker fees from AbbVie, AstraZeneca, Draeger, HiPP, Maquet, Nestle, Nutricia, and Roche; receiving grants and personal fees from Bayer; and receiving grants from Merck Sharp & Dohme and Windtree Therapeutics. Dr Moshfeghi reported receiving personal fees from Akebia Therapeutics, Bayer, Novartis, Regeneron Pharmaceuticals, and the Shapiro Law Group; serving as a consultant to Akceso Advisors AG, Bayer, Congruence Medical Solutions, Ocular Surgery News, Praxis Inc, Retina Technologies LLC, Retina Today/Pentavision, Regeneron Pharmaceuticals, and SLACK Inc; receiving grants from Aldeyra Therapeutics, Apellis, Genentech, the National Institutes of Health, and Regeneron Pharmaceuticals; serving on a steering committee or scientific advisory or data and safety monitoring board for Akebia, Alcon, Allegro, Iconic Therapeutics, Irenix, Novartis, Pykus, and Visunex; having equity in DSentz, Grand Legend Technology, Linc, Pr3vent, Promisight, Pykus, VersI, and Visunex; serving on the board of directors for DSentz, Linc, Pr3vent, and Promisight; and receiving nonfinancial support from CMEOutfitters.com, Cole Eye Institute, Northwell Health, Prime Medical Education, Regeneron Pharmaceuticals, University of Miami, Vindico, and Visunex. Drs Vitti and Athanikar and Ms Chu reported being full-time employees of Regeneron Pharmaceuticals. Dr Athanikar reported owning stock in Regeneron Pharmaceuticals. Drs Iveli, Zhao, Schmelter, Leal, and Köfüncü reported being full-time employees of Bayer. Drs Schmelter and Leal reported owning stock in Bayer. Dr Azuma reported receiving grants from Bayer and Novartis. No other disclosures were reported.

Funding/Support: The FIREFLEYE study was sponsored by Bayer AG and co-funded by Regeneron Pharmaceuticals.

Role of the Funder/Sponsor: In collaboration with the authors and investigators, Bayer AG participated in the design and conduct of the study, collection, management, analysis, and interpretation of the data. Also in collaboration with the authors, the sponsor Bayer AG and the co-funder Regeneron Pharmaceuticals participated in the preparation, review, or approval of the manuscript and decision to submit the manuscript for publication. Bayer AG and Regeneron Pharmaceuticals did not have the right to veto publication or to control the decision regarding to which journal the manuscript was submitted.

Group Information: The FIREFLEYE Study Group appears in Supplement 4.

FIREFLEYE Steering Committee Members: Noriyuki Azuma, MD, PhD, Domenico Lepore, MD, Jan Mazela, MD, PhD, Hidehiko Nakanishi, MD, PhD, Andreas Stahl, MD, Emine Sukgen, MD, and Wei-Chi Wu, MD, PhD.

Meeting Presentations: Presented in part at the 21st European Society of Retina Specialists Congress; September 12, 2021; virtual presentation. Presented in part at Hot Topics in Neonatology; December 7, 2021; virtual presentation. Presented in part at the 14th Asia-Pacific Vitreo-retina Society Congress; December 12, 2021; virtual presentation. Presented in part at the 126th Annual Meeting of the Japanese Ophthalmological Society; April 16, 2022; Osaka. Presented in part at the Annual Meeting of the Association for Research in Vision and Ophthalmology; May 3, 2022; Denver, Colorado, and virtual presentation.

Data Sharing Statement: See Supplement 5.

Additional Contributions: We thank all the investigators, patients, and their parents who participated in the FIREFLEYE trial and particularly for their involvement during the unprecedented times of the global COVID-19 pandemic, which allowed the study to be conducted without interruption. Medical writing and editorial support was provided by Sarah Feeny, BMedSci (ApotheCom), and funded by Bayer Consumer Care AG Pharmaceuticals. We also acknowledge the contributions of Torsten Zimmermann, MD (Bayer), for clinical pharmacological support and Alexander Pieper, MSc (Chrestos Concept), for statistical support. The named individuals made their contributions as part of their paid employment and did not receive other compensation for their role in the trial.

References
1.
Fierson  WM; American Academy of Pediatrics Section on Ophthalmology; American Academy of Ophthalmology; American Association for Pediatric Ophthalmology and Strabismus; American Association of Certified Orthoptists.  Screening examination of premature infants for retinopathy of prematurity.   Pediatrics. 2018;142(6):e20183061. doi:10.1542/peds.2018-3061PubMedGoogle ScholarCrossref
2.
Blencowe  H, Lawn  JE, Vazquez  T, Fielder  A, Gilbert  C.  Preterm-associated visual impairment and estimates of retinopathy of prematurity at regional and global levels for 2010.   Pediatr Res. 2013;74(suppl 1):35-49. doi:10.1038/pr.2013.205PubMedGoogle ScholarCrossref
3.
Early Treatment for Retinopathy of Prematurity Cooperative Group.  Revised indications for the treatment of retinopathy of prematurity: results of the Early Treatment for Retinopathy of Prematurity randomized trial.   Arch Ophthalmol. 2003;121(12):1684-1694. doi:10.1001/archopht.121.12.1684PubMedGoogle ScholarCrossref
4.
Good  WV; Early Treatment for Retinopathy of Prematurity Cooperative Group.  Final results of the Early Treatment for Retinopathy of Prematurity (ETROP) randomized trial.   Trans Am Ophthalmol Soc. 2004;102:233-248.PubMedGoogle Scholar
5.
Tan  QQ, Christiansen  SP, Wang  J.  Development of refractive error in children treated for retinopathy of prematurity with anti-vascular endothelial growth factor (anti-VEGF) agents: a meta-analysis and systematic review.   PLoS One. 2019;14(12):e0225643. doi:10.1371/journal.pone.0225643PubMedGoogle ScholarCrossref
6.
Lashkari  K, Hirose  T, Yazdany  J, McMeel  JW, Kazlauskas  A, Rahimi  N.  Vascular endothelial growth factor and hepatocyte growth factor levels are differentially elevated in patients with advanced retinopathy of prematurity.   Am J Pathol. 2000;156(4):1337-1344. doi:10.1016/S0002-9440(10)65004-3PubMedGoogle ScholarCrossref
7.
Sukgen  EA, Koçluk  Y.  Comparison of clinical outcomes of intravitreal ranibizumab and aflibercept treatment for retinopathy of prematurity.   Graefes Arch Clin Exp Ophthalmol. 2019;257(1):49-55. doi:10.1007/s00417-018-4168-5PubMedGoogle ScholarCrossref
8.
Salman  AG, Said  AM.  Structural, visual and refractive outcomes of intravitreal aflibercept injection in high-risk prethreshold type 1 retinopathy of prematurity.   Ophthalmic Res. 2015;53(1):15-20. doi:10.1159/000364809PubMedGoogle ScholarCrossref
9.
Huang  CY, Lien  R, Wang  NK,  et al.  Changes in systemic vascular endothelial growth factor levels after intravitreal injection of aflibercept in infants with retinopathy of prematurity.   Graefes Arch Clin Exp Ophthalmol. 2018;256(3):479-487. doi:10.1007/s00417-017-3878-4PubMedGoogle ScholarCrossref
10.
Mintz-Hittner  HA, Kennedy  KA, Chuang  AZ; BEAT-ROP Cooperative Group.  Efficacy of intravitreal bevacizumab for stage 3+ retinopathy of prematurity.   N Engl J Med. 2011;364(7):603-615. doi:10.1056/NEJMoa1007374PubMedGoogle ScholarCrossref
11.
Stahl  A, Lepore  D, Fielder  A,  et al.  Ranibizumab versus laser therapy for the treatment of very low birthweight infants with retinopathy of prematurity (RAINBOW): an open-label randomised controlled trial.   Lancet. 2019;394(10208):1551-1559. doi:10.1016/S0140-6736(19)31344-3PubMedGoogle ScholarCrossref
12.
International Committee for the Classification of Retinopathy of Prematurity.  The International Classification of Retinopathy of Prematurity revisited.   Arch Ophthalmol. 2005;123(7):991-999. doi:10.1001/archopht.123.7.991PubMedGoogle ScholarCrossref
13.
Smith  LEH, Hellström  A, Stahl  A,  et al; Retinopathy of Prematurity Workgroup of the International Neonatal Consortium.  Development of a Retinopathy of Prematurity Activity Scale and clinical outcome measures for use in clinical trials.   JAMA Ophthalmol. 2019;137(3):305-311. doi:10.1001/jamaophthalmol.2018.5984PubMedGoogle ScholarCrossref
14.
Hu  J, Blair  MP, Shapiro  MJ, Lichtenstein  SJ, Galasso  JM, Kapur  R.  Reactivation of retinopathy of prematurity after bevacizumab injection.   Arch Ophthalmol. 2012;130(8):1000-1006. doi:10.1001/archophthalmol.2012.592PubMedGoogle ScholarCrossref
15.
Snyder  LL, Garcia-Gonzalez  JM, Shapiro  MJ, Blair  MP.  Very late reactivation of retinopathy of prematurity after monotherapy with intravitreal bevacizumab.   Ophthalmic Surg Lasers Imaging Retina. 2016;47(3):280-283. doi:10.3928/23258160-20160229-12PubMedGoogle ScholarCrossref
16.
Wong  RK, Hubschman  S, Tsui  I.  Reactivation of retinopathy of prematurity after ranibizumab treatment.   Retina. 2015;35(4):675-680. doi:10.1097/IAE.0000000000000578PubMedGoogle ScholarCrossref
17.
Chiang  MF, Quinn  GE, Fielder  AR,  et al.  International Classification of Retinopathy of Prematurity, Third Edition.   Ophthalmology. 2021;128(10):e51-e68. doi:10.1016/j.ophtha.2021.05.031Google ScholarCrossref
×