The goal of this research project is to identify the long-term outcome of neurodevelopment in patients with retinopathy of prematurity(ROP) and the treatment of anti-vascular endothelial growth factor (VEGF) such as intravitreal injection of bevacizumab (IVB), ranibizumab, or aflibercept.Investigators propose this study hopefully to have a better understanding of the long-term safety of anti-VEGF on the treatment of ROP.

Studies in both animalsand humans have found evidence of systemic bevacizumab exposure after IVB. In an animal study, IVB at an early age could result in more systemic bevacizumab exposure. Our study has further shown that VEGF levels in ROP infants were depressed for 8 weeks after IVB.

VEGF plays an important role in neurogenesis in embryos and preterm newborns. In previous reports, blocking VEGF-A expression has been shown to impair brain vascularization and lead to neuron apoptosis in the retina. In addition, VEGF has been found to be lower in preterm pups compared to term pups, and this has been proposed to relate to the neurodevelopmental delay and reduced growth of the cerebral cortex in premature infants. Since neurogenesis may continue in the third trimester, further inhibition of serum VEGF in preterm newborns may have long-term effects on the development of the central nervous system and other systems.

Currently, most studies reported neurodevelopmental outcomes in anti-VEGF treated premature infants before 2 years of age, and only one study reported 5 year outcomes. Our recent study also found that the neurodevelopmental outcomes at the mean age of 1.52 ± 0.59 years after birth were similar between ROP patients who did not require treatment and ROP patients with IVB treatment.

Unfortunately, the value of early assessments of cognition in predicting cognitive functioning at school age and older is questionable.Many developmental deficits in cognition, emotional and behavioral development, and social adaptive functioning may emerge at older ages in the absence of neurodevelopmental impairment in toddlerhood. Visuomotor function deficit are also noted at school age in children who had normal development at 3 years of age.

The above studies demonstrate a need for longer follow-up of the preterm infants to fully comprehend their neurodevelopmental outcomes. To our knowledge, currently there are no reports of neurodevelopmental outcomes in anti-VEGF treated premature infants beyond 5 years of age. Therefore, investigators propose this study hopefully to have a better understanding of the long-term safety of anti-VEGF on the treatment of ROP.

This study will aim at (1) Understanding the long-term neurodevelopmental outcomes of intravitreal injection of anti-VEGF comparing to standard laser treatment for ROP in premature infants. (2) Compare the long-term neurodevelopmental outcomes in premature infants with ROP treated by different anti-VEGF agents. (3) Analysis the long-term ocular morphological and functional outcomes in premature infants with ROP with prior treatments.

Investigators plan to recruit patients from our previous ROP cohort, who now aged 3 to12-years-old. Thepatients will be divided to six groups:premature without ROP (Group 0); ROP without treatment (Group 1); ROP with laser photocoagulation treatment (Group 2); ROP with anti-VEGF treatment (Group 3); ROP with laser photocoagulation + anti-VEGF treatment (Group 4); Fullterm (Group 5).Serialneurodevelopmental tests, such as Chinese Child Development Inventory (CCDI), Child Behavior Checklist (CBCL), The Berry-Buktenica Developmental Test of Visual-Motor Integration, Bayley Scales of Infant Development, Wechsler children's intelligence test- fourth editionand other neurocognitive tests and questionnaires, will be performed yearly in all patients. The detailed visual tests, such as best-corrected visual acuity, slit lamp examination, indirect ophthalmoscopy,and optical coherence tomography (OCT) will be performed every 6 months.

Main outcome measures will be neurodevelopmental outcomes. The neurodevelopmental outcomes will be analyzed longitudinally and in the cross-section fashion. These outcomes will be compared between the five groups, and in the subgroup analysis. Secondary outcomes will include ocular morphological and functional results of these children. Finally, the correlation of ocular resultswith neurodevelopment outcomes will be analyzed.

Investigators are fortunate to have the opportunity of following a longitudinal ROP cohort and monitor their long-term outcomes. In the long-term, this studywill improve understanding the long-term safety of anti-VEGF treatment for ROP, which is a heatedly debated topic. Investigators will also have a better knowledge which anti-VEGF might be safer than the other. Understanding these facts will help us to come up with a better treatment strategy for ROP in the future.

Retinopathy of Prematurity and Vascular Endothelial Growth Factor (VEGF) Retinopathy of prematurity (ROP) is a retinal disease in preterm infants due to retinal vessels immaturity, characterized by retinal hypoxia in early stage and neovascularization in late stage leading to retinal traction and retinal detachment.3, 4 Being one of the primary causes of childhood blindness,5 prompt diagnosis and treatment of ROP are of paramount importance. The treatment of ROP aims halting the neovascularization process driven by elevated intraocular vascular endothelial growth factor (VEGF).6 Laser photocoagulation of peripheral avascular retina remains the standard treatment of ROP. Though effective, laser photocoagulation destroys a sizable portion of retina, and is associated with the narrower anterior chamber angle,7 development of myopia,8 and reduction in visual field. A new treatment modality using intravitreal injection (IVI) of anti-vascular endothelial growth factor (anti-VEGF) to treat ROP patients has been increasingly used after the Bevacizumab Eliminates the Angiogenic Threat of Retinopathy of Prematurity (BEAT-ROP)study showed a significant benefit of intravitreal injections of bevacizumab (IVB) for zone I stage 3+ (i.e., stage 3 with plus disease) ROP compared with conventional laser treatment.9 The advantages of IVI of anti-VEGF include a less time-consuming procedure, fewer risks from general anesthesia in a physically compromised preterm newborn,10-16 and a potentially lower chance of unfavorable outcomes in zone 1 ROP.9, 17-19 For these reasons, the use of IVI of anti-VEGF for treatment of ROP has gained in popularity.

Intravitreal Injection of Anti-VEGFs for ROP Anti-VEGF agents for ROP include bevacizumab (Avastin; Genentech Inc., South San Francisco, CA, USA), ranibizumab (Lucentis; Genentech Inc., South San Francisco, CA, USA), pegaptanib (Macugen; Eyetech Inc., Cedar Knolls, NJ, USA), and aflibercept (Eylea; Regeneron Pharmaceuticals, Tarrytown, NY, USA). Each of them has different pharmacokinetic effects, molecular sizes, structures, and half-lives.8, 9, 20-23 Pegaptanib is a single strand nucleotide that binds specifically VEGF165and it was the first VEGF inhibitor approved by the United States Food and Drug Administration (FDA) for the treatment of neovascular age-related macular degeneration (AMD) in December 2004.24, 25 However, it is rarely used in most countries now because this drug was proven to be less effective than the other anti-pan-VEGF compounds.

Bevacizumab is a humanized murine antiangiogenic monoclonal antibody, originally developed for the treatment of metastatic colorectal cancer, and used with good results in treating many retinopathies. Currently, bevacizumab is the most commonly studied intravitreal anti-VEGF therapy for the treatment of severe ROP and has demonstrated promising results, particularly in severe cases of posterior disease.9, 19, 26-28 A much smaller humanized monoclonal antibody Fab fragment, ranibizumab was later used with the aim of being a safer treatment option. Ranibizumab bind to all VEGF-A isoforms and have demonstrated its efficacy in the treatment of Type 1 ROP.21 It is rapidly eliminated from the blood stream because of its shorter half-life and thus has the potential of decreased systemic toxicity.29Although ranibizumab has been included in an increasing number of studies in recent year, the rate of ROP recurrence following its use has varied somewhat between different studies.21, 30 Aflibercept is a "VEGF trap"-a fusion protein that binds VEGF-A with high affinity. It is approved by FDA for the treatment of neovascular AMD in 2011 and it is the only anti-angiogenic therapy that inhibits VEGF-A and placental growth factor.31Intravitreal administration of aflibercept should have a more potent and prolonged clinical action because of its high binding affinity and estimated intraocular half-life.32, 33 Although none of these drugs have been approved for intraocular use in children to date, many investigators have evaluated the off-label use of these agents in infants with ROP. Bevacizumab and ranibizumab are the more commonly used agents and the other two are less frequent used for treating ROP.

Systemic VEGF Suppression after Intravitreal Injection of Anti-VEGFs in Adults Because of structural differences, a much longer systemic half-life has been noted with bevacizumab than that with ranibizumab in adult patients (20 days vs. 2 hours for bevacizumab and ranibizumab, respectively).29, 34 In addition, the median plasma level of VEGF has been noted to be reduced by 42% in patients with age-related macular degeneration 28 days after receiving the third monthly IVI of bevacizumab, in contrast to no changes in patients treated with ranibizumab.35 In the "alternative treatments to inhibit VEGF in age-related choroidal neovascularization" (IVAN) study,36 the serum level of VEGF in patients receiving bevacizumab was suppressed to about one half of that in patients receiving ranibizumab.

Systemic VEGF Suppression after Intravitreal Injection of Anti- VEGF s in Newborns The pharmacokinetics of IVB may be different between newborns and adults.37, 38In an animal study, IVB at an early age could result in more systemic bevacizumab exposure.39Sato et al.37 found that systemic VEGF levels were depressed for at least 2 weeks after the administration of either 0.25 mg or 0.5 mg IVB in patients with stages 3, 4, and 5 ROP. Kong et al.40 demonstrated that Serum free VEGF levels decreased 2 days following treatment of either 0.25 mg or 0.625 mg IVB or laser photocoagulation, and the reductions were more significant in both IVB-treated groups. They also found that clearance of bevacizumab from the bloodstream in premature infants takes at least 2 months after IVB.40 Our previous studies had further shown that VEGF levels in type 1 ROP infants were significantly decreased up to 12 weeks after administration of 0.625 mg IVB.41-43 As for the effect of systemic VEGF suppressioninnewborns receiving intravitreal injection of ranibizumab (IVR), Zhou et al.44 found that IVR reduced plasma VEGF levels 1 day after injection in infants with ROP. This phenomenon disappeared 1 week after the injection. They concluded that IVR did not induce prolonged systemic VEGF suppression. Recently, Wu et al.42 also reported that IVR for ROP resulted in no or barely detected suppression of systemic VEGF, compared with IVB.

Role of VEGF in Neurodevelopmental Outcomesof Newborns VEGF plays an important role in neurogenesis in embryos and preterm newborns. In previous reports, blocking VEGF-A expression has been shown to impair brain vascularization45 and lead to neuron apoptosis in the retina.46 In addition, hypoxia-induced factors, including VEGF, have been demonstrated to be lower in preterm pups compared to term pups, and this has been proposed to partially explain the neurodevelopmental delay and reduced growth of the cerebral cortex in premature infants.47Since neurogenesis may continue in the third trimester,47further deprivation of serum VEGF in preterm babies may have long-term effects on the development of the central nervous system and other systems. However, more researches are needed to verify these speculations because anti-VEGF treatments are usually used for once or short term in these ROP patients.

Early Neurodevelopmental Outcomes inPremature Patients Following Anti-VEGF Treatment Since VEGF plays a key role in neurogenesis in embryos and preterm newborns, investigation of the impact of anti-VEGF treatment for ROP on neurodevelopment is needed for better understanding of its safety on those infants. In a prospective non-comparative case series by Martínez-Castellanos et al.,48 infants who received IVB for ROP were evaluated annually using the standardized Denver Developmental Screening Test II, and the majority of patients showed normal neurodevelopmental scores 5 years after the use of IVB. However, the study included a relatively small number of patients (18 eyes).Morin et al.49 retrospectively compared the neurodevelopmental outcomes(NDO) of 125 preterm infants treated with IVB (n=27) and laser ablation (n=98) at 18 months' corrected age, using Bayley-III as assessment tool.The study showed higher odds of severe neurodevelopmental disabilities in patients receiving IVB comparing to laser ablation. However, the neurodevelopmental disabilities in that study may not be a reliable indicator for NDO, since some of the items included were questionable.16 Our earlier study by Lien et al.50 retrospectively compared the NDO of 61 preterm infants treated with IVB only (n=12), laser treatment only (n=33) and combination of laser and IVB treatment (n=16), with assessments performed at the corrected age of 6, 12, 18, and 24 months using the Bayley-II scoring system. The results showed that patients treated with laser treatment alone and IVB alone did not differ significantly in mental or psychomotor development up to 2 years of follow-up. Recently, Kennedy et al. reported a randomized prospective studyshowingno significant differences between IVB group and laser group in the Bayley-III scores at the corrected age of 18-22 months.51However, the study included a relatively small number of patients (16eyes in total), and therefore might not be powered enough to identify small but important differences. More recently, our study found that the NDO at the mean age of 1.52 ± 0.59 years after birth were similar between ROP patients who did not require treatment and ROP patients with IVB treatment (Fan et al., manuscript in submission).With 128 patients, it is the current largest prospective case series, which offered us a stronger evidence regardingthe effect of anti-VEGF on later NDO.

Late Neurodevelopmental Outcomes in Premature Patients Following Anti-VEGF Treatment Currently, most studies reported NDO in anti-VEGF treated premature infants before 2 years of age,49-51 and only one study reported 5-year outcomes.48Most studies onNDO in children born prematurely use the Bayley-III for assessment of cognitive outcomes before 3 years of age. Unfortunately, the value of early assessments of cognition in predicting cognitive functioning at school age and older is questionable.52Many developmental deficits in cognition, emotional and behavioral development, and social adaptive functioning may emerge at older ages in the absence of neurodevelopmental impairment in toddlerhood.1, 53 Also, the severity of cerebral palsy is difficult to accurately assess in infants younger than 2 years, because their motor skills are developing, the presence/absence of hypertonicity may change over time.54 In addition, visuomotor function deficit are commonly diagnosed at older ages, when such tasks as copying, pegboard completion, spatial processing, visuosequential memory, and spatial organization are affected.1, 2, 55-58These problems have been noted at school age in children who had normal development at 3 years of age.1, 2Such as Attention deficit hyperactivity disorder (ADHD), learning disorder and Tourette …. that included deficiencies such as inattention, decreased learning ability, behavioral problems and social skills disturbance are usually found and diagnosed after school age. Therefore, it will be important to follow these premature infants after school age.

The above studies demonstrate a need for longer follow-up of the preterm infants to fully comprehend theirNDO. To our knowledge, currently there are no reports of NDO in anti-VEGF treated premature infants beyond 5 years of age. Therefore, investigators propose this study hopefully to have a better understanding of the long-term safety of anti-VEGF on the treatment of ROP.

Study Aim

• Understanding the long-term neurodevelopmental outcomes of intravitreal injection of anti-VEGF comparing to standard laser treatment for ROP in premature infants.
• Compare the long-term neurodevelopmental outcomes in premature infants with ROP treated by different anti-VEGF agents (bevacizumab, ranibizumab, aflibercept).
• Analysis the long-term ocular morphological and functional outcomes in premature infants with ROP with prior treatments.

Importance of the Proposed Study Due to the advantages of anti-VEGFtreatments,the use of IVI of anti-VEGF for treatment of ROP has gained in popularity. However, VEGFplays an important role in neurogenesis in embryos and preterm newborns, and IVI anti-VEGF was shown to have systemic suppression of VEGF in infants. Therefore, the neurodevelopmental outcomes of the anti-VEGF treated preterm babies are of concernand a highly debated topic.Although our current data showed there were no difference inneurodevelopmentaloutcomesbetween the ROP patients who received IVB treatment and ROP patients who did not require treatment at the mean age of 1.52 ± 0.59 years after birth(Fan et al., manuscript in submission), thelong-term neurodevelopmental outcomesbeyond this age remains uncertain.The proposed prospective, longitudinal cohort study could hopefully establish the long-term safety profiles of anti-VEGF regarding vision and neurodevelopmental outcomes. With this knowledge, the management guidelines in ROP could change and have influence on millions of preterm babies with this condition.

Inclusion Criteria:

1. Patients now aged 3 to12-years-old
2. Premature without ROP
3. ROP without treatment
4. ROP with laser photocoagulation treatment
5. ROP with anti-VEGF treatment
6. ROP with laser photocoagulation + anti-VEGF treatment
7. Fullterm

Exclusion Criteria:

1. Patients have brain's disease
2. Patients have Scleral buckling which is a type of eye surgery
3. Patients have Trans pars plana vitrectomy which is a type of eye surgery
4. Patients have other eye surgery
5. Patients have nystagmus
6. Patients can not Cooperate to complete the text

We recruited patients from our previous ROP cohort now aged 3 to12-years-old. The patients were divided to six groups: Premature without ROP (Group 0); ROP without treatment (Group 1); ROP with laser photocoagulation treatment (Group 2); ROP with anti-VEGF treatment (Group 3); ROP with laser photocoagulation + anti-VEGF treatment (Group 4); Fullterm (Group 5)