Treatment for Nontype 1 Retinopathy of Prematurity by Intravitreal Injection of Antivascular Endothelial Growth Factor Drugs

Background To explore clinical characteristics and treatment reasons for intravitreal injection of antivascular endothelial growth factor (anti-VEGF) drugs in the treatment of nontype 1 retinopathy of prematurity (ROP). Methods A retrospective study was conducted to screen the nontype 1 ROP from the collected ROP patients who received intravitreal injections of anti-VEGF drugs in Henan Eye Hospital from September 2018 to June 2021. Results A total of 138 ROP cases (262 eyes) were included in this study, including 39 cases (28.3%), 65 eyes (24.8%) that were the nontype 1 ROP. Compared with the type 1 ROP group, the nontype 1 ROP group had slightly later treatment time (39.8 ± 2.7 weeks vs 38.1 ± 2.6 weeks, P < 0.05) and a higher proportion of fusion protein drugs (87.2% vs 54.5%, P < 0.05). After intravitreal injection of anti-VEGF drugs, 27 eyes (41.5%) were cured and 38 eyes (58.5%) improved in the nontype 1 ROP group, without recurrence and aggravation cases. There were more lesions in zone II (63 eyes, 96.9%), with stage 2 (40 eyes, 61.5%) and stage 3 (23 eyes, 35.4%), and 58 eyes (89.2%) showed preplus in the nontype 1 ROP group. Treatment reasons included preplus in 58 eyes (89.2%), ridge aggravation in 22 eyes (33.8%), simultaneous treatment of the contralateral eye in 9 eyes (13.8%), no regression of lesions in the persistent stage 2 or 3 over PMA 44 weeks of follow-up in 8 eyes (12.3%), and logistical considerations in 4 eyes (6.2%). Conclusions Considering some peculiar clinical characteristics, treatment by intravitreal injection of anti-VEGF drugs may be considered carefully for some nontype 1 ROP in critical conditions.


Introduction
Retinopathy of prematurity (ROP) is a proliferative retinal vascular disease that occurs in premature and low-weight infants, and it mostly aficts both eyes and may cause retinal detachment and blindness. Current treatments of ROP include laser photocoagulation and intravitreal injection of antivascular endothelial growth factor (anti-VEGF) drugs. Te Early Treatment for Retinopathy of Prematurity (ETROP) and the latest expert consensus or guidelines recommend treatment for type 1 ROP and close follow-up for type 2 ROP [1,2]. However, some ROP eyes have special manifestations in clinical practices, although they have not reached the level of type 1 ROP, manifested as gradually thickened or widened ridges without typical manifestation of plus disease or no regression of the retinal lesions for a long time of follow-up. Tis kind of ROP is more serious than type 2 ROP, and it is difcult to decide whether to continue follow-up or give treatment. Tese nontype 1 ROP eyes may be recommended for treatment to avoid unnecessary medical disputes or irreversible vision damage caused by the aggravation of the disease.
Previous studies have reported that 9-27% afected eyes of patients with nontype 1 ROP receive treatment in different countries [3][4][5][6]. In recent years, intravitreal injection of anti-VEGF drugs for the treatment of ROP, with advantages of simple operation, minimal invasion, and continuous growing of retinal vessels to the periphery after injection, has gradually become an important treatment method [7][8][9]. Tere is a lack of relevant research on the treatment of intravitreal injection of anti-VEGF drugs for nontype 1 ROP in critical conditions. In this study, the nontype 1 ROP cases who had received intravitreal injection of anti-VEGF drugs in our hospital were enrolled to analyze the treatment efect, clinical characteristics, and treatment reasons and to explore the personalized diagnosis and treatment for ROP with special clinical characteristics.

Subjects and Data.
Tis was a retrospective study. ROP infants who had received intravitreal injection of anti-VEGF drugs were collected in the Department of Ophthalmology of Henan People's Hospital (Henan Eye Hospital) from September 2018 to June 2021, of whom, type 1 and nontype 1 ROP cases were subgrouped. Tis study was conducted with the approval of the Medical Ethics Committee of Henan Eye Hospital (Approval Number HNEECKY-2021 (49)) and performed in accordance with the Declaration of Helsinki. All methods were confrmed to be performed with relevant regulations. All infant's parents or legal guardians signed the written informed consent prior to treatment.

Inclusion and Exclusion Criteria.
Diagnostic criteria for type 1 ROP were as follows [1]: (1) ROP at stages 1-3 with plus in zone I; and ROP at stage 3 without plus in zone I; (2) stage 2 or 3 ROP with plus in zone II; (3) aggressive ROP (A-ROP). Tose who did not meet any above conditions and received intravitreal injection of anti-VEGF drugs were defned as nontype 1 ROP. Infants with unstable vital signs caused by systemic diseases in the heart, brain, lung, or accompanied by other fundus lesions were excluded.

Study Selection.
Te retinal examination of ROP infants was performed in the neonatal intensive care unit (NICU) and ophthalmic clinic by two experienced doctors under topical anesthesia after mydriasis with RetCam 3. All abnormal retinal images were judged by a senior pediatric retinal professor (Haitao Zhang, associated professor), and then, diagnosis and treatment suggestions were made. Information of name, gender, gestational age (GA), birth weight (BW), and the time of examination and injection were recorded after the retinal examination. Te zone, stage, range, and plus disease of binocular retinal lesions were recorded according to ICROP3 [10].
Anti-VEGF drugs used in the study were ranibizumab (0.25 mg/0.025 ml), conbercept (0.25 mg/0.025 ml), and afibercept (1 mg/0.025 ml). Te latter two drugs were fusion proteins. Intravitreal injection for ROP was performed under topical anesthesia. After local disinfection, a 29G needle-equipped syringe was used to penetrate the eyeball wall at 1.0-1.5 mm posterior of the limbus to the vitreous cavity parallel to the optical axis. Antibiotic eye drops were used for 3-5 days to prevent ocular infection, and the frst eye review was conducted within 7 days. Te interval of the next review (1-3 weeks) should be determined according to the retinal manifestations, and patients should be followed up until complete retinal vascularization or for at least 24 weeks.
Te characteristics of included subjects were analyzed according to the number of cases. Te grouping criteria were as follows: ROP cases were included in the type 1 ROP group if both the eyes met the above criteria and received intravitreal injection of anti-VEGF drugs, or one eye met the criteria and received intravitreal injection while the contralateral eye with mild lesions did not receive treatment. ROP infants whose one eye or both eyes did not meet the criteria of type 1 ROP and received intravitreal injection were included in the nontype 1 ROP group. Te curative efect, retinal pathological characteristics, and causes for the treatment of nontype 1 ROP were analyzed according to the number of eyes. Te characteristics of included subjects were analyzed according to the number of cases.

Quality Assessment.
Te treatment efect was evaluated as follows: (1) Cured, complete retinal vascularization. Retinal vessels gradually grew to the ora serrata or less than 1 PD (papillary diameter) away from the ora serrata around zone III; (2) Improved, the retina was not completely vascularized. Te retinal vessels had reached zone III, and there were still nonvascular areas, but without obvious active lesions at the last follow-up. (3) Recurrence. Te tortuous dilation of retinal vessels was relived and the ridge became fattened in the early stage after the operation, but then the tortuous dilation of vessels, ridge aggravation, and neovascularization occurred again in the retina. (4) Aggravation. Te tortuous dilation of retinal vessels was not signifcantly reduced, and the proliferation and traction were aggravated, even leading to retinal detachment. Te improved cases still required regular examination. Te recurrent cases were treated with intravitreal injection of anti-VEGF drugs again or retinal laser photocoagulation. Te aggravating cases were treated with laser photocoagulation according to the retinal manifestations or surgery in case of retinal detachment.
Te reasons for the treatment of nontype 1 ROP were as follows: (1) preplus diseases in the retina; (2) ridge aggravation, shown as more obvious ridges or ridge extension, or locally thickened and widened ridges, with a risk of increased proliferation; (3) simultaneous treatment of the contralateral eye; (4) no regression of lesions in the persistent stage 2 or 3 for over PMA (postmenstrual age) 44 weeks of follow-up; (5) logistical considerations, follow-up might not be timely due to various reasons (such as living far away, parents' poor understanding, and epidemic control policy).

Data
Analyses. Data analyses were performed using the SPSS 19.0 statistical software. Te diferences in birth gestational age (GA), birth weight (BW), hospitalization days in NICU, frst injection time, and follow-up time were compared by the t-test, and the diferences in gender and drug types were compared by the χ 2 test. Te treatment efect was analyzed according to the number of aficted eyes and compared by the χ 2 test. P < 0.05 was considered statistically signifcant.
Te characteristics of subjects with the nontype 1 ROP and type 1 ROP are shown in Table 1. Tere was no signifcant diference in GA, BW, gender proportion, hospitalization days in NICU, and follow-up time between the two groups (P > 0.05). While, the time of the frst treatment of the nontype 1 ROP group was slightly later than that of the type 1 ROP group (39.8 ± 2.7 weeks vs 38.1 ± 2.6 weeks, (P < 0.01). Te diference in the types of anti-VEGF drugs was signifcant (P < 0.05), with a higher proportion of fusion protein drugs in the nontype 1 ROP group (87.2%) than that in the type 1 ROP group (54.5%).

Efect Assessment.
According to the number of eyes (n � 262), 65 eyes (24.8%) were nontype 1 ROP. After treatment, 27 eyes (41.5%) were cured and 38 eyes (58.5%) were improved. Tere was no recurrence and aggravation. Tere were 197 eyes of type 1 ROP. After treatment, 79 eyes were cured (40.1%), 100 eyes were improved (50.8%), and 18 eyes recurred (9.1%), without aggravation cases. Te difference in treatment efects between the two groups was signifcant (P < 0.05 Table 2). All 18 recurrent eyes in the type 1 ROP group received intravitreal injection of the same anti-VEGF drugs and then the retinal conditions improved.

Reasons for the Treatment.
Of all treated eyes with nontype 1 ROP (n � 65) (Table 3), 63 eyes (96.9%) had ROP lesions in zone II, of which there were 5 eyes (3.1%) in posterior zone II and 2 eyes (3.1%) with ROP in zone I. In terms of stage, there were more lesions in stages 2 and 3, with 40 eyes (61.5%) and 23 eyes (35.4%), respectively. In terms of plus diseases, 58 eyes (89.2%) showed preplus. As for treatment reasons, the main reason was preplus in 58 eyes (89.2%), followed by ridge aggravation in 22 eyes (33.8%), simultaneous treatment in 9 eyes (13.8%) due to contralateral eye treatment, no regression of lesions in stage 2 or 3 for over PMA 44 weeks of follow-up in 8 eyes (12.3%), and logistical considerations in 4 eyes (6.2%). Te above reasons could exist simultaneously ( Figure 1). Te preoperative features and specifc reasons for treated nontype 1 ROP cases are shown in Table 4.

Discussion
Some previous studies analyzed the treatment reasons of nonype 1 ROP (Table 5) [3,5,11,12], but the main treatment method in these studies was laser photocoagulation. No literature had been retrieved to explore the efect of intravitreal injection of anti-VEGF drugs on the treatment of nontype 1 ROP currently. Laser photocoagulation and anti-VEGF therapy are the main options for ROP currently. Laser therapy led to permanent destruction of the peripheral retina, and peripheral retinal vessels continued to develop after anti-VEGF agents' treatment [13]. As to efcacy compared to laser therapy, anti-VEGF agents as primary treatments had potential advantages for the eyes with posterior ROP (zone I type 1 ROP and A-ROP), and for the eyes with zone II type 1 ROP, anti-VEGF agents therapy showed similar efcacy; however, there was a signifcantly higher rate of reactivation [14]. Laser-treated eyes had a greater trend to myopia and astigmatism than anti-VEGF therapy [14,15]. For the above reasons, anti-VEGF therapy performed under topical anesthesia was preferred for treatment with ROP during our clinical practice, and laser therapy was the used option for ROP with the risk of obvious fbrosis or the recurrent eyes. In addition, laser coagulation for ROP needed to be operated under general anesthesia which most parents of neonates were reluctant to choose in China.
Of the 263 eyes treated in this study, 65 eyes (24.8%) were nontype 1 ROP, showing a higher proportion than that in previous studies (9.5-13.7%) [3,5,11]. Because of that, the treatment for ROP cases in this study was an intravitreal injection of anti-VEGF drugs, which are simpler operated and more minimally invasive than laser therapy [13]. In this study, lesions in all eyes with nontype 1 ROP were relieved after treatment, which was similar to the results of previous studies [11]. It was also found that the treatment efect of the nontype 1 ROP eyes was better than that of type 1 ROP eyes, which may mainly be related to the milder condition. In this study, the proportion of fusion protein drugs used was higher in the nontype 1 ROP group (87.2%) than that in the type 1 ROP group (54.5%). Tis might be another potential reason for the diference in treatment efects between the two groups. Some retrospective studies found that the recurrence rate of the ROP eyes treated with fusion protein drugs (conbercept or afibercept) was lower than that of the ROP eyes treated with ranibizumab [16,17]. But a multicentral prospective trial comparing clinical outcomes of conbercept vs ranibizumab treatment for ROP found there was no signifcant statistical diference in the recurrence rate between the two anti-VEGF agents [18]. It is still controversial whether there is a diference between the efcacy of ranibizumab and fusion protein drugs in the treatment of ROP. Te diferent proportion of drug selection in our study was associated with the time to market in China.
Of the eyes with nontype 1 ROP treated in this study, 63 cases (96.9%) had more lesions in zone II, and 40 eyes (61.5%) and 23 eyes (35.4%) had more lesions in stages 2 and 3, respectively. Te characteristics of pathological manifestation were similar to those of the fnding of Gupta et al. [3], in which 11 eyes (84.6%) had lesions in zone II and 12 eyes (92.3%) had lesions in stages 2 and 3. While in the study of Liu T et al. [5], most eyes (66%) had preplus lesions in zone II stage 3. Te above evidence suggests that ROP should be checked carefully about the changes of lesions at stage 2 or 3 in zone II.
Te main treatment reason for the nontype 1 ROP eyes in this study was the preplus disease (89.2%), which was diferent from the previous studies [3,5,11]. In the study of Gupta et al. and Rajan et al., the most important treatment reason was structural changes in the fundus caused by the traction of the ridge (69.2% and 72.7%, respectively) [3,11]. Preplus disease (33.3%) was the second reason in the study of Rajan et al. [11]. Te major reason in Liu et al.'s study was the contralateral eye with type 1 ROP (43%), followed by stage 3 ROP with preplus (30%) [5]. As for preplus and plus disease, ICROP3 defned it as a continuous spectrum of retinal vascular changes from normal to preplus and fnally to plus disease. Consistent judgments of diferent scholars are only in the normal and last plus stages [10]. Tis suggests a high possibility of clinical disagreement over preplus lesions, resulting in no typical plus lesions in some ROP eyes and a further risk of retinal traction with progressive worsening of the ridge. Due to the use of anti-VEGF drugs, we paid more attention to the judgment of preplus in the ROP examination.
Te second cause of treatment in this study was ridge aggravation (33.8%), shown as more obvious or/and more extension, or thickened and widened locally. Actually, the ridge aggravation was often accompanied by preplus (Figure 1). Koucheki et al. confrmed that preplus was signifcantly correlated with increased ridges (≥2 continuous clock hours of the persistent stage 3 crossing the temporal horizontal midline) in the eyes with stage 3 ROP persisting ≥40 weeks of PMA [12]. Te ridge aggravation in this study was slightly milder than the structural changes such as macular traction, retinal traction, or folds produced due to       Journal of Ophthalmology the tangential traction caused by the straightening of the temporal vessels in the fundus mentioned in previous studies [3,5,11]. Under these fbrosis conditions, intravitreal injection of anti-VEGF drugs may not be recommended because of the risk of aggravated traction [19,20]. In this study, the simultaneous treatment of the contralateral eyes accounted for 13.8%. Most previous studies considered that acute ROP commonly occurs in both the eyes. For example, 79.1% of ROP infants have high-risk prethreshold disease in both the eyes at the time of enrollment in an ETROP study [1]. A study on telemedicine approaches to evaluating of acute-phase retinopathy of prematurity (e-ROP) found that 72.7% of infants had the same severity of ROP in both the eyes among ROP image sessions [21].
However, our study does not recommend arbitrary early treatment for nontype 1 ROP. Previous studies suggested that ROP with stage 3 can be treated when no regression is found after 41 weeks of PMA [3] or continuous 6 weeks of follow-up [5,11]. In this study, the average time of the frst treatment for the eyes with type 1 ROP was 38.1 weeks, while the follow-up of another 6 weeks was 44 weeks for the nontype I ROP eyes with some of the above particular retinal features. Meanwhile, due to the use of anti-VEGF drugs, in  1w (a, b). After 2 weeks (PMA 41.1w), the temporal ridge was thickened and widened with preplus in the right eye, which was ROP in zone II, stage 3 with preplus (c). As the contralateral eye, the ridge was more obvious and extended, which was still ROP in zone II, stage 2 with plus (−) (d). Te bilateral retinal vessels grew to the periphery of zone III at 27.9 weeks (PMA 69.0w) after intravitreal injection of conbercept (e, f ).  Journal of Ophthalmology order to avoid obvious fbrosis, we paid more attention to the progression of lesions at stage 2 and stage 3. Terefore, our study considered a treatment for ROP infants with lesions at persistent stages 2 and 3 without regression at PMA 44 weeks or more and whether there were other retinal manifestations were also taken into consideration. In this study, 4 cases (8 eyes) (12.3%) were followed up for ≥ PMA 44 weeks and then received anti-VEGF treatment, and the fundus was simultaneous with preplus lesions or ridge aggravation or logistical considerations before treatment. During the struggling follow-up period, more attention should be paid to the changes of extraretinal neovascular proliferation, and the anti-VEGF therapy should be performed in time before the fbrosis is obvious. Once obvious fbrosis has formed, laser coagulation will be recommended because retinal traction may be aggravated after anti-VEGF treatment [19,20]. Unlike other ocular neovascular conditions (e.g., wetAMD), in which VEGF is continually released, there is a single burst of VEGF that promotes neovascularization in ROP [22]. Te delayed anti-VEGF therapy given at a period when VEGF levels are decreasing may promote fbrosis driven by transforming growth factor-β (TGF-β) and connective tissue growth factor (CTGF) [23][24][25]. Traction from fbrosis may cause retinal detachments. In this study, 4 eyes (6.2%), 2 cases (Nos. 4 and 5, Table 4) were treated for the logistical considerations, that is, follow-up might not be timely due to various reasons. Teir parents lived far away or afected by epidemic control reasons (all eyes accompanied by other reasons, such as preplus or follow-up time ≥ PMA44w) and might not be followed up in time, and treatment was chosen considering that the retinal lesions tended to aggravate at the same time. Te logistical considerations of Liu et al.'s study were the difculty in follow-up or general anesthesia for non-ROP surgery (3%) [5]. Te intravitreal injections for ROP infants in our study were performed under topical anesthesia, and there was no treatment under general anesthesia due to other diseases. For some ROP cases with difculty in follow-up, detailed communication with the parents before treatment was recommended to emphasize the importance of follow-up, especially after anti-VEGF drug treatment that requires longer follow-up. In fact, the 2 cases were followed up in our hospital within 4 weeks after treatment. Ten, the following examinations from the 6th week were started in the local hospital and the regular examination results including some retinal images would be transmitted to our research group through WeChat or the network of telemedicine.
Our study had some limitations. First, to avoid medical disputes caused by delayed treatment during clinical practice, there was no control group set. So, it was unable to accurately judge the progression of nontype 1 ROP with aggravating tendency if not treated. Te sample in the study group will gradually be increased to collect and follow up carefully some untreated cases with nontype 1 ROP so as to investigate reasonable methods in the future. Second, as for the systemic efect of anti-VEGF drugs on premature infants, previous studies have confrmed that the intravitreal injection of anti-VEGF drugs in the ROP eyes has a certain inhibitory efect on the level of VEGF in blood [26][27][28][29]. But diferent anti-VEGF drugs have diferent efects. Tere were no diferences in plasma-free VEGF concentrations of ROP infants after bilateral intravitreal injection of ranibizumab from the RAINBOW Trial [30]. Cheng et al. found that the serum VEGF levels in ROP infants were suppressed for a short time after intravitreal injection of conbercept and returned to the preoperative level at 4 weeks [28]. Huang CY found that serum VEGF levels in type 1 ROP infants were suppressed for 3 months after treatment with afibercept or bevacizumab, but the suppression of systemic VEGF was more pronounced in infants treated with bevacizumab than those treated with afibercept [29]. Te suppression of VEGF in blood may afect the development of important organs of premature infants. Considering these potential risks, we are very cautious in the treatment of ROP and do not arbitrarily expand the treatment indications. In clinical practice, we always follow the professional guidelines and recommendations. Type 1 ROP eyes need treatment, while type 2 ROP eyes need close observation. However, continuous observation or treatment of ROP infants with aggravating tendencies but not reaching the typical fundus manifestations of type 1 was confused and should be discussed carefully. At the meanwhile, we have explored a lower dose of anti-VEGF drug intravitreal injection for ROP to reduce the possible systemic adverse reactions (published in Chinese, http:// www.coretina.com/article/10.3760/cma.j.cn511434-20200219-00066).

Conclusions
Nontype 1 ROP with some characteristics, such as preplus, ridge aggravation, treatment with contralateral eyes, no regression in the persistent stage 2 or 3 for follow-up ≥ PMA 44 weeks, and logistical considerations, can be considered carefully to receive intravitreal injection of anti-VEGF drugs based on current expert consensus or guidelines.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request.

Disclosure
A preprint has previously been published [31].

Conflicts of Interest
Te authors declare that they have no conficts of interest.