Background

Intravitreal injection of vascular endothelial growth factor (VEGF) inhibitors is increasingly used to treat retinopathy of prematurity (ROP) in the absence of evidence about long-term efficacy or safety. In this prespecified interim analysis of the RAINBOW extension study, we aimed to prospectively assess outcomes at age 2 years.

Methods

RAINBOW was an open-label, randomised trial that compared intravitreal ranibizumab (at 0·1 mg and 0·2 mg doses) with laser therapy for the treatment of ROP in very low birthweight infants (<1500 g). Families of the 201 infants that completed the RAINBOW core study were approached for consent to enter the extension study, which evaluates treatment outcomes prospectively through to 5 years of age. At age 20–28 months corrected for prematurity, participants had ophthalmic, development, and health assessments. The primary outcome was the absence of structural ocular abnormalities; secondary outcomes included vision-related quality of life (reported by parents using the Children's Visual Function Questionnaire), development (assessed with the Mullen Scales of Early Learning), motor function, and health status. Investigator-determined ocular and non-ocular serious and other adverse events were recorded. This study is registered with ClinicalTrials.gov, NCT02640664.

Findings

Between June 16, 2016, and Jan 22, 2018, 180 infants were enrolled in the RAINBOW extension study, and 153 (85%) were evaluated at 20–28 months of age. No child developed new ocular structural abnormalities. Structural abnormalities were present in one (2%) of 56 infants in the ranibizumab 0·2 mg group, one (2%) of 51 infants in the 0·1 mg group, and four (9%) of 44 infants in the laser therapy group. The odds ratio of no structural abnormality was 5·68 (95% CI 0·60–54·0; p=0·10) for ranibizumab 0·2 mg versus laser therapy, 4·82 (0·52–45·0; p=0·14) for ranibizumab 0·1 mg versus laser therapy, and 1·21 (0·07–20; p=0·90) for ranibizumab 0·2 mg vs 0·1 mg. High myopia (–5 dioptres or worse) was less frequent after 0·2 mg ranibizumab (five [5%] of 110 eyes) than with laser therapy (16 [20%] of 82; odds ratio 0·19, 95% CI 0·05–0·69; p=0·012). Composite vision-related quality of life scores seemed higher among the ranibizumab 0·2 mg group (mean 84, 95% CI 80–88) compared with laser therapy (77, 72–83; p=0·063). Mullen Scales T-scores for visual reception, receptive and expressive language were distributed similarly between the three trial groups and there were similar proportions of infants with motor and hearing problems among treatment groups. The proportion of infants with respiratory symptoms and Z scores of standing height, weight, and head circumference were similarly distributed in the treatment groups. There were no adverse events considered by the investigator to be related to the study intervention.

Interpretation

2-year outcomes following ranibizumab 0·2 mg for the treatment of ROP confirm the ocular outcomes of the original RAINBOW trial and show reduced high myopia, with possibly better vision-related quality of life. This treatment did not appear to affect non-ocular infant development.

Funding

Novartis Pharma AG.

中文翻译：

---

雷珠单抗与激光疗法治疗早产儿视网膜病变的极低出生体重婴儿的 2 年结局（RAINBOW 扩展研究）：一项开放标签、随机对照试验的前瞻性随访

背景

在缺乏长期疗效或安全性证据的情况下，玻璃体内注射血管内皮生长因子 (VEGF) 抑制剂越来越多地用于治疗早产儿视网膜病变 (ROP)。在 RAINBOW 扩展研究的这项预先指定的中期分析中，我们旨在前瞻性地评估 2 岁时的结局。

方法

RAINBOW 是一项开放标签、随机试验，比较玻璃体内雷珠单抗（0·1 毫克和 0·2 毫克剂量）与激光疗法治疗极低出生体重婴儿（<1500 克）的 ROP。与完成 RAINBOW 核心研究的 201 名婴儿的家庭进行了接触，以同意进入扩展研究，该研究前瞻性地评估直至 5 岁的治疗结果。在校正早产的 20-28 个月大时，参与者进行了眼科、发育和健康评估。主要结果是没有结构性眼部异常；次要结果包括与视觉相关的生活质量（由父母使用儿童视觉功能问卷报告）、发育（使用马伦早期学习量表评估）、运动功能和健康状况。记录研究人员确定的眼部和非眼部严重不良事件和其他不良事件。本研究已在 ClinicalTrials.gov 注册，NCT02640664。

发现

2016 年 6 月 16 日至 2018 年 1 月 22 日期间，180 名婴儿参加了 RAINBOW 扩展研究，其中 153 名（85%）在 20-28 个月大时接受了评估。没有孩子出现新的眼部结构异常。雷珠单抗 0·2 毫克组 56 名婴儿中有 1 名 (2%)、0·1 毫克组 51 名婴儿中有 1 名 (2%) 以及激光治疗 44 名婴儿中有 4 名 (9%) 存在结构异常治疗组。雷珠单抗 0·2 mg 与激光治疗相比，无结构异常的优势比为 5·68（95% CI 0·60–54·0；p=0·10），4·82（0·52-45·0 ; p=0·14) 雷珠单抗 0·1 毫克 vs 激光治疗，1·21 (0·07–20; p=0·90) 雷珠单抗 0·2 毫克vs0·1毫克。0·2 毫克雷珠单抗（110 只眼中的 5 只 [5%]）与激光治疗（82 只眼中的 16 只 [20%]；优势比 0·19，95% CI）相比，高度近视（–5 屈光度或更差）的发生率更低0·05–0·69；p=0·012）。与激光治疗相比，雷珠单抗 0·2 毫克组（平均 84, 95% CI 80-88）的综合视力相关生活质量评分似乎更高（77, 72-83；p=0·063）。Mullen Scales 在视觉接收、接受和表达语言方面的 T 分数在三个试验组之间的分布相似，并且在治疗组中存在运动和听力问题的婴儿比例相似。有呼吸道症状的婴儿比例和站立高度、体重和头围的 Z 评分在治疗组中分布相似。

解释

雷珠单抗 0·2 mg 治疗 ROP 后的 2 年结果证实了原始 RAINBOW 试验的眼部结果，显示高度近视减少，视力相关的生活质量可能更好。这种治疗似乎不会影响非眼部婴儿的发育。

资金

诺华制药公司。