Bevacizumab or Aflibercept for ROP? Comparing Treatment Response and Retreatment Patterns

Introduction: Aflibercept is an anti-vascular endothelial growth factor (anti-VEGF) agent recently approved for retinopathy of prematurity (ROP) and may cause less systemic VEGF suppression than bevacizumab. This study compared clinical outcomes following primary intravitreal aflibercept versus bevacizumab for ROP, focusing on regression and retreatment.

Methods: Infants treated with intravitreal anti-VEGF therapy between 2016 and 2025, with at least four post-injection examinations, were included; ranibizumab-treated patients were excluded. The aflibercept group comprised of 18 infants (36 eyes) and the bevacizumab group 28 infants (56 eyes). ROP regression, retreatment, timing and reason for retreatment, and retinal detachment were compared.

Results: Baseline characteristics were comparable between groups. Mean gestational age was 24.9 ± 1.7 weeks with aflibercept and 25.2 ± 1.5 weeks with bevacizumab (P=0.45), and mean birth weight was 641.3 ± 146.2 g and 650.4 ± 189.1 g, respectively (P=0.96). Both agents induced ROP regression. Retreatment was required in 25.7% of aflibercept-treated and 17.6% of bevacizumab-treated eyes (P=0.37). Among retreated eyes, non-response was more frequent after aflibercept (62.1% vs 35.7%), whereas reactivation was more frequent after bevacizumab (38.5% vs 65.6%; P=0.46). Retreatment occurred significantly earlier after aflibercept (median 12 days) than bevacizumab (median 47 days; P=0.03). Retinal detachment requiring surgery occurred in 2.1% and 2.9%, respectively (P=0.84).

Conclusion: Both agents effectively induced ROP regression, with no significant difference in overall retreatment rates. Aflibercept retreatment occurred earlier and was more often associated with non-response, whereas bevacizumab retreatment more commonly reflected later reactivation, suggesting different post-treatment surveillance patterns.