Quantitative evaluation of pigment epithelial encapsulation using polarization-sensitive optical coherence tomography after anti-VEGF therapy for myopic choroidal neovascularization.

Abstract

Purpose: We aimed to investigate the association between post-treatment changes in polarimetric entropy (PE), which is a noninvasive surrogate marker of structural heterogeneity associated with pigment epithelial remodeling and measured by polarization-sensitive optical coherence tomography (PS-OCT), and the need for multiple anti-vascular endothelial growth factor (VEGF)-injections in myopic choroidal neovascularization (mCNV).

Methods: We retrospectively analyzed 11 eyes that received intra-vitreal anti-VEGF injections for treatment-naïve mCNV. Based on treatment response during the first 3 months, eyes were categorized into the single-injection (n = 9) and multiple-injection (n = 2) groups. PS-OCT en face PE maps (6 × 6 mm) were obtained at baseline and 3 months after treatment, with the analysis area manually centered over the CNV using a 3-mm Early Treatment Diabetic Retinopathy Study (ETDRS) grid. PE values were extracted from the central 1-mm circle (CNV core) and surrounding 1-3-mm annulus (periphery), and the differences before and after treatment were calculated. Intergroup comparisons were conducted using Wilcoxon rank-sum test.

Results: PE within the CNV core increased after treatment (p = 0.0037). The multiple-injection group showed a greater PE increase than the single-injection group (p = 0.0347), although this finding should be interpreted cautiously given the small and imbalanced sample size. Exploratory analysis suggested a positive association between the magnitude of PE change in the CNV core and injection number during the 3-month period.

Conclusions: High post-treatment entropy may reflect structural heterogeneity associated with pigment epithelial remodeling during the early healing phase. These findings suggest that PS-OCT entropy imaging may serve as a retrospective indicator of tissue response rather than a predictive biomarker in the present dataset. Given the very small and imbalanced sample size, the results should be interpreted as exploratory and hypothesis-generating.