OCT Angiography and Structural Biomarker Outcomes of Faricimab versus Aflibercept in Treatment-Naïve Polypoidal Choroidal Vasculopathy.

PURPOSE: Comparative studies of faricimab and aflibercept 2mg in polypoidal choroidal vasculopathy (PCV) have relied on conventional optical coherence tomography (OCT) biomarkers, leaving their effects on choriocapillaris perfusion uncharacterized. We compared post-loading visual and anatomical outcomes of the two agents in treatment-naïve patients with PCV, incorporating OCT angiography (OCTA) assessment of choriocapillaris vessel density.

METHODS

This retrospective cohort study consisted of treatment-naïve PCV patients receiving three monthly loading injections of faricimab or aflibercept 2 mg. After loading, best-corrected visual acuity (BCVA), macular and choroidal thickness, OCT biomarkers, and OCTA vessel density were compared between the two agents in the propensity-score matching cohort and also compared using logistic regression model to adjust for variables associated with each outcome in the original cohort.

RESULTS

Of the total 134 eyes (28 faricimab, 106 aflibercept), 26 and 52 remained after matching for balancing patient demographics and eye characteristics at baseline. Compared with aflibercept, faricimab achieved greater resolution of subretinal fluid (SRF; 0% vs 25.0% persistence; p=0.003), pigment epithelial detachment (PED; 26.9% vs 53.8%; p=0.024), and hyperreflective foci (HRF; 61.5% vs 100% retention; p<0.001) in the matched cohort. Based on multivariable analysis, faricimab treatment was independently associated with lower PED persistence and HRF retention, with a similar trend for SRF. Choriocapillaris vessel density increased in both groups without difference between groups.

CONCLUSIONS

Faricimab and aflibercept produced similar functional outcomes, reductions in macular and choroidal thickness, and increase in choriocapillaris vessel density in treatment-naïve PCV patients, whereas faricimab showed superior resolution of PED, HRF and a similar trend for SRF.