Introduction

Faricimab (Vabysmo®) and aflibercept (Eylea®) are approved therapies for diabetic macular oedema (DMO) and inhibit VEGF-A; faricimab additionally targets angiopoietin-2, while aflibercept inhibits VEGF-B and placental growth factor1-3. We compared their real-world outcomes in treatment-naïve DMO patients.

 

Methods

Retrospective study at Oxford Eye Hospital of treatment-naïve DMO patients initiating faricimab or aflibercept (November 2022-March 2024). Included patients had ≥12 months follow-up. Visual acuity (VA), central subfield thickness (CST), and treatment interval were evaluated at baseline, 6 months (±8 weeks), 12 months (±8 weeks), and 18 months (±12 weeks). Linear mixed-effects modelling with subject as random intercept accounted for non-independent, longitudinal data. Missing data was handled through the linear mixed-effects model.

 

Results

99 eyes received faricimab and 78 aflibercept; baseline characteristics were comparable. VA remained stable in both groups with no significant between-group differences. Both agents reduced CST from baseline to six months, sustained thereafter, but faricimab achieved significantly greater CST reduction at 12 (p=0.02) and 18 months (p<0.01), a 2.75-week longer mean treatment interval at 12 months (p<0.01), and fewer injections across 18 months (p=0.04).

 

Discussion

Both agents stabilised VA and reduced CST in treatment-naïve DMO. Faricimab was associated with greater CST reduction and fewer injections over 18 months.

 

Table 1. Baseline characteristics of treatment-naïve patients with DMO at Oxford Eye Hospital treated with faricimab or aflibercept. Continuous variables: mean ± SD (Welch’s unpaired t-test); categorical variables: n (%) (Fisher’s exact test). No significant between-group differences at baseline. Early Treatment Diabetic Retinopathy Study (ETDRS), HbA1c (glycated haemoglobin).

 

Figure 1. Outcomes from linear mixed-effects modelling for faricimab (orange) and aflibercept (green), adjusted for baseline covariates. Error bars represent SEM. Between-group comparisons shown above each time point: ns = not significant (p≥0.05), *p<0.05, **p<0.01.