A total of 37 eyes with nAMD were included in the analysis; baseline demographic and clinical characteristics are summarized in Table 1. At 12 months after switching to brolucizumab, BCVA was maintained (0.49 ± 0.30 to 0.48 ± 0.34 logMAR, p = 0.429), whereas CST significantly decreased from 282.27 ± 71.56 to 231.80 ± 69.85 μm (p = 0.004). Subfoveal choroidal thickness did not change significantly (240.86 ± 89.10 to 228.22 ± 84.36 μm, p = 0.463). PED was identified in 24 eyes (64.9%), and maximal PED height decreased from 311.04 ± 159.71 to 245.96 ± 159.55 μm (p = 0.009).

Table 2 shows the number of patients with exudation on OCT before and after switching. At 12 months, 17 out of 37 eyes (45.9%) achieved a dry macula, despite only one eye having achieved a dry macula during prior anti-VEGF therapy. Among the remaining 20 eyes with exudation, 15 eyes showed SRF, 3 eyes showed IRF, and 2 eyes showed both SRF and IRF.

The mean injection interval significantly increased from 13.18 ± 6.66 to 16.39 ± 7.85 weeks (p = 0.004). Improvement in interval length was observed in 22 eyes (59.5%), no change in 10 eyes (27.0%), and worsening in 5 eyes (13.5%).

Quantitative OCTA analysis demonstrated a significant reduction in MNV area after the switch (3.60 ± 3.28 mm² at baseline to 3.21 ± 2.97 mm² at 12 months, p = 0.017). FD also significantly decreased from 1.612 ± 0.054 to 1.592 ± 0.060 (p = 0.006). VD decreased slightly (34.3% ± 5.4% to 33.1% ± 6.3%, p = 0.186), and lacunarity increased from 0.723 ± 0.210 to 0.792 ± 0.285 (p = 0.078), but neither difference was statistically significant.

In the additional analysis including 12 months prior to baseline, during treatment with other anti-VEGF agents, MNV area decreased from 4.42 ± 4.45 mm² at 12 months before to 3.21 ± 2.97 mm² at 12 months after switching (p = 0.011), and VD decreased from 38.1% ± 4.7% to 33.1% ± 6.3% (p < 0.001). FD decreased from 1.635 ± 0.056 to 1.592 ± 0.060 (p < 0.001), and lacunarity increased from 0.665 ± 0.186 to 0.792 ± 0.285 (p = 0.010) (Fig. 1A-1D). In Bonferroni post hoc analysis, FD showed a statistically significant reduction not only from 12 months before to baseline (p = 0.002) but also from baseline to 12 months after switching (p = 0.019).

A 79-year-old female patient with nAMD showed no SRF recurrence after switching to brolucizumab (Fig. 2A-2E). A new PED had developed after 7 years of anti-VEGF therapy (31 injections of aflibercept and ranibizumab) (Fig. 2B), and despite three additional aflibercept injections, there was no anatomical improvement (Fig. 2C). One week after the first brolucizumab injection, PED height and SRF decreased slightly, and OCTA-derived parameters (MNV area, VD, and FD) also declined (Fig. 2D). At the 12-month follow-up, OCT showed a dry macula (Fig. 2E). Compared with 12 months before switch and baseline, both MNV area and FD demonstrated a progressive decrease (MNV area: 6.108 at 12 months before switch, 4.719 at baseline, and 3.706 at 12 months after switch; FD: 1.669, 1.658, and 1.642, respectively). In contrast, VD and lacunarity did not show a consistent directional change.

A 67-year-old male patient with nAMD showed recurrent SRF after switching to brolucizumab (Fig. 3A-3E). The patient had previously received eight injections each of aflibercept and ranibizumab with persistent SRF, which prompted the conversion (Fig. 3B). Following the switch, OCT demonstrated transient macular dryness after each injection (Fig. 3D) with recurrent SRF (Fig. 3E); nevertheless, the injection interval increased slightly from 19.3 to 22.0 weeks. Among OCTA parameters, FD decreased from 1.650 at 12 months before switch to 1.630 at 12 months after switch, with subsequent values of approximately 1.633 and 1.654.

No systemic adverse events were observed. Among 37 eyes, intraocular inflammation occurred in 1 eye after the fifth brolucizumab injection, presenting with conjunctival injection, ocular pain, and anterior chamber cells; the condition improved with topical corticosteroids, and treatment was subsequently switched to aflibercept. Floater complaints were noted in two eyes but were not classified as adverse events.

In this study, we evaluated functional, anatomical, and quantitative OCTA outcomes in eyes with nAMD switched from other anti-VEGF agents to brolucizumab. Over 12 months of follow-up, switching to brolucizumab significantly reduced CST, prolonged injection intervals, and decreased MNV area and FD while maintaining BCVA. These findings suggest that brolucizumab is effective in controlling exudation and may induce morphological remodeling of MNV lesions detectable by OCTA.

Previous randomized clinical trials, including HAWK and HARRIER, have demonstrated that brolucizumab achieves comparable visual outcomes to aflibercept with extended dosing intervals [9]. Beyond visual noninferiority, brolucizumab offers several advantages, including smaller molecular weight and higher molar concentration, which enable improved tissue penetration and more durable suppression of VEGF activity [7-9]. Real-world evidence further supports these benefits in patients switched from other anti-VEGF agents due to suboptimal responses. In a large retrospective study from the United States, switching to brolucizumab led to significantly extended injection intervals, reduced central macular thickness, and stable vision at both 12 and 18 months [15]. Similarly, a Japanese cohort study demonstrated that eyes with type 1 MNV and polypoidal choroidal vasculopathy experienced meaningful prolongation of treatment intervals after conversion from aflibercept, accompanied by reductions in lesion size and number of polypoidal lesions, while maintaining visual acuity [16]. In addition, a European real-world study reported a significant reduction in PED height after switching to brolucizumab, although PED itself was not a strong predictor of visual outcomes [17]. In our study, consistent with these findings, switching to brolucizumab significantly extended injection intervals and reduced CST and SRF, and was also associated with a notable decrease in PED height.

Previous studies have demonstrated that brolucizumab is effective in extending injection intervals and achieving a dry macula in eyes with nAMD refractory to prior anti-VEGF therapy [18]. In our cohort, the mean injection interval significantly increased from 13.18 ± 6.66 to 16.39 ± 7.85 weeks after switching, and although 36 of 37 eyes were refractory at baseline, 17 eyes achieved a dry macula at 12 months while 20 eyes continued to show recurrent or persistent fluid.

Recent studies have explored OCTA-derived biomarkers as indicators of nAMD activity and treatment response. Early work emphasized the value of MNV area and FD in distinguishing active from remission states [19]. Later analyses showed that vascular descriptors such as lacunarity may also capture differences in disease stability [20]. More recently, comparative investigations demonstrated high reproducibility of MNV area across devices but limited consistency for VD, highlighting the importance of standardized imaging protocols for longitudinal assessment [21].

Lacunarity is one of the widely used biomarkers for nAMD. It represents the heterogeneity of vascular architecture, where higher values indicate greater irregularity and larger gaps within the MNV [22]. Previous studies demonstrated that increased lacunarity after treatment was associated with irregular vascular remodeling and a favorable prognosis; however, other studies reported no significant differences across disease states. Therefore, its reliability as a biomarker remains limited due to inconsistent clinical interpretation [23,24]. In contrast, FD provides a quantitative descriptor of vascular branching complexity. Lower FD values generally reflect pruning of fine peripheral vessels and stabilization of the neovascular network after anti-VEGF treatment, and a reduction in FD has been regarded as a marker of favorable therapeutic response [25,26].

In our study, both MNV area and FD showed a significant reduction when comparing baseline at the time of switch with 12 months after brolucizumab initiation. The reduction in MNV area after switching is consistent with previous OCTA studies showing that anti-VEGF therapy can induce regression of neovascular networks [19,27]. To account for the potential confounding effect of prior anti- VEGF therapy, we also examined OCTA parameters from 12 months before the switch. Both MNV area and lacunarity showed significant changes only when comparing 12 months before and 12 months after the switch, and in our study, lacunarity tended to increase, possibly reflecting pruning of smaller vascular branches and the development of larger non-perfused areas within the MNV complex [23,24]. VD, on the other hand, had already decreased significantly during the preceding anti-VEGF treatment period and therefore did not show further reduction after switching.

Among these parameters, the consistent and significant decrease in FD is particularly noteworthy. Our Bonferroni-adjusted analysis revealed that FD declined not only during the 12 months of prior anti-VEGF therapy but also continued to decrease following the switch to brolucizumab, with the post-switch reduction remaining significant. In addition, in each case, FD values showed a marked reduction at time points when the retina was anatomically stable. This suggests that the complexity of the MNV vasculature can progressively diminish with ongoing anti-VEGF therapy, and that brolucizumab may further contribute to the simplification of lesion architecture beyond the effect of previous agents.

Previous real-world studies and case reports have highlighted that brolucizumab, while effective in achieving anatomical improvement and extending treatment intervals, may be associated with adverse events such as intraocular inflammation, vasculitis, or even vascular occlusion [28,29]. In our cohort, we observed a single case of nonocclusive intraocular inflammation after the fifth brolucizumab injection, presenting with conjunctival injection, ocular pain, and anterior chamber cells; the condition improved with topical corticosteroids, and treatment was subsequently switched to aflibercept. No systemic adverse events occurred, and no cases of retinal vasculitis or vascular occlusion were identified. Floater complaints were noted in two eyes but were not classified as adverse events. These findings underscore the need for vigilant monitoring while suggesting a generally acceptable safety profile in this switch population.

This study has several limitations. First, its retrospective design, relatively small sample size, and single-center setting may limit the generalizability of the findings. Second, the variability in prior anti-VEGF treatment regimens and the fact that patients received individualized treat-and-extend approaches rather than a standardized protocol before switching could have influenced the outcomes. In addition, because patients who were switched to brolucizumab typically had recurrent or refractory disease, there is an inherent selection bias. While OCTA provides valuable noninvasive insights into MNV morphology, projection artifacts and segmentation errors remain potential sources of bias, despite careful manual correction. The quantitative OCTA parameters are also influenced by image quality and analysis methods, which may limit comparability across studies. Finally, the follow-up period was limited to 12 months, which may not capture long-term treatment effects or disease progression.

In conclusion, switching to brolucizumab in eyes with nAMD previously treated with anti-VEGF agents resulted in anatomical improvement and extended treatment intervals while maintaining visual acuity. Quantitative OCTA analysis demonstrated significant reductions in MNV area and FD after the switch, with FD showing a significant decline even after accounting for prior injections. These findings support the potential of FD as a sensitive indicator of vascular remodeling under sustained anti-VEGF therapy. Future prospective studies with larger cohorts and standardized imaging protocols are warranted to validate these results and to establish optimal treatment strategies for eyes refractory to prior therapy.