PurposeBevacizumab, ranibizumab, and aflibercept are commonly used to treat neovascular age-related macular degeneration (nAMD). The results of various interventional, mostly randomized head-to-head studies, indicate statistical non-inferiority of these three drugs. The results of these studies are often interpreted as the three drugs being freely interchangeable, resulting in some health systems to pressure ophthalmologists to preferentially use the less expensive bevacizumab. This study analyzes switching from aflibercept or ranibizumab to bevacizumab and back under real-world conditions in order to investigate the assumption of interchangeability of the drugs.MethodsTreatment data of IVT patients with diagnosed nAMD were extracted from the clinical Berlin Macular Registry database. Patients who underwent a drug switch from aflibercept or ranibizumab to bevacizumab were subject of this study. Statistical comparisons were pre-planned for best corrected visual acuity, central retinal thickness, macular volume, and length of injection interval. Additional endpoints were analyzed descriptively.ResultsMean visual acuity decreased from 0.57 ± 0.05 under aflibercept/ranibizumab to 0.68 ± 0.06 logMAR after the switch (P = 0.001; N = 63). CRT increased from 308 ± 11 µm to 336 ± 16 µm (P = 0.011; N = 63). About half of the subjects were switched back: visual acuity increased from 0.69 ± 0.08 logMAR to 0.58 ± 0.09 logMAR (N = 26). CRT decreased from 396 ± 28 to 337 ± 20 µm (N = 28).ConclusionThe data provides real-world evidence that there is loss of visual acuity and an increase in retinal edema after switching to bevacizumab. Thus, the assumption of free interchangeability cannot be confirmed in this cohort.