Use of Bevacizumab Injections in Medicare Fee For Service in the IRIS® Registry

The Effect of Sample Medication Use on Subsequent Anti-VEGF Agent Selection for Neovascular Age-Related Macular Degeneration

PURPOSE

Medication samples of anti-VEGF agents can represent a good option for retina specialists to provide timely treatment for newly converted neovascular age-related macular degeneration (nvAMD) while prior-authorizations (PA) are pending. Our study examines the effect of medication sample use (ranibizumab or aflibercept) on future anti-vascular endothelial growth factor (VEGF) agent selection in nvAMD.

DESIGN

Retrospective cohort study.

PARTICIPANTS

nvAMD patients who underwent an initial anti-VEGF injection with a sample medication were compared to nvAMD control patients who never received a medication sample.

METHODS

Charts from 2017 through 2020 were reviewed for data regarding demographics, anti-VEGF agent selection, and visual acuity outcomes for both groups. The utilization of different anti-VEGF agents in each group was compared at various time points using chi-square tests for independence of proportions.

MAIN OUTCOME MEASURES

Anti-VEGF agent selection for the first four injections and at one year were examined.

RESULTS

Adherence to the initial agent was high between first and subsequent injections (2^nd, 3^rd, 4^th injection, and 1 year) in sample (96.2%, 95.9%, 91.9%, 93.4%, respectively), and control groups (98.1%, 94.2%, 94.9%, 87.8%, respectively). Bevacizumab usage was significantly lower among eyes receiving samples relative to controls at the second (1.9% vs. 38.7%, p < .001), third (3.1% vs. 41.3%, p < .001), fourth injections (4.7% vs. 40.4%, p < .001), and at 1 year (0% vs. 33.8%, p < .001). Aflibercept usage was significantly higher in sample eyes relative to controls at the second (78.3% vs. 43.4%, p < .001), third (76.3% vs. 41.5%, p < .001), and fourth injections (76.7% vs. 43.4%, p < .001), and at 1 year (77.0% vs. 52.7%, p < .001).

CONCLUSIONS

Sample medications in nvAMD may be initiated for many reasons, including awaiting PA approval. Our study found that eyes receiving a sample anti-VEGF agent (ranibizumab or aflibercept) for their initial injection were less likely to receive bevacizumab at future visits relative to eyes that did not receive an anti-VEGF sample, even after one year of treatment. Given the persistent use of more expensive medications at subsequent injections for patients who were initiated on samples, insurance payors may consider waiving PA requirements for bevacizumab to avoid a paradoxical increase in health-care costs.

Suprachoroidal delivery of bevacizumab in rabbit in vivo eyes: Rapid distribution throughout the posterior segment

PURPOSE

To test the in-vivo bio-distribution and safety of bevacizumab delivery into the suprachoroidal space (SCS) using a novel injection system in a large eye model.

METHODS

Bevacizumab (1.25 mg) was injected into the vitreous (IVT, 50 µL, n = 12) or the SCS, (150 µL, n = 37) of live rabbits. Immunofluorescence and ELISA were used to assess bevacizumab distribution. Intraocular pressure (IOP) measurements, SD-OCT and fundus imaging, electroretinogram, and histology analysis were performed for safety assessment.

RESULTS

Bevacizumab was observed throughout the choroid layers up to the retinal pigment epithelium (RPE), within 1 h following SCS injection. The C_max of bevacizumab in the retina/choroid was 1043 ± 597 μg/gr tissue (mean ± standard error), 40-fold higher than in IVT injected eyes (p = 0.0339). One day following SCS injection, bevacizumab was detected throughout the posterior pole with a two-fold lower concentration. One week post-SCS injection, bevacizumab concentration in the retina/choroid dropped to 2.36 ± 1.32 μg/gr tissue (p = 0.034 vs. 1 h), with a half-life of 20 h. No suprachoroidal blebs, retinal detachment, hemorrhages, inflammation or changes in retinal function were observed up to 2 months following SCS injection. Elevated IOP (+16 mmHg) was observed two minutes post-SCS injection and spontaneously returned to baseline levels within 10 minutes.

CONCLUSIONS

The novel injection system enabled a minimally invasive, safe, and consistent delivery of bevacizumab with rapid distribution throughout the choroid layers up to the RPE in large eyes. Large volumes of anti-angiogenic are delivered in close proximity to the retina due to the high volume distribution.