Rate of bimatoprost implant biodegradation in the phase 3 ARTEMIS studies

Abstract

Purpose : Bimatoprost implant 10 µg (Durysta) is a biodegradable, intracameral implant (cylindrical diameter ~200 µm, length ~1.1 mm) that releases bimatoprost steadily to lower intraocular pressure and eventually converts to water and CO₂. The implant elutes bimatoprost for ~4 months but the polymers in the implant matrix may last longer. We evaluated the rate of biodegradation of the implant in our randomized, controlled phase 3 clinical trials.

Methods : Two 20-month, phase 3 trials (NCT02247804, NCT02250651) randomized 1 eye per patient with open-angle glaucoma or ocular hypertension to an intracameral 10- or 15-µg bimatoprost implant administered on Day 1, Weeks 16 and 32 or topical timolol 0.5% BID. Estimated implant size on gonioscopy as a percentage of the initial size was recorded. We pooled data for the 10-µg implants placed on Day 1 in both trials for analysis. The population mean implant size (expressed as the percentage of the initial size) over time was modeled with a nonlinear mixed-effects model with repeated measures (MMRM) approach. The rate of biodegradation was defined as the decrease in implant size (expressed as a percentage of the initial size) per month.

Results : A total of 230 eyes were included in the analysis. Implant size data were best fit to a model that considered inter-patient variability and included an exponential function for the implant size decrease (Figure). The estimated mean (± standard error) percentage decrease in implant size from baseline was 24.1% ± 2.3% at Week 28 and 80.2% ± 1.5% at Week 52. The estimated mean rate of implant biodegradation was 3.7%/month through Week 28, 10.2%/month after Week 28 through Week 52, and 2.4%/month after Week 52 through Month 20.

Conclusions : The estimated mean size of 10-µg bimatoprost implant administered on Day 1 in phase 3 trials decreased according to an exponential function, which allows one to predict the size of a single or multiple implants in the AC. Clinical studies are in progress to further understand implant biodegradation and the ideal timing for implant re-administration.

This is a 2021 ARVO Annual Meeting abstract.

 

Figure. MMRM model of mean implant size relative to initial size. The model parameters were estimated using the SAS NLMIXED procedure where y_it is patient i’s implant size at time t; β₁, β₂, and β₃ are fixed-effects parameters for the nonlinear mean implant size profile over time; b_i is the random effect corresponding to patient i; and ε_it is the error term.