Oral Emixustat Hydrochloride Ineffective for GA Risk Reduction in AMD Patients

Philip Rosenfeld, MD, PhD

A randomized, 24-month, 49-site, phase 2b/3 clinical trial has determined that oral emixustat hydrochloride is ineffective at reducing the growth rates of geographic atrophy (GA) in patients with GA secondary to age-related macular degeneration (AMD).

The Safety and Efficacy Assessment Treatment Trials of Emixustat hydrochloride (SEATTLE) study, found that GA lesions in study and control eyes grew at similar rates, and did so despite the presence of higher and lower risk alleles for AMD. Although the study failed to achieve its primary (mean annual growth rate of total GA area in study eyes) and secondary efficacy (change from baseline best corrected visual acuity [BCVA]) endpoints, researchers were able determine the presence of 5 specific baseline characteristics associated with more rapid GA growth and greater loss of vision.

Emixustat is a compound that inhibits the development of retinal pigment epithelium-specific 65 kDa protein (RPE65) associated with atrophy. Philip Rosenfeld, MD, PhD, study lead author and director of the Bascom Palmer Eye Institute at the University of Miami's Miller School of Medicine, wrote that the "progressive dysfunction of the RPE is a key feature" in the complex and multifactor pathogenesis of AMD" and that emixustat had been proven in previous preclinical analysis to slow the visual cycle and inhibit the activity of RPE65 and associated loss of photoreceptors associated with RPE dysfunction.

In order to determine whether emixustat could reduce the rate of GA enlargement in AMD patients, 503 healthy subjects (≥ 55 years of age) diagnosed with GA (total area 1.25 - 18 mm²) secondary to nonexudative AMD in 1 or both eyes; a BCVA score in the study eye of ≥ 35 Early Treatment Diabetic Retinopathy Study letters (ETDRS) and no history of choroidal neovascularization (CNV) were enrolled in the study.

Subjects were randomized 1:1:1:1 to receive emixustat orally once-daily for the 24-month period at 2.5 mg (n= 133), 5 mg (n= 134), 10 mg (n= 103), or placebo (n= 133).

Study data were collected on BCVA in normal and low luminance conditions, and slit-lamp biomicroscopy, dilated ophthalmoscopy, and spectral domain optical coherence tomography (SD OCT) were performed. Rosenfeld and colleagues also collected data through "color fundus photography, fundus autoflourescence (FAF) imaging, and intraocular pressure (IOP) measurements" performed at baseline, 6-, 12-, 18-, and 24- months, and flourescein angiography at screening and 24-months.

Rosenfeld and colleagues reported that the mean total area of GA among the 320 subjects completing the 24-month study was equal among all 4 treatment arms. Study data showed that the mean annual growth rate remained similar to the expected average growth rate of 1.75 mm² per year (standard deviation; 1.20 mm² per year) among treatment groups during the clinical trial (emixustat 2.5 mg: 1.69 [0.12] mm²/ year; 5 mg: 1.83 [0.12] mm²/year; 10 mg: 1.84 [0.14] mm²/year; placebo: 1.69 [0.11] mm²/year; P ≥ 0.81).

Similarly, despite that mean changes in BCVA from baseline ranged from more than 25 letters to less than 84 letters at baseline, the mean change in BCVA remained similar across treatment groups.

Although the study failed to meet its primary endpoints, Rosenfeld and colleagues did determine that 5 specific baseline characteristics were associated with increased GA growth and greater BCVA losses. Subjects with greater baseline GA saw more rapid growth in GA lesion size (≥ 6.0 mm²) than those with smaller or multifocal lesions at baseline.

Similarly, subjects with a BCVA EDTRS of ≥ 20 letters saw a greater mean change (-10.8 letters; 95% CI) from baseline in comparison to those subjects with < 20 letters (-1.3 letters; 95% CI). Other baseline characteristics which were associated with rapid GA growth were a lack of "RPE atrophy under the foveal center, presence of reticular pseudodrusen, and a larger LDD [low luminescent VA deficit]."

Baseline characteristics associated with a greater loss of BCVA were "larger LLD and a shorter history of GA," researchers wrote.

Rosenfeld bited that although the results of the clinical trial do not support modulation of the visual cycle by emixustat as an efficacious treatment strategy for GA secondary to AMD, the study remains significant "because it confirms the natural history of GA growth rate in a large population" and confirms associations of baseline GA characteristics with GA growth rate.

The study, "Emixustat Hydrochloride for Geographic Atrophy Secondary to Age-Related Macular Degeneration: A Randomized Clinical Trial," was published online in Ophthalmology.