Abstract
Purpose :
The development of effective therapies for non-advanced age-related macular degeneration (AMD) has been hindered by a lack of sensitive, reversible endpoints to be used in clinical trials. Our group has previously reported (1) on the Ora Variable Contrast Flicker (Ora-VCF™) test which has successfully identified differential contrast sensitivity (CS) thresholds for non-advanced AMD patients as compared to matched controls. The present study analyzed the longitudinal data from a real-world cohort (2) to estimate the required sample size for a short duration clinical trial (2 years) in patients with non-advanced AMD.
Methods :
33 non-advanced AMD patients with visual acuity of 20/25 or better (at the entry point) and age- and visual acuity- matched controls were included. On a single study visit, all participants completed a battery of visual function assessments, including Ora-VCF™ tests. During the Ora-VCF™ test, patients viewed flickering targets presented via a customized software.(1) All patients received baseline and annual testing for up to 4 years. Rates of change were measured with linear mixed effects models and the parameters were used for sample size calculations, which were based upon the work of Gelman.(3) Type-1 error was set at 5% and the effect size at 30%. We compared 3 testing paradigms: (A) 1 test every 3 months, (B) 3 tests every 3 months, (C) 1 test every week, and (D) 3 tests every week. All paradigms assumed a 2-year trial, one study eye per patient, and 1:1 randomization.
Results :
At 80% power, paradigms A, B, C, and D required a minimum of approximately 4000, 2000, 1400, 1000 patients (half in each group), respectively. The Figure the depicts the relationship between power and sample size for paradigm D. Intense, home-based testing can therefore reduce sample size requirements to one quarter of conventional, office-based testing.
Conclusions :
The Ora-VCF™ test could be used as primary endpoint for clinical trials in non-advanced dry AMD with as few as 500 patients per arm followed for 2 years. To put in perspective, the AREDS trial investigated 4757 participants followed for 5 years. These findings have implications for the design of shorter, less expensive trials for a disease stage that currently lacks better treatment options.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.