Purchase this article with an account.
M. Fleckenstein, A. Bindewald–Wittich, S. Schmitz–Valckenberg, H.P. N. Scholl, J. Dreyhaupt, U. Mansmann, F.G. Holz, FAM study group; Patterns of Abnormal Fundus Autofluorescence in Advanced Atrophic AMD – Impact on Progression and Sample Size Calculation for Interventional Clinical Trials . Invest. Ophthalmol. Vis. Sci. 2006;47(13):5687.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
We recently identified distinct patterns of abnormal FAF in the junctional zone of geographic atrophy (GA) (BJO 2005;89:874–8). Here we investigated the impact of such FAF phenotypes on GA progression rates. Based on these analyses we performed sample size calculations for interventional trials in this subset of AMD patients aiming at slowing GA progression and, thus, prevent additional visual loss and scotoma enlargement.
In the prospective, multicenter natural history FAM study FAF images were recorded with a confocal scanning laser ophthalmoscope (cSLO) in 225 eyes from 147 patients over a median of 21 months. GA areas were quantified with digital image analysis and the impact of distinct patterns of abnormal FAF in the junctional zone of GA on progression rate was evaluated. Based on a linear regression model sample size was calculated for an interventional, randomized, double blind, parallel, clinical trial aimed to reduce speed of GA growth of 0,5 mm²/year at a significance level of α < 0.05 with a power of 80%.
Areas of GA showed a mean enlargement of 1.92±1.89 mm2/year. Eyes with diffuse and patchy FAF patterns had a significantly higher mean rate of progression (2.05±1.93 mm2/year) compared to eyes with no FAF abnormalities, focal or band patterns outside the GA (1.29±1.02 mm2/year; P=0.0009). In consideration of differential progression rates, a mean change of 1.85 mm² per year of the GA area with a total variability of 2.96 was determined. Sample size calculation based on these data gave186 patients per study arm.
Based on cSLO FAF phenotyping we identified patterns of abnormal FAF in the junctional zone of GA associated with high and low progression rates of GA, respectively. Consideration of high–risk characteristics and, thus, prognostic factors is helpful to design a clinical trial in patients with GA. Based on our data we performed sample size calculations that considered ‘fast progressers’ and were thus able to reduce required sample sizes, which in turn would also increase the likelyhood to detect a therapeutic effect e.g. of a pharmocological agent to slow down rates of atrophy spread.
This PDF is available to Subscribers Only