Purchase this article with an account.
Gui-Shuang Ying, Maureen G Maguire, Jiayan Huang, Daniel F Martin, ; Visual Acuity Response Pattern and Prediction in the Comparison of AMD Treatments Trials (CATT). Invest. Ophthalmol. Vis. Sci. 2014;55(13):866.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To describe the response patterns of visual acuity (VA) and determine whether the early VA response predicts VA after 2 years of anti-VEGF treatment of neovascular age-related macular degeneration.
1185 CATT participants were randomly assigned to ranibizumab or bevacizumab with dosing regimens of monthly or PRN for 2 years or 1 year of monthly and 1 year of PRN. VA results during follow-up were used in a trajectory analysis to identify 3 VA response groups. The ability of early VA response (at Week 4 or 12) to predict late VA response at Year 2 was assessed by the agreement between early and late VA change categories, and R2 from multivariate linear regression for VA change.
Trajectory analysis: The VA response groups (Figure 1) identified were: (1) large loss (n=76, mean change -30 letters); (2) stable (n=684, mean change ≤4 letters),; (3) large gain (n=417, mean change +20 letters). VA response groups differed in baseline VA, type of lesion, OCT fluid and thickness (p<0.05). Prediction analysis: Among 440 eyes with VA gain of ≥5 letters at Week 4, 80% had a ≥5 letter gain at Year 2, while among 133 eyes with VA loss of ≥5 letters at Week 4, 38% had a ≥5 letter gain at Year 2. Among 553 eyes with VA gain of ≥5 letters at Week 12, 78% had a ≥5 letter gain at Year 2, while among 127 eyes with loss of ≥5 letters, 27% had a ≥5 letter gain at Year 2 (Table 1). Among 360 eyes with VA gain of ≥5 letters at both Week 4 and 12, 87% had a gain of ≥5 letters at Year 2, while among 62 eyes with VA loss of ≥ 5 letters at both Week 4 and 12, 22% had ≥5 letter gain at Year 2. VA response at Year 2 was better predicted by response at Week 12 (R2=0.31) than by VA response at Week 4 (R2=0.17) or by baseline VA predictors alone (R2=0.12). Combining VA responses at Week 4 and 12 and baseline predictors moderately improved prediction of VA response at Year 2 (R2=0.38).
The VA response to anti-VEGF treatment varied substantially among subjects. Early VA response only moderately predicted late response. The majority of eyes with early VA gain had the gain at Year 2. However, some eyes with an initial decline in VA had gains late, supporting continuation of treatment.
This PDF is available to Subscribers Only