A Phase III trial examined how well different measurements predict vision changes in people with neovascular age-related macular degeneration. The study included 290 participants, covering both those new to treatment and those previously treated. Researchers compared AI-based quantification of fluid volumes against standard central subfield thickness measurements.
The findings showed that AI models tracking fluid in a 6-millimeter area explained the largest proportion of vision variance. In patients who had received prior treatment, larger fluid volumes were linked to higher thickness values. However, the AI model fit was poorer in patients who had not yet received treatment compared to the standard thickness model.
The study highlights that standard thickness measurements may miss important details about disease activity. The researchers emphasize the need for distinct fluid volume quantification to accurately track visual function loss or benefit. This approach is particularly relevant for understanding fluid in regions beyond the central 1-millimeter area and across different treatment durations.