Assessment of Visual Field Loss Progression for DOA and Wolfram Syndrome employing a DOA-OPA1 Archetypal Analysis Model

Abstract

Purpose : Analyze and quantify the progression of visual field (VF) loss in Dominant Optic Atrophy (DOA) and in Wolfram Syndrome based on DOA-OPA1 Archetypes

Methods : Twenty patients (40 eyes) affected by molecular confirmed OPA1 gene mutation (dominant optic neuropathy form) with multiple VFs performed at different time points were considered. For the Wolfram Syndrome (recessive optic atrophy), 16 patients (32 eyes) were enrolled. SITA standard 30-2 or 24-2 Humphrey VF analyzer (Carl Zeiss Meditec, Dublin, CA, USA) tests were collected. VFs from both optic atrophy groups were decomposed into patterns by employing the unsupervised machine learning Archetypal Analysis (AA) model developed by our group for DOA OPA1, composed of 9 archetypes (ATs). Progression was assessed by evaluation AT weight change and considering an archetype change score according to Doshi (Ophthalmology, 2022).

Results : The VF-decomposition for the OPA1 group was mostly made up of ATs resembling central or ceco-central defects, whereas Wolfram Syndrome VF-decomposition had a higher frequency of more severe ATs as total loss. A slow progression was detected for the OPA1 group, with a low change in the ATs weight over time. In turn, the Wolfram Syndrome group showed a more pronounced change of weight over time, with a decreasing weight for the AT resembling a normal VF and an increase of the more severe vision loss resembling VFs.

Conclusions : The implementation of the DOA-OPA1 AA-model as basis for progression evaluation of the same optic neuropathy form (OPA1) and a recessive form (Wolfram Syndrome) allowed insights of the distinct disease’s course. Whereas the first appears to have a slow progress, the second seems to worsen over time. Furthermore, for both DOA-OPA1 and Wolfram Syndrome, the course and severity of disease progression might be age dependent.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.