Early Macular Retinal Ganglion Cell Loss in Dominant Optic Atrophy: Genotype-Phenotype Correlation

PURPOSE: To assess the peripapillary retinal nerve fiber and macular retinal ganglion cell (RGC) loss in patients with dominant optic atrophy (DOA) stratified by OPA1 mutation type. DESIGN: Cross-sectional study. METHODS: We studied 39 patients from 28 pedigrees with DOA harboring heterozygous mutations in the OPA1 gene along with 45 age-matched healthy subjects. The retinal nerve fiber layer (RNFL) and ganglion cell-inner plexiform layer (GC-IPL) of patients with DOA were evaluated by optical coherence tomography (OCT) and compared to those of controls. Patients' eyes were divided into 4 groups based on increasing severity of visual loss (DOA1 to DOA4) and were stratified by OPA1 mutation type. RESULTS: The average thicknesses of the RNFL and GC-IPL were smaller in patients with DOA than in healthy controls (P < 0.0001). RNFL analysis showed a significant reduction of the average, superior and inferior quadrants thicknesses in the DOA4 group compared to the DOA1 group (P = 0.001, P = 0.002 and P = 0.001, respectively). GC-IPL analysis showed a significant thinning in the superotemporal and superior sectors in the patients with DOA2 compared to those with DOA1 (P = 0.046 and P = 0.04, respectively). Stratifying by mutation type, average, superior and nasal RNFL thinning was significantly more severe in missense mutations and had a presumed dominant-negative effect compared to mutations causing haploinsufficiency. CONCLUSIONS: The present study demonstrates that in DOA, loss of macular RGCs is the earliest pathologic event, better reflected by GC-IPL measurements, whereas RNFL thickness is a measure of spared axons in late stages of the disease. Thus, mild cases (DOA2) show significant macular RGC loss as opposed to substantial maintenance of RNFL thickness, which is significantly decreased only in severe cases (DOA4). A clear genotype/phenotype correlation emerged, stratifying OCT measures by OPA1 mutation type, missense mutations being the most severe. (C) 2014 by Elsevier Inc. All rights reserved.