While anti-retinal autoantibodies (ARA) are thought to be a marker of autoimmune retinopathy (AIR), recent studies have suggested that the specificity of many of the ARAs make them less helpful in clinical practice. This study evaluated the profiles of several anti-retinal antibodies in patients with autoimmune retinopathy, with retinitis pigmentosa, and in normal control patients.
Study design
This retrospective multicenter study examined plasma samples of 49 patients with a clinical diagnosis of AIR who visited 7 tertiary care hospitals in Korea. Twenty patients with typical retinitis pigmentosa (RP) served as disease controls while 30 healthy participants served as normal controls. Plasma samples of patients and controls were acquired from previous studies. Western blot was used to detect the presence of 6 ARAs in each sample.
Outcomes
Autoantibodies against CRX/CORD2, HSP60, and aldolase C were found in higher rates in patients with AIR than in patients with RP and normal controls, while α-enolase and CAII were seen at similar rates. Antibodies against recoverin were found in 3 patients with AIR, and not seen in normal controls or patients with RP. The area under the ROC curve was 0.531 for anti-recoverin, 0.479 for anti-α-enolase, 0.489 for anti-CAII, 0.737 for anti-CRX/CORD2, 0.637 for anti-HSP60, and 0.664 for anti-aldolase C. A higher number of ARAs was associated with AIR (≥4 antibodies was seen in 32.6% of patients with AIR, 5% of patients with RP, and 3% of controls). A higher number of ARAs were slightly associated with photodisruption on OCT and severe dysfunction on ERG.
Limitations
This was a retrospective study that used Western blotting to detect the ARAs at various time points in the disease, which could have impacted the results. In addition, a 2-tier design with confirmatory immunohistochemistry of the human retina was not employed.
Clinical significance
This study demonstrates that patients with AIR have a higher number of ARAs than heathy controls and patients with RP. However, it is important to note that more than 80% of normal individuals were found to have ARAs (other than anti-recoverin). In addition, the study shows that antibodies against α-enolase and CAII, 2 of the most commonly cited ARAs, were found in similar percentages among those with AIR and normal individuals. Therefore, ARAs alone are not sufficient to make a diagnosis of AIR, which must be made in the correct clinical context.