Recent advances in the diagnosis and management of intraocular inflammatory diseases (uveitis) have led to an increase in the use of polymerase chain reaction (PCR) testing using intraocular fluid and elevated the diagnostic rate of infectious uveitis. Such advances have also resulted in the introduction of biologics, such as tumor necrosis factor (TNF) α inhibitors, into the treatment of noninfectious uveitis. However, there are still some patients diagnosed with unclassified uveitis and some having poor visual outcomes. Therefore, the development of methods for the objective and quantitative assessment of the disease activity of uveitis, an understanding of the pathogenesis of uveitis, and the development of novel treatment strategies based on molecular pathogenesis are needed in the management of uveitis. This study describes the following three parts.
I. Development of objective and quantitative assessment methods for uveitis based on imaging analysis
1) Quantitative analysis of the anterior segment inflammation of uveitis using anterior segment optical coherence tomography (OCT)
Currently, the Standardization of Uveitis Nomenclature (SUN) grading system is widely used to evaluate the disease activity of anterior segment inflammation of uveitis. However, this is a subjective, semiquantitative system; therefore, a highly objective and reproducible evaluation system needs to be established. In the present study, we attempted the automatic quantification of anterior chamber cells in patients with uveitis using anterior segment OCT images. The approximated anterior chamber grading value was calculated based on the logarithmic approximation curve between the number of cellular spots in the anterior chamber and the SUN grading score of anterior segment inflammation by an ophthalmologist. Cross-validation analysis demonstrated a significant correlation between the SUN grading score provided by the ophthalmologist and the approximated anterior chamber grading value based on anterior segment OCT images. These findings suggest the usefulness of this automated method using anterior segment OCT images for the objective and quantitative assessment of anterior segment inflammatory cells in patients with uveitis.
2) Quantitative assessment of fluorescein leakage in fluorescein angiography (FA) images using machine learning
FA is currently performed in identifying fundus lesions and monitoring responses to therapy in patients with uveitis. However, the evaluation of angiographic findings is subjective and qualitative, resulting in measurement variations between assessors. In the present study, we acquired FA images of patients with uveitis using Optos^Ⓡ, a system to capture a widefield retinal ima ge, to quantitatively evaluate the leakage of dye from retinal vessels. Therefore, we developed retinal vessel segmentation algorithms that detected retinal vessels from FA images using machine learning, identified regions of retinal vessels and capillaries using images in the late FA images, and automatically extracted the leakage region of fluorescein dye by identifying differences between the former and later FA images. The mean precision was 43.4% and the mean recall was 52.9%, and there was a significant positive correlation between the automated segmented area (pixels) of retinal vascular leakage and the FA vascular leakage score using the international FA scoring system.
3) Quantitative assessment of disease activity in acute Vogt-Koyanagi-Harada disease using OCT
The present study focused on the subfoveal choroidal thickness (SFCT) at 1 week after initiating corticosteroid pulse therapy using OCT to evaluate the post-treatment clinical course. The frequency of development of sunset glow fundus and cataract at 1 year was significantly higher for the large SFCT group (SFCT≥500 μm) and visual outcomes of this group was poor. Quantitative assessment of SFCT in patients with acute Vogt-Koyanagi-Harada disease who receive corticosferoid pulse therapy may be an important biomarker for evaluating the severity of this disease.
II. New classification of unclassified (idiopathic) uveitis through serum microRNA (miRNA) analysis and understanding its pathogenesis
A recent study has revealed that there are some patients with idiopathic uveitis showing ocular clinical sings for probable ocular sarcoidosis proposed by the International Workshop on Ocular Sarcoidosis (IWOS) (3 or more out of 7 ocular sings). There may be some patients who have a similar phenotype to ocular sarcoidosis. In the present study, we focused on miRNA, which has been gaining attention as a biomarker that can be used to determine disease progression and the diagnosis of malignant disease, and compared the miRNA profile in serum obtained from healthy subjects, patients with ocular sarcoidosis, and patients with idiopathic uveitis with ocular manifestations of sarcoidosis (suspected ocular sarcoidosis). Principal component and hierarchical clustering analyses showed a similar miRNA expression pattern between patients with ocular sarcoidosis and those with suspected ocular sarcoidosis compared with healthy subjects. Pathway analysis performed simultaneously demonstrated that common pathways, including the wingless/integrated (WNT) signaling pathway and the transforming growth factor (TGF)-β signaling pathway, were involved in the two groups. These findings suggest that both groups may share a similar underlying molecular pathology.
III. Novel treatment strategy for uveitis via regulation of the nuclear factor-erythroid 2-related factor (Nrf2) pathway
The present study focused on Nrf2, a stress-activated transcription factor, and used experimental autoimmune uveitis (EAU) which is known as an animal model of uveitis, to examine whether oxidative stress was associated with autoimmune uveitis. EAU was more severe in Nrf2 knockout (Nrf2-KO) mice than in wild-type mice, suggesting that the increased expression of ocular cytokines, retinal gliosis progression, and the activation of microglia in Nrf2-KO mice may contribute to the exacerbation of EAU. Oral administration of the Nrf2 activator 2-cyano-3,12 dioxooleana-1,9 dien-28-oyl imidazoline (CDDO-Im) ameliorated EAU. These findings indicate that Nrf2 may be a potential target for the treatment of refractory uveitis.
Nippon Ganka Gakkai Zasshi (J Jpn Ophthalmol Soc) 127: 367-401, 2023.