The work, published July 25 in Translational Vision Science and Technology, shows that being of African heritage is an independent risk factor for this drastic decline in vision, and should prompt more eye screening in this population for early glaucoma detection.
This is the first study to use an artificial intelligence algorithm to break down visual field loss in new-onset glaucoma cases among United States-based population groups.
The work is important because glaucoma is the leading cause of blindness in the United States and primary open angle glaucoma is the most common type. POAG is the leading cause of optic nerve degeneration that is related to the pressure level inside the eye, but other factors also contribute to this condition.
“This study has tremendous implications for glaucoma screening of Blacks, who we already knew were a population at increased risk of glaucoma,” says senior author Louis R. Pasquale, MD, FARVO, Deputy Chair for Ophthalmology Research at the Icahn School of Medicine at Mount Sinai and Director of the NYEE Eye and Vision Research Institute.
“Screening earlier in life could significantly increase the chance of detecting glaucoma and slowing down progression before it reaches one of the advanced patterns shown in our research.”
A team of researchers analyzed nearly 210,000 participants from three population-based databases of nurses and health professionals from the Nurses’ Health Study (enrolled between 1980 and 2018, and 1989-2019), and the Health Professionals Follow-up Study (enrolled between 1986 and 2018). Participants were over the age of 40 and their data was collected during comprehensive eye exams—none had glaucoma at baseline.
They were followed biennially and provided updated information on their lifestyle, diet, and medical status, including glaucoma diagnosis.
Within the study group, 1,946 patients developed glaucoma. Researchers analyzed their earliest record of visual field loss using archetype analysis, a form of artificial intelligence. The algorithm identified 14 archetypes: four representing advanced loss patterns, nine of early loss, and one of no visual field loss.
Black patients made up 1.3 percent of the study, but had a nearly twofold increased risk of early visual field loss archetypes, and a sixfold higher risk for advanced field loss archetypes, when compared to white patients.
Hispanic patients made up 1.1 percent of the study population, and did not have an increased risk of any archetypes compared to white patients; however, the study showed they were at risk of an archetype showing initial loss near the center of their visual field.
The results were controlled for a number of variables including socioeconomics, frequency of eye exams, heart disease, diabetes, and exercise.
“This study started decades ago in three health professional cohorts that were not as diverse as current numbers—and if we collected more representation of people of color, the results would likely be even more profound,” Dr. Pasquale adds.
“This emphasizes the importance of early screening strategies in Blacks to identify early-onset glaucoma so that visual disability in this population is averted.”
Dr. Pasquale says the next step for this work is to figure out the specific risk factors for the different patterns of visual loss seen in glaucoma patients—including genetic and environmental factors—so as to fully disentangle the pathogenesis of primary open angle glaucoma.
This study was done with collaborators at Brigham and Women’s Hospital and Massachusetts Eye and Ear.
The overall objective of our study was to test the hypothesis that immune activation is associated with glaucoma. To this end, we tested whether SNPs in TNF-α and IL-2 genes are associated with g laucoma. We also measured cytokine TNF-α and IL-2 levels in the plasma of clinically well-defined POAG cases and compared them to controls to evaluate if these plasma cytokines serve as biomarkers for POAG. The practicality of aqueous humor sampling as a clinical diagnostic tool is dubious.
Though previous studies have obtained aqueous humor or anterior chamber tissues from POAG patients to study differences in cytokine concentrations, this method is invasive and carries risks and complications. [23,34,35,36] Blood is easily accessible and offers a potential screening modality for diagnosing POAG. Hence, we isolated and screened plasma from blood samples collected from our POAAGG cohort. Patients with Alzheimer’s disease have differing plasma concentrations of certain signaling proteins compared to healthy subjects, suggesting that dysfunction in the central nervous system is accompanied by the systemic changes found in plasma [37]. A similar relationship could extend to POAG.
Patients with glaucoma are known to have atypical T-cell subsets and higher levels of serum antibodies against proteins in the retina and optic nerve [38]. The fact that the cellular milieu is altered in POAG implies that the immune system contributes strongly to the onset and development of optic neuropathy. Additionally, research has shown that the expression patterns of genes in the TM and Schlemm’s canal are like those of peripheral leukocytes circulating in the blood [39].
IL-2, which is regulated by Th1 cells and triggers the proliferation of T-cells, has previously been studied as a biomarker. However, it has never been investigated in an African American cohort. Prior studies have measured IL-2 levels in aqueous humor, vitreous humor, tear film, and peripheral serum samples without targeting a specific ethnicity. Serum IL-2 concentrations in POAG cases from other ethnicities were much lower than those reported in our plasma findings, which could be due to differences in ethnicity, environment, underlying comorbidities, and procurement and the processing of blood samples [20,38]. Additionally, the prior studies had smaller sample sizes (~ 50 patients), likely contributing to the lack of significant differences between groups.
Our findings did not reveal a statistically significant difference in plasma IL-2 levels between POAG cases and controls. Mean concentrations were similar among the two groups. Due to the lack of longitudinal data, it is not known how IL-2 titers change throughout the course of POAG. IL-2 in cases may be lower than in controls in the early stages of disease before increasing, or vice-versa. Additionally, we observed a mild, but significant negative correlation between IL-2 concentration and mean IOP.
Although it is difficult to infer any causality, this suggests that IL-2 levels may contribute to a decreased IOP by affecting a mechanism such as outflow facility or endothelial permeability. This could also explain why IL-2 levels might be lower in POAG patients than controls. However, there is no evidence in the literature documenting such a physiology. There was no correlation between IL-2 levels and the other endophenotypic values documented in our population, such as CCT, CDR, RNFL mean thickness, or baseline MD.
While TNF-α has been studied as a biomarker in various populations, it has yet to be explicitly measured in an African American population [30,40,41]. Previous studies have measured TNF-α in the aqueous humor of POAG and normal patients without specifically measuring levels within a particular racial or ethnic group [28,29]. TNF-α is reported to be significantly higher in the plasma of POAG cases in a Saudi Arabian population.
Compared to our findings, the overall TNF-α levels measured by ELISA were much lower than those reported in our findings [30], which could be due to differences in population, environment, modifying genes or other factors. It has previously been reported that increased levels of TNF-α and TNF-α-1 receptor, which result in an apoptotic response, have been found in higher quantities in retinal sections of glaucomatous eyes when compared to control donor eyes [42]. Although our results did not show a significant difference between plasma TNF-α levels in controls when compared to POAG cases, we did observe higher expression levels for plasma TNF-α in POAG cases when compared to normal subjects, which agrees with previous findings [30].
Our results may additionally be altered due to undiagnosed systemic diseases affecting TNF-α levels. The correlation between CDR, RNFL, MD, PSD, CCT and IOP and TNF-α levels did not reach significance with any tested endophenotypes. Analysis of a larger POAAGG cohort, POAG subtypes such as early and advanced POAG and a common set of diagnostic criteria for both cases and controls may provide a better understanding of the role of TNF-α and POAG. Once a better understanding of TNF-α and its effect on POAG is obtained, it would be interesting to further elucidate the mechanism by which it contributes to RGC death by focusing on the receptors that are activated by TNF-α.
The results from our study could also be influenced by underlying systemic diseases that may influence IL-2 production. Cytokines, such as TNF-α, have been shown to change based on age, infection, or other system conditions including hypertension and diabetes [43]. Patients with infectious or autoimmune disease were not excluded from our study. Our patient exclusion criteria were confounding ocular conditions. They did not include chronic inflammatory systemic illnesses such as systemic lupus erythematosus, sarcoidosis, or vasculitides. These diseases are known to upregulate inflammatory pathways, and likely impact cytokine levels.
A strength of our study was that samples were obtained from a single racial population, and this race is self-reported. However, self-reporting alone has previously been shown to be an imprecise measure of genetic ancestry. A final limitation is the lack of longitudinal data. It may be clinically relevant to record the change in TNF-α and IL-2 expression with disease progression or effective glaucoma treatment.
Although IL-2 is a major activator of T-cell proliferation, our data did not show increased levels in POAG patients. This suggests that IL-2 alterations in POAG plasma are complex and warrant further investigation. Cytokines such as IL-2 may serve a role beyond stimulating an inflammatory response. As glaucoma is on a continuum and POAG may not be just one disease but a multitude of diseases, a panel of biomarkers would be useful for categorizing the various subtypes of POAG.
It is possible that plasma IL-2 levels may not be significantly altered between POAG cases and controls, but there might be a difference when the population is divided into more specific subgroups. Further studies will be helpful to explore and understand the precise mechanism of IL-2 and its pathway that is involved in the pathogenesis of POAG. Future research can also be directed toward investigating proteins that are not involved in inflammatory or oxidative stress pathways as potential biomarkers for POAG in African Americans.
Previous research has indicated specific parameters for which plasma was obtained from patients, such as at the time of cataract surgery while in a fasted state [44]. However, in our study, the plasma was not obtained at set times during the day, nor were patients in controlled satiety states. To account for these potential factors, we normalized our data to the corresponding total protein concentration for a given sample. However, we still demonstrated a large standard deviation in both the control and case populations.
Alternatively, a panel of biomarkers may be needed to fully characterize the various endophenotypes of glaucoma, as this disease demonstrates a spectrum of clinical presentations. It is also possible that cytokines, IL-2 and TNF-α plasma levels may not be significantly different between cases and controls without further separating and dividing the POAG population into more specific groups. In future, other proteins, outside inflammatory and apoptotic pathways, could also be investigated as potential biomarkers to better understand this disease [45].
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8774833/
Original Research: The findings will appear in Translational Vision Science & Technology
