ICI Use Tied to Lower Non-Neovascular AMD Risk; Diabetic Retinopathy Screening Disparities Found

Two recent ophthalmology studies published in peer-reviewed journals present distinct findings: one suggests immune checkpoint inhibitors may lower risk of non-neovascular age-related macular degeneration, while a systematic review identifies significant sociodemographic disparities in diabetic retinopathy screening adherence.

A study published in the American Journal of Ophthalmology used the TriNetX Global Collaborative Network, which includes more than 180 million patients from 151 healthcare organizations, to investigate the association between immune checkpoint inhibitor (ICI) use and age-related macular degeneration (AMD). Researchers identified 49,337 patients treated with ICIs and 4,357,830 patients with cancer who never received ICIs. After 1:1 propensity score matching on demographics, cardiovascular and metabolic comorbidities, cancer-related factors, smoking, obesity, and health-service utilization, 36,037 patients remained in each group. Patients were followed for up to 5 years with a 3-month lag.

The analysis found that ICI exposure was associated with a significantly lower incidence of non-neovascular AMD. Early AMD showed a hazard ratio (HR) of 0.70, early-intermediate AMD an HR of 0.75, and non-neovascular AMD overall an HR of 0.77. No association was observed with neovascular AMD. In melanoma-specific sensitivity analyses, the protective association was stronger, with HRs of 0.43 for dry AMD, 0.48 for intermediate AMD, 0.44 for early-intermediate AMD, and 0.50 for overall non-neovascular AMD. In patients with metastatic malignancies, ICI treatment was likewise associated with lower risks of non-neovascular AMD (HR, 0.55) and degeneration of the macula and posterior pole (HR, 0.69). The positive control outcome of colitis was more frequent in ICI-treated patients (HR, 2.14), while the negative control outcome of femoral fracture did not differ significantly (HR, 0.78), supporting the matching strategy's robustness. Study limitations included the retrospective design, unavailability of adherence and drug dosage data, and loss to follow-up.

Separately, a systematic review published in Retina examined factors influencing adherence to diabetic retinopathy (DR) screening. The review analyzed 33 studies involving more than 100,000 participants across North America, Europe, the Middle East, and Asia. It confirmed that age, income, education, employment, ethnicity, and geography significantly influence screening participation. Screening participation rates were 76% to 86% for patients older than 60 years compared to 32% to 68% for those younger than 40. Participants with higher education and socioeconomic status, including private health insurance, were significantly more likely to engage in routine screening.

Low-income individuals, those on Medicaid, or those without health insurance demonstrated adherence rates between 25% and 48%. Patients with Black or Hispanic ethnicities showed lower screening rates overall compared to non-Hispanic White and Asian groups, particularly among participants in the US and UK. Working individuals, especially those with rigid schedules or engaging in manual labor, had lower participation rates, as did those experiencing transportation and distance barriers to care. For example, in the UK and Ireland, those living beyond 2-9 kilometers and 60 kilometers, respectively, demonstrated significant declines in screening adherence. Study limitations included hindering efforts to conduct a meta-analysis, inclusion of self-reported screening data potentially introducing recall bias, and social desirability bias.