Interim Analysis of the Retinal Toxicity and Hydroxychloroquine Therapy (INTACT): A Prospective Population-Based Cohort Study

Objectives To estimate the risk of Hydroxychloroquine-related Retinopathy (HCQ-R) among long-term HCQ users (≥5 years) in British Columbia (BC). Methods We conducted a prospective cohort study in BC, using a standardized retina-screening protocol for HCQ-R through a network of 20 ophthalmologic/retina clinics all-over BC, starting in October 2022. Sample: we included patients with rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE) who were on HCQ at enrollment and have been using it for ≥5 years. Over 40 rheumatology practices in BC were engaged to identify eligible participants through their electronic medical record (EMR) system, which were flagged in EMRs. Then, informed consent by their rheumatologist was obtained to be referred to an ophthalmologic/retina clinic closer to them for annual HCQ-R screening using macular Spectral-Domain Optical Coherence Tomography (SD-OCT). Normal scans were booked for the next year’s annual screening. Outcome: We assessed HCQ-R events defined as equivocal or abnormal scans based on the eye specialist diagnosis, which then were uploaded to a secure cloud platform for independent, masked review and staging by 2 retina specialists, with discrepancies resolved through consensus. For quality control, 30% of normal scans were also randomly selected for review. Statistical analyses: We estimated risk of HCQ-R using the cumulative incidence function (CIF), accounting for right censoring and competing risk of death. Results There were about 3,000 eligible participants detected and flagged in the EMRs of BC rheumatologists. From October 1, 2022, to September 31, 2025, 1,300 referrals (~40% of flagged) were received; 952 participants completed a first visit, 254 a second, and 54 a third. Participants were predominantly female (84.3%) with mean age of 58.2±15.0 years, and 43.5% had SLE (Table 1). Sixteen HCQ-R events (early or moderate) were confirmed; 13 equivocal cases remain under review pending confirmatory testing. CIF estimates (risk (95% CI) at X years since HCQ initiation) were as following: 0 at 5 years; 0.36% (0.10-1.00) at 10 years; 0.72% (0.07-1.60) at 15 years; 1.25% (0.53-2.55) at 20 years; 4.23% (2.04-7.65) at 25 years; and 6.54% (3.17-11.58) at 30 years. There was no adequate risk set to estimate CIF beyond 30 years. Table 1- Characteristics of the 952 enrolled participants in the INTACT study who completed their 1st (or 2 nd /3 rd ) HCQ-R screening visit (01-10-2022 to 31-09-2025) Conclusion In this interim analysis of the INTACT study, we identified that cumulative risk of pre-clinical HCQ-R detected through screening using SD-OCT roughly doubled across 5-year intervals from 10-20 years, then tripled from 20-25 years. Next, we will expand enrollment and link pharmacy claims to capture time-varying dose for dose-stratified risk estimates.