Reporting of safety events across anti-VEGF indications and the Pharmacovigilance approach: Findings from the non-interventional OCEAN study

Abstract

Purpose : The number and characteristics of reported adverse events (AEs) with anti-VEGF treatments vary across publications. Randomized controlled trials (RCTs) offer reliable safety data, but are usually not powered to detect rare AEs and prone to selection bias (less sick patients compared to RCTs). "Big data" from patient registries, although becoming more available, face other limitations due to the quality of documentation (retrospective character, limited standardization monitoring), uncontrolled selection (regional/socioeconomic parameters) and absence of control groups. The dataset from the non-interventional OCEAN trial might therefore provide further insights on the safety profile of ranibizumab in a real-life setting.

Methods : In the non-interventional, prospective OCEAN study (NCT02194803) 5,781 patients treated with ranibizumab were observed and documented over 24 months. Demographic data, safety events, visit- and injection frequency were analyzed for 3,726 neovascular age-related macular degeneration (nAMD), 1,250 diabetic macular edema (DME) and 764 retinal vein occlusion (RVO) patients.

Results : In the OCEAN study, 451 (8%) of patients had a documented non-serious AE, and 417 (7%) a serious AE. A non-serious adverse drug reaction was reported for 498 (9%) and a serious adverse drug reaction for 213 patients (4%).
Regarding ocular AE, the incidence of an increase in intraocular pressure was the highest in RVO patients (5.4%) compared to nAMD (1.9%) and DME (2.6%), but this rate was lower than in some RCTs (e.g. CRYSTAL 7.8%). The combined rate of cerebrovascular accidents and heart failures was slightly higher in the DME cohort (1.3%) compared with nAMD (0.7%) and RVO (0.7%). Despite different age distributions, the number of deaths were comparable across indications (0.6% nAMD, 1.0% DME, 0.8% RVO).

Conclusions : Although some rates of AEs found in the OCEAN study are lower and maybe less consistent than in the pivotal trials, the study cohorts of non-interventional trials offer complimentary information, e.g. on how ophthalmologists evaluate and deal with AEs in routine clinical practice. Projecting the heterogeneous risk profile of patients, a NIS has relevant limitations (underreporting and patients lost to follow-up). In our opinion, future risk assessements e.g. by the use of check-cards might help to detect patients at an increased risk.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.