Report Provides Insight Into Inflammation After Anti-VEGF Injections

A review of 2 clinical trials shows a continued need for both vigilance and monitoring for intraocular inflammation-related adverse events associated with brolucizumab therapy in neovascular age-related macular degeneration (nAMD), says a study published in Ophthalmology Retina.  

Phase 3 HAWK and HARRIER (ClinicalTrials.gov identifiers NCT02307682 and NCT02434328) are the most complete datasets for brolucizumab therapy in nAMD to date. Therefore, researchers conducted a post-hoc analysis to report on the timing of presentation, management, and outcomes of investigator-reported, intraocular inflammation-related adverse events that occurred during these trials. 

HAWK and HARRIER were 96-week, randomized, double-masked, phase 3, multicenter trials that included 1817 eyes with treatment-naïve nAMD (mean age, 74±9.1 years; 73.5% women). Participants were randomized to receive either intravitreal brolucizumab 3 mg (HAWK only) or 6 mg (n=1088) or aflibercept 2 mg (n=729). Investigators administered loading injections at weeks 0, 4, and 8 for both groups, with brolucizumab injections every 12 weeks and aflibercept injections every 8 weeks. 

Safety outcomes included incidence of both ocular and nonocular adverse events and serious adverse events during the 96-week study period. 

The current post-hoc analysis includes only adverse events in brolucizumab 3 mg and 6 mg treated eyes from both trials.

Within the patient cohort, 4.5% of eyes (49 eyes from 49 patients) experienced at least 1 investigator-reported, intraocular inflammation-related adverse event. Of these, 71.4% of eyes had only 1 adverse event, while 28.6% had 2 or more. No reports of multiple intraocular inflammation events were noted at a single visit during the course of the trial. 

A total of 70 intraocular inflammation-related adverse events coded to a specific MedDRA preferred term were reported; of these, 54.3% were mild, 40% were moderate, and 5.7% were severe. 

The mean and median number of brolucizumab injections prior to the onset of the first adverse event was 3.9±2.21 and 3.0, respectively. Based on the severity, mean and median number of injections before the adverse event onset was 4.0±2.38 and 3 for mild, 3.8±2.20 and 3 for moderate, and 3.7±0.58 and 4 for severe. 

Mean and median time to onset of the first adverse event from baseline was 165.6±153.6 days and 100 days, respectively. From the last brolucizumab injection, mean and median times were 20.3±17.8 and 18 days each. Among 70 adverse events, 50 were reported at scheduled visits and 20 were reported at unscheduled visits. 

Mean and median duration of the adverse events were 77.8±104.9 and 36 days. By severity, mean and median durations were 53.5±68.7 and 29 for mild, 87.5±94.3 and 39 for moderate, and 236.8±255.5 and 133 for severe. 

Of the adverse events, 87.1% were treated, primarily with corticosteroids. Sixteen adverse events were treated through a combination of topical corticosteroids and topical antibiotics. In 3 adverse events, systemic corticosteroids or intraocular corticosteroids were provided in combination with topical corticosteroids and/or topical antibiotics. No adverse events were treated with either systemic or intraocular corticosteroids alone. 

Within 61 treated adverse events, 85.3% were reported by investigators as resolved, 6.5% were reported as resolved with sequelae, and 8.2% were not resolved by the end of the study. Of the 70 adverse events, 8 were considered mild were observed, and no treatment was administered. 

Within the cohort, inflammation resolved in 79.6% of eyes; it resolved with sequelae in 10.2% of eyes and did not resolve in 10.2% of eyes. 

The majority of adverse events were treated within 3 days of diagnosis. 

Mean best-corrected visual acuity (BCVA) change for all eyes with intraocular inflammation-related adverse events between baseline and study conclusion was -0.84±20.6 ETDRS letters. Investigators found that 24.5% of eyes gained ≥15 ETDRS letters, while 18.4% lost at least 15 ETDRS letters and 8.2% lost at least 30 ETDRS letters. 

Mean BCVA change from before the adverse event to the lowest BCVA in 3 months following the event was -16.31±17.6 ETDRS letters and was -0.22±18.9 ETDRS letters from before the adverse event to the end of study. 

In total, 73.5% of eyes continued brolucizumab therapy following the first adverse event. Of these eyes (n=36), 24 finished the study on brolucizumab treatment only and showed a mean BCVA gain of 7.8±13.21 ETDRS letters by the conclusion of the study. The remaining eyes discontinued brolucizumab treatment later due to another adverse event, physician decision, patient withdrawal, lack of efficacy, or other undocumented reasons (n=8, 1, 1, 1, and 1, respectively). Two of these eyes were switched to other anti-vascular endothelial growth factor (VEGF) agents during the study. 

In each trial, 10 brolucizumab-treated eyes experienced a retinal artery occlusion (RAO), retinal artery embolism, or retinal artery thrombosis. Seven of these eyes had both RAO and intraocular inflammation. Of these eyes, 5 experienced a loss of at least 15 ETDRS letters and 1 experienced a gain of at least 15 ETDRS letters by the end of the study. Following the RAO event, treatment was discontinued in 4 of these eyes. 

The study’s limitations include those inherent to the retrospective, post-hoc approach, the use of a controlled population of treatment naïve patients, and the exclusion of infectious endophthalmitis events from the analysis. 

“Findings from this post hoc analysis provide insights to clinicians on the timing of presentation, management, and outcomes of [intraocular inflammation-related adverse events] following brolucizumab treatment,” the researchers concluded. “In clinical practice, clinicians should exercise continued vigilance for [intraocular inflammation-related adverse events] and educate patients to return promptly should symptoms or visual loss develop.” 

Disclosure: This clinical trial was supported by Novartis Pharma AG. Please see the original reference for a full list of authors’ disclosures. 

Reference

Singer M, Albini TA, Seres A, et al. Clinical characteristics and outcomes of eyes with intraocular inflammation after brolucizumab: post hoc analysis of HAWK and HARRIER. Ophthalmol Retina. Published online May 7, 2021. doi:10.1016/j.oret.2021.05.003