FIRST PRIZE, COS AWARDS FOR EXCELLENCE IN OPHTHALMIC RESEARCH

Authors: Jacob Rullo, Steven Bae, Wilma Hopkins, Isabella Irrcher, Manpartap Bal, Tom Gonder, Sanjay Sharma

Author Disclosure Block: J. Rullo: Funded grants or clinical trials; Retina Foundation of Canada – Awarded Junior Clinical Research Grant. S. Bae: None. W. Hopkins: None. I. Irrcher: None. M. Bal: None. T. Gonder: None. S. Sharma: Clinical Trials and Advisory Board; Bayer, Novartis.

Abstract Body:

Purpose: Neovascular diseases of the retina are the leading cause of legal blindness in the developed world. Treatment with anti-VEGF biologic agents has been shown to both prevent further deterioration and improve visual function. Considering up to one half of patients receiving anti-VEGF agents fail treatment, a greater understanding of failure is paramount. The development of systemic anti-drug antibodies against biologic agents has been proposed as one mechanism.
Considering the focal nature of VEGF-mediated diseases in the eye, sampling of the intraocular milieu would improve the diagnostics of disease and our understanding of treatment failure. Therefore, we hypothesized that the aqueous humour in patients with VEGF-mediated eye diseases receiving intravitreal anti-VEGF injections contain antibodies that reflect disease activity and may be responsible for treatment failure.
Study Design: Prospective, cross-sectional study
Methods: Aqueous humour was obtained from participants undergoing intravitreal injections for retinal edema and neovascularization (cases) or routine cataract surgery (controls). Antibody isotyping of aqueous humour and serum was performed using MILLIPLEX® MAP Human Isotyping Multiplex Assay. Variables were tested using the one-way ANOVA (Tukey’s post-hoc test) and the Kruskal-Wallis test as appropriate. Pearson correlations were used to test associations. P<0.05 was used as the criteria for statistical significance.
Results: Nintey-seven samples of aqueous humour were removed from 106 eyes (45 male, 61 female). Mean number of intravitreal injections was 14 (range 0-77). Mean duration of treatment was 46 months (range 0-120). Antibodies were detected in the aqueous humour of participants in microgram per milliliter concentrations. The aqueous to serum antibody ratio was 2 to 12-fold higher in cases as compared to controls. IgM titres were 19-fold higher in cases, p<0.0007. Higher titres of intraocular IgG_1-3 and IgA antibodies were detected in eyes with retinal edema or neovascular disease undergoing treatment compared to those undergoing cataract surgery (p<0.0001). Number of intravitreal injections and the type of anti-VEGF agent injected correlated with higher titres of antibodies (p<0.001). Intraocular antibodies also correlated with patient outcomes. In all diseased eyes receiving intravitreal injections, there was a medium strength of association between IgG_1-3, IgA and worse BCVA [R=0.47 (CI:0.30 to 0.62), p<0.001). Diabetic retinopathy eyes had the largest strength of association [R=0.84 (CI: 0.57 to 0.94), p<0.001] between aqueous antibodies and worse visual acuity. Persistent intra/sub- retinal fluid was associated with higher mean titres of antibodies, but statistical significance was only achieved in the IgG₃ group.
Conclusions: The presence of intraocular antibodies in patients receiving intravitreal injections represent a novel mechanism to help explain poor visual outcomes in patient’s receiving anti-VEGF agents. This study highlights an important potential cause of treatment failure, namely the creation of antibodies that might affect drug efficacy. The measurement of peak and trough levels of intraocular antibodies may allow for more personalized dosing of medications; further study is recommended.