Abstract
Purpose :
Genome-wide association studies have implicated multiple members of the complement pathway in the risk of progressing to advanced stages of age-related macular degeneration (AMD). Better understanding of the relationship between complement and AMD progression may help with strategies for patient selection and timing of intervention with complement inhibitors. In this study, aqueous humor (AH) complement levels and activity were compared across patient cohorts with different stages of dry AMD and between eyes with different stages of disease severity from the same patient.
Methods :
AH and plasma (n=64) were collected during ocular procedures (n=49 unique patients for unilateral collection, n=5 for bilateral; cataract, intermediate AMD (iAMD), and geographic atrophy from AMD (GA) patients). 9 patients had matched AH and plasma (4 unilateral collection patients and 5 bilateral collection patients, including one patient with bilateral GA). Custom immunoassays to measure full-length (C3) and processed (PC3) complement component 3 in AH were developed on the Quanterix Simoa platform.
Results :
Demographics were well balanced across patient cohorts. AH C3 and PC3 levels and the PC3:C3 ratio were not significantly different in eyes with different stages of AMD across patients (Table 1). Consistent with this result, patients with GA in one eye and iAMD in their other were well correlated with each other (r^2 ≥ 0.9, 4 donors, GA eye vs iAMD eye) and did not display significantly different C3, PC3, levels or PC3:C3 ratio. The similarity in complement levels between eyes from the same patient was unlikely to be driven by peripheral complement activity, as AH and plasma C3 and PC3 levels in patients with GA showed no meaningful correlations (r^2 ≤ 0.067, 9 patients tested, AH from GA eyes only, plasma vs AH).
Conclusions :
This study suggests that AH complement levels are not indicative of AMD severity. The optimal timing of intervention with complement inhibitors to slow AMD progression requires further investigation.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.