Aqueous humor cytokine profile in insufficient responder to aflibercept for neovascular age-related macular degeneration.

Purpose: To identify cytokines associated with insufficient response to aflibercept against neovascular age-related macular degeneration.

Methods: This prospective, comparative control study enrolled 40 eyes of 40 patients with nAMD. Aqueous humor (AH) samples were collected at the baseline before the intravitreal administration of aflibercept. The patients were further classified into responder and non-responder groups based on the clinical course. Patients were classified as "responders" if they required three or fewer additional injections after the three initial monthly loading doses within one year, and as non-responders, if they required four or more injections after the initial three-monthly loading doses or were switched to alternative anti-VEGF agents or treatments such as photodynamic therapy. The concentration of Angiopoietin 1, angiopoietin like 4 (ANGPTL4), interferon gamma-induced protein 10, hepatocyte growth factor, interleukin 10, platelet derived growth factor BB, plasminogen activator inhibitor 1 (PAI1), vascular endothelial growth factor A, angiopoietin 2, monocyte chemotactic protein 1, IL8, IL12, platelet-derived growth factor (PlGF), and vascular cell adhesion molecule 1 in AH samples were analyzed using a multiplex immunoassay, in order to compare between responders and non-responders.

Results: 21 eyes were defined as responders, and 19 eyes were defined as non-responders. There were no significant differences in baseline characteristics. Multiple variate analysis using logistic regression analysis found that PAI1 (p = 0.023, coefficient = 0.025), PlGF (p = 0.016, coefficient = - 1.4), and ANGPTL4 (p = 0.032, coefficient = - 0.00070) at the baseline were significantly associated with the resistance to aflibercept.

Conclusion: Baseline higher PAI1 and lower PlGF and ANGPTL4 were associated with insufficient response to aflibercept in 1 year. These cytokines can potentially predict the treatment effect against nAMD.