Age-related macular degeneration: using morphological predictors to modify current treatment protocols

To assess predictors of treatment response in neovascular age-related macular degeneration (AMD) in an attempt to develop a patient-centric treatment algorithm. We conducted a systematic search using PubMed, EMBASE and Web of Science for prognostic indicators/predictive factors with the key words: age related macular degeneration', neovascular AMD', choroidal neovascular membrane (CNV)', anti-vascular endothelial growth factor (anti-VEGF)', aflibercept', ranibizumab', bevacizumab', randomized clinical trials', post-hoc', prognostic', predictive', response' injection frequency, treat and extend (TAE), pro re nata (PRN)', bi-monthly' and quarterly'. We only included studies that had an adequate period of follow-up (>1year), a single predefined treatment regimen with a predetermined re-injection criteria, an adequate number of patients, specific morphological [optical coherence tomography (OCT)] criteria that predicted final visual outcomes and injection frequency and did not include switching from one drug to the other. We were able to identify seven prospective studies and 16 retrospective studies meeting our inclusion criteria. There are several morphological and demographic prognostic indicators that can predict response to therapy in wet AMD. Smaller CNV size, subretinal fluid (SRF), retinal angiomatous proliferation (RAP) and response to therapy at 12weeks (visual, angiographic or OCT) can all predict good visual outcomes in patients receiving anti-VEGF therapy. Patients with larger CNV, older age, pigment epithelial detachment (PED), intraretinal cysts (IRC) and vitreomacular adhesion (VMA) achieved less visual gains. Patients having VMA/VMT required more intensive treatment with increased treatment frequency. Patients with both posterior vitreous detachment (PVD) and SRF require infrequent injections. Patients with PED are prone to recurrences of fluid activity with a reduction in visual acuity (VA). A regimen that involves less intensive therapy and extended follow-up intervals (4weekly) can be suggested for patients who show adequate visual response and have both SRF and PVD at baseline. In addition, patients with poor prognostic indicators such as IRC, VMA, large CNV size, older age and poor response at 12weeks should be extended very cautiously with the possibility of fixed monthly/bimonthly (every 2months) treatments if they fail to achieve dryness. Patients with PED at baseline should receive monthly/bimonthly injections of anti-VEGF therapy or can be extended very cautiously (two weekly intervals) using a TAE protocol.