Outcomes of AVF treatment switch in DME patients in US clinical practice: Analysis of the AAO IRIS Database

Abstract

Purpose : In diabetic macular edema (DME), anti-vascular endothelial growth factor (AVF) agents have demonstrated efficacy in multiple RCTs and are recommended for first-line treatment. However, a substantial proportion of patients in clinical practice still experience suboptimal vision outcomes and switch to a different AVF. The objective of this study was to assess treatment patterns and vision outcomes in DME patients who initiate treatment with AVF agents and subsequently switch AVF agents.

Methods : Retrospective analysis of electronic health records from the AAO IRIS Registry for eyes with DME initiating AVF therapy with bevacizumab (BEV), ranibizumab (RAN) or aflibercept (AFL) between January 2014 and March 2016, with 12 months of follow-up. Outcomes of interest were corrected visual acuity (CVA) – absolute CVA and CVA change from baseline over 12 months of follow-up – overall and among AVF switchers. Switchers were further stratified by time to switch and 1^st to 2^nd AVF agent pairing. Eyes were followed per doctor recommendations and contributed to a given month’s data if observed during that month.

Results : 34,972 eyes met inclusion criteria. Patients had a mean 9.4 ophthalmology encounters and 7.7 retina specialist encounters in the first 12 months. Mean number of injections at 12 months was low, ranging from 3.3 (BEV, RAN) to 3.7 (AFL). N=4,241 (12.1%) eyes switched AVF agents within the first year, predominantly from BEV to AFL (n=2,256) or RAN (n=597). N=19,858 (56.8%) eyes had recorded CVA at baseline. The number of eyes contributing data each month ranged from n=5,955 (17.0%, month 12) to n=11,884 (34.0%, month 2). Mean absolute CVA at baseline was 0.50 logMAR and improved minimally over the first 12 months ranging from 0.48 logMAR at month 1 to 0.43 logMAR at month 12. Mean CVA change from baseline peaked at -0.07 logMAR over months 3-5, down to -0.05 at month 12. Among AVF switchers, mean CVA at baseline was 0.49 (N=2,160); mean CVA at treatment switch was 0.46 logMAR (n=1,841); and mean CVA change from baseline at 12 months was -0.05 logMAR (N=876).

Conclusions : In this large, national cohort of DME patients initiating AVF treatment, mean frequency of injections was significantly lower than treatment patterns employed in pivotal trials. Visual acuity outcomes were correspondingly lower. Among patients who switched AVF treatment, minimal VA improvement was seen after switching.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.