Early response to anti-VEGF therapy predictive of longer-term results

Reviewed by Scott M. Whitcup, MD

Los Angeles-Patients with diabetic macular edema (DME) are looking for prompt, consistent, and long-lasting improvement in vision.

A new analysis of an old study suggests that early response to agents that block vascular endothelial growth factor can predict longer-term improvement in visual acuity in patients with DME.

“The key finding is that early change in visual acuity, after three injections of ranibizumab, is highly predictive of how patients will do over the remaining three year study and how consistent their vision response will be,” said Scott M. Whitcup, MD, clinical faculty member of the Jules Stein Eye Institute of the David Geffen School of Medicine at the University of California, Los Angeles. “The clinical implication is that patients who don’t respond well after three anti-VEGF injections are generally the patients who will not do well over the longer term and the patients where additional therapy or a change in therapy may be warranted.”

Dr. Whitcup presented a post hoc analysis of Diabetic Retinopathy Clinical Research Network (DRCRnet) Protocol I data. He was chief scientific officer at Allergan during the DRCRnet Protocol I study that compared ranibizumab plus prompt or deferred laser photocoagulation or triamcinolone plus prompt laser for DME. He is currently founder and chief executive officer of Akrivista and Whitecap Biosciences.

The original analysis of DRCRnet Protocol I found that intravitreal ranibizumab plus laser photocoagulation either within a week of the initial injection or later during treatment provided better anatomic and functional outcomes than laser alone. The original study did not attempt to tease out factors that might predict success or failure of treatment. Until this post hoc analysis, physicians often continued treating with anti-VEGF without any guide as the likelihood of long-term improvement in vision.

“Our initial analysis of the data from Protocol I focused on mean values, how the group as a whole did,” Dr. Whitcup said. “Patients don’t care about mean improvements, they want to know how am I going to do. Physicians want to know the same thing so they can either consider additional interventions or tell patients they are likely to do well on the current treatment.”

This analysis included only the two Protocol I cohorts that had been randomized to ranibizumab plus either prompt or delayed laser photocoagulation with observed visual acuity at twelve weeks and three anti-VEGF injections. A total of 340 patients from the original study population of 691 were included. The patients were stratified into three cohorts based on improvement in best corrected visual acuity at 12 weeks, less than five letters, five to nine letters and ten or more letters compared to baseline. Patients were followed for another two years and nine months for a total study duration of three years.

Patients are looking for consistent improvement in vision rather than dramatic swings from better to worse to better that average out to better vision, Dr. Whitcup noted. Researchers assessed consistency of improvement by using a measure of ten or more letters improvement from baseline at 50 percent or more of visits, 75 percent or more of visits or all but one visit over the succeeding 33 months. Differences in rates of consistent suboptimal response of less than five letters improvement from baseline was also assessed.

At 12 weeks, 135 patients (39.7 percent) had limited early response of less than five letters, 79 patients (23.2 percent) had intermediate response of five to nine letters improvement and 126 patients (37.1 percent) had strong early response of ten or more letters improvement from baseline. Subsequent rates of consistently good vision were significantly lower in patients with limited early response compared to strong early response, 21.5 percent versus 81.0 percent for ten or more letters improvement on at least 50 percent of subsequent visits.

The intergroup differences for ten or more letters improvement were even more striking for more consistent improvement. Only 11.1 percent of limited early responders showed ten or more letters of improvement on at least 75 percent of subsequent visits versus 66.7 percent for strong early responders. Only 0.7 percent of limited early responders showed ten or more letters improvement at all but one subsequent visit versus 37.3 percent of strong early responders. Suboptimal response rates showed similar and inverse patterns.

“These data show that by the third anti-VEGF injection, you get a very good idea of which patients will do well and which will not,” Dr. Whitcup said. “Early response is not a perfect predictor, but based on how well you do by the third injection, there was a very strong indication of how well you would do over the rest of the study period.” 

Scott M. Whitcup, MD

E: swhitcup@cox.net

This article was adapted from Dr. Whitcup’s presentation at the 2016 meeting of the Association for Research in Vision and Ophthalmology. The source of the data is the DRCRnet, but the analyses, content and conclusion presented herein are solely the responsibility of the authors and have not been reviewed or approved by DRCRnet.