Real-World Use of Anti-VEGF Injections Significantly Less Than Trial Regimens in AMD

Arghavan Almony, MD

The frequency of both anti-vascular endothelial growth factor (anti-VEGF) injections and the number of injection-eligible visits for patients in real-world clinical settings appears to be significantly less than those encountered in controlled trial environments.

According to the findings of an analysis presented at the American Society of Retina Specialists (ASRS) annual meeting in Vancouver, British Columbia, patients with neovascular age-related macular degeneration (nAMD) may be generally undertreated in clinical practice.

Led by Arghavan Almony, MD, an assistant professor at Campbell University, the retrospective analysis included electronic health record data from 57 private practices in the United States on a total of 51,496 eyes with nAMD. Patients were treated with aflibercept (Eylea, Regeneron; n = 12,635; 24%), ranibizumab (Lucentis, Genentech; n = 19,387; 38%), or bevacizumab (Avastin, Genentech; n = 19,474; 38%) and followed for 1 year.

Almony and colleagues noted that the analysis’s findings that the frequencies of injections in clinical practice, injection-eligible visits, and optical coherence tomography (OCT) scans being lower than in trials but similar to one another, suggests “that the lower injection frequency in clinical practice was not a result of disease inactivity, but due to fewer clinic visits to monitor disease activity. The undertreatment seen in clinical practice is differentiated from patients appropriately receiving as-needed dosing.”

"Over the last 6 years, our group has analyzed data from numerous electronic health record databases in the United States which has covered a wide range of practices, and this particular study was 1 of the largest cohorts of neovascular AMD patients for which real-world treatment patterns were assessed," Almony told MD Magazine^®. "The results have all led to similar conclusions: patients with active neovascular AMD are regularly undertreated in clinical practice as compared with pivotal clinical trials. In this study, we wanted to specifically focus on the temporal relationship between encounters and injections to shed light on patient undertreatment in clinical practice. Our findings suggest that the lower injection frequency in clinical practice is a result of fewer clinic visits to monitor disease activity rather than a result of disease inactivity. While we were not able to analyze clinical data to explain the reason for these findings in this study, we certainly know that treatment burden, patient co-morbidities, and follow-up scheduling can be factors in patient undertreatment. Alternative therapies that target different proteins and drug-delivery systems that require fewer injections are some of the exciting options coming down the pipeline that will soon be within reach."

Interestingly, the authors also noted that all 3 therapies included in the examination showed similar frequencies of injections, injection-eligible visits, and OCT scans, suggesting that their usage is comparable in clinical practice despite differences in labeled dosing recommendations.

Over the 12-month follow-up period, patients were administered an average of 5.9 (standard deviation [SD], 2.9) anti-VEGF injections, 5.3 (SD, 4.1) OCT scans, and 8.3 (SD, 4.1) encounters with their providers, of which a mean of 5.5 (SD, 2.7) were injection-eligible visits. The mean re-injection intervals for patients receiving aflibercept, ranibizumab, and bevacizumab injections were 63 (median, 49) days, 66 (median, 42) days, and 67 (median, 42) days, respectively

Therapy interruption, defined as a duration ≥12 weeks after prior injection, was observed most in eyes receiving bevacizumab (58.6%), followed by 56.8% of those receiving ranibizumab, and 49.8% of those receiving aflibercept.

“About one-third of encounters within the 12 months post-index date were not injection-eligible owing to the close proximity in time to the prior injection,” Almony and colleagues wrote. “This suggests that patients may have had follow-up visits to monitor for safety or initial response, and possibly even treatment or monitoring of the contralateral eye.”

The authors also noted that the majority of the treated eyes did not experience a loading dose injection patterns observed in clinical trials. In total, 56.3% of eyes had ≤2 injection-eligible encounters during the loading phase, defined as the index date to day 105, while the mean number of injections during that period was 2.5 (SD, 0.9) and the average number of OCT encounters was 2.0 (SD, 1.5).

In the subsequent dosing phase, defined as day 106 and on, 7.7% of eyes had no physician visits and 9.1% had no injection-eligible encounters during that period. Additionally, the average number of injections during this phase was 2.6 (SD, 2.4), while 31.8% of eyes received no injections during that period. OCT encounters occurred an average of 3.4 (SD, 3.1) times.

In total, 34% (n = 17,722) of eyes were treated bilaterally—40% of those given aflibercept, 35% of those given ranibizumab, and 30% of those given bevacizumab. It was significantly more likely for patients to receive aflibercept for bilateral treatment than unilateral compared to ranibizumab (risk ratio [RR], 1.15; 95% CI, 1.11 to 1.18) and bevacizumab (RR, 1.31; 95% CI, 1.27 to 1.35). Patients that were treated bilaterally also received 36% more physician visits (10.1 compared to 7.4) than those treated unilaterally.

"Treatment differences between unilateral and bilateral patients were not a focus of this study but is an important factor we took into consideration in conducting the analysis," Almony said. "Generally speaking, these treatment differences may be related to patient or doctor preference for treating unilaterally versus bilaterally, resulting in office visits in which one eye may not be injection-eligible."

The study was limited in its use of electronic health records, as potential errors in data could occur. Additionally, no clinical outcomes were assessed, limiting the ability to draw conclusions about the implications of these treatment patterns.

Almony and colleagues concluded that “comparing the frequency of injections to injection-eligible visits and OCT scans provides additional insight on the undertreatment of patients in clinical practice and provides additional evidence differentiating undertreated patients versus patients who receive regular monitoring but require fewer injections owing to disease inactivity.”

"This study supports our previous analysis and conclusions that patients with neovascular AMD are undertreated in clinical practice as compared with pivotal clinical trials," she said. "We look forward to presenting further analysis with clinical data, specifically visual acuity outcomes in the near future."

The study, “Anti-VEGF Injection Patterns and Healthcare Resource Utilization Among Neovascular Age-Related Macular Degeneration Patients in Clinical Practice,” was presented as a poster at ASRS’s 36^th annual meeting.

Click here to sign up for more MD Magazine content and updates.