Head-to-head comparison of anti-VEGF agents was unable to show non-inferiority of bevacizumab, while showing stable VA outcomes of aflibercept with fewer injections

The LEAVO (Lucentis, Eylea, Avastin in Vein Occlusion) phase III, multicenter, randomized prospective trial, conducted in the UK to evaluate the non-inferiority of aflibercept and bevacizumab to the original standard National Health Service treatment with ranibizumab, was unable to demonstrate that bevacizumab was non-inferior, while aflibercept led to comparable outcomes with a lower number of injections.

The trial was conducted over a period of 100 weeks in 44 UK NHS hospitals.  A total of 463 patients with visual impairment due to CRVO-related macular edema were randomly assigned to receive ranibizumab, aflibercept or bevacizumab injections.  In all three arms the treatment consisted of a loading dose of 4 monthly injections followed by a PRN retreatment regimen with monitoring every 4 weeks up to week 20 extended from 4 to 8 weeks if retreatment criteria were not met at 3 consecutive visits. 

The mean BCVA letter gain at 100 weeks was 12.5 for ranibizumab, 15.1 for aflibercept and 9.8 for bevacizumab.  The proportion of patients with at least 15 BCVA letter gain was 47% for ranibizumab, 52% for aflibercept and 45% for bevacizumab.  The mean change in OCT CST from baseline to week 100 was similar in the three groups, but there was a significantly greater proportion of patients with an OCT CST of less than 320 μ in the aflibercept group. 

Aflibercept also allowed for a lower number of injections over the entire study period, which was significantly less by half an injection at 6 months, by 1 injection at 1 year and by approximately 2 injections at 2 years.   

The authors pointed out that “the significantly greater proportion of patients with a CST OCT of less than 320 µm at weeks 52 and 100 with aflibercept compared with ranibizumab is a novel finding in CRVO.” Equally new in CRVO-related macular edema was the observation that fewer injections were needed for aflibercept as compared with ranibizumab, and this was seen as early as at week 24. 

 “Such a difference in anti-VEGF agent injection has not been previously reported in macular edema due to CRVO and would be a potential advantage to aflibercept use in similar populations,” the authors wrote. 

Cost-effectiveness considerations

According to Philip Hykin, MD, principal author of the study, this was an important project from the cost perspective in the UK. 

“The National Institute for Health Research (NIHR) were interested to know whether bevacizumab was non-inferior to the other two agents and could possibly be used in routine clinical practice as a more affordable alternative.  However, from a clinical effectiveness standpoint the study could not rule out the possibility that bevacizumab was worse by 5 visual acuity letters than the licensed medications at 100 weeks,” he said.

Quality of life analysis of the study by health economists showed minimal differences between the three agents.  Consequently, from a cost-effectiveness standpoint, thanks to the significantly lower per-patient cost, bevacizumab would be the most cost-effective choice in terms of quality-adjusted life-years (QALY).  However, the VA outcomes suggest that bevacizumab, if used as in the LEAVO study, would not be interchangeable with the two licensed agents.

“Patients should be fully informed of these findings before bevacizumab could be considered as a routine treatment,” Hykin said.  “Vision was slightly less good, although in terms of day to day visual function and quality of life there was little difference. Nevertheless you could not say to the patient that the drug is equivalent to the licensed medications.”

In a small survey, a group of patients were asked if in principle, given these results, they would be prepared to consider bevacizumab as an option.  Two thirds said they would, provided they could be switched to aflibercept in case they did not respond well, while about one third said they would want the best drug from the start.

The results of the LEAVO study will be likely the topic for further discussion and review by the UK health regulatory authorities. 

“ We were optimistic that the results of the study would enable us to draw more definitive conclusions, and perhaps confirm the findings of the SCORE2 study, which showed bevacizumab when given by monthly injection was non-inferior to aflibercept for treating macular edema secondary to CRVO in terms of VA changes.  On the other hand, the fewer injections required when using aflibercept versus ranibizumab would be better for the patient but may not result in NHS savings, considering the higher cost of the drug.  We are submitting the final version of our report to the NIHR, when it will be up to the government to make the decisions,” Hykin said.

Comparing outcomes in the long term

A direct comparison of all three anti-VEGF agents was done in the DRCR.net Protocol T trial for diabetic macular edema (DME), which demonstrated the superior outcomes of aflibercept in patients with lower baseline vision.  No prospective, randomized trial has compared the three drugs in AMD. 

“In CRVO, our study was the first to perform a head-to-head randomized comparison with a longer follow up of 2 years,” Hykin said. 

Many of the studies using anti-VEGF agents have had their primary outcome at 6 months.  SCORE2 categorized the patients at 6 months as good responders or poor/marginal responders and re-randomized them. Thus it is not possible to compare the initial cohort in SCORE2 with LEAVO outcomes beyond six months.

“What the LEAVO study clearly showed, is that anti-VEGF treatment improves vision for the vast majority of patients with CRVO, and that improvement can be maintained at 2 years with relatively few injections if patients are regularly monitored and treated promptly if needed. With aflibercept, we had only 10 injections on average in two years, but vision remained stable,” he noted. 

References

Hykin P, Prevost AT, Vasconcelos JC, Murphy C, Kelly J, Ramu J, Hounsome B, Yang Y, Harding SP, Lotery A, Chakravarthy U, Sivaprasad S; LEAVO Study Group. Clinical Effectiveness of Intravitreal Therapy With Ranibizumab vs Aflibercept vs Bevacizumab for Macular Edema Secondary to Central Retinal Vein Occlusion: A Randomized Clinical Trial. JAMA Ophthalmol. 2019 Aug 29. doi: 10.1001/jamaophthalmol.2019.3305. Contact information: Philip Hykin, MD, is a Consultant in Ophthalmology and former director of the medical retina service at Moorfields Eye Hospital. He is responsible for the Clinical Trial Research programme, which seeks to evaluate new therapies for age-related macular degeneration and diabetic retinopathy.

Contact information: Philip Hykin, MD, is a Consultant in Ophthalmology and former director of the medical retina service at Moorfields Eye Hospital. He is responsible for the Clinical Trial Research programme, which seeks to evaluate new therapies for age-related macular degeneration and diabetic retinopathy.