By targeting neovascularization and edema, intravitreal anti-VEGF injections are rapidly becoming a first-line therapy for certain posterior segment conditions like central retinal vein occlusion (CRVO). And as such, there are now a variety of anti-VEGF formulations to choose from like aflibercept, ranibizumab and bevacizumab.
At the recent Association for Research in Vision and Ophthalmology (ARVO) Congress in Vancouver, Canada, delegates gathered to hear results from the recent Phase III SCORE2 and LEAVO noninferiority trials, both of which compared different anti- VEGF agents as treatment for macular edema secondary to CRVO.
Both of the government-sponsored SCORE2 and LEAVO trials have similar characteristics, including a 5-letter inferiority margin and a primary outcome of mean change in visual acuity (VA) letter scores (VALS) from baseline. However, there are also differences between the two, including treatments used, that provide complementary and additive details. According to SCORE2 Study Chair Dr. Ingrid Scott, a professor of ophthalmology and public health sciences at Penn State College of Medicine (Pennsylvania, USA), the results of these trials will assist clinicians in selecting the type, frequency and duration of treatment for their patients.
SCORE2: The first six months
Although aflibercept is FDA approved for the treatment of macular edema due to CRVO, bevacizumab is often used off-label due to cost – despite a lack of supporting evidence. Therefore, SCORE2 investigated whether bevacizumab was noninferior to aflibercept in treating this condition.
SCORE2 included 362 patients with macular edema due to central retinal or hemiretinal vein occlusion, who were randomized 1:1 to receive either intravitreal injections of bevacizumab (1.25 mg; n = 182) or aflibercept (2.0 mg; n = 180) every 4 weeks through month 6.
At month 6, the primary outcome of mean change from baseline in VALS between groups was assessed: The bevacizumab arm VALS was 69.3 (a mean increase from baseline of 18.6); the aflibercept arm VALS was also 69.3 (a mean increase from baseline of 18.9). This led the investigators to conclude that at month 6, bevacizumab was noninferior to aflibercept based on a VALS margin of 5. In addition, both groups showed statistically significant SD-OCT central subfield thickness (CST) decreases from baseline through month 6.
These results led investigators to conclude that, “among patients with macular edema secondary to central retinal or hemiretinal vein occlusion, intravitreal bevacizumab was noninferior to aflibercept with respect to VA after 6 months of treatment”.
Keeping Score: From months 6 to 12
According to SCORE2 Co-Chair Dr. Michael Ip, at six months eyes were divided into having had either a good or poor response to initial therapy. This was based on a defined protocol that relied on VA and OCT and that was followed until month 12.
“Eighty-four percent (84%) fell into the good responding group, versus 16% for the poor response group,” said Dr. Ip, who is a professor in the department of ophthalmology at the David Geffen School of Medicine at the University of California Los Angeles.
Eyes in both groups with a good response were re-randomized to continue treatment with the originally assigned drug, either on a monthly or treat-and-extend (T&E) basis. Those with a poor response were changed to an alternate treatment.
“From baseline to month 5, there was excellent adherence to the treatment protocol in both the aflibercept and bevacizumab monthly arms, with both receiving a mean monthly 5.8 injections,” said Dr. Ip. In the good responder groups, from months 6 to 12, the monthly aflibercept group received 5.8 injections and the T&E group received 3.8. In the bevacizumab treated eyes, the monthly group had 5.8 injections, versus 4.5 for T&E – which was statistically significant.
From months 6 to 12, there were no statistically significant differences in VALS between the monthly and T&E regimens in both arms, and no meaningful difference in OCT central subfield thickness between groups.
So, what about those who responded poorly? Few eyes met the protocol for a “poor response” definition. Fifteen eyes that responded poorly to aflibercept were reassigned to receive intravitreal dexamethasone implant treatment. The 39 eyes in the poor responding bevacizumab group were reassigned to aflibercept.
“In the bevacizumab group, there was an immediate rise in visual acuity from months 6 to 9 when switched to aflibercept . . . and this stayed stable up to month 12,” said Dr. Ip, noting that there was no difference in the aflibercept group’s VALS after treatment with dexamethasone.
He continued: “In the aflibercept poor response group, there was no change to central subfield thickness. In distinction, those in the bevacizumab group, when switched to aflibercept, had an immediate reduction in residual edema which stayed stable up to month 12.”
SCORE2: Months 12 to 24
During the first 12 months of the SCORE2 trial, participants followed a protocol designed treatment schedule. However, after month 12, there was no treatment protocol and physicians could treat participants at their own discretion, using any commercially available drug (or no drug).
At month 12, 330 (of 362 enrolled) participants completed the SCORE2 study and were targeted for month 24 follow-up. Of those, 236 of the original cohort completed a visit at month 24 (called ‘completers’), with similar retention among both the aflibercept and bevacizumab groups.
“What we see from months 12 to 24 translates to between 3-4 injections per participant during that time period,” explained Dr. Scott, noting that about a quarter of completers received no treatment from months 12 to 24.
“Among the remainder, the mean number of treatments between months 12 to 24 was 3.6 in the aflibercept group and 4.5 to those originally assigned to bevacizumab.”
How did VALS fare? In both groups VALS improved significantly during the first six months, when patients received monthly treatment. There was a much smaller improvement from months 6 to 12, when the groups were randomized to continue monthly or T&E. However, between months 12 to 24, when there was no treatment protocol, VALS worsened in both groups – although they were still significantly better than baseline.
Dr. Scott said that similar results were observed in CST: “Central subfield thickness improved significantly in the first six months in both groups. Between months 6 and 12 when they were randomized, there was a much smaller improvement in CST in the bevacizumab group, and a small worsening in CST in the aflibercept group. Between months 12 and 24, CST worsened in both groups – although the CST at month 24 was still significantly better than at baseline.”
Overall, these results led investigators to conclude that with respect to VALS and CST, participants improved significantly compared from baseline to month 6 (the period of monthly injections). There were lesser improvements from months 6 to 12 when half of the good responders were treated according to a T&E regimen; and worsened outcomes when patients were treated by investigator discretion and fewer treatments were received.
“The outcomes worsened more notably in those patients who received no treatment from months 12 to 24. Nonetheless, month 24 outcomes were significantly better than at baseline,” explained Dr. Scott.
“Results of SCORE2 indicate the chronic nature and variability in disease course in CRVO and HRVO. This in turn points out the importance of continued monitoring and individualized treatment to optimize visual acuity outcomes,” she said. “We will continue to follow participants out to five years to continue to collect outcome information.”
The LEAVO Clinical Trial
LEAVO is a multi-center, Phase III, double-masked, randomized, controlled, noninferiority trial comparing the clinical and cost-effectiveness of intravitreal therapy with ranibizumab versus aflibercept and bevacizumab for macula edema due to central retinal vein occlusion.
When the LEAVO study occurred in 2014, ranibizumab was the only UK-licensed treatment for CRVO at that time – and therefore it became the comparator versus aflibercept and bevacizumab, which were unlicensed and considered to be the investigative interventions within the trial protocol.
According to Dr. Philip Hykin a consultant ophthalmologist at Moorfields Eye Hospital, the primary objective was to compare the clinical effectiveness, side effect profile and relative cost effectiveness of the three selected anti-VEGF agents. The study randomized 463 patients with macular edema due to CRVO, with the primary endpoint in best corrected visual acuity from baseline to 100 weeks in the study eye of all participants.
“There were mandated visits at weeks 4, 8 and 12,” said Dr. Hykin. “Then PRN (pro re nata or as needed) injections at weeks 16 and 20. Following that, we went to 4 to 8 weekly visits with PRN injections, going to 8 weekly if stability was met.”
“The best corrected ETDRS visual acuity at baseline was approximately 54 letters across arms,” said Dr. Hykin, noting that 87.9% of patients completed the 100-week analysis. The mean gain in BCVA at 100 weeks was 12.5 letters for the ranibizumab group, 15.1 for aflibercept and 9.8 for bevacizumab.
“Looking at the primary outcome, during the initial phase of the study there was a rapid improvement in vision with mandated injections,” he said.
“With the early PRN injection phase at 16 and 20 weeks, there was a small decrease in visual acuity . . . and out to 100 weeks visual acuity remained relatively stable.”
At 100 weeks, the difference between ranibizumab and bevacizumab was -1.73 letters, falling above the noninferiority scale of -5 letters, therefore bevacizumab was noninferior to ranibizumab. Between aflibercept and ranibizumab, the difference was +2.23 letters in favor of aflibercept; making it noninferior to ranibizumab, but not superior.
“The post hoc analysis of aflibercept versus bevacizumab at 100 weeks was -3.96 letters, below the preset noninferiority margin of -5 letters,” said Dr. Hykin. These results led the investigators to conclude that compared to baseline, aflibercept, bevacizumab and ranibizumab substantially improved and maintained VA up to 100 weeks, however they do not support bevacizumab as interchangeable for treating macular edema due to CRVO in the longer term.
For OCT CST across study arms at 100 weeks, there was a decrease in thickness initially during the injection mandated phase, a slight increase during the PRN phase (that was consistent and mirrored well the VA data) and then slowly decreased again out to one hundred weeks. Differences at 100 weeks were not statistically significant.
The number of injections per treatment arm was significantly less for aflibercept at week 100 (-1.8), at week 52 (-1.1) and at week 24 (-1.4).
Dr. Hykin summarized: “For the management of macular edema due to CRVO, ranibizumab, aflibercept and bevacizumab provided substantial and sustaining proof of visual acuity throughout the study. At least eight weekly follow-ups and prompt treatment in the second year maintained first year visual acuity gains and no huge safety concerns were identified.”
“In conclusion, in the UK, for routine treatment of ME due to CRVO, these data support that the EMA (European Medicines Agency) license medications aflibercept and ranibizumab for syndication, and do not support bevacizumab for being interchangeable for this indication,” concluded Dr Hykin.
Among patients with macular edema secondary to central retinal or hemiretinal vein occlusion, intravitreal bevacizumab was noninferior to aflibercept with respect to VA after 6 months of treatment.
Results of SCORE2 indicate the chronic nature and variability in disease course in CRVO and HRVO. This in turn points out the importance of continued monitoring and individualized treatment to optimize visual acuity outcomes.
In conclusion, in the UK, for routine treatment of ME due to CRVO, these data support that the EMA (European Medicines Agency) license medications aflibercept and ranibizumab for syndication, and do not support bevacizumab for being interchangeable for this indication.
Editor’s Note: The ARVO 2019 Congress was held in Vancouver, Canada, from April 28 to May 2. Reporting for this story also took place at ARVO 2019.
