Retina specialists receptive but cautious regarding use of Beovu
While further analysis of trial data confirms efficacy and safety events in real-world practice continue to be monitored, the attitude of retina specialists toward the newest FDA-approved anti-VEGF agent remains open but cautious.
“Brolucizumab is, at this stage, a risk-benefit question,” Nancy M. Holekamp, MD, said.
Welcomed with high expectations, Beovu (brolucizumab, Novartis) had positive initial feedback from early users, until cases of severe sight-threatening side effects were reported, derailing the drug’s initial launch.
“The question is whether it is worth taking the very small risk of intraocular inflammation that could potentially cause irreversible vision loss when we have three well-established, effective agents that don’t carry this potential risk,” Holekamp said.
On the other hand, brolucizumab showed superior drying efficacy in clinical trials. Anecdotally, in clinical practice it has shown the ability to treat patients who did not respond well to other agents. While retina specialists make it clear that Beovu is likely not a frontline option until the safety issues are further clarified, the benefit-risk balance may still be favorable for its use in select patients as a second-line therapy.
“The practice of medicine has inherent risks, which should be weighed and managed in conjunction with patients. There are many who can probably be helped by this drug. Hopefully we will learn more on how to select them and how to treat potential complications,” Marco A. Zarbin, MD, PhD, said.
Brolucizumab is a single-chain antibody fragment with low molecular weight and high solubility, allowing for a higher molar concentration than other anti-VEGFs.
“What that means is that brolucizumab might be able to penetrate tissues better and might have better efficacy, potentially leading to a more prolonged duration of action,” Paul Hahn, MD, PhD, chair of the American Society of Retina Specialists Research and Safety in Therapeutics (ReST) committee, said.
Data from the HAWK and HARRIER trials demonstrated comparable visual outcomes and a more rapid and superior drying effect in comparison with Eylea (aflibercept, Regeneron). However, shortly after the approval of Beovu, the ReST committee received multiple reports of safety events, including intraocular inflammation alone or associated with vasculitis and vascular occlusion.
“We analyzed the reports and promptly informed the retina community. We recommended to stay alert for signs of vasculitis, to examine the eye for inflammation prior to the injection, to obtain appropriate informed consent and to judiciously select patients,” Hahn said.
He specified that it is not the role of ASRS to make decisions on whether a drug should or should not be used but to provide physicians with the information they need to make good decisions. An editorial recently published in American Journal of Ophthalmology by Philip Rosenfeld, MD, PhD, and David J. Browning, MD, called for an official moratorium on the use of Beovu “until the results of further investigations are concluded and remedies are implemented.”
“ASRS does not try to restrict access or choice for physicians. We generate alerts and provide information and recommendations so that physicians can make the best choices for themselves,” Hahn said.
Simultaneously, Novartis set up a Safety Review Committee (SRC), which conducted an unmasked review on a subset of imaging data from the trials and identified some previously overlooked adverse events leading to loss of vision. They found a 4.6% rate of intraocular inflammation, with 3.3% of eyes presenting retinal vasculitis and 2.1% presenting concomitant vascular occlusion. Most of the patients (74%) developed symptoms within the first 6 months, but some (12%) experienced them up to 12 to 18 months later.
“There appears to be perception that Novartis was not forthright with these events. I had the opportunity to interact with the Novartis leadership during the identification of these cases. They worked hand in hand in a transparent manner with ASRS and recruited a top-notch group of people, independent of Novartis’ oversights, to autonomously investigate and report their findings, and that is admirable,” Hahn said.
Zarbin also praised Novartis for its “admirable discipline and honesty.”
“They looked at all their data, involved external experts in the analysis, worked closely with the ASRS, found errors in the trial data and reported them, and are looking into real-word data. They are behaving as a model corporate partner,” he said.
A rare but dangerous complication
According to Holekamp, the 4.6% rate of inflammatory reactions seen in the trials, which was higher than the rate seen in contemporary trials for currently available Lucentis (ranibizumab, Genentech) and aflibercept, was likely a safety signal that may have been overlooked by the FDA approval process.
“And it turned out to be a significant one, as after the drug was launched, we started seeing cases of severe, irreversible vision loss,” she said.
“Cases of vasculitis were missed because, in some cases, it was peripheral and did not affect the patient’s vision,” Zarbin said. “When you go back and look, you see the problem.”
Although these severe events were uncommon — the incidence of retinal artery occlusion with Beovu was 1% in HAWK and 0.5% in HARRIER — they represent a concern.
“The reason is that some of those patients have profound vision loss, even down to no light perception, and you may have started treating them with a vision of 20/40. Severe intraocular inflammation is an important complication never seen with ranibizumab and almost never with aflibercept. Just looking at the frequency of the complication does not tell you the whole story,” Zarbin said.
Again, just looking at numbers, the proportion of eyes that lost more than 15 letters by week 96 was comparable, 8% with Beovu vs. 7% with Eylea in HAWK and 7% with both agents in HARRIER. However, the risk for vasculitis was higher with Beovu, he said.
Real-world data presented at the Euretina congress showed that, based on the cases reported until August, the rate of retinal vasculitis plus vascular occlusion was 4.7 per 10,000 injections, and the collective rate of retinal vasculitis and vascular occlusion was 10.67 per 10,000. However, real-world incidence should be taken “with some healthy degree of skepticism because voluntary reporting of adverse events is inherently subject to underreporting,” Zarbin said.
In addition, because more than 10% of intraocular inflammation cases develop this complication after 12 months from injection, more time is needed to gain reliable real-world incidence data.
According to Holekamp, a key point of the HAWK and HARRIER trials was that only brolucizumab was subjected to disease activity assessment, determining which patients could be switched from every 12-week to every 8-week dosing intervals.
“Aflibercept was not subjected to treatment adjustments based on disease activity assessment. Therefore, we cannot speak of a head-to-head comparison beyond the first three doses. Strictly speaking, we can say that brolucizumab is a better drying agent only in the first 3 months, but from week 16, the results become confounded,” she said.
Better drying, however, does not necessarily mean better results overall, in her opinion. Post hoc analyses of the CATT, IVAN and HARBOR trials suggested that subretinal fluid may be protective against atrophy. HARBOR data with ranibizumab and AERIE data with aflibercept also showed an association between residual subretinal fluid and better vision.
“Our goal is to get patients dry, but some subretinal fluid can be tolerated. It is not harmful to vision, and I believe it is protective against atrophy,” Holekamp said.
Better drying is not as critical as visual acuity as a criterion to judge the efficacy of a drug. In this respect, the visual acuity results of aflibercept were similar to those of brolucizumab, Zarbin said. However, most physicians aim to eliminate fluid with treatment.
“This is because we believe that at least over a 5-year period this is the best way to preserve vision. It could turn out that making the fluid go away completely over year 6 to 10 may cause macular atrophy. On the other hand, many of the patients we treat may no longer be alive in 6 to 10 years,” he said.
At present, there is no clear evidence that there is more than a coincidental relationship between persistent subretinal fluid and less atrophy, and association should not be confused with causation, he said.
Conversely, Holekamp said that “when you have multiple post hoc analyses of different studies saying the same thing, then we begin to believe that there is some biologic rationale for this to be true.”
Overall, higher dehydration properties are an advantage in this COVID era, Zarbin said.
“If you have a longer period of dehydration, the patients have to come less frequently. Those injected with a 3-month drug back in February would not have had to come back until the end of May, when the curve of COVID was down,” he said.
There are a variety of attitudes toward the use of Beovu at this stage.
“Some universities and teaching hospitals were just getting Beovu on formulary when the reports about inflammation came out, and then they stopped the process and never used it. There are several large, private retina groups that heard of the reports and, as a group, decided not to use it. There are smaller groups and individual specialists who were early adopters, like myself, and feel that since we have safe drugs, Beovu should be used in selected patients until we gain more confidence that it is safe or the complications can be managed. Finally, there are some specialists who feel the risk is low and still use it, but that is a minority of the colleagues I have talked to,” Holekamp said.
Hahn works at NJRetina, a large practice with 16 locations. Early on, the practice deferred using Beovu until reimbursement issues were figured out.
“And once we figured this out, these safety events happened, and then COVID happened. Now we have decided to consider using it in selected cases. Even taking into account the side effects, I believe there is a role for Beovu, but that role still needs to be defined. I bring it up in my conversation with patients who are refractory to other treatments, and I am willing to use it for those who decide to,” he said.
Alexander Melamud, MD, MA, works in a retina-only subspecialty group and is a member of his practice’s medical advisory committee. He was an early adopter of Beovu and continues to use it on a routine basis in a group of patients.
“I agree with the recommendations of ASRS and ReST committee that risks and benefits in each patient should be discussed and that we have to proceed with caution. There are very legitimate, real concerns, but there are cases where the advantages of using Beovu definitely outweigh the potential disadvantages,” he said.
Personally, he has not seen any severe complications, and in his practice overall, the complication rate was low.
“They were mainly mild inflammatory reactions with no vasculitis, which resolved with topical steroid treatment,” he said.
He added that efficacy and better drying make Beovu the best drug for some patients, but not yet as a first-line therapy.
“The difficulty today is that we don’t know the mechanisms of this inflammation nor the predictive factors that may predispose to it. My personal protocol is to try patients on the other VEGF inhibitors first, and if they fail, try brolucizumab. I have good success in these cases but would not feel comfortable using it in treatment-naive patients,” Melamud said.
A frank discussion and a request to sign a specific informed consent that identifies the risk related to inflammation are important preliminary steps when proposing Beovu to patients.
“On top of this, I perform a careful slit lamp examination prior to injection to rule out the presence of any intraocular inflammation,” Holekamp said.
Overall, she uses Beovu “in selected patients” because the two extra steps of special informed consent and a careful slit lamp examination are time-consuming in a busy office.
“After discussing the pros and cons, about 80% of the patients chose to go back to their prior anti-VEGF agent, and 20% continued Beovu because they were seeing benefits they had not had with other drugs,” Holekamp said.
Risk tolerance is an important factor when patients are involved in the decision-making process, Zarbin said.
“I have patients who have complete equanimity about a 1% risk of a disastrous complication and other patients who absolutely cannot tolerate that risk,” he said.
Other criteria for patient selection can be inferred from the currently available data, which show that the overwhelming majority of complications, around 88%, occurred in women.
“I would not use it in an elderly woman with a history of intraocular inflammation, as initial therapy in a person who is one-eyed or in patients who are doing well with other agents. But I would consider using it for those who do not fall into these categories and are not adequately controlled with other drugs, as well as in patients whom I am not sure I will be able to see in 2 months, for whatever reason, because I can reasonably rely on a more prolonged effect,” Zarbin said.
Understanding and treating complications
“If we could identify in advance the patients who are more at risk for severe inflammatory reactions, or if we could find an effective strategy to promptly treat them, Beovu could have a new takeoff and conquer a large share of the market,” Zarbin said.
One of his patients recognized she was having a problem, promptly called him and the next day received intensive steroid therapy.
“Thank God she did fine, and it could be that that kind of approach really works. If I were sure it did, this would really change the concerns I have about that particular complication. But we don’t have enough experience to be sure that that would be generally the case,” he said.
Caroline R. Baumal, MD, is one of the experts involved in a think tank set up by Novartis with several goals, one of which is management considerations for inflammatory complications. Recently a paper was published in Ophthalmology Retina, based on the review of early reports of inflammation and vasculitis by a panel of external international experts and Novartis internal experts.
“One thing that we noted was that there was no uniform paradigm for management, but there were cases noted where management had led to improvement and even resolution. We reviewed the scientific literature, looked at phase 3 trial data, at all of the post-marketing publications and SRC assessments, and used our expertise to develop a preliminary treatment algorithm,” Baumal said.
Of first importance is to rule out endophthalmitis, she said. Some factors that may help to differentiate infection from inflammation include the time of onset in relation to the injection and the clinical features. Presentation with delayed and milder symptoms is more typical of inflammation, which may have an immune pathogenesis. In one publication, the mean time to present with inflammation associated with occlusive vasculitis after intravitreal brolucizumab was 30 days.
“The treatment paradigm outlined in the expert opinion manuscript is divided as to whether the clinical presentation consists of intraocular inflammation alone or whether there is associated retinal vasculitis or retinal vascular occlusion. There are imaging recommendations to guide this determination, and overall the recommendations are to use potent steroids topically with a low threshold to add intravitreal or oral steroids,” Baumal said.
“As the pathogenesis of inflammatory events becomes better understood, this may direct us to more targeted approaches. The MERLIN study evaluated eyes that had received other anti-VEGF agents before switching to Beovu. This will provide additional information about the efficacy of Beovu in non-treatment-naive eyes with persistent subretinal fluid, as well as data on adverse events in the context of a clinical trial.
“As data continues to accumulate, there will be more evidence from which to formulate a plan to best treat our patients,” she said.
- Baumal CR, et al. Ophthalmology. 2020;doi:10.1016/j.ophtha.2020.04.017.
- Baumal CR, et al. Ophthalmol Retina. 2020;doi:10.1016/j.oret.2020.09.020.
- Dugel PU, et al. Ophthalmology. 2020;doi:10.1016/j.ophtha.2019.04.017.
- Dugel PU, et al. Ophthalmology. 2020;doi:10.1016/j.ophtha.2020.06.028.
- Grunwald JE, et al. Ophthalmology. 2017;doi:10.1016/j.ophtha.2016.09.012.
- Haug SJ, et al. Am J Ophthalmol Case Rep. 2020;doi:10.1016/j.ajoc.2020.100680.
- Holekamp NM. Am J Manag Care. 2019;25(10 Suppl):S172-S181.
- Jain A, et al. Am J Ophthalmol Case Rep. 2020;doi:10.1016/j.ajoc.2020.100687.
- Kondapalli SSA. JAMA Ophthalmol. 2020;doi:10.1001/jamaophthalmol.2020.2810.
- Nguyen QD, et al. Ophthalmology. 2020;doi:10.1016/j.ophtha.2019.12.031.
- Rosenfeld PJ, et al. Am J Ophthalmol. 2020;doi:10.1016/j.ajo.2020.05.012.
- Sadda SR, et al. Ophthalmology. 2020;doi:10.1016/j.ophtha.2020.03.031.
- Sharma A, et al. Eye (Lond). 2020;doi:10.1038/s41433-019-0582-0.
- Sharma A, et al. Eye (Lond). 2020;doi:10.1038/s41433-020-01227-w.
- Sharma A, et al. Eye (Lond). 2020;doi:10.1038/s41433-020-0831-2.
- Study of safety and efficacy of brolucizumab 6 mg dosed every 4 weeks compared to aflibercept 2 mg dosed every 4 weeks in patients with retinal fluid despite frequent anti-VEGF injections (MERLIN). clinicaltrials.gov/ct2/show/NCT03710564.
- Witkin AJ, et al. J Vitreoretin Dis. 2020;doi:10.1177/2474126420930863.
- Yannuzzi NA, et al. Clin Ophthalmol. 2019;doi:10.2147/OPTH.S184706.
- Ying GS, et al. Ophthalmol Retina. 2018;doi:10.1016/j.oret.2017.10.003.
- Zarbin M, et al. Ophthalmology. 2019;doi:10.1016/j.ophtha.2019.01.003.
- For more information:
- Caroline R. Baumal, MD, can be reached at Tufts University School of Medicine, New England Eye Center, 260 Tremont St., Boston, MA 02111; email: firstname.lastname@example.org.
- Paul Hahn, MD, PhD, can be reached at NJRetina, 628 Cedar Lane, Teaneck, NJ 07666; email: email@example.com.
- Nancy M. Holekamp, MD, can be reached at Center for Macular Degeneration, Pepose Vision Institute, 1815 Clarkson Rd, Chesterfield, MO 63017; email: firstname.lastname@example.org.
- Alexander Melamud, MD, MA, can be reached at The Retina Group of Washington, 8270 Willow Oaks Corporate Drive, 6th Floor, Fairfax, VA 22031-4621; email: email@example.com.
- Marco A. Zarbin, MD, PhD, can be reached at the Institute of Ophthalmology and Visual Science, New Jersey Medical School, 90 Bergen St., Room 6155, Newark, NJ 07103-2499; email: firstname.lastname@example.org.
Click here to read the Point/Counter to this Cover Story.