Cornea
July 2021

by Ellen Stodola
Editorial Co-Director

Numerous studies have looked at different approaches to using bevacizumab and other anti-VEGF agents for anterior segment surgery. Several experts discussed their use and study of bevacizumab, particularly relating to corneal neovascularization and in corneal transplants. Different approaches—subconjunctival, intrastromal, and topical—were discussed.

Subconjunctival bevacizumab use for corneal graft survival

Alanna Nattis, DO, and Puneet Singh, MS, recently examined subconjunctival bevacizumab for corneal graft survival. Though a number of patients do well after corneal transplants, there are some who have a lot of inflammation, Dr. Nattis said. This could be due to infection or possibly preexisting corneal neovascularization, she said, adding that neovascularization can portend a poor prognosis. “We know steroids can be helpful, and clinically, I’ve had positive experience using subconjunctival bevacizumab for patients with and without corneal transplants with significant neovascularization.”

However, there has not been a large review or many controlled trials in this area. “We thought it was important to look at the literature that was out there,” she said.

A literature review found little research on the subconjunctival use of bevacizumab, Ms. Singh said, adding that most research was done outside the U.S.

Despite this, Ms. Singh said the use of bevacizumab to prevent corneal graft rejection has increased, with corneal neovascularization being a major risk factor for rejection. Its use limits graft rejection and the need for further invasive procedures that may result from neovascularization.

“We’re seeing that it’s being used, but there are no set guidelines about the dosage, how to administer it, how to get maximum effect, or whether there’s direct adverse effects seen with the drug,” Ms. Singh said.

Dr. Nattis and Ms. Singh are hoping that their literature review and exploration into this topic may serve as a baseline for future research.

Small sample size seemed to be an issue in the papers reviewed. Ms. Singh noted that several were just case reports, while other studies on the topic were limited to 10–50 patients. “We’re seeing a large disparity in sample size as well,” she said. “We were hoping that this would set a baseline to get a bigger project rolling and see if this would translate to a larger population for a study.”

Dr. Nattis added that additional data could be gathered from corneal specialists in a retrospective study. She noted that she has some personal experience using subconjunctival bevacizumab in fellowship training, as well as topical. She found the subconjunctival approach was more efficacious and was well tolerated by patients.

Ms. Singh noted that in her review, she found a study comparing the use of subconjunctival vs. topical bevacizumab. The patients receiving topical treatment ended up going back to subconjunctival because there was suboptimal regression seen with corneal neovascularization. That study suggested it was probably due to a tissue barrier that couldn’t be penetrated with topical use.¹

Subconjunctival and intrastromal approaches

Allan Slomovic, MD, has explored both subconjunctival and intrastromal approaches for using bevacizumab.

A paper published in Cornea in 2008² looked at 10 patients receiving subconjunctival injections of 2.5 mg/0.1 mL. Dr. Slomovic said this yielded very good results. He noted that imaging with a camera and computer software was used to identify the density and the number of clock hours the neovascularization extended.

The study results noted that “corneal neovascularization covered, on average, 14.8%±2.5% (SD) of the corneal surface before the injections, compared with 10.5%±2.8% (P=0.36, t test) after bevacizumab injection. Therefore, bevacizumab decreased corneal neovascularization by 29%.” This led to the conclusion that bevacizumab was well tolerated with a subconjunctival approach and was effective for partial regression of corneal neovascularization.

Dr. Slomovic mentioned an additional study in Cornea from 2011³ that introduced an intrastromal approach as well as subconjunctival injection. “We found that was very effective,” he said, noting that there were no complications in either study. The second study included 12 patients with the same total dosage.

“With our first study, we didn’t think there was enough of the bevacizumab by giving subconjunctival at the limbus to reach the distal part of the vessel,” Dr. Slomovic said. “So we combined that with an intrastromal injection, and we got a bit more regression.” He added that he thinks using an injection approach may be better than topical because of the longer duration of contact.

Dr. Slomovic said that using bevacizumab for these patients may be most efficacious when the neovascularization is localized to a small number of clock hours, like in those with herpes zoster, herpes simplex, or infectious keratitis. He said it’s not as helpful in diffuse vascularization, like limbal stem cell disease or a vascularized corneal transplant.

Subconjunctival and topical bevacizumab in corneal transplantation

Thomas Dohlman, MD, highlighted his work using bevacizumab for corneal transplantation. Dr. Dohlman is working as part of a team at Massachusetts Eye and Ear along with Reza Dana, MD, and others to analyze the data of a multicenter trial evaluating the safety and efficacy of bevacizumab in high-risk corneal transplantation. The preliminary data was presented late last year.

“Corneal transplantation is the most commonly performed form of solid organ transplantation, and we know that one of the greatest predictors of corneal transplant rejection and failure is host bed neovascularization,” Dr. Dohlman said. “These vascularized, ‘high-risk’ grafts have uniformly worse outcomes when compared to ‘low-risk’ transplants performed in avascular host beds.” Dr. Dohlman noted that he has investigated the role of VEGF inhibition in animal models of corneal transplantation, which fueled his interest in the current research.

While there have been some small studies and case reports previously on this topic, Dr. Dohlman said there have been no prospective, randomized, placebo-controlled studies in the literature evaluating bevacizumab in corneal transplantation. “The study we are working on is a prospective, randomized, double-masked, placebo-controlled study that, to our knowledge, is the largest clinical study on this topic to date,” he said.

In the study, Dr. Dohlman and his colleagues compared bevacizumab to placebo in high-risk corneal transplantation, with the bevacizumab group receiving a subconjunctival injection of 2.5 mg/0.1 ml (2.5%) at the time of surgery and topical 1% bevacizumab four times per day for 4 weeks. The control group received a subconjunctival injection of 0.9% saline at the time of surgery and carboxymethylcellulose four times per day for 4 weeks. Patients were followed for 52 weeks for any adverse effects and to track efficacy, which included rate of endothelial immune rejection and rate of overall graft failure. There were 43 patients in each group.

Dr. Dohlman noted that bevacizumab has been used for corneal neovascularization at different concentrations both topically and subconjunctivally. “Subconjunctival injection allows for higher and more lasting drug availability, while the advantage of topical application is that it is easy to administer, although the penetration of bevacizumab into the corneal stroma is lower than when given subconjunctivally,” he said. “Both methods of administration are effective at reducing corneal neovascularization, but in an animal model of corneal transplantation where subconjunctival treatment was directly compared to topical treatment, the subconjunctival route showed a greater effect in reducing corneal neovascularization and improved graft survival.” In this study, a hybrid approach was used with subconjunctival injection at the time of surgery followed by topical treatment for 4 weeks postoperatively.

While Dr. Dohlman’s recent research specifically looked at bevacizumab in corneal transplantation, he said it could be an effective treatment in a variety of corneal conditions characterized by pathologic neovascularization, including after chemical injuries and infectious keratitis. He said that there may also be a role for pretreating corneal neovascularization prior to corneal transplantation.

Dr. Dohlman said results from his study showed that bevacizumab decreased the 52-week rate of endothelial rejection and graft failure, and although the results did not reach statistical significance, they showed that treatment with bevacizumab had a modest effect on improving the rate of endothelial rejection and overall graft failure in high-risk corneal transplantation, making bevacizumab a potential adjunctive therapy for improving outcomes in high-risk corneal transplantation.

“Bevacizumab was safe, as there was no difference in treatment-related adverse effects between groups,” Dr. Dohlman said. “A potential concern with bevacizumab is delayed epithelial healing, but there was no difference in the incidence of delayed epithelial healing at 7 days postop (p=0.16), and the control group actually had a higher number of patients with delayed epithelial healing (16% in the control group versus 5% in the bevacizumab group).”

PROSE with bevacizumab

Jia Yin, MD, PhD, has researched the use of bevacizumab in Prosthetic Replacement of the Ocular Surface Ecosystem (PROSE), a custom-designed, fabricated prosthetic device developed by BostonSight. The device is a gas permeable lens with a diameter of 17.5–23.0 mm that rests on the bulbar conjunctiva and vaults over the entire cornea and bathes the cornea in artificial tears. 

She said it is used in the management of a distorted cornea, as well as complex ocular surface diseases. In addition to protecting the ocular surface, PROSE can serve as a novel drug delivery system, Dr. Yin said, which is where the idea of using PROSE to deliver an anti-VEGF agent originated. “I worked with the former director of BostonSight Deborah Jacobs, MD, to examine the long-term outcome of using PROSE to deliver bevacizumab to [treat] corneal neovascularization,” she said.

Dr. Yin noted that patients were already fitted and using the PROSE device routinely before starting this treatment. They applied a drop of 1% preservative-free bevacizumab to the reservoir of the PROSE device twice daily, and patients continued daily wear of the device for an average of 3 months, she said.

In a cohort of 13 patients,⁴ there was an observed reduction in corneal neovascularization and improvement in vision with no significant adverse effects, Dr. Yin said. However, she did note that the treatment is off label and best suited for select patients. 

Comparing use of PROSE to other delivery methods for bevacizumab, Dr. Yin said that it provides higher bioavailability of the anti-VEGF agent, specifically compared to topical application. Compared to a subconjunctival injection, PROSE is less invasive and used daily. “We observed no adverse events, even in patients who used this method intermittently for many years,” she said.

PROSE with bevacizumab may be an effective treatment for corneal neovascularization. Dr. Yin noted that PROSE has also been used to deliver ophthalmic antibiotics. “PROSE alone has been used in many ocular surface diseases, such as severe corneal ectasia, dry eye disease, Stevens-Johnson syndrome, and persistent corneal epithelial defect,” she said, noting that a limitation could be the needed customization of the device. Thus, it may not be a first-line choice for the sole purpose of treating corneal neovascularization.

---

About the physicians

Thomas Dohlman, MD
Massachusetts Eye and Ear
Boston, Massachusetts

Alanna Nattis, DO
SightMD
Babylon, New York

Puneet Singh, MS
Medical Student
New York Institute of Technology College of Osteopathic Medicine
Old Westbury, New York

Allan Slomovic, MD
Marta and Owen Boris Endowed Chair in Cornea and Stem Cell Research
Department of Ophthalmology and Vision Sciences
University of Toronto
Toronto, Canada

Jia Yin, MD, PhD
Massachusetts Eye and Ear
Department of Ophthalmology
Harvard Medical School
Boston, Massachusetts

References

1. Kim SW, et al. The effect of topical bevacizumab on corneal neovascularization. Ophthalmology. 2008;115:e33–38.
2. Bahar I, et al. Subconjunctival bevacizumab injection for corneal neovascularization. Cornea. 2008;27:142–147.
3. Yeung S, et al. Combined use of subconjunctival and intracorneal bevacizumab injection for corneal neovascularization. Cornea. 2011;30:1110–1114.
4. Yin J, Jacobs DS. Long-term outcome of using Prosthetic Replacement of Ocular Surface Ecosystem (PROSE) as a drug delivery system for bevacizumab in the treatment of corneal neovascularization. Ocul Surf. 2019;17:134–141.

Relevant disclosures

Dohlman: None
Nattis: None
Singh: None
Slomovic: Abbott, Alcon, Allergan, Johnson & Johnson Vision, Labtician Thea, Novartis, Santen, Sun Pharmaceutical
Yin: None

Contact

Dohlman: Thomas_Dohlman@meei.harvard.edu
Nattis: asn516lu@gmail.com
Singh: psingh59@nyit.edu
Slomovic: allan.slomovic@utoronto.ca
Yin: Jia_Yin@meei.harvard.edu