Short-term topical bevacizumab in the treatment of stable corneal neovascularization

Abstract

Purpose: To evaluate the safety and efficacy of topical bevacizumab in the treatment of corneal neovascularization.

Design: Prospective, nonrandomized, interventional case series.

Methods: setting: Institutional, multicenter clinical trial. study population: Twenty eyes from 20 patients with stable corneal neovascularization. intervention procedures: Patients were treated with topical 1.0% bevacizumab for 3 weeks and were monitored for a total of 24 weeks. main outcome measures: Primary outcome measures included: neovascular area, defined as the area of the corneal vessels themselves; vessel caliber, defined as the mean corneal vessel diameter; and invasion area, defined as the fraction of the total cornea into which the vessels extended. The occurrence of ocular and systemic adverse events was monitored closely.

Results: As compared with the baseline visit, patients exhibited a statistically significant improvement in neovascular area by week 6 (P = .007) and in vessel caliber by week 12 (P = .006). At the final visit, neovascular area, vessel caliber, and invasion area were reduced by 47.5%, 36.2%, and 20%, respectively. The decreases in neovascular area and vessel caliber were statistically significant (P < .001 and P = .003, respectively); however, the reduction in invasion area did not reach statistical significance (P = .06). There were no significant changes in the secondary outcomes, and there were no adverse events.

Conclusions: Short-term topical bevacizumab treatment reduced the extent of stable corneal neovascularization as measured by neovascular area and vessel caliber with no associated adverse events. Interestingly, the degree of treatment efficacy was inversely proportional to the baseline invasion area.

Trial registration: ClinicalTrials.gov NCT00559936.

Copyright © 2012 Elsevier Inc. All rights reserved.

Figures

Figure 1

Quantification of corneal neovascularization. A graphics editing software (Photoshop CS2; Adobe Systems Inc.; Berkeley, CA) and a mathematical program (MATLAB; MathWorks Inc.; Natick, MA) were used to analyze slit lamp digital corneal photographs. These metrics were computed as follows: neovascular area, which measures the area of the corneal vessels themselves; vessel caliber, which determines an approximate mean diameter of the corneal vessels; and invasion area, which measures the fraction of corneal area in which neovascular incursion is seen.

Figure 2

Summary of changes for all patients at various time points using three metrics of corneal neovascularization. The metrics used were: neovascular area, vessel caliber, and invasion area. Compared to the baseline visit, the neovascular area showed statistically significant decrease at week 6 (P = .007) and the vessel caliber at week 12 (P = .006). By week 24, the mean changes were −47.5% ± 8.8% (n = 18) for neovascular area, −36.2% ± 10.4 (n = 18) for vessel caliber, and −20.0% ± 9.9% (n = 18) for invasion area. The decreases in neovascular area and vessel caliber were statistically significant (P < .001 and P = .003, respectively), but invasion area was not (P = .06). Error bars represent SEM (standard error of mean). * P < .05, ** P < .001

Figure 2

Summary of changes for all patients at various time points using three metrics of corneal neovascularization. The metrics used were: neovascular area, vessel caliber, and invasion area. Compared to the baseline visit, the neovascular area showed statistically significant decrease at week 6 (P = .007) and the vessel caliber at week 12 (P = .006). By week 24, the mean changes were −47.5% ± 8.8% (n = 18) for neovascular area, −36.2% ± 10.4 (n = 18) for vessel caliber, and −20.0% ± 9.9% (n = 18) for invasion area. The decreases in neovascular area and vessel caliber were statistically significant (P < .001 and P = .003, respectively), but invasion area was not (P = .06). Error bars represent SEM (standard error of mean). * P < .05, ** P < .001

Figure 2

Summary of changes for all patients at various time points using three metrics of corneal neovascularization. The metrics used were: neovascular area, vessel caliber, and invasion area. Compared to the baseline visit, the neovascular area showed statistically significant decrease at week 6 (P = .007) and the vessel caliber at week 12 (P = .006). By week 24, the mean changes were −47.5% ± 8.8% (n = 18) for neovascular area, −36.2% ± 10.4 (n = 18) for vessel caliber, and −20.0% ± 9.9% (n = 18) for invasion area. The decreases in neovascular area and vessel caliber were statistically significant (P < .001 and P = .003, respectively), but invasion area was not (P = .06). Error bars represent SEM (standard error of mean). * P < .05, ** P < .001

Figure 3

Correlation between the extent and reduction of corneal neovascularization. Patients with less invasion area at baseline experienced greater reductions in neovascular area and vessel caliber: reductions in neovascular area (Top) and vessel caliber (Bottom) showed statistically significant correlations with baseline invasion area (P = .049 and P = .015, respectively).

Figure 3

Correlation between the extent and reduction of corneal neovascularization. Patients with less invasion area at baseline experienced greater reductions in neovascular area and vessel caliber: reductions in neovascular area (Top) and vessel caliber (Bottom) showed statistically significant correlations with baseline invasion area (P = .049 and P = .015, respectively).

Figure 4

Decrease in corneal neovascularization over 24 weeks of follow-up. The penetrating keratoplasty (PKP) group (n = 5) showed less improvement than the limbal stem cell deficiency (LSCD) group (n = 4) and herpetic simplex virus keratitis (HSV)/ herpetic zoster ophthalmicus (HZO) group (n = 4) either in (Top) neovascular area (P = .004, ANOVA; HSV/HZO vs PKP: P < .01, LSCD vs PKP: P < .05, Bonferroni’s multiple comparison test ) or in (Bottom) vessel caliber (P = .005, ANOVA; HSV/HZO vs PKP: P < .01, LSCD vs PKP: P < .05, Bonferroni’s multiple comparison test).

Figure 4

Decrease in corneal neovascularization over 24 weeks of follow-up. The penetrating keratoplasty (PKP) group (n = 5) showed less improvement than the limbal stem cell deficiency (LSCD) group (n = 4) and herpetic simplex virus keratitis (HSV)/ herpetic zoster ophthalmicus (HZO) group (n = 4) either in (Top) neovascular area (P = .004, ANOVA; HSV/HZO vs PKP: P < .01, LSCD vs PKP: P < .05, Bonferroni’s multiple comparison test ) or in (Bottom) vessel caliber (P = .005, ANOVA; HSV/HZO vs PKP: P < .01, LSCD vs PKP: P < .05, Bonferroni’s multiple comparison test).

Figure 5

Representative cases of stable corneal neovascularization treated with topical bevacizumab. Patient 11 was a 75-year-old female with a failed corneal graft. (Top left) The baseline photograph revealed a corneal opacity with circumferential (360 degrees) neovascularization (NV) in the recipient bed. (Top right) After topical bevacizumab treatment, the NV showed only modest changes. Patient 4 was a 40-year-old male with a history of laser in situ keratomileusis and herpes simplex keratitis. (Bottom left) A relatively small area of corneal NV was present at baseline and by week 24 the NV had decreased significantly (Bottom right).

Figure 5

Representative cases of stable corneal neovascularization treated with topical bevacizumab. Patient 11 was a 75-year-old female with a failed corneal graft. (Top left) The baseline photograph revealed a corneal opacity with circumferential (360 degrees) neovascularization (NV) in the recipient bed. (Top right) After topical bevacizumab treatment, the NV showed only modest changes. Patient 4 was a 40-year-old male with a history of laser in situ keratomileusis and herpes simplex keratitis. (Bottom left) A relatively small area of corneal NV was present at baseline and by week 24 the NV had decreased significantly (Bottom right).

Figure 5

Representative cases of stable corneal neovascularization treated with topical bevacizumab. Patient 11 was a 75-year-old female with a failed corneal graft. (Top left) The baseline photograph revealed a corneal opacity with circumferential (360 degrees) neovascularization (NV) in the recipient bed. (Top right) After topical bevacizumab treatment, the NV showed only modest changes. Patient 4 was a 40-year-old male with a history of laser in situ keratomileusis and herpes simplex keratitis. (Bottom left) A relatively small area of corneal NV was present at baseline and by week 24 the NV had decreased significantly (Bottom right).

Figure 5

Representative cases of stable corneal neovascularization treated with topical bevacizumab. Patient 11 was a 75-year-old female with a failed corneal graft. (Top left) The baseline photograph revealed a corneal opacity with circumferential (360 degrees) neovascularization (NV) in the recipient bed. (Top right) After topical bevacizumab treatment, the NV showed only modest changes. Patient 4 was a 40-year-old male with a history of laser in situ keratomileusis and herpes simplex keratitis. (Bottom left) A relatively small area of corneal NV was present at baseline and by week 24 the NV had decreased significantly (Bottom right).