- Email: [email protected]
Benefit from Bevacizumab for Macular Edema in Central Retinal Vein Occlusion: Twelve-Month Results of a Prospective, Randomized Study
Benefit from Bevacizumab for Macular Edema in Central Retinal Vein Occlusion: Twelve-Month Results of a Prospective, Randomized Study David L. Epstein, MD, Peep V. Algvere, MD, PhD, Gunvor von Wendt, MD, PhD, Stefan Seregard, MD, PhD, Anders Kvanta, MD, PhD Purpose: To evaluate the efficacy of intraocular injections with bevacizumab over 12 months in patients with macular edema (ME) secondary to central retinal vein occlusion (CRVO). Design: A prospective study including a randomized 6-month, sham injection-controlled, double-masked clinical trial followed by a 6-month open-label extension. Participants: Sixty patients with ME secondary to CRVO. Methods: At baseline, patients were randomized 1:1 to receive intraocular injections of bevacizumab or sham injections every 6 weeks for 6 months. From month 6, all patients received intraocular injections of bevacizumab every 6 weeks for 6 months. Main Outcome Measures: The primary outcome measure was the proportion of patients gaining at least 15 letters at 12 months. Secondary outcome measures included mean change from baseline best-corrected visual acuity (BCVA), change in foveal thickness, and development of neovascular glaucoma. Results: At the end of follow-up, 18 of 30 patients (60.0%) in the bevacizumab/bevacizumab (bz/bz) group had gained ⱖ15 letters compared with 10 of 30 patients (33.3%) in the sham/bevacizumab (sh/bz) group (P ⬍ 0.05). The BCVA improved by 16.0 letters at 12 months in the bz/bz group compared with 4.6 letters in the sh/bz group (P ⬍ 0.05). In an unplanned retrospective analysis, patients aged ⬎70 years had a significantly worse outcome when receiving delayed treatment, losing 1.4 letters (95% confidence interval [CI], ⫺9.7 to 8.4) in the sh/bz group compared with a gain of 20.1 letters (95% CI, 13.9 –26.3) in the bz/bz group in patients aged ⬍70 years (P ⬍ 0.003). The mean decrease in central retinal thickness (CRT) was 435 m in the bz/bz group compared with 404 m in the sh/bz group (P ⫽ not significant). No patients developed iris rubeosis during the 6-month open-label extension period. There were no events of endophthalmitis, retinal tear, or retinal detachment during the 12-month treatment period. No serious nonocular adverse events were reported. Conclusions: Intraocular injections of bevacizumab given every 6 weeks for 12 months improve visual acuity (VA) and reduce ME significantly. Patients receiving delayed treatment have a limited visual improvement. Financial Disclosure(s): Proprietary or commercial disclosure may be found after the references. Ophthalmology 2012;119:2587–2591 © 2012 by the American Academy of Ophthalmology.
Central retinal vein occlusion (CRVO) is a common retinal vascular disorder and often presents with a severely reduced visual acuity (VA) caused by macular edema (ME).1– 4 Vascular endothelial growth factor A (VEGF-A) is produced in response to retinal ischemia secondary to the increased resistance to venous blood flow. The intravitreal level of VEGF-A in CRVO is higher than in any other ischemic retinal vascular disease and several times higher than the levels found in branch vein occlusion.5,6 Vascular endothelial growth factor A increases vascular permeability by disrupting junctional proteins in the blood vessel wall, causing uncoupling of the endothelial cell– cell junctions.7 By targeting VEGF-A, these changes can be reversed and the ME reduced. Different antibodies that block the vascular permeability effect of VEGF-A have been studied recently. © 2012 by the American Academy of Ophthalmology Published by Elsevier Inc.
In the Treatment of Macular Edema after Central Retinal Vein Occlusion (CRUISE) study, intravitreal ranibizumab (Lucentis, anti-VEGF Fab; Genentech Inc., South San Francisco, CA) improved vision significantly compared with sham, with approximately half of the patients gaining 3 or more Early Treatment Diabetic Retinopathy Study (ETDRS) lines.8 After 6 months, all patients could receive ranibizumab injections if ME was present and best-corrected visual acuity (BCVA) was ⱕ20/40. The visual benefit was sustained after 12 months, and patients originally treated with sham injections showed an improvement in BCVA but significantly less than in the ranibizumab groups.9 The CRUISE study neither answered whether even better visual outcomes would have been achieved by continuing monthly injections through 12 months nor addressed the possible impact of giving the ISSN 0161-6420/12/$–see front matter http://dx.doi.org/10.1016/j.ophtha.2012.06.037
2587
Ophthalmology Volume 119, Number 12, December 2012 group originally designated to sham treatment monthly ranibizumab injections during the final 6 months. In a sham-controlled prospective study, we recently showed that intravitreal bevacizumab (IVB) (Avastin, antiVEGF IgG; Genentech, Inc) significantly improved VA and ME over 6 months.10 After 4 injections every 6 weeks of 1.25 mg bevacizumab in patients with CRVO, mean improvement in BCVA score was 14.1 letters compared with a loss of 2.0 letters in the sham group. After 6 months, all patients were enrolled in the present open-label extension study and received an additional 4 treatments of bevacizumab every 6 weeks for 6 months. The objectives of this study were to investigate whether this regimen can maintain VA and to elucidate the impact of delaying treatment for 6 months in patients with ME secondary to CRVO.
received 4 IVB injections every 6 weeks. The 2 treatment groups were designated bevacizumab/bevacizumab (bz/bz) and sham/bevacizumab (sh/bz). The total follow-up period was 48 weeks. Study patients were masked to the treatment given during the first 24 weeks. Staff performing VA testing, OCT, fundus photography, and follow-up investigators were also masked to treatment group.
Outcome Measures The primary outcome measure was the proportion of patients gaining ⱖ15 ETDRS letters at 12 months. The secondary outcome measures were the change in BCVA, change in foveal thickness as measured by OCT, and number of patients with neovascular glaucoma defined as increased intraocular pressure due to the formation of new vessels in the angle as diagnosed by gonioscopy.
Power Calculation and Statistical Analysis
Patients and Methods We conducted a prospective randomized study on 60 eyes of 60 patients with ME secondary to CRVO. Informed consent was obtained from all patients. From April 2009 to December 2010, the patients were enrolled at St. Eriks Eye Hospital in Stockholm, Sweden. These patients comprised the intent-to-treat population in whom the last observation carried forward method was used for missing data for all efficacy parameters. The study adhered to the guidelines of the Declaration of Helsinki. The protocol was approved by the local ethics committee and the Swedish Medical Products Agency. The study is listed on www.clinicaltrials.gov, under identifier NCT00906685 (accessed May 23, 2012).
Study Protocol The details of the protocol, including eligibility criteria, randomization, and procedures for examination, optical coherence tomography (OCT), and IVB injections, have been described.10 The OCT images were obtained by a certified technician using the Zeiss Cirrus OCT machine (Carl Zeiss Meditec, Inc, Dublin, CA). There were no significant differences between the study groups at baseline (Table 1). Initially, each study participant was randomized 1:1 to IVB injections (study group) or sham injections (control group) every 6 weeks for 6 months. During months 6 to 12, all patients
The primary hypothesis was based on a difference between the treatment groups at 6 months in the proportion of patients achieving the main outcome measure. We assumed that 35% of IVBtreated patients and 5% of sham-treated patients would achieve the primary end point (gain of at least 15 ETDRS letters). With a statistical power of 80% and the level of statistical significance set at P ⬍ 0.05, we estimated that a minimum of 24 patients would be required (MedCalc Software, Mariakerke, Belgium). For statistical analyses, the independent Student t test and the Fisher exact test (to compare differences in distributions between the groups) were used.
Results Sixty eyes of 60 patients were originally assigned randomly to IVB (30 eyes) or sham injections (30 eyes). Demographic and baseline characteristics have been reported.10 Some 93% of the patients completed the study through month 12. Four patients discontinued the study during the open-label extension phase (3 patients in the bz/bz group and 1 patient in the sh/bz group). In the bz/bz group, the patients discontinued the study because of difficulties returning to control visits caused by metastatic lung carcinoma, metastatic prostate carcinoma, and old age. One patient in the sh/bz group had a transient ischemic attack while receiving sham treatment and was therefore not offered IVB injections.
Table 1. Patient Demographics and Baseline Ocular Characteristics Parameter
sh/bz n ⴝ 30
bz/bz n ⴝ 30
All
Age (yrs⫾SD) Gender ratio m:f (%) Time from diagnosis to inclusion (w⫾SD) ⬍90 days n (%) ⬎90 days n (%) BCVA (ETDRS letters⫾SD) BCVA distribution n (%) ⬍34 ⬎34 CRT (m⫾SD) Hypertension n (%) Diabetes mellitus n (%)
70.4⫾10.4 57:43 9.4⫾6.5 22 (73.3) 8 (26.7) 43.9⫾16.0 10 (33.3) 20 (66.7) 729⫾195 16 (53.3) 3 (10)
70.6⫾12.6 63:37 8.3⫾4.8 21 (70) 9 (30) 44.4⫾15.3 9 (30) 21 (70) 712⫾330 13 (43.3) 1 (3.3)
70.5⫾12.6 60:40 8.8⫾5.7 43/60 (71.7) 17/60 (28.3) 44.1⫾15.5 19 (31.7) 41 (58.3) 721⫾269 29 (48.3) 4 (6.7)
BCVA ⫽ best corrected visual acuity; bz/bz ⫽ bevacizumab/bevacizumab; CRT⫽ central retinal thickness; ETDRS ⫽ Early Treatment Diabetic Retinopathy Study; SD ⫽ standard deviation; sh/bz ⫽sham/bevacizumab. P value was nonsignificant between treatment groups for all parameters.
2588
Epstein et al 䡠 Bevacizumab for ME in CRVO
Figure 1. Mean change from baseline best-corrected visual acuity (BCVA) over time to month 12. The difference in BCVA between the treatment groups was statistically significant at week 48 (P ⬍ 0.05). Error bars represent ⫾ standard error of the mean. The last observation carried forward method was used to compute missing data. bz/bz ⫽ bevacizumab/ bevacizumab; ETDRS ⫽ Early Treatment Diabetic Retinopathy Study; sh/bz ⫽ sham/bevacizumab.
Visual Acuity In the bz/bz group, the VA improved by 16.1 letters at week 48 (Fig 1). During weeks 24 to 48, a further gain of 2.0 letters was seen with this fixed injection schedule. In the sh/bz group, VA decreased by 2.0 letters at 24 weeks but improved by 4.6 letters at week 48. The difference in VA between the treatment groups was statistically significant from week 12 and beyond (P ⬍ 0.05). At
Figure 3. Mean change from baseline central retinal thickness (CRT) over time to month 12. The mean decrease in CRT was similar in the study groups (P ⫽ not significant). The CRT decreased from 729 to 325 m in the sham/bevacizumab (sh/bz) group and from 712 to 277m in the bevacizumab/bevacizumab (bz/bz) group. The last observation carried forward method was used to compute missing data. Error bars represent ⫾ standard error of the mean.
the end of follow-up, 18 of 30 patients (60.0%) in the bz/bz group had gained ⱖ15 letters compared with 10 of 30 patients (33.3%) in the control group (P ⬍ 0.05). In both groups, 2 of 30 patients (6.7%) lost ⬎15 ETDRS letters after 48 weeks. Snellen BCVA equivalent of ⱕ20/200 is considered a poor visual outcome. In the bz/bz group, 1 of 30 patients (3.3%) had this outcome compared with 5 of 30 patients (16.7%) in the control group (P ⬎ 0.05). An unplanned retrospective analysis of the impact of age and response to therapy was done. Younger patients had a better visual outcome. Overall, patients aged ⬍70 years improved 14.2 letters (95% confidence interval [CI], 6.3–22.0) compared with 7.4 letters (95% CI, 0.2–14.6) in patients aged ⬎70 years (P ⬎ 0.05) (Fig 2). Patients aged ⬎70 years had a significantly worse outcome when receiving delayed treatment, losing 1.4 letters (95% CI, ⫺9.7 to 8.4) in the sh/bz group compared with a gain of 20.1 letters (95% CI, 13.9 –26.3) in the bz/bz group in patients aged ⬍70 years (P ⬍ 0.003). No significant confounders were found in the older group regarding the baseline characteristics.
Anatomic Outcomes at 48 Weeks
Figure 2. Change in best-corrected visual acuity (BCVA) (Early Treatment Diabetic Retinopathy Study [ETDRS] letters) according to age and treatment group. Patients aged ⬎70 years in the sham/bevacizumab (sh/bz) group had a significantly worse visual outcome, losing 1.4 letters (95% CI, ⫺9.7 to 8.4) compared with a gain in the bevacizumab/bevacizumab (bz/bz) group in patients ⬍70 years of age of 20.1 letters (95% CI, 13.9 –26.3) (P ⬍ 0.003). P ⬎ 0.05 for comparison between all other groups. Error bars represent 95% CI. The last observation carried forward method was used to compute missing data. y.o. ⫽ years of age.
The mean change in central retinal thickness (CRT) was similar in both groups, with a mean decrease of 435 m (bz/bz) and 404 m (sh/bz) at week 48 (P ⬎ 0.05; Fig 3). Assessment of residual ME, defined morphologically and evaluated manually as any residual fluid at 48 weeks, was performed. No ME was found in 25 of 30 patients (83.3%) in the bz/bz group compared with 18 of 30 patients (60%) in the sh/bz group (P ⫽ not significant).
Neovascularization As reported earlier, 5 of 30 patients (16.7%) in the sham group had developed iris rubeosis at week 24.10 The iris neovascularization had regressed in all these patients at 48 weeks. No new cases of iris rubeosis were seen during months 6 to 12 in either group.
2589
Ophthalmology Volume 119, Number 12, December 2012 Ocular Safety and Systemic Adverse Events There were no events of endophthalmitis, retinal tear, or retinal detachment during the 48-week treatment period. No serious nonocular adverse events were reported. However, this prospective study was not powered to provide definite data on safety.
Discussion The present study showed that IVB given on a fixed dosing regimen for 12 months provided a substantial visual benefit in patients with ME secondary to CRVO. We also showed that the visual gain was limited in patients who initially received sham treatment followed by subsequent IVB injections. Our results compare favorably with other studies. In the CRUISE study, patients receiving 0.5 mg ranibizumab gained a mean of 13.9 letters at 12 months after intravitreal injections of ranibizumab every 4 weeks for 6 months and then treated as needed. During the last 6 months when the flexible dosing regimen was used, patients lost 1.0 letters. Fifty percent of the patients showed improved BCVA of at least 3 ETDRS lines.9 In the Global Evaluation of Implantable Dexamethasone in Retinal Vein Occlusion with Macular Edema (GENEVA) study, improvement peaked at day 60 with a BCVA of 9.7 letters.11 Direct comparisons between investigations always should be made with caution, and in these studies disease duration, for example, was different. Patients in the GENEVA study had a longer disease duration with 84% of subjects treated more than 90 days after disease onset versus 30% in our study and 28% in the CRUISE study. Disease duration seems to have a critical impact on visual outcome. In our study, patients in the sh/bz group showed a substantial improvement in BCVA when treated with IVB. However, there remained a significant difference in mean improvement from baseline BCVA at 12 months between the sh/bz and bz/bz groups. Patients with treatment delayed for 6 months gained only 4.6 letters compared with a gain of 16.1 letters in the bz/bz group. This difference could not be attributed to changes in CRT because the mean improvement from baseline CRT was similar in both groups. This is the first study to show that patients with ME in CRVO receiving delayed IVB treatment on a fixed schedule still have a significantly worse visual outcome than patients treated without delay. Recently published data from a subset of patients with CRVO in the GENEVA study showed that at 12 months each monthly increase of ME duration was associated with a decreased likelihood of achieving an improvement of at least 3 ETDRS lines. However, in the subset of patients with CRVO, the data were not statistically significant.12 This could be due to the long mean duration of ME in this subset of patients of approximately 8 months compared with 2 months in the present study. In our study, the likelihood of improving 3 ETDRS lines when therapy was delayed for 6 months was significantly lower (33%) than when treatment was given at study start (60%). Cystoid ME is a severe complication in CRVO, resulting in disorganization of the retinal cellular structure, causing severe damage to the neural cells. It has been shown
2590
that structural abnormality to the photoreceptor layer reducing foveal function and retinal sensitivity may remain after resolved ME secondary to retinal vein occlusion.13 An association has been reported in patients with ME between the reduction of visual pigment density and their visual function.14 This suggests that a long-lasting edema may lead to irreversible macular damage, which cannot be restored solely by resolution of the ME. Therefore, early therapeutic intervention to reduce the ME is important for improving the odds of obtaining a good visual outcome. Age has been reported as a risk factor for retinal vein occlusion.15 In the present study, we found that age also might be a prognostic factor in response to IVB injections. Especially in the group aged ⬎70 years, delayed IVB treatment seemed to have a negative impact on the visual outcome. These results should be interpreted with caution because of the small subgroup size. Similar findings have been shown in a retrospective study in which younger patients showed a more complete resolution of ME and a better visual outcome when treated with IVB injections.16 Neovascularization is a severe complication in CRVO. Of note, in the present study none of the patients developed neovascular glaucoma when treated with IVB, supporting previous observations suggesting a protective effect of bevacizumab.17,18 However, it is important to consider that by injecting anti-VEGF we are changing the natural course of the disease. This may in turn delay the onset of neovascular complications, especially when anti-VEGF is withdrawn. The reduction in CRT seems to be somewhat slower when applying treatment every 6 weeks compared with the monthly treatments initially given in the CRUISE study.8 The vitreous half-life of bevacizumab and ranibizumab in humans is approximately 10 days, suggesting that their therapeutic duration should not differ much.19,20 Some ME may have returned by week 6 after IVB, requiring further injections to remove the recurrent ME. Nevertheless, given the good overall anatomic and visual outcomes at week 48, a dosing regimen of IVB every 6 weeks seems adequate. Also, this would have the advantage of limiting the number of injections and follow-up visits. In conclusion, the present study shows that IVB injections given every 6 weeks for 48 weeks significantly improve vision and reduce ME. Early treatment is important in these patients to optimize the visual outcome. Further studies are needed to evaluate the long-term outcome of IVB injections in CRVO and whether the need for reinjections can be reduced over time.
References 1. Mitchell P, Smith W, Chang A. Prevalence and associations of retinal vein occlusion in Australia: the Blue Mountains Eye Study. Arch Ophthalmol 1996;114:1243–7. 2. Klein R, Klein BE, Moss SE, Meuer SM. The epidemiology of retinal vein occlusion: the Beaver Dam Eye Study. Trans Am Ophthalmol Soc 2000;98:133– 41. 3. Rogers S, McIntosh RL, Cheung N, et al. International Eye Disease Consortium. The prevalence of retinal vein occlusion: pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology 2010;117:313–19.
Epstein et al 䡠 Bevacizumab for ME in CRVO 4. McIntosh RL, Rogers SL, Lim L, et al. Natural history of central retinal vein occlusion: an evidence-based systematic review. Ophthalmology 2010;117:1113–23. 5. Ehlken C, Rennel ES, Michels D, et al. Levels of VEGF but not VEGF(165b) are increased in the vitreous of patients with retinal vein occlusion. Am J Ophthalmol 2011;152:298 –303. 6. Aiello LP, Avery RL, Arrigg PG, et al. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med 1994;331:1480 –7. 7. Weis SM, Cheresh DA. Pathophysiological consequences of VEGF-induced vascular permeability. Nature 2005;437:497– 504. 8. Brown DM, Campochiaro PA, Singh RP, et al. CRUISE Investigators. Ranibizumab for macular edema following central retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology 2010;117:1124 –33. 9. Campochiaro PA, Brown DM, Awh CC, et al. Sustained benefits from ranibizumab for macular edema following central retinal vein occlusion: twelve-month outcomes of a phase III study. Ophthalmology 2011;118:2041–9. 10. Epstein D, Algvere PV, von Wendt G, et al. Bevacizumab for macular edema in central retinal vein occlusion: a prospective randomized double masked clinical study. Ophthalmology 2012;119:1184 –9. 11. Haller JA, Bandello F, Belfort R Jr, et al. OZURDEX GENEVA Study Group. Randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with macular edema due to retinal vein occlusion. Ophthalmology 2010;117:1134 – 46. 12. Yeh WS, Haller JA, Lanzetta P, et al. Effect of the duration of macular edema on clinical outcomes in retinal vein occlusion
13.
14.
15.
16.
17.
18.
19.
20.
treated with dexamethasone intravitreal implant. Ophthalmology 2012;119:1190 – 8. Ota M, Tsujikawa A, Ojima Y, et al. Retinal sensitivity after resolution of the macular edema associated with retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol 2012;250:635– 44. Lardenoye CW, Probst K, DeLint PJ, Rothova A. Photoreceptor function in eyes with macular edema. Invest Ophthalmol Vis Sci 2000;41:4048 –53. Klein R, Moss SE, Meuer SM, Klein BE. The 15-year cumulative incidence of retinal vein occlusion: the Beaver Dam Eye Study. Arch Ophthalmol 2008;126:513– 8. Ach T, Hoeh AE, Schaal KB, et al. Predictive factors for changes in macular edema in intravitreal bevacizumab therapy of retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol 2010;248:155–9. Beutel J, Peters S, Lüke M, et al. Bevacizumab Study Group. Bevacizumab as adjuvant for neovascular glaucoma. Acta Ophthalmol 2010;88:103–9. Wakabayashi T, Oshima Y, Sakaguchi H, et al. Intravitreal bevacizumab to treat iris neovascularization and neovascular glaucoma secondary to ischemic retinal diseases in 41 consecutive cases. Ophthalmology 2008;115:1571– 80. Krohne TU, Eter N, Holz FG, Meyer CH. Intraocular pharmacokinetics of bevacizumab after a single intravitreal injection in humans. Am J Ophthalmol 2008;146:508 –12. Lucentis (ranibizumab) [package insert]. New South Wales; Novartis Australia; 2011:3. Available at: http://www.novartis.com. au/PI_PDF/luc.pdf. Accessed May 23 2012.
Footnotes and Financial Disclosures Originally received: March 12, 2012. Final revision: June 11, 2012. Accepted: June 25, 2012. Available online: August 17, 2012.
Financial Disclosure(s): The author(s) have made the following disclosure(s): D.L.E. is a consultant for Allergan and Novartis; A.K. is a consultant for Alcon, Allergan, and Bayer. Manuscript no. 2012-362.
Department of Ophthalmology, Karolinska Institutet, St. Eriks Eye Hospital, Stockholm, Sweden.
Correspondence: David L. Epstein, MD, Department of Ophthalmology, Karolinska Institutet, St. Eriks Eye Hospital, Polhemsgatan 50, Stockholm 11282 Sweden. E-mail: [email protected].
2591
Central retinal vein occlusion (CRVO) is a common retinal vascular disorder and often presents with a severely reduced visual acuity (VA) caused by macular edema (ME).1– 4 Vascular endothelial growth factor A (VEGF-A) is produced in response to retinal ischemia secondary to the increased resistance to venous blood flow. The intravitreal level of VEGF-A in CRVO is higher than in any other ischemic retinal vascular disease and several times higher than the levels found in branch vein occlusion.5,6 Vascular endothelial growth factor A increases vascular permeability by disrupting junctional proteins in the blood vessel wall, causing uncoupling of the endothelial cell– cell junctions.7 By targeting VEGF-A, these changes can be reversed and the ME reduced. Different antibodies that block the vascular permeability effect of VEGF-A have been studied recently. © 2012 by the American Academy of Ophthalmology Published by Elsevier Inc.
In the Treatment of Macular Edema after Central Retinal Vein Occlusion (CRUISE) study, intravitreal ranibizumab (Lucentis, anti-VEGF Fab; Genentech Inc., South San Francisco, CA) improved vision significantly compared with sham, with approximately half of the patients gaining 3 or more Early Treatment Diabetic Retinopathy Study (ETDRS) lines.8 After 6 months, all patients could receive ranibizumab injections if ME was present and best-corrected visual acuity (BCVA) was ⱕ20/40. The visual benefit was sustained after 12 months, and patients originally treated with sham injections showed an improvement in BCVA but significantly less than in the ranibizumab groups.9 The CRUISE study neither answered whether even better visual outcomes would have been achieved by continuing monthly injections through 12 months nor addressed the possible impact of giving the ISSN 0161-6420/12/$–see front matter http://dx.doi.org/10.1016/j.ophtha.2012.06.037
2587
Ophthalmology Volume 119, Number 12, December 2012 group originally designated to sham treatment monthly ranibizumab injections during the final 6 months. In a sham-controlled prospective study, we recently showed that intravitreal bevacizumab (IVB) (Avastin, antiVEGF IgG; Genentech, Inc) significantly improved VA and ME over 6 months.10 After 4 injections every 6 weeks of 1.25 mg bevacizumab in patients with CRVO, mean improvement in BCVA score was 14.1 letters compared with a loss of 2.0 letters in the sham group. After 6 months, all patients were enrolled in the present open-label extension study and received an additional 4 treatments of bevacizumab every 6 weeks for 6 months. The objectives of this study were to investigate whether this regimen can maintain VA and to elucidate the impact of delaying treatment for 6 months in patients with ME secondary to CRVO.
received 4 IVB injections every 6 weeks. The 2 treatment groups were designated bevacizumab/bevacizumab (bz/bz) and sham/bevacizumab (sh/bz). The total follow-up period was 48 weeks. Study patients were masked to the treatment given during the first 24 weeks. Staff performing VA testing, OCT, fundus photography, and follow-up investigators were also masked to treatment group.
Outcome Measures The primary outcome measure was the proportion of patients gaining ⱖ15 ETDRS letters at 12 months. The secondary outcome measures were the change in BCVA, change in foveal thickness as measured by OCT, and number of patients with neovascular glaucoma defined as increased intraocular pressure due to the formation of new vessels in the angle as diagnosed by gonioscopy.
Power Calculation and Statistical Analysis
Patients and Methods We conducted a prospective randomized study on 60 eyes of 60 patients with ME secondary to CRVO. Informed consent was obtained from all patients. From April 2009 to December 2010, the patients were enrolled at St. Eriks Eye Hospital in Stockholm, Sweden. These patients comprised the intent-to-treat population in whom the last observation carried forward method was used for missing data for all efficacy parameters. The study adhered to the guidelines of the Declaration of Helsinki. The protocol was approved by the local ethics committee and the Swedish Medical Products Agency. The study is listed on www.clinicaltrials.gov, under identifier NCT00906685 (accessed May 23, 2012).
Study Protocol The details of the protocol, including eligibility criteria, randomization, and procedures for examination, optical coherence tomography (OCT), and IVB injections, have been described.10 The OCT images were obtained by a certified technician using the Zeiss Cirrus OCT machine (Carl Zeiss Meditec, Inc, Dublin, CA). There were no significant differences between the study groups at baseline (Table 1). Initially, each study participant was randomized 1:1 to IVB injections (study group) or sham injections (control group) every 6 weeks for 6 months. During months 6 to 12, all patients
The primary hypothesis was based on a difference between the treatment groups at 6 months in the proportion of patients achieving the main outcome measure. We assumed that 35% of IVBtreated patients and 5% of sham-treated patients would achieve the primary end point (gain of at least 15 ETDRS letters). With a statistical power of 80% and the level of statistical significance set at P ⬍ 0.05, we estimated that a minimum of 24 patients would be required (MedCalc Software, Mariakerke, Belgium). For statistical analyses, the independent Student t test and the Fisher exact test (to compare differences in distributions between the groups) were used.
Results Sixty eyes of 60 patients were originally assigned randomly to IVB (30 eyes) or sham injections (30 eyes). Demographic and baseline characteristics have been reported.10 Some 93% of the patients completed the study through month 12. Four patients discontinued the study during the open-label extension phase (3 patients in the bz/bz group and 1 patient in the sh/bz group). In the bz/bz group, the patients discontinued the study because of difficulties returning to control visits caused by metastatic lung carcinoma, metastatic prostate carcinoma, and old age. One patient in the sh/bz group had a transient ischemic attack while receiving sham treatment and was therefore not offered IVB injections.
Table 1. Patient Demographics and Baseline Ocular Characteristics Parameter
sh/bz n ⴝ 30
bz/bz n ⴝ 30
All
Age (yrs⫾SD) Gender ratio m:f (%) Time from diagnosis to inclusion (w⫾SD) ⬍90 days n (%) ⬎90 days n (%) BCVA (ETDRS letters⫾SD) BCVA distribution n (%) ⬍34 ⬎34 CRT (m⫾SD) Hypertension n (%) Diabetes mellitus n (%)
70.4⫾10.4 57:43 9.4⫾6.5 22 (73.3) 8 (26.7) 43.9⫾16.0 10 (33.3) 20 (66.7) 729⫾195 16 (53.3) 3 (10)
70.6⫾12.6 63:37 8.3⫾4.8 21 (70) 9 (30) 44.4⫾15.3 9 (30) 21 (70) 712⫾330 13 (43.3) 1 (3.3)
70.5⫾12.6 60:40 8.8⫾5.7 43/60 (71.7) 17/60 (28.3) 44.1⫾15.5 19 (31.7) 41 (58.3) 721⫾269 29 (48.3) 4 (6.7)
BCVA ⫽ best corrected visual acuity; bz/bz ⫽ bevacizumab/bevacizumab; CRT⫽ central retinal thickness; ETDRS ⫽ Early Treatment Diabetic Retinopathy Study; SD ⫽ standard deviation; sh/bz ⫽sham/bevacizumab. P value was nonsignificant between treatment groups for all parameters.
2588
Epstein et al 䡠 Bevacizumab for ME in CRVO
Figure 1. Mean change from baseline best-corrected visual acuity (BCVA) over time to month 12. The difference in BCVA between the treatment groups was statistically significant at week 48 (P ⬍ 0.05). Error bars represent ⫾ standard error of the mean. The last observation carried forward method was used to compute missing data. bz/bz ⫽ bevacizumab/ bevacizumab; ETDRS ⫽ Early Treatment Diabetic Retinopathy Study; sh/bz ⫽ sham/bevacizumab.
Visual Acuity In the bz/bz group, the VA improved by 16.1 letters at week 48 (Fig 1). During weeks 24 to 48, a further gain of 2.0 letters was seen with this fixed injection schedule. In the sh/bz group, VA decreased by 2.0 letters at 24 weeks but improved by 4.6 letters at week 48. The difference in VA between the treatment groups was statistically significant from week 12 and beyond (P ⬍ 0.05). At
Figure 3. Mean change from baseline central retinal thickness (CRT) over time to month 12. The mean decrease in CRT was similar in the study groups (P ⫽ not significant). The CRT decreased from 729 to 325 m in the sham/bevacizumab (sh/bz) group and from 712 to 277m in the bevacizumab/bevacizumab (bz/bz) group. The last observation carried forward method was used to compute missing data. Error bars represent ⫾ standard error of the mean.
the end of follow-up, 18 of 30 patients (60.0%) in the bz/bz group had gained ⱖ15 letters compared with 10 of 30 patients (33.3%) in the control group (P ⬍ 0.05). In both groups, 2 of 30 patients (6.7%) lost ⬎15 ETDRS letters after 48 weeks. Snellen BCVA equivalent of ⱕ20/200 is considered a poor visual outcome. In the bz/bz group, 1 of 30 patients (3.3%) had this outcome compared with 5 of 30 patients (16.7%) in the control group (P ⬎ 0.05). An unplanned retrospective analysis of the impact of age and response to therapy was done. Younger patients had a better visual outcome. Overall, patients aged ⬍70 years improved 14.2 letters (95% confidence interval [CI], 6.3–22.0) compared with 7.4 letters (95% CI, 0.2–14.6) in patients aged ⬎70 years (P ⬎ 0.05) (Fig 2). Patients aged ⬎70 years had a significantly worse outcome when receiving delayed treatment, losing 1.4 letters (95% CI, ⫺9.7 to 8.4) in the sh/bz group compared with a gain of 20.1 letters (95% CI, 13.9 –26.3) in the bz/bz group in patients aged ⬍70 years (P ⬍ 0.003). No significant confounders were found in the older group regarding the baseline characteristics.
Anatomic Outcomes at 48 Weeks
Figure 2. Change in best-corrected visual acuity (BCVA) (Early Treatment Diabetic Retinopathy Study [ETDRS] letters) according to age and treatment group. Patients aged ⬎70 years in the sham/bevacizumab (sh/bz) group had a significantly worse visual outcome, losing 1.4 letters (95% CI, ⫺9.7 to 8.4) compared with a gain in the bevacizumab/bevacizumab (bz/bz) group in patients ⬍70 years of age of 20.1 letters (95% CI, 13.9 –26.3) (P ⬍ 0.003). P ⬎ 0.05 for comparison between all other groups. Error bars represent 95% CI. The last observation carried forward method was used to compute missing data. y.o. ⫽ years of age.
The mean change in central retinal thickness (CRT) was similar in both groups, with a mean decrease of 435 m (bz/bz) and 404 m (sh/bz) at week 48 (P ⬎ 0.05; Fig 3). Assessment of residual ME, defined morphologically and evaluated manually as any residual fluid at 48 weeks, was performed. No ME was found in 25 of 30 patients (83.3%) in the bz/bz group compared with 18 of 30 patients (60%) in the sh/bz group (P ⫽ not significant).
Neovascularization As reported earlier, 5 of 30 patients (16.7%) in the sham group had developed iris rubeosis at week 24.10 The iris neovascularization had regressed in all these patients at 48 weeks. No new cases of iris rubeosis were seen during months 6 to 12 in either group.
2589
Ophthalmology Volume 119, Number 12, December 2012 Ocular Safety and Systemic Adverse Events There were no events of endophthalmitis, retinal tear, or retinal detachment during the 48-week treatment period. No serious nonocular adverse events were reported. However, this prospective study was not powered to provide definite data on safety.
Discussion The present study showed that IVB given on a fixed dosing regimen for 12 months provided a substantial visual benefit in patients with ME secondary to CRVO. We also showed that the visual gain was limited in patients who initially received sham treatment followed by subsequent IVB injections. Our results compare favorably with other studies. In the CRUISE study, patients receiving 0.5 mg ranibizumab gained a mean of 13.9 letters at 12 months after intravitreal injections of ranibizumab every 4 weeks for 6 months and then treated as needed. During the last 6 months when the flexible dosing regimen was used, patients lost 1.0 letters. Fifty percent of the patients showed improved BCVA of at least 3 ETDRS lines.9 In the Global Evaluation of Implantable Dexamethasone in Retinal Vein Occlusion with Macular Edema (GENEVA) study, improvement peaked at day 60 with a BCVA of 9.7 letters.11 Direct comparisons between investigations always should be made with caution, and in these studies disease duration, for example, was different. Patients in the GENEVA study had a longer disease duration with 84% of subjects treated more than 90 days after disease onset versus 30% in our study and 28% in the CRUISE study. Disease duration seems to have a critical impact on visual outcome. In our study, patients in the sh/bz group showed a substantial improvement in BCVA when treated with IVB. However, there remained a significant difference in mean improvement from baseline BCVA at 12 months between the sh/bz and bz/bz groups. Patients with treatment delayed for 6 months gained only 4.6 letters compared with a gain of 16.1 letters in the bz/bz group. This difference could not be attributed to changes in CRT because the mean improvement from baseline CRT was similar in both groups. This is the first study to show that patients with ME in CRVO receiving delayed IVB treatment on a fixed schedule still have a significantly worse visual outcome than patients treated without delay. Recently published data from a subset of patients with CRVO in the GENEVA study showed that at 12 months each monthly increase of ME duration was associated with a decreased likelihood of achieving an improvement of at least 3 ETDRS lines. However, in the subset of patients with CRVO, the data were not statistically significant.12 This could be due to the long mean duration of ME in this subset of patients of approximately 8 months compared with 2 months in the present study. In our study, the likelihood of improving 3 ETDRS lines when therapy was delayed for 6 months was significantly lower (33%) than when treatment was given at study start (60%). Cystoid ME is a severe complication in CRVO, resulting in disorganization of the retinal cellular structure, causing severe damage to the neural cells. It has been shown
2590
that structural abnormality to the photoreceptor layer reducing foveal function and retinal sensitivity may remain after resolved ME secondary to retinal vein occlusion.13 An association has been reported in patients with ME between the reduction of visual pigment density and their visual function.14 This suggests that a long-lasting edema may lead to irreversible macular damage, which cannot be restored solely by resolution of the ME. Therefore, early therapeutic intervention to reduce the ME is important for improving the odds of obtaining a good visual outcome. Age has been reported as a risk factor for retinal vein occlusion.15 In the present study, we found that age also might be a prognostic factor in response to IVB injections. Especially in the group aged ⬎70 years, delayed IVB treatment seemed to have a negative impact on the visual outcome. These results should be interpreted with caution because of the small subgroup size. Similar findings have been shown in a retrospective study in which younger patients showed a more complete resolution of ME and a better visual outcome when treated with IVB injections.16 Neovascularization is a severe complication in CRVO. Of note, in the present study none of the patients developed neovascular glaucoma when treated with IVB, supporting previous observations suggesting a protective effect of bevacizumab.17,18 However, it is important to consider that by injecting anti-VEGF we are changing the natural course of the disease. This may in turn delay the onset of neovascular complications, especially when anti-VEGF is withdrawn. The reduction in CRT seems to be somewhat slower when applying treatment every 6 weeks compared with the monthly treatments initially given in the CRUISE study.8 The vitreous half-life of bevacizumab and ranibizumab in humans is approximately 10 days, suggesting that their therapeutic duration should not differ much.19,20 Some ME may have returned by week 6 after IVB, requiring further injections to remove the recurrent ME. Nevertheless, given the good overall anatomic and visual outcomes at week 48, a dosing regimen of IVB every 6 weeks seems adequate. Also, this would have the advantage of limiting the number of injections and follow-up visits. In conclusion, the present study shows that IVB injections given every 6 weeks for 48 weeks significantly improve vision and reduce ME. Early treatment is important in these patients to optimize the visual outcome. Further studies are needed to evaluate the long-term outcome of IVB injections in CRVO and whether the need for reinjections can be reduced over time.
References 1. Mitchell P, Smith W, Chang A. Prevalence and associations of retinal vein occlusion in Australia: the Blue Mountains Eye Study. Arch Ophthalmol 1996;114:1243–7. 2. Klein R, Klein BE, Moss SE, Meuer SM. The epidemiology of retinal vein occlusion: the Beaver Dam Eye Study. Trans Am Ophthalmol Soc 2000;98:133– 41. 3. Rogers S, McIntosh RL, Cheung N, et al. International Eye Disease Consortium. The prevalence of retinal vein occlusion: pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology 2010;117:313–19.
Epstein et al 䡠 Bevacizumab for ME in CRVO 4. McIntosh RL, Rogers SL, Lim L, et al. Natural history of central retinal vein occlusion: an evidence-based systematic review. Ophthalmology 2010;117:1113–23. 5. Ehlken C, Rennel ES, Michels D, et al. Levels of VEGF but not VEGF(165b) are increased in the vitreous of patients with retinal vein occlusion. Am J Ophthalmol 2011;152:298 –303. 6. Aiello LP, Avery RL, Arrigg PG, et al. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med 1994;331:1480 –7. 7. Weis SM, Cheresh DA. Pathophysiological consequences of VEGF-induced vascular permeability. Nature 2005;437:497– 504. 8. Brown DM, Campochiaro PA, Singh RP, et al. CRUISE Investigators. Ranibizumab for macular edema following central retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology 2010;117:1124 –33. 9. Campochiaro PA, Brown DM, Awh CC, et al. Sustained benefits from ranibizumab for macular edema following central retinal vein occlusion: twelve-month outcomes of a phase III study. Ophthalmology 2011;118:2041–9. 10. Epstein D, Algvere PV, von Wendt G, et al. Bevacizumab for macular edema in central retinal vein occlusion: a prospective randomized double masked clinical study. Ophthalmology 2012;119:1184 –9. 11. Haller JA, Bandello F, Belfort R Jr, et al. OZURDEX GENEVA Study Group. Randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with macular edema due to retinal vein occlusion. Ophthalmology 2010;117:1134 – 46. 12. Yeh WS, Haller JA, Lanzetta P, et al. Effect of the duration of macular edema on clinical outcomes in retinal vein occlusion
13.
14.
15.
16.
17.
18.
19.
20.
treated with dexamethasone intravitreal implant. Ophthalmology 2012;119:1190 – 8. Ota M, Tsujikawa A, Ojima Y, et al. Retinal sensitivity after resolution of the macular edema associated with retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol 2012;250:635– 44. Lardenoye CW, Probst K, DeLint PJ, Rothova A. Photoreceptor function in eyes with macular edema. Invest Ophthalmol Vis Sci 2000;41:4048 –53. Klein R, Moss SE, Meuer SM, Klein BE. The 15-year cumulative incidence of retinal vein occlusion: the Beaver Dam Eye Study. Arch Ophthalmol 2008;126:513– 8. Ach T, Hoeh AE, Schaal KB, et al. Predictive factors for changes in macular edema in intravitreal bevacizumab therapy of retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol 2010;248:155–9. Beutel J, Peters S, Lüke M, et al. Bevacizumab Study Group. Bevacizumab as adjuvant for neovascular glaucoma. Acta Ophthalmol 2010;88:103–9. Wakabayashi T, Oshima Y, Sakaguchi H, et al. Intravitreal bevacizumab to treat iris neovascularization and neovascular glaucoma secondary to ischemic retinal diseases in 41 consecutive cases. Ophthalmology 2008;115:1571– 80. Krohne TU, Eter N, Holz FG, Meyer CH. Intraocular pharmacokinetics of bevacizumab after a single intravitreal injection in humans. Am J Ophthalmol 2008;146:508 –12. Lucentis (ranibizumab) [package insert]. New South Wales; Novartis Australia; 2011:3. Available at: http://www.novartis.com. au/PI_PDF/luc.pdf. Accessed May 23 2012.
Footnotes and Financial Disclosures Originally received: March 12, 2012. Final revision: June 11, 2012. Accepted: June 25, 2012. Available online: August 17, 2012.
Financial Disclosure(s): The author(s) have made the following disclosure(s): D.L.E. is a consultant for Allergan and Novartis; A.K. is a consultant for Alcon, Allergan, and Bayer. Manuscript no. 2012-362.
Department of Ophthalmology, Karolinska Institutet, St. Eriks Eye Hospital, Stockholm, Sweden.
Correspondence: David L. Epstein, MD, Department of Ophthalmology, Karolinska Institutet, St. Eriks Eye Hospital, Polhemsgatan 50, Stockholm 11282 Sweden. E-mail: [email protected].
2591