Aflibercept (Eylea; Regeneron, Tarrytown, NY and Bayer Healthcare, Leverkusen, Germany) is a commonly used anti-vascular endothelial growth factor (VEGF) agent, has higher affinity for VEGF isoforms with greater duration of action, and similar efficacy to ranibizumab or bevacizumab.1 The costs of Food and Drug Administration (FDA) approved drugs such as aflibercept and ranibizumab are high and this compels ophthalmologists to use bevacizumab, a cheaper alternative.2
Ziv-aflibercept (Ziv-aflibercept, Sanofi-Aventis US, LLC, Bridgewater, NJ and Regeneron Pharmaceuticals, Inc, Tarrytown, NY) is an approved drug for treatment of metastatic colorectal carcinoma. It is similar to aflibercept with respect to affinity to bind both VEGF isoforms and placental growth factor.3 The higher osmolarity and potential cytotoxicity were initial concerns, but subsequent studies have shown that ziv-aflibercept has no toxicity on the retina and retinal pigment epithelium. Ever since, many studies have tested the safety and efficacy of ziv-aflibercept in various retinal and choroidal disorders and proven its safety and efficacy.4,5 In addition, the cost of ziv-aflibercept ($138) is also significantly less (6 times) than aflibercept ($833) (Indian Market).
Considering the similar molecular structure, mechanism of action and the differences in cost-effectiveness, a comparative study between aflibercept and ziv-aflibercept is essential. There are no head-to-head comparative studies between aflibercept and ziv-aflibercept. In the present study, we compared the efficacy of 3 monthly doses of aflibercept and ziv-aflibercept for treatment-naïve neovascular age-related macular degeneration (nAMD).
This was a retrospective chart review in patients with newly diagnosed nAMD attending tertiary care center(s) in India from October 2015 to October 2018. Treatment-naïve nAMD, that is, type 1, type 2, or mixed choroidal neovascular membrane (CNVM), was included. Previous institutional review board approval was obtained. The study conformed to all tenets of Declaration of Helsinki. All patients signed a written consent form which included the description of off-label use of ziv-aflibercept. Baseline visual acuity (BCVA) in Snellen and logarithm of minimum angle of resolution (logMAR) were recorded. Fundus photo, fundus fluorescein angiography, indocyanine green angiography, and spectral domain optical coherence tomography were done in all patients at baseline to substantiate the diagnosis of CNVM. All patients received 3 monthly doses of either intravitreal ziv-aflibercept (1.25 mg/0.05 mL) (Group 1) or aflibercept (2 mg/0.05 mL) (Group 2). BCVA was noted at monthly follow-up. On spectral domain optical coherence tomography, central subfoveal thickness (CSFT), presence of subretinal hyperreflective material (SHRM) with SHRM height, pigment epithelial detachment (PED) height, neurosensory detachment (NSD) height were noted at baseline and at 1st, 2nd, and 3rd month follow-up in both groups.
SPSS version 23 (IBM corporation, New York, NY) was used for statistical analysis. Repeated measures analysis of variance with a Greenhouse-Geisser correction was performed to determine the change of BCVA, central macular thickness (CMT), SRHM, NSD, and PED height at baseline, 1, 2, and 3 months within individual groups. Post hoc tests for pair wise comparison were performed using the Bonferroni correction. For categorical data, χ2 test was used. For comparison between the 2 groups, independent t test was employed. Level of significance was set at 0.05 at 95% confidence interval.
Twenty-three eyes of 23 patients were included. There were 14 males and 9 females. Right eye was affected in 14 patients and left in 9 patients. Mean age of patients was 73.69 ± 7.09 years. Mean duration of symptoms of our patients was 3.3 ± 2. 1 months. Five eyes were diagnosed to have type 2 CNVM, whereas remaining 18 eyes had presence of type 1 CNVM.
Twelve eyes received ziv-aflibercept (group 1), whereas 11 eyes received aflibercept (Group 2). Both groups were matched for age (Group 1, 71.81 ± 8.57 years; Group 2, 75.81 ± 5.29 years, P = 0.089). There were 8 females in group 2 and 1 female patient in group 1. Five patients in group 2 had type 2 CNVM and 1 patient in group 1 had type 2 CNVM.
Group 1 (Ziv-aflibercept Group)
Visual Acuity. Baseline BCVA in the ziv-aflibercept group was logMAR 1.006 ± 0.39 [Snellen equivalent (SE) 20/200]. Mean BCVA at first month was 0.67 ± 0.23 (SE 20/80), 0.54 ± 0.21 (SE 20/60) at second month and 0.49 ± 0.26 (SE 20/60) at third month. The change in BCVA from baseline to 1st month was 0.33 ± 0.12 (approximately 3 lines improvement, P = 0.007). From baseline to 2nd month, it was 0.46 ± 0.12 (approximately 4 lines improvement, P = 0.002) and from baseline to 3rd month 0.51 ± 0.12 (approximately 5 lines improvement, P = 0.008). Overall, the change in BCVA at each month was significant compared with baseline (P < 0.001).
Central Subfoveal Thickness
At baseline, CSFT was 409.18 ± 166.89 μm. At first month, it was 307.45 ± 86.55 μm, and 289.36 ± 63.72 μm and 288.27 ± 80.37 μm in the 2nd and 3rd month, respectively. Overall, there was statistically significant reduction in CSFT at 3 months compared with baseline (P = 0.007). Pairwise comparisons of CSFT between baseline and 3rd month were significant (P = 0.04), whereas the difference in CSFT was not significant over follow-ups viz. baseline to 1st month (P = 0.09), baseline to 2nd month (P = 0.08).
At baseline, mean SHRM height was 82.54 ± 63.82 μm. This changed to 95.00 ± 99.43 μm at first month, 110.45 ± 101.77 μm at 2nd month, and then reduced to 90.72 ± 78.94 μm at the end of 3 months. However, there was no significant difference in the change of SHRM height (P = 0.48).
At baseline NSD height was 110.54 ± 168.16 μm. At first month, it reduced to 63.72 ± 87.15 μm, 54.72 ± 73.75 μm at second month and 38.81 ± 44.33 μm at third month. Repeated measures analysis of variance showed no significant difference in NSD height (P = 0.22).
At baseline, PED height was 242.54 ± 190.73 μm. With treatment, PED height at 1, 2, and 3 months’ follow-up was 182.18 ± 137.04, 186.36 ± 159.31, and 211.90 ± 143.32 μm, respectively (P = 0.26).
Group 2 (Aflibercept Group)
In the aflibercept group, there was no significant improvement in BCVA from baseline through 3 months (P = 0.26). Similarly, CSFT (P = 0.43), SHRM height (P = 0.47), NSD height (P = 0.44), and PED height (P = 0.08) also did not show significant differences over monthly follow-up visits compared with baseline values. Table 1 highlights the comparison of parameters between baseline and monthly follow-ups. Table 2 shows the change in parameters over follow-ups compared with baseline.
Comparison Between the Groups
At baseline, group 1 was seen to have worse visual acuity than eyes in group 2 (1.00 ± 0.39 vs 0.52 ± 0.30; P = 0.003) (Table 3). There were no differences in baseline CSFT, SHRM height, NSD height, and PED height between the 2 groups.
At follow-ups, there were no significant differences between the groups with regards to change in BCVA, CSFT, SHRM height, and NSD height at month 3 compared with baseline. Tables 3–6 show the comparison of BCVA and OCT parameters of each group at baseline and 1st, 2nd, and 3rd month follow-ups, respectively. PED height was seen to be lower in eyes at 3rd month follow-up that received aflibercept (Table 6). Representative images are shown as Figures 1 and 2.
The concerns about the ziv-aflibercept such as hyperosmolarity and cytotoxicity have been countered by number of studies published by our group and others.6,7 Although the osmolarity (1000 mOsm/kg) is 3 times higher than that of aflibercept (300 mOsm/kg), the effective intravitreal osmolarity of ziv-aflibercept is 312 mOSm/ kg (toxicity seen at ≥500 mOsm/kg).8 Singh et al9 have shown that ocular and systemic adverse effects are infrequent in a study involving 1704 eyes of 1562 patients (total of 5914 injections). de Oliveira Dias et al studied the 6- month safety and efficacy of ziv-aflibercept. They noted significant improvements in BCVA, central retinal thickness, and mean macular volume. The multifocal electroretinography in the central 15 degrees showed increased average P1 amplitudes as well. No retinal toxicity on full-field electroretinography was noted in their study.10
Three initial monthly doses are usual treatment protocols regardless of dosing schedules for treatment-naïve nAMD in many randomized controlled trials. In the real-world setting, the further treatments after the 3 loading doses are pro-re-nata/ treat and extend/8 weekly (q 8) weeks depending on multiple factors like affordability and compliance to follow ups. Hence in this retrospective study, we compared the efficacy of 3 loading doses of aflibercept versus ziv-aflibercept in treatment of naïve eyes, considering the different dosing schedules for retreatments after 3 months.
Even the prospect of 3, monthly doses of aflibercept can seem prohibitive in terms of cost in low- and middle-income countries. The annual Medicare part B spent $1.08 billion in 2013 and $1.81 billion in 2015 for aflibercept. This was a result of spending $9934 per beneficiary per year for aflibercept injections. Similarly, $9719 was spent annually per beneficiary for ranibizumab injections from 2011 to 2015.11 Such costs cause a significant burden on the economy. In countries without a centralized medical insurance, it necessitates out-of-pocket payment by patients receiving anti-VEGF treatment and thus results in poor compliance to follow ups and retreatments. Another analysis on costs for bevacizumab and ranibizumab has estimated that changing from ranibizumab to bevacizumab can potentially eliminate 18 billion dollars from Medicare Part B budget.12 Hence, switching to lower-cost drugs can significantly impact utilization of anti VEGF therapies and better healthcare for age–related macular degeneration. Both bevacizumab and ziv-aflibercept cost $30 to $60 per injection, making them affordable.13 Considering, ziv-aflibercept has a similar mechanism of action as aflibercept and may provide longer duration of action and efficacy, 3 loading doses of this cheaper alternative to aflibercept is cost-effective. From our study, we learnt that ziv-aflibercept performed better than aflibercept after 3 loading doses. This reflected in improvement of BCVA, and change in BCVA from baseline at first, second, and third month. Even CSFT showed significant reduction at the end of 3 months and there was a significant change in CSFT at 3 months from baseline. However, at 3 months, SHRM, NSD, and PED height did not improve significantly in either group.
To evaluate outcomes of ziv-aflibercept at 3 months, Mansour et al studied 30 patients including 11 treatment-naïve and 19 treatment-resistant cases, and noted improvements in CMT and BCVA in both naïve and treated cases. Treatment-naïve eyes showed a better response in this study.14 The same group studied 2-year outcomes of ziv-aflibercept pro-re-nata dosing and found that in 48 eyes with nAMD, there was statistically significant change in BCVA at 1 year, 1.5 years, and 2 years. The mean number of injections at 1 year, between 12 and 18 months, and between 18 and 24 months were 9.2, 2.3, and 1.7, respectively.15
The dosages of aflibercept and ziv-aflibercept are different (2 mg/0.05 mL vs 1.25 mg/0.05 mL) raising concerns about comparing their efficacies. Baghi et al compared the efficacy of 2 doses of intravitreal ziv-aflibercept (1.25 mg/0.05 mL and 2.5 mg/ mL) with intravitreal bevacizumab (1.25 mg/ mL) in 123 treatment-naïve diabetic macular edema (DME) eyes at 12 weeks. Both 1.25 mg and 2.5 mg doses of ziv-aflibercept and bevacizumab demonstrated significant BCVA improvement over baseline in the treatment of center-involved DME. Both doses of intravitreal zev-aflibercept were equally efficacious as demonstrated by the VIEW trial. However, ziv-aflibercept was superior to intravitreal zev-aflibercept in terms of BCVA and CSFT reduction in eyes with baseline BCVA <20/50.16
Providência et al17 studied the effect of intravitreal aflibercept and bevacizumab using a fixed dosing schedule in 72 eyes with naïve nAMD. All eyes received 3 loading doses followed by bimonthly injections up to 18 months. Ophthalmological examination was done at 3, 6, 12, and 18 months. In their study, they observed that there was no significant difference in letters gained between the groups after 3 loading doses. However, the CSFT reduction was significant.17 Our study compared the results of 3 loading doses of 2 similar molecules and saw that there were significant improvements in BCVA and CSFT in the ziv-aflibercept group and but not in the aflibercept group.
To the best of our knowledge, a study comparing outcomes of loading doses for 2 similar molecules has not been done. The limitations of our study include a small sample size, lack of randomization and double blinding, and the retrospective nature of the study.
In our study, the ziv-aflibercept group showed significant improvements in BCVA and CSFT at 3 months. The aflibercept group did not show significant improvements. This could be because of the worse visual acuity in ziv-aflibercept group at baseline. However, the efficacy of ziv-aflibercept in reduction of CSFT was clearly shown. The anatomical and visual improvements seen with the ziv-aflibercept group along with the affordability make it a viable and cost-effective option for the crucial 3 initial loading doses in nAMD.
1. Trichonas G, Kaiser PK. Aflibercept for the treatment of age-related macular degeneration. Ophthalmol Ther
2. van Asten F, Michels CTJ, Hoyng CB, et al. The cost-effectiveness of bevacizumab, ranibizumab and aflibercept for the treatment of age-related macular degeneration—a cost-effectiveness analysis from a societal perspective. PLoS One
3. de Oliveira Dias JR, de Andrade GC, Novais EA, Farah ME, Rodrigues EB. Fusion proteins for treatment of retinal diseases: aflibercept, ziv-aflibercept
,;1; and conbercept. Int J Retina Vitreous
4. Mansour AM, Ashraf M, Dedhia CJ, Charbaji A, Souka AAR, Chhablani J. Long-term safety and efficacy of ziv-aflibercept
in retinal diseases. Br J Ophthalmol
5. Safety of intravitreal ziv-aflibercept
in choroido-retinal vascular diseases: a randomised double-blind intervention study. - PubMed - NCBI [Internet]. [cited 2019 Nov 9]. Available from: https://www.ncbi.nlm.nih.gov/pubmed/?term=Safety+of+intravitreal+ziv-aflibercept+in+choroido-retinal+vascular+diseases%3A+A+randomised+double-blind+intervention+study
6. Chhablani J, Narayanan R, Mathai A, Yogi R, Stewart M. Short-term safety profile of intravitreal ziv-aflibercept
7. Mansour AM, Al-Ghadban SI, Yunis MH, El-Sabban ME. Ziv-aflibercept
in macular disease. Br J Ophthalmol
8. Chhablani J. Intravitreal ziv-aflibercept
for recurrent macular edema secondary to central retinal venous occlusion. Indian J Ophthalmol
9. Singh SR, Stewart MW, Chattannavar G, et al. Safety of 5914 intravitreal ziv-aflibercept
injections. Br J Ophthalmol
10. de Oliveira Dias JR, de Andrade GC, Kniggendorf VF, et al. Clinical and electrophysiological evaluation after intravitreal ziv-aflibercept
for exudative age-related macular degeneration. Retina
11. Patel S. Medicare spending on anti-vascular endothelial growth factor medications. Ophthalmol Retina
12. Hutton D, Newman-Casey PA, Tavag M, Zacks D, Stein J. Switching to less expensive blindness drug could save medicare part B $18 billion over a ten-year period. Health Aff (Millwood)
13. Singh SR, Dogra A, Stewart M, Das T, Chhablani J. Intravitreal ziv-aflibercept
: clinical effects and economic impact. Asia Pac J Ophthalmol (Phila)
14. Mansour AM, Chhablani J, Antonios RS, et al. Three-month outcome of ziv-aflibercept
for exudative age-related macular degeneration. Br J Ophthalmol
15. Mansour AM, Ashraf M, Charbaji A, et al. Two-year outcomes of intravitreal ziv-aflibercept
. Br J Ophthalmol
16. Baghi A, Jabbarpoor Bonyadi MH, Ramezani A, et al. Two doses of intravitreal ziv-aflibercept
versus bevacizumab in treatment of diabetic macular edema: a three-armed, double-blind randomized trial. Ophthalmol Retina
17. Providência J, Rodrigues TM, Oliveira M, et al. Real-world results of aflibercept versus ranibizumab for the treatment of exudative AMD using a fixed regimen. Biomed Res Int