Article Text

Download PDFPDF
Three-month outcome of intravitreal ziv-aflibercept in eyes with diabetic macular oedema
  1. Ahmad M Mansour1,2,
  2. Chintan Dedhia3,
  3. Jay Chhablani3
  1. 1Department of Ophthalmology, American University of Beirut, Beirut, Lebanon
  2. 2Department of Ophthalmology, Rafik Hariri University Hospital, Beirut, Lebanon
  3. 3Smt. Kanuri Santhamma Centre for Vitreo Retinal Diseases, LV Prasad Eye Institute, Hyderabad, India
  1. Correspondence to Dr Jay Chhablani, Smt. Kanuri Santhamma Centre for Vitreo Retinal Diseases, LV Prasad Eye Institute, Hyderabad, TS 500034, India; jay.chhablani{at}


Purpose We report the 3-month efficacy of monthly intravitreal ziv-aflibercept in patients with diabetic macular oedema (DME).

Methods Prospectively, consecutive patients with DME underwent intravitreal injection of 0.05 ml of compounded ziv-aflibercept (1.25 mg) from March 2015 to November 2015. Monitoring of best-corrected visual acuity (BCVA), intraocular inflammation, cataract progression and retinal structure by spectral domain optical coherence tomography was carried out at baseline, 1 week, 1 month, 2 months and 3 months after 3 monthly injections.

Results A total of 17 eyes (11 right eyes and 6 left eyes) were treated. The participants were divided into 10 Caucasians and 6 Indians, 11 men and 5 women, and had a mean age of 61.5 years. Five eyes were treatment-naïve cases and 12 eyes were treatment non- naïve with last treatment received at least more than 4-month interval. Mean BCVA in log MAR (equivalent Snellen visual acuity) improved from baseline 0.70 (20/100) to 0.49 (20/60) at 1 month, 0.43 (20/50) at 2 months and 0.42 (20/50) at 3 months (p ≤ 0.003). Central macular thickness decreased from mean baseline 517.5 to 388.1 μm at 1 week, 355.4 μm at 1 month, 351.4 μm at 2 months and 322.2 μm at 3 months (p ≤ 0.001).

Conclusions Off-label use of intravitreal ziv-aflibercept improves visual acuity, without detectable ocular toxicity or systemic side effects in DME. It offers a less expensive alternative to the approved intravitreal aflibercept (Eylea), especially in the low/middle-income countries and in countries where Eylea is not available.

  • Retina

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Diabetic macular oedema (DME) refers to the accumulation of fluid in the centre of the retina resulting in substantial progressive visual impairment if left untreated. DME affects nearly 21 million individuals worldwide and is a major cause of visual loss.1 In the past three decades, focal laser photocoagulation was the treatment of choice.2 ,3 The Early Treatment Diabetic Retinopathy Study (ETDRS) was the first study to provide a treatment paradigm in DME using laser therapy to reduce moderate vision loss by approximately 50%.2 ,3 Although prevention of vision loss is important, visual improvement would be preferable.2 Recently, pharmacotherapy allows visual improvement and has replaced laser after the discovery of the role of inflammation, oxidative damage and vascular endothelial growth factor (VEGF) in the pathogenesis of the disease.3 ,4 Corticosteroid, anti-VEGF agents, dexamethasone implant and fluocinolone insert have also been sequentially added to the armamentarium of therapeutic choices.5 ,6 Currently, anti-VEGF agents constitute the preferred initial treatment for DME and include bevacizumab, ranibizumab and aflibercept (VEGF-Trap Eye or Eylea; Regeneron Pharmaceuticals, Tarrytown, New York, USA; United States Food and Drug Administration—approved 29 July 2014).5

As there are no standard protocols or recognised paradigms to follow, the choice of the anti-VEGF is very variable and often subjective relating to the severity of the visual loss suffered by the patient, affordability or availability of the drug, and personal experience of the healthcare provider. The Diabetic Retinopathy Clinical Research (DRCR) findings5 and phase III results of VISTA-DME (Study of Intravitreal Administration of VEGF Trap-Eye in Patients with DME)6 and VIVID-DME (VEGF Trap-Eye in Vision Impairment due to DME) studies7 provided strong evidence for aflibercept as an efficient anti-VEGF therapy in DME, yet its use in the low/middle-income countries has been severely limited due to its high cost and its unavailability in some countries. Hence, we used ziv-aflibercept (Zaltrap, Regeneron Pharmaceuticals), United States Food and Drug Administration-approved (since 9 August 2012) aflibercept for oncology, as intravitreal therapy. We compounded it for use in patients who cannot afford the ophthalmic version of aflibercept (Eylea) on a compassionate basis after lack of toxicity in the laboratory and in case reports8–11 and phase I clinical studies, one involving DME.12 ,13

The current clinical study reports the clinical efficacy of three monthly intravitreal injection of ziv-aflibercept in eyes with DME.


This is a prospective non-randomised two-centre study. This study was approved by the institutional review committee at both centres and adhered to the tenets of the Declaration of Helsinki. Study duration at both centres was March 2015 to November 2015. After detailed explanation of the protocol, all patients signed a formal consent before the initiation of therapy. Inclusion criteria included naïve or previously treated DME with the last injection of VEGF antagonist 4 months or more before enrolment. Exclusion criteria included vitreous haemorrhage, macular ischaemia, macular scar from subretinal fibrosis, corneal scar, infectious conjunctivitis, prior vitreous surgery and inability to commit for long-term follow-up.

Best-corrected visual acuity (BCVA) was assessed by ETDRS R chart (Precision Vision, La Salle, Illinois, USA) in Lebanon and Snellen charts in India. Central macular thickness (CMT, mean thickness in the central 1000-μm diameter area) was done using optical coherence tomography (OCT) 3D OCT-2000 FA plus Topcon (Topcon, Tokyo, Japan) in Lebanon and Cirrus (Carl Zeiss Meditec, Dublin, California, USA) in India. Intravenous fluorescein angiography was performed initially using standard protocol. The same operator (AMM) performed the vision exam, OCT test and intravitreal injection in the Lebanese series while the certified ophthalmology associates participated in the care of the patients in the Indian series. Comprehensive ocular examination was performed at baseline, week 1, month 1, month 2 and month 3. Ziv-aflibercept 0.05 mL (1.25 mg aflibercept) was prepared and injected every 4 weeks according to standard protocols (compounding under sterile conditions with storage at 4°C for 4 weeks).9 Monitoring of intraocular inflammation and cataract progression was carried out initially, 1 week, 1 month, 2 months and 3 months after injections using the slit lamp.

Statistical analyses were done using SPSS V.22 (IBM Corporation, Chicago, Illinois, USA) and Wilcoxon-signed rank test with significance set at p value of 0.05. To measure the mean CMT, measurements from Zeiss Cirrus were converted into Topcon 2000 equivalents using well-established conversion formula.14


Seventeen eyes were treated (10 Caucasians, 6 Indians; 11 men and 5 women; 11 right eyes and 6 left eyes) with a mean age of 61.5 years (range: 49–71 years). The cohort included 5 treatment-naïve eyes and 12 treatment non-naïve eyes. Seven eyes were phakic (five eyes had mild-to-moderate nuclear sclerosis and the remaining two had clear lens) and 10 eyes were pseudophakic. Diabetic retinopathy was classified as mild non-proliferative diabetic retinopathy (NPDR) in four eyes, moderate NPDR in eight eyes and proliferative in five eyes that received prior panretinal photocoagulation and were stable at presentation.15 Systemic hypertension was present in 10 subjects, coronary artery disease in 2 (one with coronary artery bypass), asthma in 2, hyperlipidaemia in 1, obesity in 1 and heavy smoker in 1. Mean duration of diabetes was 13.3 years (range 1–30 years).

Five eyes were treatment-naïve cases and 12 eyes were treatment non-naïve cases with last treatment received by more than 4-month interval. Previous treatment for DME in 12 eyes included intravitreal bevacizumab (8 eyes), intravitreal ranibizumab (1 eye), intravitreal triamcinolone (1 eye), dexamethasone implant (2 eyes), focal laser (2 eyes) and micropulse laser (1 eye). The mean number of prior injections of either bevacizumab or ranibizumab among treated eyes was 3 (range 1–7) with a mean interval from the last injection of either bevacizumab or ranibizumab was 9 months (range 4–24 months).

BCVA (log minimum angle of resolution (MAR)) significantly improved from baseline 0.70 to 0.64 at 1 week, 0.49 at 1 month, 0.43 at 2 months and 0.42 at 3 months (p≤0.003 for all time periods). CMT in microns decreased from mean baseline 517.5 to 388.1 μm at 1 week, 355.4 μm at 1 month, 351.4 μm at 2 months and 322.2 μm at 3 months (p≤0.001 for all time periods). The clinical response started few days after the intravitreal injection (figure 1).

Figure 1

A 63-year-old male with a history of diabetes for last 12 years, presented with left eye vision loss for last 5 months. He had undergone two intravitreal bevacizumab injections (1.25 mg/0.05 mL) for diabetic macular oedema with last injection 4 months back along with panretinal photocoagulation for proliferative diabetic retinopathy 2 months back. Left eye was pseudophakic with best-corrected visual acuity 20/40 with persistent cystic changes as seen on spectral domain optical coherence (SD-OCT) scan (A). Patient underwent three intravitreal ziv-aflibercept injections (1.25 mg/0.05 mL) monthly, which resulted in improvement in visual acuity to 20/20 at 3 months follow-up and gradual decrease in cystic changes as seen on SD-OCT scan at 1 month (B) and 3 months (C) follow-up.

For naïve eyes (5 eyes), BCVA (log MAR) improved from baseline 1.02 to 0.77 at 1 week, 0.66 at 1 month, 0.54 at 2 months and 0.56 at 3 months (table 1). In naïve eyes, CMT in microns decreased from mean baseline 519.2 to 384 μm at 1 week, 343.2 μm at 1 month, 334.0 μm at 2 months and 341.0 μm at 3 months.

Table 1

Central macular thickness (CMT) and best-corrected visual acuity (BCVA) before and at 1, 2 and 3 months after monthly injection of intravitreal ziv-aflibercept

For treatment non-naïve eyes, BCVA (log MAR) improved from baseline 0.57 to 0.64 at 1 week, 0.42 at 1 month, 0.39 at 2 months and 0.37 at 3 months. Two patients did not show at the 3-month follow-up because they felt too weak at that time to travel long distances. There were no signs of intraocular inflammation or change in lens status throughout the study. There were no systemic complications noted during the follow-up period.


In the past decade, VEGF emerged as a key target of DME therapy consisting of monthly intravitreal anti-VEGF medications. Aflibercept, a recombinant fusion protein, is the latest anti-VEGF approved for DME by US Federal Drug Administration based on results of VIVID-DME (phase III sponsored by Bayer in Europe, Japan and Australia with 421 participants) and VISTA (Study of Intravitreal Administration of VEGF Trap-Eye in Patients with DME) (phase III sponsored by Regeneron in USA, Canada and other countries with 461 participants). Both studies demonstrated that aflibercept (2 mg), dosed every 2 months after monthly loading dose for 5 months (gain of 11.1 letters VISTA vs 9.4 letters VIVID), was non-inferior to monthly injections of aflibercept (11.5 letters VISTA vs 11.4 letters VIVID) and significantly better than laser therapy (0.2 letters VISTA vs 0.7 letters VIVID).7 Furthermore, several DRCR network publications confirmed the efficacy of aflibercept in DME.5 ,16

Being structurally identical, aflibercept differs from ziv-aflibercept in terms of osmolarity from different buffers making aflibercept preparation as iso-osmotic solution (300 mOsm/kg) and ziv-aflibercept as an hyperosmotic solution (1000 mOsm/kg from the addition of sucrose). The issue of hyperosmolarity of ziv-aflibercept causing retinal toxicity was previously refuted:9 injecting 0.05 mL of a hyperosmotic solution like ziv-aflibercept (1000 mOsm/kg) will raise the vitreous cavity osmolarity by 4% that is within the physiologic range. Further issues relating to off-label ziv-aflibercept when used in a compounded manner include potential ocular toxicity (cataract, glaucoma, retinal damage), clinical efficacy (short term and long term) assessed by visual gain and CMT, stability profile of compounded drug and finally the cost. There was no clinical evidence of ocular toxicity in the current study throughout the 90-day observation period. Several case reports and small case series dealing with age-related macular degeneration or central retinal vein occlusion8–13 experienced visual gains without change in the electroretinogram recordings. Electroretinogram analysis in 12 subjects with wet age-related macular degeneration revealed no alterations 1 month after a single intravitreal ziv-aflibercept injection.12 The only mention of the use of ziv-aflibercept in DME comes from one phase I study:9 two patients with DME were treated in one eye and examined 1 week postinjection: the first had a decrease in CMT of 64 μm from 508 to 443 μm with visual improvement from 20/800 to 20/100; the second had a decrease in CMT of 352 μm from 720 to 368 μm with visual improvement from 20/800 to 20/200 at 1 week. Naïve cases had comparable outcome with previously treated eyes in the current study.

The dose used was 0.25 mg of ziv-aflibercept and this corresponds to 62.5% of the standard dose of 2 mg aflibercept used in VIVID and VISTA protocols. The present data support the efficacy of the dose currently used. It is not necessary to inject 0.08 mL instead of 0.05 mL to achieve the standard dose of 2 mg, as this extra volume can result in increase in intraocular pressure and jeopardise the already fragile retinal circulation. The best argument comes from the VIEW trials, where 0.5 and 2.0 mg produced quite comparable outcomes,6 ,7 similar to the READ-3 trial that showed no advantage to quadruple the dose of intravitreal ranibizumab.17

The compounded ziv-aflibercept drug was stable for 4 weeks as shown indirectly by good clinical response in the current study and directly by VEGF assays showing no loss of VEGF blockade at 1-month storage.12 The current study was carried out using polycarbonate barrel syringes that have poor binding to protein and excellent safety profile.12

There is a paucity of data on the systemic safety of repeated ziv-aflibercept as its off-label use is very recent, but we assume that it is similar to aflibercept. Subjects with DME are at a higher risk for cardiovascular disease and kidney dysfunction.18 ,19 VIVID and VISTA trials analysed systemic risk factors of repeated intraocular aflibercept injection in 466 subjects with DME at week 100. They found similar adverse effects in 154 subjects in the laser arm (3.2%) and in 307 subjects in the aflibercept arm (2.6%).7 In a meta-analysis of the anti-VEGF agents for patients with DME, assessment of the highest level exposure group (those high-risk patients with DME who received 2 years of monthly treatment) revealed a possible increased risk for death and potentially for cerebrovascular accidents using either ranibizumab or aflibercept with data missing for bevacizumab.20 Besides, ocular side effects of ziv-aflibercept are not known except absence of ocular morbidity in the small number of cases reported so far in different diseases.8–13 In an integrated safety analysis involving hundreds of eyes, the most frequent serious ocular adverse event of aflibercept was cataract (2.4%, 1.0% and 0.3% for the monthly regimen, bimonthly regimen and laser treated) at 100-week follow-up in VIVID and VISTA.7

Currently, bevacizumab is the most cost-effective medication for control of DME21 compared with ranibizumab (40 times less expensive) or aflibercept.18 ,20 ,22 The actual compounded cost is 60 times less for ziv-aflibercept than for aflibercept, if the 4 mL ziv-aflibercept phial is divided into 40 aliquots (4 mL Zaltrap 25 mg/mL phial costs US$536 while single Eylea phial costs US$1200 in Lebanon). In view of the excellent stability profile of the drug and short-term clinical safety, ziv-aflibercept would appear to be even more cost-effective than bevacizumab (100 mg of 4 mL phial costs US$680 which makes 40 aliquots). Medicare-allowable charges in 2014 were US$1961 for aflibercept (2.0 mg/0.05 mL), US$67 for bevacizumab (1.25 mg/0.05 mL) and US$1189 for ranibizumab (0.3 mg/0.05 mL).16 Considering a prior clinical trial reported a median of 17 ranibizumab injections for DME over 5 years (including 13 in the first year), costs might be much reduced if ziv-aflibercept was equally safe and effective.16 Interestingly, ophthalmologic treatment costs did not differ substantially between patients with DME, age-related macular degeneration or retinal vein occlusion due to the fact that all three patient groups were managed within the same therapeutic concept.23 Note that patients with DME were twice as expensive as patients with age-related macular degeneration with the cost excess occurring from non-ophthalmic interventions.19 Hence, the need to be even more cost-sparing in DME because diabetics are younger, missing many working days annually to treat their eyes and their diabetic non-ophthalmic complications leading to financial burden. DME is the most frequent cause of blindness in young and middle-aged adults and therapy for DME may be a very long-term therapy compared with age-related macular degeneration.23

The limitations of the current study include the small number of eyes treated, small sample bias overall, short duration of follow-up and one-armed open-label design. We did not perform any electroretinographic tests to detect retinal toxicity at 3 months.

In conclusion, off-label use of monthly ziv-aflibercept over 3-month period improves visual acuity, without detectable ocular toxicity while offering a much less expensive alternative to the identical molecule aflibercept, especially in the low/middle-income countries and in countries where aflibercept (Eylea) is not available. Further long-term safety and efficacy studies of ziv-aflibercept using multiple injections in DME are warranted.


Dr Maamoun Abdul Fattah (American University of Beirut) performed the statistical analyses.


View Abstract


  • Contributors Design (AMM, JC) and conduct (AMM, JC) of the study; collection (AMM, JC, CD), management (AMM, JC, CD), analysis (AMM) and interpretation of the data (AMM, JC); and preparation (AMM), review (AMM, JC) or approval (AMM, JC, CD) of the manuscript.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Institutional review board.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles

  • At a glance
    Keith Barton James Chodosh Jost Jonas