Antiangiogenic therapy with anti-vascular endothelial growth factor modalities for neovascular age-related macular degeneration

Abstract

Background

Age-related macular degeneration (AMD) is a common cause of severe vision loss in people 55 years and older.

Objectives

The objective of this review was to investigate the effects of anti-VEGF (vascular endothelial growth factor) modalities for treating neovascular AMD.

Search strategy

We searched CENTRAL, MEDLINE, EMBASE and LILACS. We handsearched ARVO abstracts for 2006, 2007 for ongoing trials.

Selection criteria

We included randomized controlled trials (RCTs).

Data collection and analysis

Two review authors independently extracted data. We contacted trial authors for additional data. We summarized outcomes as relative risks (RR), number needed to treat (NNT) and weighted mean differences.

Main results

We included five RCTs of good methodological quality. All five trials were conducted by pharmaceutical companies. An intention-to-treat analysis using the last observation carried forward method was done in most trials.

Two trials compared pegaptanib versus sham. One trial compared ranibizumab versus sham, another compared ranibizumab/sham verteporfin PDT versus verteporfin PDT/sham ranibizumab, and the final trial compared ranibizumab plus verteporfin PDT versus verteporfin PDT alone.

Fewer patients treated with pegaptanib lost 15 or more letters of visual acuity at one year follow-up compared to sham (pooled relative risk (RR) 0.71; 95% confidence interval (CI) 0.61 to 0.84). The NNT was 6.67 (95% CI 4.35 to 14.28) for 0.3 mg pegaptanib, 6.25 (95% CI 4.17 to 12.5) for 1 mg pegaptanib and 14.28 (95% CI 6.67 to 100) for 3 mg pegaptanib. In a trial of ranibizumab versus sham, RR for loss of 15 or more letters visual acuity at one year was 0.14 (95% CI 0.1 to 0.22) in favour of ranibizumab. The NNT was 3.13 (95% CI 2.56 to 3.84) for 0.3 mg ranibizumab and 3.13 (95% CI 2.56 to 3.84) for 0.5 mg ranibizumab. In a trial of ranibizumab versus verteporfin PDT, RR for loss of 15 or more letters at one year was 0.13 (95% CI 0.07 to 0.23) favouring ranibizumab. The NNT was 3.33 (95% CI 2.56 to 4.76) for 0.3 mg ranibizumab and 3.12 (95% CI 2.43 to 4.17) for 0.5 mg ranibizumab. In another trial of combined ranibizumab plus verteporfin PDT versus verteporfin PDT, RR for loss of 15 or more letters at one year favoured combined therapy (RR 0.3 (95% CI 0.15 to 0.60). The NNT was 4.35 (95% CI 2.78 to 11.11).

Pooled RR for gain of 15 or more letters visual acuity at one year was 5.81 (95% CI 3.29 to 10.26) for ranibizumab versus sham, 6.79 (95% CI 3.41 to 13.54) for ranibizumab/sham verteporfin PDT versus verteporfin PDT/sham ranibizumab, and 4.44 (95% CI 1.40 to 14.08) for ranibizumab plus verteporfin PDT versus verteporfin PDT.

Frequency of endophthalmitis in included studies was between 0.7% to 4.7% with ranibizumab and 1.3% with pegaptanib. Improvement in vision-specific quality of life was reported for both treatments.

Authors' conclusions

Pegaptanib and ranibizumab reduce the risk of visual acuity loss in patients with neovascular AMD. Ranibizumab causes gains in visual acuity in many eyes. Quality of life and cost will be important for treatment decisions. Other agents blocking VEGF are being tested in ongoing trials.

Background

Introduction

Age-related macular degeneration (AMD) is a progressive and de-generative disease of the retina that occurs with increasing frequency with age. There are two major types of AMD: non-neovascular, or atrophic, and neovascular AMD. The non-neovascular type is characterized by drusen (yellow spots under the retina), pigmentary changes (re-distribution of melanin within the retinal pigment epithelial (RPE) cells under the retina and migration of melanin into the retina), and geographic atrophy (loss of the RPE and choriocapillaris).

This review is concerned with neovascular AMD. The hallmark of this type of AMD is choroidal neovascularization (CNV). Choroidal neovascularization is the process by which a vascular membrane, which originates in the choroid, grows under and through the RPE and Bruch's membrane to spread beneath the retina. These vessels may leak and bleed causing exudative or hemorrhagic retinal detachments, or both. Without treatment, the process usually evolves into a fibrous scar, which replaces the outer layers of the retina, the RPE and the choriocapillaris. The scarred retina has a greatly diminished visual function.

Epidemiology

Age-related macular degeneration is the leading cause of irreversible vision loss in the elderly (Bunce 2006; Congdon 2004; Ghafour 1983; Hyman 1987; Leibowitz 1980; Tielsch 1994). While the non-neovascular type is much more common, it is the neovascular type that is responsible for the most serious loss of vision. Neovascular AMD occurs in only 10% of people with AMD, yet 80% of those with severe visual loss (less than 20/200 Snellen acuity) have the neovascular form (Leibowitz 1980). Once neovascular disease develops in one eye, the risk of developing neovascular disease in the other eye in the same person is approximately 40% (AREDS 2001; SST 20).

Age-related macular degeneration (AMD) increases in prevalence from 0% among people younger than 55 years old to 18.5% among those 85 years or older (O'Shea 1998). In the Beaver Dam Study, the prevalence of neovascular AMD was 1.2% but also increased with age (Klein 2002). The prevalence data are similar in other large population studies from the United States, Australia, and the Netherlands (Leibowitz 1980; Mitchell 1995; Vingerling 1995). In the Beaver Dam Eye Study, the 10-year incidence of neovascular AMD was 1.4% in those 43 to 86 years of age but increased to 4.1% in people over 75 years of age (Klein 2002). These findings were slightly lower in the Blue Mountain Eye Study (Mitchell 1995).

There is a similar prevalence of AMD among whites, blacks, and hispanics but whites have a higher prevalence of late AMD (Congdon 2004; Klein 2006; Friedman 1999). Past research has supported the notion that family history is a risk factor for AMD (Piguet 1993; Seddon 1997; Silvestri 1994) and more recent studies have identified genes (e.g. complement factor H (CFH), hemicentin-1, HTRA1/10q26, VEGF toll-like receptor 4, ApoE) that have been found to be associated with AMD (Haddad 2006). However with the exception of CFH, these associations have been inconsistent.

There are no studies to consistently show that modifiable factors such as lipid levels, light exposure, or alcohol intake put people at greater risk of developing AMD. One notable exception is smoking (Klein 2008; Mitchell 2002; Smith 1996). Elevated baseline levels of inflammatory biomarkers such as C reactive protein have been found to be associated with the development of early and late AMD in a large population-based cohort (Boekhoorn 2007). Furthermore, several studies have shown gene-environment interactions of CFH with smoking and C-reactive protein (Deangelis 2007; Haddad 2006; Schaumberg 2007; Seddon 2006).

High doses of vitamins C and E, beta-carotene, and zinc may provide a modest protective effect against the progression of AMD in patients with extensive drusen or fellow eyes with neovascular AMD (AREDS 2001). With the aging population, a higher prevalence of this blinding disease is expected in the future at least in the Western populations.

Presentation and diagnosis

Neovascular AMD may affect one eye or both eyes at the same time or sequentially. The symptoms of AMD are metamorphopsia (distortions while looking at objects), scotomata (missing spots) and blurry vision. Individuals with AMD may be unaware of the change in their visual acuity or may note difficulty with normal activities such as reading and writing, watching television, driving and recognizing faces. If AMD affects only one eye, visual loss may go undetected until monocular testing at routine eye examinations or by chance occlusion of the better eye. Frequently, people are unaware that their disturbed binocular vision is caused by changes in only one eye.

The diagnosis of AMD is made clinically and with the help of imaging such as angiography. At the onset of symptoms, fundus examination often reveals subretinal exudation of fluid, lipid, or blood. The area of the lesion includes area with blood, scar or atrophy and neovascularization. Fluorescein angiography may be necessary to detect subtle exudation in some patients with a recent change in visual acuity. Choroidal neovascularization has several characteristic patterns on fluorescein angiography. Classic CNV is defined as a well-demarcated area of early hyperfluorescence and increasing fluorescein leakage on late frames of the angiogram (MPSG 1991). Occult CNV occurs in two different patterns: fibrovascular pigment epithelial detachment (PED) and late leakage from an undetermined source.

Two other tests may aid in studying patients with neovascular AMD: indocyanine green angiography images the choroidal circulation better than fluorescein angiography and may show ‘hot’ spots under the RPE that are amenable to treatment. Optical coherence tomography (OCT), a non-invasive imaging modality, shows cross-sectional images of the retina, RPE, and choroid. Some studies have defined the characteristic appearance of the different stages of the disease process on OCT (Ting 2002; Van Kerckhoven 2001).

Treatment options

Treatment options are based on angiographic appearance, whether the AMD includes classic or occult lesions. Treatments most frequently used for neovascular AMD during the past 20 years included thermal laser photocoagulation and verteporfin photodynamic therapy (PDT). However, laser photocoagulation used to obliterate neovascularization causes full thickness retinal burns and, in the case of subfoveal lesions, leads to immediate visual acuity loss. Lesions that are treated with thermal laser need to be well-demarcated and should be in extrafoveal or juxtafoveal locations to avoid producing a central scotoma. Only a small percentage of patients with exudative neovascular AMD fit the criteria for argon laser treatment and those that can undergo laser treatment experience a 50% recurrence rate (MPSG 1991). A Cochrane systematic review concluded that laser photocoagulation effectively slowed progression of neovascularization in non-subfoveal lesions compared with observation alone (Virgili 2007).

Photodynamic therapy with verteporfin was designed to treat CNV without damaging the overlying retina and to treat CNV with indistinct borders. This more selective treatment is advantageous for subfoveal lesions. In verteporfin PDT, a photosensitizing drug is injected intravenously and preferentially adheres to the CNV. Focusing a non-thermal laser light on the patient's CNV activates the dye. The highlighted CNV vessels are preferentially occluded when treated with the appropriate laser dose with little or no persistent short-term damage to the adjacent choroid and retina. Photodynamic therapy with verteporfin (Visudyne) has been approved by the Food and Drugs Administration (FDA) in the United States to treat predominantly classic subfoveal CNV. However, only approximately one third of all patients with neovascular AMD have classic CNV. More than 90% of these patients require retreatment after three months and most need multiple treatments during the first year (TAP 1999). In the United States, Centers for Medicare and Medicaid Services reimburses for treatment with verteporfin PDT of minimally classic and occult lesions if they are small (less or equal to 4 standard disc areas in size) based on a subgroup analysis (Blinder 2003). A Cochrane review of verteporfin PDT has been published in The Cochrane Library (Wormald 2007).

Anti-angiogenic therapy is the latest approach to the treatment of neovascular AMD. This treatment approach aims to disrupt neovascularization and to prevent further neovascularization rather than destroy it. Angiogenesis is a complex process that results in new blood vessel formation. This process requires interactions between different factors that can be either stimulatory or inhibitory. These factors have been identified in CNV formation in animal models and human tissue (Aiello 1994; Kvanta 1996; Lopez 1996). Anti-angiogenic treatments work by either blocking stimulatory factors or promoting the inhibitory ones. One of the potential anti-angiogenic treatments is anti-vascular endothelial growth factor (anti-VEGF), a secreted polypeptide with mitogenic effects on the endothelial blood vessels. Vascular endothelial growth factor antagonists have been shown to inhibit CNV in animal models. An example of an anti-VEGF antagonist is pegaptanib (Macugen, Genentech). Pegaptanib is a chemically synthesized 28-base ribonucleic acid molecule. It is an aptamer and has a capability to change its three dimensional structure to fit a target protein, in this case VEGF. By binding to VEGF, pegaptanib blocks VEGF and inactivates its action. Thus, the process of neovascularization is halted. Ranibizumab previously known as rhuFab-VEGF (Lucentis (R), a trademark of Genentech, Inc.) is another example of an anti-VEGF medication developed for ocular administration. It is a humanized antibody fragment capable of binding to VEGF protein, preventing it from binding to its receptor, thus inhibiting angiogenic activity. Bevacizumab is another anti-VEGF agent used to treat CNV. Bevacizumab (Avastin (R), a trademark of Genentech, Inc.) is a humanized monoclonal antibody against VEGF. It is the larger parent molecule from which ranibizumab is also derived. Bevacizumab is currently indicated for treatment of other conditions such as colorectal cancer but it is also used by clinicians for treatment of CNV as an off-label use. Pegaptanib was approved by the FDA in the United States in December 2004 and ranibizumab in 2007. Pegaptanib is marketed by Pfizer and both bevacizumab and ranibizumab were developed by Genentech. Anti-VEGF agents currently are administered via periodic intravitreal injections. Anti-angiogenesis therapy modalities provide a promising means of treating the potentially devastating problem of AMD. In the remaining portion of the review, we will refer to the medications by their generic names.

Objectives

The aim of this review was to investigate the effects of, and quality of life associated with, anti-angiogenic therapy with anti-VEGF modalities for the treatment of neovascular AMD.

Methods

Criteria for considering studies for this review

Types of studies

We included randomized clinical trials only.

Types of participants

We included trials in which participants were people with neovascular AMD as defined by study investigators.

Types of interventions

We included studies in which anti-VEGF treatment was compared to another treatment, sham treatment, or no treatment.

Types of outcome measures

Primary outcomes

The primary outcome for this review was best corrected visual acuity after at least one year of follow up.

Secondary outcomes

The following secondary outcomes were considered:

1. Contrast sensitivity, reading speed or any other validated measure of visual function as available in the studies.

2. Assessment of CNV by fluorescein angiography or OCT. This may include resolution of subretinal or intraretinal fluid by OCT evaluation.

3. Any ocular/systemic adverse outcomes.

4. Quality of life measures - as assessed with any validated measurement scales.

5. Economic data such as comparative cost analyses.

Follow up

We included trials in which participants were followed for at least one year.

Search methods for identification of studies

Electronic searches

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (which contains the Cochrane Eyes and Vision Group Trials Register) (The Cochrane Library Issue 1, 2008), MEDLINE (January 1966 to February 2008), EMBASE (January 1980 to February 2008) and the Latin American and Caribbean Literature on Health Sciences (LILACS) (1982 to February 2008). There were no date or language restrictions. The databases were last searched on 1 February 2008.

See: Appendices for details of search strategies for CENTRAL, MEDLINE, EMBASE and LILACS.

Searching other resources

We contacted pharmaceutical companies conducting studies on anti-VEGF drugs for any ongoing or completed clinical trials not previously published. One author (SSV) handsearched abstracts from the annual meetings of the Association for Research in Vision & Ophthalmology for the years 2006 and 2007 for ongoing trials (http://files.abstractsonline.com/SUPT/163/1807/PresentationTitle.htm; http://files.abstractsonline.com/SUPT/163/1601/Presentation_Title_PDF_wlinks.htm accessed November 24, 2007). For a future update of this review, we will search abstracts for the following conferences: Macula Society 2006, 2007; Retina Society 2006; 2007; subspecialty meeting proceedings preceding the American Academy of Ophthalmology meetings 2006, 2007.

Data collection and analysis

Selection of studies

Two review authors independently evaluated the titles and abstracts resulting from the electronic searches. We assessed the full-text articles for those that appeared to satisfy the inclusion criteria. We selected trials that fulfilled the stated criteria. We contacted authors to clarify any details necessary to make a complete assessment of the relevance of the study.

Data extraction and management

We extracted study characteristics including details of participants, interventions, funding resources using paper data collection forms developed specifically for this purpose. We contacted the trial authors for data on primary and secondary outcomes in the individual trials when the information was not clearly available from published reports. We extracted data on visual acuity for different angiographic sub-types for the two trials forming part of the VISION study from graphs in documents available on the FDA website (http://www.fda.gov/ohrms/dockets/ac/04/slides/2004-4053S1_01_EyeTech-Main.ppt, accessed September 21, 2005). Documents available online on the FDA and European Medicines Agency websites were also utilized for data extraction for the ANCHOR and MARINA trials (http://www.fda.gov/cder/foi/nda/2006/125156s0000_LucentisTOC.htm, accessed June 20, 2007; http://www.emea.europa.eu/humandocs/Humans/EPAR/lucentis/lucentis.htm, accessed February 14, 2008). We also extracted data from figures published in the trial reports and communicated with the authors to verify the same.

Assessment of risk of bias in included studies

Two review authors assessed potential sources of systematic bias in trials according to methods set out in Section 6 of the Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 (Higgins 2006). The following parameters were considered: method of allocation concealment (selection bias), masking of participants and researchers (performance bias), masking of outcome assessment (detection bias), rates of follow up and compliance as well as analysis of all patients after randomization (attrition bias). Allocation concealment was graded as (A) Adequate or Yes, (B) Unclear or Not Reported, or (C) Inadequate or No. We contacted authors of trials categorized as B for additional information.

Data synthesis

Data analysis was guided by Section 8 of the Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 (Deeks 2006). The primary outcome for this review was visual acuity in the treated eye. We analyzed visual acuity as both a dichotomous and continuous outcome. We calculated RR of loss of 15 letters or more (same as loss of 3 or more lines) of best-corrected visual acuity at one year of follow-up and the weighted mean difference (WMD) of mean visual acuity at one year of follow-up. Where data were available, we also calculated RRs of maintenance or gain in 0 lines of visual acuity, loss of 30 letters or more (same as loss of 6 lines or more) of visual acuity and of blindness (defined as visual acuity worse than 20/200) at one year of follow-up. We also calculated the number needed to treat for loss of 15 or more letters visual acuity. We summarized other continuous outcomes as WMDs.

Statistical analyses were performed using RevMan 4.2. We assessed heterogeneity based on the chi-square test, I-square value and overlap of confidence intervals in the forest plots. We reported a narrative summary if we found substantial heterogeneity. We used a fixed-effects model where there was no heterogeneity. We summarized the outcomes stratified for different doses of administration as stated in our protocol and also report a pooled outcome for the intervention. Where possible, we also conducted a sub-group analysis for the primary outcome stratifying the data according to the angiographic sub-type of CNV using the definitions adopted in the included trials.

Sensitivity analysis

We conducted a sensitivity analysis to examine any systematic bias caused due to exclusion of patients excluded after randomization and lost to follow-up, from analysis for the primary outcome. We did this by analyzing the primary outcome assuming that participants lost to follow-up had lost 15 or more letters visual acuity (worst-case analysis) and that they did not lose 15 or more letters visual acuity at one year follow-up (best-case analysis). We planned to repeat the analysis after excluding studies of lower methodological quality (i.e. graded C on any parameter), unpublished data, and industry-funded studies.

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of ongoing studies.

The electronic searches (conducted in August 2005, October 2006, June 2007 and February 2008) resulted in a total of 1407 titles and abstracts. We identified 36 papers for full-text review. Five trials described in ten reports were identified. We excluded 16 studies, which are listed in the table: ‘characteristics of excluded studies’ along with reasons for exclusion. Seven publications, requiring further clarification, are listed as ‘Studies awaiting assessment’ along with three additional studies identified as potentially eligible from communications with experts during peer review. Two additional abstracts identified through hand searching are also listed as ‘Studies awaiting assessment’. Acronyms used to refer to the studies in this review are listed in Table 1.

Table 1. Table of Acronyms.

Acronym	Details
ABC	A randomised, double-masked phase III study of the efficacy and safety of Avastin® (Bevacizumab) intravitreal injections compared to best available therapy in subjects with choroidal neovascularisation secondary to age-related macular degeneration
ANCHOR	Anti-VEGF Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in Age-related Macular Degeneration
CATT	Comparison of Age-related macular degeneration Treatment Trials
FOCUS	RhuFab V2 Ocular Treatment Combining the Use of Visudyne® to Evaluate Safety
IVAN	A randomised controlled trial of alternative treatments to Inhibit VEGF in Age-related choroidal Neovascularisation
MARINA	Minimally Classic/Occult Trial of the Anti-VEGF Antibody Ranibizumab in the Treatment of Neovascular Age-Related Macular Degeneration
PIER	A Phase IIIb, Multicenter, Randomized, Double-Masked, Sham Injection-Controlled Study of the Efficacy and Safety of Ranibizumab in Subjects with Subfoveal Choroidal Neovascularization with or without Classic CNV Secondary to Age-Related Macular Degeneration
PrONTO	Prospective Optical coherence tomography imaging of patients with Neovascular AMD Treated with intra-Ocular ranibizumab
PROTECT	Unclear
SAILOR	Safety Assessment of Intravitreal Lucentis for Age-Related Macular Degeneration
SUMMIT	Unclear
VERITAS	Unclear

To our knowledge, two other clinical trials evaluating ranibizumab for AMD have not been published (PIER; SAILOR) and we were unable to procure any data on these studies. The data presented in this review are current as of November 2007.

One of the included studies reported an individual patient data meta-analysis of two randomized trials under the acronym VISION (Gragoudas 2004). This included an international trial (EOP 1003) and a North American trial (EOP 1004). In our review we assessed the data from the two trials separately and analyzed them according to the original protocol of the review. We obtained the data for the primary and secondary outcomes for the two trials from the information available on the FDA web site (www.fda.gov/ohrms/dockets/ac/04/briefing/2004-4053b1.htm, accessed on June 22, 2005) and by contacting the authors.

All the included trials were funded by various pharmaceutical firms with the authors reporting varying levels of financial interests in them.

Types of participants

The EOP 1004 trial included 586 patients from 58 sites in the United States and Canada and EOP 1003 recruited 622 patients from 59 sites in Austria, Belgium, Brazil, Chile, Colombia, Czech Republic, Denmark, France, Germany, Hungry, Israel, Italy, Poland, Portugal, Spain, Switzerland, The Netherlands, UK, and Sweden. Participants in the trial included men and women age 50 or older with neovascular AMD. In both the trials, the investigators stratified the patients based on angiographic subtype of the lesion at baseline. The trials were also stratified on status of prior treatment with verteporfin PDT at baseline. The inclusion criteria for both studies were the same in both the trials (EOP 1003; EOP 1004).

ANCHOR and FOCUS included patients with primary or recurrent subfoveal CNV secondary to AMD that were predominantly classic in composition. Only patients with lesion size less than or equal to 5400 microns and best-corrected visual acuity of 20/40 to 20/320 (Snellen equivalent) were eligible for inclusion.

MARINA included patients with active or recurrent subfoveal lesions with CNV secondary to AMD that involved the foveal center. Patients with lesions classified as either minimally classic or occult with no classic CNV were included. Patients with lesions with an occult CNV component were also included but for concomitant classic CNV, the area of classic CNV must have been less than 50% of the total lesion size. Other inclusion criteria specified were a total lesion area less than or equal to 12 standard disc areas (1 standard disc area is 2.54 mm²), visual acuity of 20/40 to 20/320 (Snellen equivalent) in the study eye.

The trials included in this review considered the following definitions: predominantly classic => 50% classic CNV, minimally classic - 1 to 49% classic CNV, pure occult - 0% classic CNV.

Additional details of the included trials are summarized in the table: ‘Characteristics of included studies’.

Types of intervention

Comparisons of interventions evaluated in the included trials are listed in Table 2 and are summarized here. Both the EOP 1003 and EOP 1004 trials compared the same types of interventions. They compared sham versus intravitreal injection of pegaptanib at dosages of 0.3 mg, 1.0 mg, or 3.0 mg given every six weeks over period of 48 weeks.

Table 2. Treatment comparisons in included trials.

Trial	Treatment group 1	Treatment group 2	Treatment group 3	Control
EOP 1003	0.3 mg pegaptanib	1 mg pegaptanib	3 mg pegaptanib	Sham
EOP 1004	0.3 mg pegaptanib	1 mg pegaptanib	3 mg pegaptanib	Sham
ANCHOR	0.3 mg ranibizumab plus sham verteporfin PDT	0.5 mg ranibizumab plus sham verteporfin PDT	Verteporfin PDT plus sham intravitreal injection	-
FOCUS	Verteporfin PDT plus 0.5 mg ranibizumab	Verteporfin PDT plus sham intravitreal injection	-	-
MARINA	0.3 mg ranibizumab	0.5 mg ranibizumab	-	Sham

Ranibizumab was evaluated in three included studies, ANCHOR, FOCUS, and MARINA. Different doses of ranibizumab were compared in MARINA, and ANCHOR. Patients in the treatment groups in ANCHOR received an intravitreal injection of ranibizumab (0.3 mg and 0.5 mg) along with sham verteporfin PDT therapy. Patients in the control group in this trial received sham intravitreal ranibizumab and active verteporfin PDT therapy. Patients in the MARINA trial received either 0.3 mg ranibizumab or 0.5 mg ranibizumab or sham injection. Verteporfin PDT for the study eye was allowed if the CNV evolved to become predominantly classic in composition. FOCUS trial compared verteporfin PDT and 0.5 mg ranibizumab in one group with only verteporfin PDT therapy in the control group.

Types of outcome measures

The primary outcome for this review was based on best corrected visual acuity after at least one year of follow-up. We considered loss of 15 or more letters of visual acuity as the primary outcome. Other visual acuity outcomes that were analyzed included maintenance of visual acuity (gain = 0 lines) at one year follow-up, mean visual acuity at one year, gain of 15 letters or 3 or more lines of visual acuity, loss of 30 or more letters or 6 or more lines, and blindness (defined as visual acuity worse than 20/200) at one year.

Secondary outcomes

The following secondary outcomes were available and were analyzed at one year.

1. Assessment of CNV.

2. Total CNV size.

3. Mean change in size of lesion from baseline.

All five trials included in this systematic review have been conducted by pharmaceutical companies manufacturing the drugs and thus have a potential for bias due to conflicts of interest.

Adverse reactions reported in the included trials are summarized in tables below. The trials did not study contrast sensitivities, or reading speed. Information on costs of treatment was discussed for pegaptanib and ranibizumab in a separate article (Brown 2006 - www.evrs.org/pages/2006/Scientifis%20Sessions_abstracts).

Vision-specific quality of life was evaluated in EOP 1004, ANCHOR, and MARINA using the 25-item National Eye Institute-Visual Functioning Questionnaire (NEI-VFQ) (www.evrs.org/pages/2006/Scientifis%20Sessions_abstracts).

Risk of bias in included studies

Patients in EOP 1003 and EOP 1004 trials were allocated to a treatment group based on a dynamic randomization sequence generated using a stochastic treatment allocation algorithm. The allocation sequence was concealed from the people enrolling patients. Patients who received the medication and those who received the sham were treated identically except for the scleral penetration. The ophthalmologists responsible for examining and taking care of the patients were not administering the injections. Certified refractionists measuring visual acuity were masked as well. Follow-up in the different groups were similar in both trials. 91.5% in the EOP 1003 trial and 90% in the EOP 1004 trial were followed-up at one year. It is not clear if EOP 1003 and EOP 1004 adjusted for multiple comparisons in their analyses.

ANCHOR and MARINA trials reported stratifying randomization by treatment center and baseline visual acuity. Per information provided by Genentech in an email communication, randomization in ANCHOR was done using a dynamic randomization method. Allocation concealment in both these trials was achieved by central randomization. Patients and personnel involved in assessing outcomes were masked to treatment assignment in both trials. An unmasked ophthalmologist administered the treatment. Exclusions after randomization in ANCHOR were uneven in distribution between treatment groups with three participants excluded only from the 0.3 mg ranibizumab group owing to either patient or physician decision. One participant in the 0.5 mg ranibizumab group and two in sham group in MARINA did not receive treatment. Reasons included randomization before eligibility confirmation (one in each group) and apprehensiveness (one in sham group). In MARINA, 10 participants each were lost to follow-up in the 0.3 mg ranibizumab group and the control group. Five participants were lost to follow-up in the 0.5 mg ranibizumab group. Reasons for losses to follow-up in both trials included patient and physician decisions and non-compliance. Both trials reported an intention-to-treat (ITT) analysis using a last-observation carried forward method and reported an a priori sample size/power calculation. Adjustment for multiple comparisons in both trials was by the Hochberg-Bonferroni procedure.

Randomization in FOCUS was stratified by treatment center. Method of allocation concealment, however, was not reported in the paper. In an email communication with Dr. Heier, first author of the published trial report, we were informed that “physicians and personnel were unaware of the allocation sequences prior to randomization”. Outcome assessors were masked to treatment assignment. Seven participants in the verteporfin PDT + ranibizumab group and three in verteporfin PDT alone group were lost to follow-up at 12 months. Last observation carried forward method was used to report an ITT analysis.

Further details are summarized in the table of included studies.

Effects of interventions

We did not conduct a meta-analysis of results when different treatment comparisons were evaluated in the individual trials as evident from Table 2. Meta-analysis of data from EOP 1003 and EOP 1004 was possible due to the similarity in design of the two trials. In the forest plots reported in this review, it should be noted that for outcomes such as gain in visual acuity and mean differences in visual acuity, effect estimates to the right of the vertical line favor the test treatment.

Pegaptanib versus sham

Loss of 15 or more letters of visual acuity at one year of follow-up (Figure 01.01)

To account for the participants for whom there was no data on change in visual acuity, i.e. whether they lost 15 or more letters of visual acuity at one year, we conducted an analysis using only those patients who received treatment and had data on baseline and final visual acuity; an available-case data analysis. We could not perform an ITT analysis due to unavailability of data.

An available-case data analysis showed that fewer patients treated with pegaptanib lost 15 or more letters of visual acuity. The RR was 0.71 (95% confidence interval (CI) 0.60 to 0.84). A similar treatment effect was seen with 0.3 mg and 1 mg pegaptanib. Relative risk for loss of 15 or more letters with 0.3 mg pegaptanib compared to sham was 0.66 (95% CI 0.53 to 0.83) and was 0.64 (95% CI 0.51 to 0.80) with 1 mg pegaptanib compared to sham. However, the RR for3 mg pegaptanib compared to sham was 0.84 (95% CI 0.69 to 1.04).

We calculated the number needed to treat (NNT) for this outcome from the pooled risk differences from the two trials. There was no statistical heterogeneity between the estimates of the two trials on this scale. The NNT was 6.67 (95% CI 4.35 to 14.28) for 0.3 mg pegaptanib, 6.25 (95% CI 4.17 to 12.5) for 1 mg pegaptanib and 14.28 (95% CI 6.67 to 100) for 3 mg pegaptanib.

Loss of 30 or more letters of visual acuity at one year of follow-up (Figure 01.02)

For this outcome, as is the case with the rest of the outcomes following this, we used the number of patients receiving treatment and having had baseline visual acuity measurements with last observation carried forward as the denominator (as reported in the FDA documents for the trials) to calculate RRs in our analyses. Risk of losing 30 or more letters of visual acuity at one year was lower on treatment with pegaptanib compared to sham. The pooled RR was 0.47 (95% CI 0.35 to 0.63). This means a RR reduction of 0.53 (95% CI 0.37 to 0.65). This effect was consistently observed across all doses of pegaptanib treatment. The RR of losing 30 or more letters of visual acuity at one year was 0.43 (95% CI 0.29 to 0.65) in the 0.3 mg pegaptanib group, 0.36 (95% CI0.23 to 0.57) in the 1 mg pegaptanib group, and 0.62 (95% CI 0.43 to 0.88) in the 3 mg pegaptanib group. There was no substantial statistical heterogeneity.

Blindness (visual acuity worse than 20/200) at one year of follow-up (Figure 01.03)

Pegaptanib consistently resulted in a reduction in risk of blindness at one year follow-up. The pooled RR was 0.69 (95% CI 0.59 to 0.82). This is equivalent to a RR reduction of 0.31 (95% CI 0.18 to 0.41). The RR of blindness was 0.62 (95% CI 0.49 to 0.77) in the 0.3 mg pegaptanib group, 0.75 (95% CI 0.61 to 0.92) in the 1 mg pegaptanib group, and 0.71 (95% CI 0.58 to 0.88) in the 3 mg pegaptanib group. There was no substantial heterogeneity.

Gain of 15 or more letters of visual acuity at one year of follow-up (Figure 01.04)

Very few participants in all treatment and sham groups had a gain of 15 or more letters of visual acuity. Analyses on this outcome are unreliable due to such small numbers of events. I-square values were also high for these comparisons which suggested against reporting any pooled results. As shown in the graphs, the CIs were very wide due to the small number of events and the point estimates in the two trials were also different.

Maintenance of visual acuity at one year of follow-up (Figure 01.05)

A greater number of patients treated with pegaptanib seemed to maintain visual acuity at one year. The pooled RR for pegaptanib versus sham was 1.49 (95% CI 1.19 to 1.88).

We do not report the pooled results in the two trials for RR of maintenance of visual acuity at one year for the 0.3 mg pegaptanib group since the p-value for the chi-square test was 0.07 and the I-square value was 69.8% suggesting statistical heterogeneity. The RR of maintenance of visual acuity at one year with 0.3 mg pegaptanib compared to sham was 1.18 (95% CI 0.84 to 1.67) in the EOP 1003 trial and 1.96 (95% CI 1.28 to 2. 99) in the EOP 1004 trial.

We combined the results for RR of maintenance of visual acuity at one year for the 1 mg and 3 mg pegaptanib groups compared to sham using a fixed-effect model. The p-values for the chi-square test were 0.17 and 0.82 for the 1 mg and 3 mg pegaptanib groups respectively and the corresponding I-square values were 47.4% and 0%.

The pooled RR of maintaining visual acuity at one year follow-up was 1.62 (95% CI 1.25 to 2.10) in the 1 mg pegaptanib group and 1.39 (95% CI 1.06 to 1.81) in the 3 mg pegaptanib group.

Mean visual acuity at one year of follow-up (Figure 01.06)

Visual acuity in both trials was measured using the ETDRS scale placed at 2 meters from the patient. The mean number of letters read on the ETDRS chart was reported. The final mean visual acuity in the three active treatment groups was consistently greater than in the sham group. The WMD in letters for the 0.3 mg pegaptanib group was 7.20 (95% CI 4.27 to 10.13), 1 mg pegaptanib group was 5.76 (95% CI 2.76 to 8.77), and3 mg pegaptanib group was 3.64 (95% CI 0.78 to 6.51). Converting mean change in visual acuity in letters to logMAR scores, the mean change with 0.3 mg pegaptanib is equivalent to 0.14 logMAR units. This means that the mean change in visual acuity was better in the 0.3 mg pegaptanib group than in the sham group by 0.14 logMAR units. Mean change in visual acuity was better by 0.11 logMAR units for the 1 mg pegaptanib group and by 0.07 logMAR units for the 3 mg pegaptanib group compared with the sham group.

Total CNV size at one year of follow-up (Figure 01.07)

Pegaptanib treatment, across all doses studies, resulted in a lower final mean CNV size compared to the sham group at one year of follow-up. Total CNV sizes are reported as numbers of standard disc area. The WMD for 0.3 mg pegaptanib group was-1.05 (95% CI -1.63 to -0.47), and was -0.97 (95% CI -1.54 to -0.40) for the 1 mg pegaptanib group, and -0.74 (95% CI -1.35 to -0.13) for the 3 mg pegaptanib group.

Mean size of lesion at one year of follow-up (Figure 01.08)

Pegaptanib treatment resulted in a lower mean size of lesion at one year of follow-up compared to sham. Mean size of lesion is described in terms of standard disc area. The WMD was -1.10 (95% CI -1.69to -0.52) for 0.1 mg pegaptanib group, -0.89 (95% CI -1.46 to -0.31) for the 1 mg pegaptanib group and -0.52 (95% CI -1.18 to 0.13) for the 3 mg pegaptanib group.

Analyses by angiographic sub-type (Figures 02.01 and 02.02)

There was no indication of the treatment effect noticed in the overall analysis being segregated to patients in sub-group of any particular angiographic sub-type. Data for analyses by angiographic subtype were available for only some treatment groups from the individual trials. Compared with sham, the pooled RR of loss of 15 or more letters of visual acuity at one year was 0.72 (95% CI 0.47 to 1.11), 0.52 (95% CI 0.35 to 0.78), and 0.77 (95% CI 0.54 to 1.08) in patients treated with 0.3 mg pegaptanib with predominantly classic, minimally classic and pure occult lesions, respectively. And, the pooled RR of loss of fewer than 15 letters of visual acuity in patients treated with 1 mg pegaptanib compared with sham was 0.88 (95% CI 0.61 to 1.26), 0.56 (95% CI 0.39 to 0.81), and 0.64 (95% CI 0.45 to 0.92) in predominantly classic, minimally classic and pure occult sub-types, respectively. There was no substantial statistical heterogeneity observed in any of these meta-analyses.

Sensitivity analyses (Figures 03.01; 03.02)

We conducted a sensitivity analysis for the primary outcome in the EOP 1003 and EOP 1004 trials to assess for any impact of missing data on the results. For a worst-case analysis, we assumed that all patients lost to follow-up before one year had lost 15 or more letters visual acuity at one year. The treatment effect persisted with RR (95% CI) of 0.72 (95% CI 0.59 to 0.87) and 0.72 (95% CI 0.59 to 0.86) for the 0.3 mg and 1 mg pegaptanib groups respectively. As the second part of our sensitivity analysis, we assumed that none of the patients lost to follow-up before one year had lost 15 or more letters visual acuity at one year. The RR was 0.67 (95% CI 0.53 to 0.84) for the 0.3 mg pegaptanib group and 0.63 (95% CI 0.5 to 0.8) for the 1 mg pegaptanib group.

Ranibizumab versus sham

Loss of 15 or more letters of visual acuity at one year of follow-up

Fewer patients treated with ranibizumab lost 15 or more letters visual acuity at one year follow-up compared with those treated with sham. The overall RR for ranibizumab versus sham was 0.14 (95%CI 0.10 to 0.22). This beneficial treatment was similar across both 0.3 mg and 0.5mg ranibizumab. Relative risk was 0.14 (95% CI 0.08 to 0.25) for both 0.3 mg and 0.5 mg ranibizumab. The NNT for this outcome was 3.13 (95% CI 2.56 to 3.84) for 0.3 mg ranibizumab and 3.13 (95% CI 2.56 to 3.84) for 0.5 mg ranibizumab.

Loss of 30 or more letters of visual acuity at one year of follow-up

As with the earlier outcome, fewer patients in both the ranibizumab groups lost 30 or more letters of visual acuity at one year follow-up. The overall RR for ranibizumab versus sham was 0.07 (95% CI 0.03 to 0.18). The RR was 0.06 (95% CI 0.01 to 0.24) for the 0.3 mg ranibizumab group and 0.09 (95% CI 0.03 to 0.28) for the 0.5 mg ranibizumab group.

Blindness (visual acuity worse than 20/200) at one year of follow-up

Ranibizumab resulted in fewer cases of blindness at one year follow-up. The overall RR for ranibizumab compared with sham was 0.28 (95% CI 0.21 to 0.37). This effect was uniform across both doses of ranibizumab with RR of 0.28 (95% CI 0.2 to 0.41) for 0.3 mg group and 0.27 (95% CI 0.19 to 0.40) for the 0.5 mg ranibizumab group.

Gain of 15 or more letters of visual acuity at one year of follow-up

A greater proportion of participants treated with ranibizumab gained 15 or more letters of visual acuity at one year follow-up. The overall RR for ranibizumab versus sham was 5.81 (95% CI 3.29 to 10.26). The RR of gain of 15 or more letters visual acuity at one year was 4.92 (95% CI 2.71 to 8.91) for the 0.3 mg ranibizumab group and 6.69 (95% CI 3.75 to 11.94) for the 0.5 mg ranibizumab group compared with sham.

Mean change in visual acuity at one year of follow-up

Visual acuity in all three trials involving ranibizumab included in this review was measured using ETDRS charts placed at 4 meters and reported as the mean number of letters read on the chart. Patients treated with ranibizumab had greater mean visual acuity at one year compared with those treated with sham. The WMD for the mean change was 16.9 (95% CI 14.23 to 19.57) for the 0.3 mg ranibizumab group and 17.6 (95% CI 14.80 to 20.40) for the 0.5 mg ranibizumab group compared with sham injection. Converting to logMAR scores, mean change in visual acuity was better by 0.34 logMAR units for the 0.3 mg ranibizumab group and by 0.35 logMAR units for the 1 mg ranibizumab group compared with the sham group.

Ranibizumab plus sham verteporfin PDT versus active verteporfin PDT plus sham intravitreal injection

Loss of 15 or more letters of visual acuity at one year of follow-up

Ranibizumab resulted in a smaller proportion of patients losing 15 or more letters of visual acuity at one year. The overall RR was 0.13 (95% CI 0.07 to 0.23). The effect was consistent across both 0.3 mg and 0.5 mg ranibizumab groups compared with the verteporfin PDT group. The RR was 0.16 (95% CI 0.08 to 0.33) for the 0.3mg ranibizumab group and 0.10 (95% CI 0.04 to 0.25) for the 0.5 mg ranibizumab group.

The NNT was 3.33 (95% CI 2.56 to 4.76) for 0.3mg ranibizumab and 3.125 (95% CI 2.43 to 4.17) for 0.5 mg ranibizumab.

Loss of 30 or more letters of visual acuity at one year of follow-up

No patients treated with ranibizumab lost 30 or more letters visual acuity at one year. Nineteen of 143 patients in the verteporfin PDT group lost 30 or more letters visual acuity at one year.

Blindness (visual acuity worse than 20/200) at one year of follow-up

Fewer patients treated with ranibizumab had visual acuity worse than 20/200 at the end of one year follow-up. The overall RR was 0.32 (95% CI 0.24 to 0.42). The RR was 0.37 (95% CI 0.26 to 0.52) for 0.3 mg ranibizumab group and 0.27 (95% CI 0.18 to 0.41) for the 0.5 mg ranibizumab group.

Gain of 15 or more letters of visual acuity at one year of follow-up

A greater proportion of patients treated with ranibizumab gained 15 or more letters visual acuity at one year. The overall RR was 6.79 (95% CI 3.41 to 13.54). The RR of gain of 15 or more letters visual acuity at one year was 6.38 (95% CI 3.14 to 12.97) for the 0.3 mg ranibizumab group and 7.2 (95% CI 3.57 to 14.55) for the 0.5 mg ranibizumab group.

Mean change in visual acuity at one year of follow-up

Patients treated with ranibizumab had a greater mean visual acuity at the end of one year compared with those treated with verteporfin PDT alone. The WMD for mean change in visual acuity was 18.00 (95% CI 14.38 to 21.62) for the 0.3 mg ranibizumab group and 20.80 (95% CI 17.18 to 24.42) for the 0.5 mg ranibizumab group. Converting to logMAR scores, mean change in visual acuity was better by 0.36 logMAR units for the 0.3 mg ranibizumab group and by 0.42 logMAR units for the 1 mg ranibizumab group compared with the sham group.

Mean change in size of lesion at one year of follow-up

The size of lesion was measured in terms of the number of standard disc areas in all trials included in this review that involved ranibizumab. Patients treated with ranibizumab had a greater reduction in size of lesion by the end of one year follow-up compared with those treated with verteporfin PDT alone. The WMD for the mean change in size of lesion was -2.2 (95% CI -2.74 to -1.66) for the 0.3 mg ranibizumab group and -2.28 (95% CI -2.83 to -1.73) for the 0.5 mg ranibizumab group.

Ranibizumab (0.5 mg) plus verteporfin PDT versus verteporfin PDT alone

Loss of 15 or more letters of visual acuity at one year of follow-up

Fewer patients treated with ranibizumab and verteporfin PDT together lost 15 or more letters of visual acuity at one year follow-up. The RR was 0.3 (95% CI 0.15 to 0.60). The NNT was 4.35 (95% CI 2.78 to 11.11) for the treatment group with ranibizumab.

Loss of 30 or more letters of visual acuity at one year of follow-up

Fewer patients lost 30 or more letters of visual acuity at one year follow-up. The RR was 0.11 (95% CI 0.01 to 0.89) for ranibizumab plus verteporfin PDT group compared with the verteporfin PDT group.

Blindness (visual acuity worse than 20/200) at one year of follow-up

A smaller proportion of patients treated with both ranibizumab plus verteporfin PDT had visual acuity worse than 20/200 at one year follow-up. The RR was 0.64 (95% CI 0.42 to 0.96).

Gain of 15 or more letters of visual acuity at one year of follow-up

A greater proportion of patients treated with ranibizumab plus verteporfin PDT gained 15 or more letters visual acuity at one year follow-up. The RR was 4.44 (95% CI 1.40 to 14.08).

Mean change in visual acuity at one year of follow-up

Mean change in visual acuity was greater in patients treated with ranibizumab plus verteporfin PDT than in patients treated with verteporfin PDT alone. The WMD for the mean change was 13.10 (95% CI 7.99 to 18.21). Converting to logMAR scores, mean change in visual acuity was better by 0.26 logMAR units for the ranibizumab plus verteporfin PDT group.

Mean change in size of lesion at one year of follow-up

Reduction in size of lesion was greater in patients treated with ranibizumab plus verteporfin PDT than in patients treated with verteporfin PDT alone. The WMD for mean change in the size of lesion for these two groups was -1.82 (95% CI -2.47 to -1.17). The negative effect estimate indicates that the size of lesion was smaller at the end of one year for patients treated with ranibizumab plus verteporfin PDT.

Quality of life

Quality of life was measured in one of the two VISION studies, EOP 1004 using the NEI-VFQ questionnaire (EOP 1004). Since it was validated only for United States English, this questionnaire was administered to participants from the United States and Canada alone at baseline and at weeks 30 and 54 of follow-up. Treatment with pegaptanib was associated with better scores on the NEI-VFQ questionnaire. However, we did not have standard deviations for the scores to conduct an analysis.

Improvement in quality of life scores was reported among patients treated with ranibizumab plus verteporfin PDT compared with those treated with verteporfin PDT alone. Data available from different conference abstracts were extracted (ANCHOR). At two years follow-up, mean NEI-VFQ scores were greater for patients treated with ranibizumab (0.3 mg and 0.5 mg) compared with patients treated with verteporfin PDT alone. Mean change in score for composite subscale was 8.1, 5.9, and 2.2 for 0.5 mg ranibizumab, 0.3 mg ranibizumab and verteporfin PDT only groups, respectively. The difference was consistent across sub-scales for near activities, distance activities and vision-specific dependency.

A greater proportion of patients treated with ranibizumab were reported to have had improvement in vision-specific quality of life scores (MARINA) compared with those treated with sham. MARINA included patients with minimally classic or occult CNV. Data on NEI-VFQ scores reported in a conference abstract and a consequent full publication were extracted (MARINA). Treatment with 0.5mg ranibizumab resulted in greater improvement in mean scores in both the better seeing eye and the worse seeing eye. For the better seeing eye, the mean change in score was 14.7 for near activities, 10 for distance activities and 10 for dependency in those treated with 0.5 mg ranibizumab. Mean changes in scores were -8.7 for near activities, -12.9 for distance activities and -14.7 for dependency scales for those treated with sham in the better-seeing eye. Mean change in scores for patients treated with 0.5 mg ranibizumab and sham in the worse seeing eye are 3.4 and -5.6 respectively on the near activities subscale, 1.6, -5.6 on distance activities subscale, and 2.1, -7.9 on the dependency subscale. The mean change in overall composite quality of life score in the treated eye was 5.2 (95% CI 3.5 to 6.9) with 0.3 mg ranibizumab, 5.6 (95% CI 3.9 to 7.4) with 0.5 mg ranibizumab and -2.8 (95% CI -4.6 to -1.1) with sham injection at 12 month follow-up. At 24 month follow-up, the mean change in the overall composite score was 4.8 (95% CI 2.9 to 6.8) with 0.3 mg ranibizumab, 4.5 (95% CI 2.5 to 6.5) with 0.5 mg ranibizumab and -6.5 (95% CI -8.4 to -4.6) with sham injection.

Adverse events

Adverse events reported in the ANCHOR, MARINA, FOCUS, EOP 1003 and EOP 1004 trials are summarized in Table 3, Table 4, Table 5 and Table 6. Pooling of results was not attempted for the trials studying ranibizumab due to different treatment comparisons and patient characteristics. Data on the two trials involving pegaptanib were extracted from documents available from the FDA website and listed above. Endophthalmitis, intraocular inflammatory reaction following injection, cataract and post-injection increase in intraocular pressure were the most common adverse effects with ranibizumab. Frequency of endophthalmitis with ranibizumab was from 0.7% to 4.7% in the included trials. Range for frequency of elevated intraocular pressure with ranibizumab was 4.3% to 16%. Frequency of endophthalmitis with pegaptanib was 1.3% in the included trials. These rates of potentially vision threatening problems such as endophthalmitis and retinal detachments are comparable to rates from intravitreal injections with other drugs (Jager 2004).

Table 3. Adverse events (ANCHOR).

Adverse event	Ranibizumab (N=277)	Verteporfin (N=143)
Endophthalmitis	2	0
Uveitis	1	0
Rhegmatogenous retinal detachment	1	1
Vitreous hemorrhage	1	0
Treatment emergent hypertension	12	12
Post-injection IOP of 30 mm Hg or more	24	6
Post-injection IOP of 40 mm Hg or more	8	1
Cataract	33	10
Myocardial infarction	4	1
Stroke/cerebral infarction	2	1
Death	5	2

Table 4. Adverse events (MARINA).

Adverse event	Ranibizumab (N=477)	Sham control (N=236)
Endophthalmitis	6	0
Uveitis	6	0
Rhegmatogenous retinal detachment	0	1
Vitreous hemorrhage	2	2
Treatment emergent hypertension	80	38
Post-injection IOP of 30 mm Hg or more	73	8
Post-injection IOP of 40 mm Hg or more	11	0
Cataract	74	37
MI	9	4
Stroke/cerebral infarction	9	2
Death (at 24 months)	11	6

Table 5. Adverse events (FOCUS).

Adverse event	Ranibizumab+PDT(105)	PDT (56)
Endophthalmitis	5	0
Intraocular inflammatory adverse effects	40	3
Post-injection IOP of 30 mm Hg or more	14	2
Cataract	13	6
Arterial thromboembolic events (using The Antiplatelet Trialists Collaboration criteria)	4	2
Treatment emergent hypertension	13	4
Vascular systemic adverse events	9	3
Non-vascular systemic adverse events	14	11

Table 6. Adverse effects for EOP 1003 and EOP 1004.

Adverse effect	0.3 mg pegaptanib	1 mg pegaptanib	3 mg pegaptanib	All doses pegaptanib	Sham
Endophthalmitis	6	3	3	12	0
Retinal detachment	1	2	2	5	0
Traumatic cataract	1	2	2	5	0
Retinal hemorrhage	1	0	1	2	0
Vitreous hemorrhage	0	0	1	1	0
Uveitis	0	0	1	1	0
Elevated intraocular pressure	1	0	0	1	0
Papilledema	0	0	0	0	1
Cardiac disorders	11	4	10	25	14
Neoplasms (benign, malignant, unspecified)	11	7	8	26	12
Nervous system disorders	10	5	10	25	7
Infections and infestations	2	7	11	20	5
Injury and procedural complications	10	9	8	27	3
Gastrointestinal disorders	3	6	5	14	4
Respiratory, thoracic, mediastinal disorders	2	5	5	12	4
Musculoskeletal and connective tissue	1	5	3	9	2
Renal and urinary disorders	2	3	2	7	3
Vascular disorders	3	2	2	7	3
Hepatobiliary disorders	1	0	2	3	0

The systemic serious adverse affects are difficult to interpret at this time. We have not performed an exhaustive search for a systematic review of adverse effects but provide descriptive comments about data available to date. The initial reports with MARINA and ANCHOR showed that there was more common incidence of systemic serious adverse events in treatment groups with ranibizumab such as hypertension and thromboembolic events. However, 24 month data of these trials showed no significant differences between the treatment and sham groups. In January 2007, Genentech, the company manufacturing ranibizumab sent out a letter to eye care providers in the United States. This letter had information on interim results from the SAILOR trial. They reported that patients treated with ranibizumab delivered through the intravitreal route had a higher incidence of strokes in the 0.5 mg dose group compared with the 0.3 mg dose group (1.2% versus 0.3% respectively; p = 0.02). Patients with a history of prior stroke appeared to be at higher risk for a subsequent stroke. There were no statistically significant differences between these two groups for arterial thromboembolic events such as myocardial infarction or vascular death. Complete results from the SAILOR trial have not been published yet.

Discussion

All five trials included in this systematic review were of good methodological quality. There is no direct evidence that ranibizumab or pegaptanib treatment strategies are better than the other. We found no trial directly comparing pegaptanib with ranibizumab. Moreover, pegaptanib and ranibizumab trials had different inclusion criteria; therefore, meta-analysis was not possible. All studies have been conducted by pharmaceutical companies manufacturing these agents and thus have a potential for bias due to conflicts of interest.

The results of this review indicate efficacy of pegaptanib, and ranibizumab in terms of lower proportion of patients losing 15 or more letters compared with sham at one year follow-up. Numbers needed to treat calculated above indicate that seven patients need to be treated with 0.3 mg or 0.5 mg pegaptanib to prevent loss of 15 or more letters visual acuity in one patient. This number is higher for 3 mg pegaptanib (15). Four people need to be treated with ranibizumab to prevent loss of 15 or more letters visual acuity in one patient. Monthly administration of ranibizumab also results in a greater proportion of patients gaining 15 or more letters visual acuity compared with sham. The beneficial effects of these two agents with visual acuity are consistent with their effects on changes in lesion size evaluated by fluorescein angiography. Available information on adverse effects with each medication do not suggest higher incidence of potentially vision threatening complications with intravitreal injection compared with intravitreal injection of other drugs. We found no trials comparing ranibizumab with pegaptanib. Both compounds were associated with improved vision-specific quality of life assessed using the NEI-VFQ scale. Below, we discuss the results for each agent in further detail. This is followed by a detailed discussion of ongoing trials evaluating different dosing regimens with ranibizumab and with other anti-VEGF compounds.

Vision and angiography outcomes with pegaptanib

The two randomized clinical trials (EOP 1003 and EOP 1004) have shown a statistically significant benefit for treatment with pegaptanib for patients with neovascular AMD. In other words, patients on pegaptanib were 29% less likely to lose 15 or more letters visual acuity atone year follow-up overall. Patients in the 0.3 mg pegaptanib group were 34% less likely and those in the 1 mg pegaptanib group 36% less likely to lose 15 or more letters visual acuity at one year follow-up. This effect, observed with an analysis including patients who received treatment and had baseline and final visual acuity measurement, was retained even in sensitivity analyses evaluating bias introduced by any systematic exclusion of patients who were doing worse or inclusion of only patients doing clinically well. Patients considered for sensitivity analyses included those excluded after randomization, those lost to follow-up either due to death or other reasons, and those for whom either baseline or final visual acuity or both measurements were not recorded. The actual numbers for each group as calculated from data tables in the FDA documents cited above are noted in the table of included studies.

The beneficial effect observed in the overall analysis (pegaptanib versus sham) was retained in the pooled analyses for 0.3 mg and 1 mg pegaptanib groups. The point estimates for the RR for all the analyses were different between the worst-case and best-case analyses though the 95% CIs overlapped each other. This difference in RR estimates was greater in the 1 mg, 3 mg and overall pegaptanib versus sham comparisons than the comparison between 0.3 mg pegaptanib and sham.

A greater number of patients were lost to follow-up or discontinued treatment in the 3 mg pegaptanib group compared to the other groups. 11 % in EOP 1003 and 14% in EOP 1004 discontinued treatment as compared to 8%, 12%, and 8% and in 0.3 mg, 1 mg and sham groups in EOP 1003 respectively and 7%, 8%, and 8% in 0.3 mg, 1 mg, and sham groups in EOP 1004 respectively. This may influence the conclusions on treatment effect for the 3 mg pegaptanib group.

Although the number of patients who were lost to follow-up or who discontinued treatment for other reasons between 0.3 mg and 1 mg pegaptanib groups (12% in 1 mg pegaptanib group as compared to 8% in 0.3 mg pegaptanib group in EOP 1003), the effect estimates were not much different between the 0.3 mg pegaptanib and 1 mg pegaptanib comparisons. This was the case with the best-case and worst-case analyses in which the CIs overlapped each other. This suggests that using a higher dose of 1 mg pegaptanib does not confer any additional treatment benefit.

Patients receiving pegaptanib had a lower risk of losing 30 or more letters or becoming blind. The final mean visual acuity in the three treatment groups was consistently greater than in the sham group.

Patients treated with pegaptanib also seemed to have a slightly higher chance of visual acuity stability as assessed by the number of participants with maintenance of visual acuity at one year, although the results differed slightly in the EOP 1003 and EOP 1004 trials. There was a higher chance for visual acuity gain or maintenance for participants in the EOP 1004 trial than those in the EOP 1003 trial. The point estimates for maintenance of visual acuity in the 0.3 mg and 1 mg groups in the two trials differed markedly with consequent high I-square values (69.8% and 47.4% respectively). However, there was some overlap in the 95% CIs.

There was significant statistical heterogeneity for the comparison of 0.3 mg and 1 mg pegaptanib groups versus sham. This indicates that EOP 1003 and EOP 1004 trials were estimating different treatment effect for the outcome maintenance of visual acuity. When we evaluated similarity of baseline characteristics between the two groups in both trials we noticed that a greater proportion of patients in the 0.3 and 1 mg pegaptanib groups in EOP 1004 had a history of prior treatment with verteporfin PDT as well as concomitant treatment with verteporfin PDT in comparison to groups in EOP 1003 trial. Our results suggest that previous or concomitant treatment with verteporfin PDT may have resulted in an increase in the risk of maintenance of visual acuity. It is possible that the number of patients treated with verteporfin PDT before and/or during the trial in EOP 1003 was too small to detect a difference. On the other hand, it is also possible that the number of patients receiving verteporfin PDT before or during the trial was too small in the EOP 1004 trial to detect a true outcome especially in view of the wide CIs. Therefore, these data suggest that previous and/or concurrent treatment with verteporfin PDT and pegaptanib may have a role in visual acuity maintenance and improvement. However, the data do not provide conclusive evidence regarding the effect of pegaptanib on maintenance or gain in visual acuity. Further, the number of participants with this outcome was small in both trials which make these comparisons unreliable. Our observations differ from the authors of the VISION trial who combined the results of the EOP 1003 and EOP 1004 in an individual patient data meta-analysis (Gragoudas 2004) and did not detect any influence of verteporfin PDT on the results.

In the individual patient data meta-analysis of the two trials (Gragoudas 2004), the authors reported that the proportion of participants with a prior history of treatment with verteporfin PDT was similar in all treatment groups since they combined data from the two trials and did not consider the differences in participant characteristics between the two trials. Also, they did not report an analysis examining possible interaction between treatment status with verteporfin PDT and the outcomes of maintenance of visual acuity or gain in 15 or more letters visual acuity at one year. We, on the other hand, evaluated the two trials separately and looked at the differences in participant characteristics between the two trials.

Sub-group analyses based on angiographic sub-types show that the overall beneficial effects of pegaptanib do not seem to have been contributed solely by any particular sub-type. Pegaptanib seems to be uniformly beneficial over the different angiographic sub-types considered. Though patients were stratified by angiographic subtype at baseline, the trial did not allow for adequate sample sizes to conclusively address questions on efficacy of pegaptanib in these sub-groups. Even in our meta-analysis, no definitive conclusions of benefit to any particular sub-type can be made from these subgroup analyses.

Visual acuity outcomes of our review have been further supported by the angiographic results. The fluorescein angiographic results of patients treated with pegaptanib showed a decrease in size of total lesion or choroidal neovascular complex. In addition, they also had less leakage compared with patients treated with sham injections

Vision and angiography outcomes with ranibizumab

Patients treated with ranibizumab were 86% less likely to lose 15 or more letters visual acuity than those treated with sham at one year follow-up. The effect was similar at both doses (0.3 mg and 0.5 mg). There was also a 72% reduction with ranibizumab in RR of blindness defined as visual acuity worse than 20/200 when compared with sham. Consistent with these results, patients treated with ranibizumab had higher mean final visual acuity compared with those treated with sham.

When compared with patients treated with verteporfin PDT, those treated with ranibizumab were 87% less likely to lose 15 or more letters at one year (ANCHOR). This effect was consistent with the final mean visual acuity levels, which were greater for patients in the ranibizumab groups (0.3 mg and 0.5 mg). Ranibizumab also resulted in smaller lesion sizes at the end of one year. It appears that the RR reduction with ranibizumab compared with sham is similar to the RR reduction with ranibizumab when compared with verteporfin PDT. However, this could be because of differences in types of participants included in the two trials.

Patients treated with ranibizumab were almost six times more likely to gain 15 or more letters of visual acuity at one year follow-up. This effect was apparent when compared with sham as well as with verteporfin PDT. Patients treated with 0.5 mg ranibizumab plus verteporfin PDT were four times more likely to gain 15 or more letters of visual acuity at one year compared with those treated with verteporfin PDT alone.

Combined therapy with 0.5 mg ranibizumab and verteporfin PDT resulted in a 70% reduction in RR of loss of 15 or more letters at one year when compared with patients treated with verteporfin PDT alone. Patients treated with 0.5 mg ranibizumab plus verteporfin PDT also had higher final mean visual acuity and smaller lesion sizes at one year follow-up.

Clinical improvement with ranibizumab was correlated with OCT in a study labeled PrONTO by Fung, et al (Fung 2007). In this study, the investigators found reduction in central retinal thickness by OCT on treatment with ranibizumab. Requirement for further injections of ranibizumab was guided by either loss of 5 or more letters of visual acuity or increase in retinal thickness by OCT. This study provides information on feasibility of using OCT to guide administration of ranibizumab at variable intervals as needed.

Quality of life and cost utility analyses

Improved scores (overall and different subscales) on the NEI-VFQ scale were reported with both pegaptanib and ranibizumab in analyses of the included trials. Comparison between treatments is facilitated by cost utility analyses.

Brown et al. conducted a cost utility analysis using data from the MARINA trial (Brown 2007b). They used standardized utility values correlating with visual acuity in the better seeing eye and costs from the 2006 National Physician Fee Schedule (available online at http://www.cms.hhs.gov/apps/ama/license.asp?file=/physician-feesched/downloads/pe_townhall_hcpcs_level.zip; http://www.cms.hhs.gov/apps/ama/license.asp?file=/ascpayment/downloads/ascpuf2006.zip). Calculating total value conferred by ranibizumab using a 12-year model generated by extrapolation of data from MARINA, the authors calculated a 10.4% improvement in quality of life on average. For patients treated with ranibizumab for affliction in the second eye where visual acuity in the first eye is already compromised due to some reason, the authors report a 15.8% improvement in quality of life.

By standardizing utilities of degree of visual loss Brown et al. (Brown 2007b), were able to compare ranibizumab with pegaptanib. They report that ranibizumab results in 169% greater quality of life when compared with pegaptanib based on comparison of standardized utilities. And, when compared with verteporfin PDT, ranibizumab delivers 95% greater value for treatment of occult and minimally classic subfoveal CNV.

However, all randomized trials included in this review have used fluorescein angiography and clinical examination to monitor lesion activity. It is unknown whether similar findings would have resulted if monitoring had been via OCT.

Using data provided by manufacturers on all trials included in this review, the National Institute for Health and Clinical Excellence (NICE) in the UK calculated incremental cost effectiveness ratios for pegaptanib and ranibizumab. This analysis is available on the World Wide Web at http://www.nice.org.uk/guidance/index.jsp?action=article&o=35000.

Dosing of therapy

Several factors encourage evaluation of efficacy of alternative and less frequent dosing regimens with anti-VEGF compounds. Some of these include concerns about ocular and systemic toxicity, patient and physician convenience with fewer intravitreal injections, and costs of treatment. Trials evaluating alternate dose strategies include PIER, CATT, IVAN and ABC. CATT and ABC are ongoing trials discussed in further detail below. Non-randomized studies that support less frequent dosing regimen guided by OCT include the PrONTO study (Fung 2007) as discussed above. Many retinal specialists in the United States perform intravitreal injections of ranibizumab every month for three months and then treat a patient as needed using worsening of visual acuity, presence of subretinal fluid and/or macular edema on the OCT, and occasionally changes on the fluorescein angiography as guidelines. The European Medicines Agency also recommended monthly injections for three months and then assessing a patient on the subsequent monthly visits. However, EMA recommended another injection if vision worsened (available online at http://www.emea.europa.eu/humandocs/PDFs/EPAR/lucentis/H-715-en1.pdf. Accessed February 19, 2008).

Ongoing trials evaluating ranibizumab and bevacizumab

Ranibizumab

PIER

Results from a randomized trial with quarterly administration of ranibizumab (PIER) are awaited (accepted for publication, per communication with Genentech). This trial compared 0.3 mg and 0.5 mg ranibizumab administered monthly for three months and then quarterly with sham treatment. Patients included in this trial had subfoveal CNV secondary to AMD. Analyses of visual acuity reported using unpublished data (Brown 2007a) indicate that fewer patients treated with ranibizumab lost 15 or more letters at 12 months compared with sham. Although a greater proportion gained 15 or more letters at 12 months compared with sham, the difference was less than that seen with monthly dosing in the ANCHOR and MARINA trials. However, full publication of results from this trial is awaited and will be included in an update of this review. Further details of this trial are listed in the table of ongoing studies.

SAILOR

This is described as a phase IIIb trial evaluating two doses of ranibizumab, 0.3 mg and 0.5 mg in patients with progressive subfoveal CNV secondary to AMD. Two cohorts of patients, 2000 in the first and 3000 in the second are being followed. While no results are available on visual acuity, analysis for systemic side effects revealed a statistically significant higher risk of stroke with 0.5 mg ranibizumab compared with the 0.3 mg dose of the medication (http://www.fda.gov/medwatch/safety/2007/Lucentis_DHCP_01-24-2007.pdf - accessed November 25, 2007). Patients with a previous history of stroke were reported to be at higher risk for a subsequent stroke according to this report.

Bevacizumab

Many uncontrolled studies have reported use of bevacizumab (Avastin) for AMD, as an off-label use of this medication (Avery 2006; Emerson 2007; Spaide 2006). Bevacizumab was originally approved for use in colorectal cancer. Significant savings in costs with bevacizumab have been reported in the literature (Raftery 2007). Ongoing studies evaluating bevacizumab compared with ranibizumab for AMD are discussed below.

CATT (Comparison of Age-Related Macular Degeneration Treatments Trials) compares bevacizumab and ranibizumab and is funded by the National Eye Institute of the National Institutes of Health. Starting January 2008, this trial will enroll 1200 patients aged 50 or older with subfoveal choroidal neovascularization secondary to AMD in at least one eye. Visual acuity at enrolment should be between 20/40 and 20/320. Participants will be followed monthly for 2 years. Participants will be randomized to one of the following four groups:

1. Ranibizumab on a fixed schedule every four weeks for one year with re-randomization at one year to schedule four week dosing or to variable dosing.

2. Fixed schedule bevacizumab of every four weeks for one year and re-randomization one year to scheduled four week dosing or to variable dosing.

3. Ranibizumab with variable dosing for two years based on monthly evaluation of signs of lesion activity.

4. Bevacizumab with variable dosing for two years based on monthly evaluation of signs of lesion activity.

The primary outcome for this study is change in visual acuity at one year. Secondary outcomes include number of treatments, change in visual acuity by 3 lines or 15 letters on ETDRS chart, change in subretinal fluid on OCT, change in lesion size by fluoresce in angiography, adverse effects and costs of treatment. Additional details are listed in the table of ongoing studies.

ABC

Currently recruiting, this trial includes patients with subfoveal CNV secondary to AMD. 130 participants will be randomized to intravitreal bevacizumab (1.25 mg or 0.05 ml) or to standard therapy with either verteporfin PDT or pegaptanib or sham. Intravitreal bevacizumab will be administered at six weekly intervals for 3 doses with retreatment every six weeks guided by standardized criteria. Gain in visual acuity will be the primary outcome for this trial. Additional details are listed in the table of ongoing studies.

VERITAS

This trial will include 340 patients at least 50 years of age with subfoveal CNV secondary to AMD. Participants will be randomized 1:1:1 to verteporfin PDT with either pegaptanib or one of two doses of intravitreal corticosteroid. The adjunct treatment will be administered the same day. Follow-up will be for 24 months. An abstract presented at a conference in May 2007 reported enrolment of 111 participants. Additional details of this ongoing trial are reported in the table of ongoing studies.

SUMMIT

This includes two clinical trials, DENALI in the United States and MONT BLANC in Europe. Both trials include participants 50 years of age or older with subfoveal CNV secondary to AMD. Participants with previous treatments for AMD will be excluded. Participants will be randomized to monthly 0.5 mg ranibizumab or verteporfin PDT standard fluence (light dose 50J/cm²) or to verteporfin PDT reduced fluence (light dose 25 J/cm²) at baseline. This baseline treatment will be supplemented with intravitreal 0.5 mg ranibizumab at months 1 and 2, and as needed from month 3 and with verteporfin PDT standard or reduced fluence as needed. Earlier studies have reported on safety of same-day treatment with 0.5 mg ranibizumab and verteporfin PDT (PROTECT).

IVAN

This randomised controlled trial of alternative treatments to Inhibit VEGF in Age-related choroidal Neovascularisation (IVAN), compares bevacizumab to ranibizumab. It is designed as a factorial trial sponsored by the Health Technology Assessment (HTA) Clinical Trials programme of the UK National Institute for Health Research (NIHR). Patients in this trial will be randomized to one of the two anti-VEGF agents with allocation to either continuous treatment or monthly treatment for the first three months followed by monthly review to detect disease relapse/progression.

Other anti-VEGF treatments in clinical trials

Aflibercept

Also called VEGF-Trap, it acts as a decoy receptor that binds VEGF preventing effects of the latter in diseases such as AMD. VEGF-Trap is generated by combining three portions of a VEGF receptor to a section of the human immunoglobulin IgG chain (Holash 2002). One randomized phase II study of this compound for neovascular AMD, CLEAR-IT-2, has been reported and is listed along with other studies awaiting assessment (CLEAR-IT-2; Nguyen 2007; Hariprasad 2007). Participants in this trial were randomized to 3 dose levels of VEGF-Trap at four or 12 week intervals. Additional details of this trial and any other trials evaluating this medication will be included in an update of this review.

Small Interfering RNA (siRNA)

This is a compound that shuts down a production of VEGF by interfering with communication between RNA and DNA. Also called bevasirinab or Cand5, one preliminary trial was presented at the American Academy of Ophthalmology (AAO) annual meeting in 2006 (Thompson 2006). 129 patients in this trial were randomized 1:1:1 to 0.2 mg, 1.5 mg and 3 mg Cand5 administered six weeks apart. While no details of visual acuity or safety results were reported in this abstract, published data on these outcomes are awaited. This is listed as a trial awaiting assessment.

Summary

Treatments for neovascular AMD with anti-VEGF compounds are developing at a rapid pace. The results of this systematic review indicate efficacy of pegaptanib and ranibizumab for treating neovascular AMD. Beneficial effects with these two medications are evident in terms of proportion with loss of 15 or more letters. Ranibizumab resulted in a greater proportion gaining 15 or more letters visual acuity at one year follow-up. These effects were consistent with angiographic findings following treatment with these medications. Considerations of costs and adverse effects were limited in the trials included in this systematic review. However, additional analyses conducted by other teams indicate a favourable cost utility ratio for ranibizumab. This systematic review also included a search for ongoing trials on anti-VEGF modalities for AMD. Questions that are being addressed by ongoing trials are evaluating alternate dosing regimens with ranibizumab, combined therapy with ranibizumab and verteporfin PDT, comparison of bevacizumab with ranibizumab, and other anti-VEGF modalities including VEGF-Trap and siRNA.

Authors' Conclusions

Implications for practice

Pegaptanib and ranibizumab are associated with reduced risk of visual acuity loss in patients with neovascular AMD. Ranibizumab also results in improved visual acuity in a substantial proportion of eyes. There is no direct evidence that one of these treatment strategies is better than the other. We found no trial directly comparing pegaptanib with ranibizumab. Although ranibizumab is associated with more improvement in visual acuity and vision-specific quality of life, it is more costly compared to pegaptanib. However, cost utility analyses conducted so far are in favour of ranibizumab over pegaptanib. Off-label use of bevacizumab for neovascular AMD has been studied in many uncontrolled trials and bevacizumab is a potentially attractive alternative to ranibizumab because of lower costs. Trials comparing bevacizumab with ranibizumab are currently under way and should provide data on efficacy, quality of life and cost-effectiveness of bevacizumab.

Implications for research

Pegaptanib introduced a new treatment strategy for neovascular AMD: ocular pharmacology. Ranibizumab has been shown to have a significant benefit in preventing loss of visual acuity as well as improving visual acuity. Adverse effects with these two medications with longer term follow-up must be quantified and documented to provide a better understanding of their risk-benefit profile to patients with AMD. Trials comparing ranibizumab with be-vacizumab will provide information on possible alternatives that are more cost-effective for patients. These and other ongoing trials evaluating various other anti-VEGF compounds such as VEGF-Trap, siRNA and tyrosine kinase inhibitors should focus on gain in visual acuity, quality of life, and cost-effectiveness compared with the evolving best standard therapies.

Plain Language Summary.

Antiangiogenic therapy with anti-vascular endothelial growth factor modalities for neovascular age-related macular degeneration

Age-related macular degeneration (AMD) is a common cause of severe vision loss in people 55 years and older. Neovascular AMD, which involves abnormal growth of blood vessels in the back of the eye, accounts for most AMD-related severe vision loss. Medications such as pegaptanib and ranibizumab that block this abnormal growth of blood vessels in the back of the eye are one way to treat this condition. Fewer patients treated with pegaptanib lost 15 or more letters visual acuity at one year. Ranibizumab alone and when combined with verteporfin photodynamic therapy (PDT) resulted in fewer patients losing 15 or more letters visual acuity at one year. Approximately seven patients need to be treated with 0.3 mg or 0.5 mg pegaptanib to prevent loss of 15 or more letters visual acuity in one patient. This number is about 14 for 3 mg pegaptanib. In contrast just over three patients need to be treated with either 0.3 mg or 0.5 mg doses of ranibizumab to prevent loss of 15 or more letters visual acuity. Very few patients treated with pegaptanib gained visual acuity. A greater proportion of patients treated with ranibizumab gained 15 of more letters visual acuity at one year compared with sham or verteporfin PDT. No trial directly compared pegaptanib and ranibizumab. Pegaptanib and ranibizumab are beneficial for treatment of neovascular AMD and their use is associated with few adverse effects. Trials on other agents that block abnormal growth of blood vessels in this condition are ongoing and will be included in updates of the review.

Characteristics of Studies

Characteristics of included studies [ordered by study ID]

ANCHOR.

Methods	Method of randomization: Method of randomization: Using a dynamic randomisation method, stratified by study center and visual acuity scores on day 0 (<45 letters vs >=45 letters) Method of allocation concealment: Not clearly reported, authors communicated it was by central randomization Masking Participants – yes Care-givers – yes Outcome assessors - yes Number randomized: 140 to 0.3 mg ranibizumab group, 140 to 0.5 mg ranibizumab group and 143 to verteporfin group Exclusions after randomization: 3 participants in the 0.3 mg group did not receive treatment after randomization, one because of patient's decision and two based on physician's decision Number analyzed: 140 in 0.3 mg ranibizumab group, 139 in 0.5 mg ranibizumab group and 143 in verteporfin group Losses to follow-up: 10 in 0.3 mg group, 5 in 0.5 mg and 10 in verteporfin group. Reasons included death, adverse events, loss to follow-up, patient's decision, physician's decision and patient noncompliance Intention to treat analysis: yes, using last observation carried forward for missing data Unit of analysis: Individuals Reported power calculation: yes
Participants	Country: USA, France, Germany, Hungary, Czech Republic, and Australia Age: Mean (SD, range) was 77.4 (7.5, 54-97) in 0.3 ranibizumab group, 76 (8.6, 54-93) in the 0.5 mg ranibizumab group and 77.7 (7.8, 53-95) in verteporfin PDT group Gender: 47.9 in 0.3 mg group, 46.4 in the 0.5 mg group and 55.3 in verteporfin PDT group were females Inclusion criteria: At least 50 years of age with subfoveal CNV secondary to AMD determined independently based on fluorescein angiography and fundus photography to be predominantly classic in composition and suitable for treatment with verteporfin PDT, a lesion whose total size was not greater than 5400 microns in greater linear dimension in the study eye, best-corrected visual acuity of 20/40 to 20/320 Snellen equivalent ETDRS charts), no permanent structural damage to the central fovea, no previous therapy that might compromise an assessment of study treatment including previous treatment with verteporfin PDT. Patients with juxta- or extra-foveal photocoagulation in the study eye more than 1 month prior to day 0 and prior verteporfin PDT in the nonstudy eye more than 7 days before study day 0 were included Exclusion criteria: Previous treatment with verteporfin PDT, external-beam radiation therapy, or transpupillary thermotherapy in the study eye, previous participation in a clinical trial involving anti-angiogenic drugs, treatment with verteporfin PDT in the nonstudy eye less than 7 days preceding study day 0, prior intravitreal drug delivery or subfoveal focal laser photocoagulation in the study eye, laser photocoagulation in the study eye within 1 month preceding study day 0, history of vitrectomy surgery or submacular surgery or other surgical intervention for AMD in study eye, participation in any studies of investigational drugs within 1 month preceding study day 0, subretinal hemorrhage in study eye involving center of the fovea if the size of hemorrhage is either 50 % or more of the total lesion area or 1 or more standard disc areas in size, subfoveal fibrosis or atrophy in study eye, choroidal neovascularization in either eye due to other causes, retinal pigment epithelia tear involving the macula in the study dye, any concurrent intraocular condition in study eye that in the opinion of the investigator could require medical or surgical intervention during the 24-month study period to prevent or treat visual loss resulting from that condition, or if allowed to progress untreated, may contribute to loss of at least 2 Snellen equivalent lines of BCVA over the 24-month study period, active intraocular inflammation current vitreous hemorrhage or history of rhegmatogenous retinal detachment or macular hole (stage 3 or 4) in the study eye, history of idiopathic or autoimmune-associated uveitis in either eye, history of infectious conjunctivitis, keratitis, scleritis, or endophthalmitis in either eye, intraocular surgery in study eye within 2 months preceding study day 0, aphakia or absence of posterior capsule in study eye, spherical equivalent of refractive error in study eye greater than -8 diopters of myopia, uncontrolled glaucoma in study eye, history of glaucoma filtering surgery or corneal transplant in study eye, premenopausal women not using adequate contraception, history of other disease, metabolic dysfunction, physical examination of laboratory finding giving reasonable suspicion of a condition that contraindicates use of an investigational drug or that might affect interpretation of the results of the study or place the participant at a high risk for complications, current treatment for active systemic infection, history of allergy to fluorescein not amenable to treatment, inability to obtain fundus photographs or fluorescein angiogram of sufficient quality to be analyzed and graded, inability to comply with study or follow-up procedures Equivalence of baseline characteristics: A slightly higher percentage of patients in 0.3 mg group were aged 75-84 (60% compared with 45.7% in 0.5 mg group and 51.7% in verteporfin PDT group)
Interventions	Intervention 1: 0.3 mg ranibizumab plus sham verteporfin (intravenous infusion of saline followed by laser irradiation of macula) Intervention 2: 0.5 mg ranibizumab plus sham verteporfin Intervention 3: Sham intravitreal injection plus active verteporfin therapy (laser irradiation of macula following intravenous administration of verteporfin) Ranibizumab was injected into the study eye at a monthly interval (ranging from 23 to 37 days) for a total of 12 injections in the first year beginning on day 0. Either verteporfin or sham verteporfin was administered on day 0 and then if needed on the basis of investigators' evaluation of angiography at months 3, 6, 9, or 12 Length of follow-up: 12 months
Outcomes	Primary outcome: Proportion of patients who at 12 months lost fewer than 15 letters from baseline visual acuity in study eye Other outcomes reported: Proportion of patients who gained 15 or more letters from baseline, proportion of patients with a Snellen equivalent of 20/200 or worse and mean change from baseline (letters over time); mean change from baseline to month 12 in the size of the classic CNV component and total area of leakage from CNV Exploratory efficacy end points: Loss of 30 or more letters visual acuity, mean changes in area of CNV and area of the entire lesion Safety assessments: IOP measurement before and 50-70 minutes after each study treatment, ocular and nonocular adverse events, changes and abnormalities in clinical laboratory parameters and vital signs, and immunoreactivity to ranibizumab Reported quality of life indicators: yes Intervals at which outcome were assessed: “at regularly scheduled study visits” 12 and 24 months, angiography evaluation was performed at months 3, 6, 9, 12
Notes	Funding: Genentech and Novartis Pharma
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A-Adequate

EOP 1003.

Methods	Method of randomization: stochastic treatment allocation algorithm based on the variance method Method of allocation concealment: Centralized randomization where the study coordinator was instructed the code of the medication for the patient after determining her eligibility. The medication packet was not opened until just before administering the injection Masking: Participants- yes Care providers- examiner- yes; injector- no Outcome assessors- yes Number randomized: 144 to 0.3 mg pegaptanib, 146 to 1 mg pegaptanib, 143 to 3 mg pegaptanib and 145 to placebo groups Exclusions after randomization: None Number analyzed: 144 in 0.3 mg pegaptanib, 146 in 1 mg pegaptanib, 143 in 3 mg pegaptanib and 145 in placebo groups for the primary outcome alone Losses to follow-up: 11 in placebo group, 12 in 0.3 mg pegaptanib group, 17 in 1 mg pegaptanib group, 20 in 3 mg pegaptanib group discontinued therapy during the trial Intention to treat analysis: Reported an intention to treat analysis only for the primary outcome Unit of analysis: Individuals Reported power calculations: yes
Participants	Country: USA, Canada Age: Mean age was 78, 76.5, 77.1, and 76.7 years in 0.3 mg pegaptanib, 1 mg pegaptanib, 3 mg pegaptanib and placebo groups respectively Gender: 56%, 53%, 69%, and 57% in 0.3 mg pegaptanib, 1 mg pegaptanib, 3 mg pegaptanib and placebo groups respectively, were females Inclusion criteria: age greater than or equal to 50 years; subfoveal CNV secondary to AMD; best corrected visual acuity of 20/40 to 20/320 in the treated eye and greater than 20/800 in the fellow eye; CNV lesion may be predominantly classic, minimally classic, occult with no classic; size of lesion < 12 disc areas (including blood, scar/atrophy, neovascularization); no greater than 50% of lesion could be due to subretinal hemorrhage and 50% of lesion had to be due to CNV; for occult lesions, lesions had to be subretinal and no greater than 50% of total lesion area, or presence of lipid or loss of 15 or more letters of visual acuity during previous 12 weeks; patients were eligible even if they received 1 PDT treatment if it was atleast 8-12 weeks prior to enrollment; intraocular pressure < 23mmHg; adequate pupil dilation; clear media Exclusion criteria: atrophy exceeding 20% of total lesion or subfoveal scarring; previous thermal laser; therapy with another investigational drug; likelihood of requiring cataract removal within 2 years; other potential causes of CNV including high myopia, ocular histoplasmosis, angiod streaks, choroidal rupture, multifocal choroiditis, any intraocular surgery within 3 months or extrafoveal/juxtafoveal laser within 2 weeks of study entry or posterior vitrectomy or scleral buckle or presence of intraretinal tears or rips; concomitant presence of diabetic retinopathy, severe cardiac disease, myocardial infarction within 6 months, ventricular tachycardia requiring treatment, unstable angina, evidence of peripheral vascular disease, stroke within 12 months, acute or chronic periocular infection, previous therapeutic radiation to eye/head/neck; any treatment with any investigational agent within last 30 days; serious allergies to fluoroscein dye or indocyanin green or components of pegaptanib Equivalence of baseline characteristics: The treatment groups were similar with respect to age, gender, race, smoking status, angiographic subtypes, prior treatment status with photodynamic therapy, and ETDRS visual acuity scores
Interventions	Treatment: Intravitreal injection of pegaptanib at dosages of either 0.3 mg, 1.0 mg, or 3.0 mg given every 6 weeks over period of 48 weeks Control: sham injection with patients treated identically with exception of scleral penetration with the needle Length of follow-up: 54 weeks
Outcomes	Primary outcome: proportion of patients losing fewer than 15 letters of visual acuity between baseline and week 54 Other outcomes reported: Gain of 3 or more lines visual acuity, maintenance of visual acuity or gain of 0 lines visual acuity, mean visual acuity, legal blindness, loss of 30 or more letters visual acuity, size of lesion, and total CNV size Reported quality of life indicators: yes Intervals at which outcome assessed: every 6 weeks before treatment with main assessment analyzed after 54 weeks
Notes	Funding: Eyetech Pharmaceuticals and Pfizer
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A-Adequate

EOP 1004.

Methods	Method of randomization: stochastic treatment allocation algorithm based on the variance method Method of allocation concealment: Centralized randomization where the study coordinator was instructed the code of the medication for the patient after determining her eligibility. The medication packet was not opened until just before administering the injection Masking: Participants- yes Care providers- examiner- yes; injector- no Outcome assessors- yes Number randomized: 151 to 0.3 mg pegaptanib, 155 to 1 mg pegaptanib, 153 to 3 mg pegaptanib and 153 to placebo groups Exclusions after randomization: None Number analyzed: 151 in 0.3 mg pegaptanib, 155 in 1 mg pegaptanib, 153 in 3 mg pegaptanib and 153 in placebo groups for the primary outcome alone Losses to follow-up: 12 in placebo group, 11 in 0.3 mg pegaptanib group, 13 in 1 mg pegaptanib group, 17 in 3 mg pegaptanib group discontinued therapy during the trial Intention to treat analysis: yes except don't know why 18 patients were excluded after randomization Unit of analysis: Individuals Reported power calculations: yes
Participants	Country: US, Canada, Europe, Israel, Australia, South America Age: Mean age was 74.9, 74.5, 75.4, and 74.9 years in 0.3 mg pegaptanib, 1 mg pegaptanib, 3 mg pegaptanib and placebo groups, respectively Gender: 54%, 56%, 61%, and 63% in 0.3 mg pegaptanib, 1 mg pegaptanib, 3 mg pegaptanib and placebo groups, respectively were females Inclusion criteria: age greater than or equal to 50 years; subfoveal CNV secondary to AMD; best corrected visual acuiyt of 20/40 to 20/320 in the treated eye and greater than 20/800 in the fellow eye; CNV lesion may be predominantly classic, minimally classic, occult with no classic; size of lesion < 12 disc areas (including blood, scar/atrophy, neovascularization); no greater than 50% of lesion could be due to subretinal hemorrhage and 50% of lesion had to be due to CNV; for occult lesions, lesions had to be subretinal and no greater than 50% of total lesion area, or presence of lipid or loss of 15 or more letters of visual acuity during previous 12 weeks; patients were eligible even if they received 1 PDT treatment if it was atleast 8-12 weeks prior to enrollment; intraocular pressure < 23mmHg; adequate pupil dilation; clear media Exclusion criteria: atrophy exceeding 20% of total lesion or subfoveal scarring; previous thermal laser; therapy with another investigational drug; likelihood of requiring cataract removal within 2 years; other potential causes of CNV including high myopia, ocular histoplasmosis, angiod streaks, choroidal rupture, multifocal choroiditis, any intraocular surgery within 3 months or extrafoveal/juxtafoveal laser within 2 weeks of study entry or posterior vitrectomy or scleral buckle or presence of intraretinal tears or rips; concomitant presence of diabetic retinopathy, severe cardiac disease, myocardial infarction within 6 months, ventricular tachycardia requiring treatment, unstable angina, evidence of peripheral vascular disease, stroke within 12 months, acute or chronic periocular infection, previous therapeutic radiation to eye/head/neck; any treatment with any investigational agent within last 30 days; serious allergies to fluoroscein dye or indocyanin green or components of pegaptanib Equivalence of baseline characteristics: The treatment groups were similar with respect to age, gender, race, smoking status, angiographic subtypes, prior treatment status with photodynamic therapy, and ETDRS visual acuity scores
Interventions	Treatment: Intravitreal injection of pegaptanib at dosages of either 0.3 mg, 1.0 mg, or 3.0 mg given every 6 weeks over period of 48 weeks Control: sham injection with patients treated identically with exception of scleral penetration with the needle Length of follow-up: 54 weeks
Outcomes	Primary outcome: proportion of patients losing fewer than 15 letters of visual acuity between baseline and week 54 Other outcomes reported: Gain of 3 or more lines visual acuity maintenance of visual acuity or gain of 0 lines visual acuity mean visual acuity legal blindness, loss of 30 or more letters visual acuity size of lesion, and total CNV size Reported quality of life indicators: yes Intervals at which outcome assessed: every 6 weeks before treatment with main assessment analyzed after 54 weeks
Notes	Funding: Eyetech Pharmaceuticals and Pfizer
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A-Adequate

FOCUS.

Methods	Method of randomization: Randomization schedule was generated using a static randomization method stratified by study site and a block size to maintain the 2:1 ratio between ranibizumab + PDT versus PDT alone Method of allocation concealment: “Physicians and personnel were unaware of the allocation sequences prior to randomization.” (email communication with Dr. Heier) Masking Participants - no Care-givers - no Outcome assessors - yes Number randomized: 106 to verteporfin (PDT) + ranibizumab and 56 to PDT alone group Exclusions after randomization: 1 patient in control group (ranibizumab + PDT) was excluded after randomization, no reason for exclusion reported Number analyzed: 105 in verteporfin (PDT) + ranibizumab group and 56 in PDT alone group for the primary outcome Losses to follow-up: 7 in PDT +ranibizumab group and 3 in PDT alone group. Reasons included adverse events, unavailability to follow-up, patient's decision and patient's condition requiring other therapy Intention to treat analysis: yes, using last-observation-carried-forward method Unit of analysis: Individuals Reported power calculation: Yes, but it is unclear if this was an a priori calculation
Participants	Country: USA Age: 50-93 years; mean (SD) was 74.7 (7.2) in PDT + ranibizumab group and 73 (8.7) in PDT alone group Gender: 56.6% in PDT + ranibizumab group and 46.4% in PDT alone group were females Inclusion criteria: At least 50 years of age with primary or recurrent subfoveal CNV secondary to AMD determined to be predominantly classic in composition, lesion total size not greater than 5400 microns in greater linear dimension in the study eye and suitable for treatment with PDT, best-corrected visual acuity of 20/40 to 20/320 Snellen equivalent ETDRS charts) Exclusion criteria: History of verteporfin PDT 3 months preceding study for the study eye or 7 days preceding study for non-study eye, more than 3 previous verteporfin PDT treatments in 12 months preceding study, juxtafoveal or extrafoveal laser photocoagulation within 1 month, previous subfoveal laser photocoagulation, external beam radiation therapy or transpupillary thermotherapy at any time, vitrectomy, submacular surgery or other surgical intervention for AMD, prior participation in a clinical trial involving anti-angiogenic treatment of either eye, or any investigational drug in the preceding 1 month, permanent structural damage to center of fovea in the study eye, concurrent ocular or systemic condition that can contraindicate administration of investigational drug, verteporfin or fluorescein, or place the patient at high risk for treatment complications. Equivalence of baseline characteristics: yes
Interventions	Treatment: Verteporfin was administered on day 0, 0.5 mg intravitreal ranimizumab was injected starting on day 7 and every month for up to a total of 24 injections in 2 years Control: Verteporfin was administered on day 0, and sham injection was administered on day 7 and monthly Length of follow-up: 12 months
Outcomes	Primary outcome: Visual acuity measured as proportion of patients with loss of fewer than 15 letters in the study eye from baseline Other outcomes reported: Proportion of patients with gain of 15 or more letters, visual acuity 20/200 or worse, mean change in visual acuity from baseline; total areas of CNV lesion, fluorescein leakage from CNV lesion plus intense, progressive staining of retinal pigment epithelium, secondary sensory retinal detachment/subretinal fluid, and number of verteporfin PDT repeated treatments of study eye during the first treatment year. Safety outcomes included ocular and nonocular adverse events and proportion of patients developing immunoreactivity to ranimizumab, intraocular inflammation, and IOP Reported quality of life indicators: none Intervals at which outcome assessed: Months 3, 6, 9, 12, 18, and 21.
Notes	Funding: Genentech and Novartis Pharma
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A-Adequate

MARINA.

Methods	Method of randomization: Using a dynamic randomization algorithm, stratified according to study center, by visual acuity scores on day 0 (<55 letters vs >=55 letters), and by choroidal neovascularization subtype (minimally classic or occult with no classic). Method of allocation concealment: Central randomization Masking: Participants- yes Care providers- yes, except the injecting ophthalmologist who was unmasked Outcome assessors- yes Number randomized: 238 to 0.3 mg ranibizumab group, 240 to 0.5 mg ranibizumab group and 238 to sham control group Exclusions after randomization: 1 participant in the 0.5 mg ranibizumab group and 2 participants in the sham control group did not receive treatment after randomization. Reasons included patient apprehensiveness about intravitreal injection and randomization ahead of schedule before safety eligibility confirmation Losses to follow-up: 10 in 0.3 mg ranibizumab group, 5 in 0.5 mg ranibizumab and 10 in control group. Reasons included death, adverse events, loss to follow-up, patient's decision, physician's decision and patient noncompliance Intention to treat analysis: yes, using the last observation carried forward method Unit of analysis: Individuals Reported power calculations: yes
Participants	Country: USA Age: Mean (SD, range) was 77 (8, 52-95) in 0.3 ranibizumab group, 77 (8, 52-93) in the 0.5 mg ranibizumab group and 77 (7, 56-94) in control group Gender: 64.3% in 0.3 mg group, 63.3 in the 0.5 mg group and 66.8 in control group were females Inclusion criteria: At least 50 years of age with active primary or recurrent subfoveal lesions with CNV secondary to AMD in the study eye. An active lesion was defined by any of the below criteria: (1) exhibiting at least a 10% increase in lesion size determined by comparing a fluorescein angiogram performed within 1 month preceding study day 0 with a fluorescein angiogram performed within 6 months preceding study day 0, (2) resulting in a visual acuity loss of greater than 1 Snellen any time within the prior 6 months, or (3) subretinal hemorrhage associated with CNV within 1 month preceding study day 0. Patients with lesions with an occult CNV component were also included but for patients with concomitant classic CNV, the area of classic CNV must have been less than 50% of the total lesion size. Other inclusion criteria were the total area of CNV encompassed within the lesion must be 50% or more of the total lesion area, the total lesion area must be 12 disc areas or less in size, best-corrected visual acuity, using ETDRS chars of 20/40 to 20/320 (Snellen equivalent)in the study eye Exclusion criteria: Major exclusion criteria included prior treatment with verteporfin, external-beam radiation therapy, or transpupillary thermotherapy in the study eye, previous participation in a clinical trial involving anti-angiogenic drugs, treatment with verteporfin in the nonstudy eye less than 7 days preceding study day 0, previous intravitreal drug delivery or subfoveal focal laser photocoagulation in the study eye, laser photocoagulation in the study eye within 1 month preceding study day 0, history of vitrectomy surgery in the study eye, history of submacular surgery or other surgical intervention for AMD in study eye, participation in any studies of investigational drugs within 1 month preceding study day 0, subretinal hemorrhage in study eye involving center of the fovea if the size of hemorrhage is either 50 % or more of the total lesion area or 1 or more disc areas in size, subfoveal fibrosis or atrophy in study eye, CNV in either eye due to other causes, retinal pigment epithelia tear involving the macula in the study eye Equivalence of baseline characteristics: yes
Interventions	Intervention 1: 0.3 mg ranibizumab Intervention 2: 0.5 mg ranibizumab Control: Sham injection Verteporfin photodynamic therapy for the study eye was allowed if the choroidal neovascularization converted to a predominantly classic pattern. Length of follow-up: 2 years
Outcomes	Primary outcome: Proportion of patients who at 12 months lost fewer than 15 letters from baseline visual acuity in study eye Other outcomes reported: Proportion of patients who gained 15 or more letters from baseline, proportion of patients with a Snellen equivalent of 20/200 or worse and mean change from baseline (letters over time); mean change from baseline to month 12 in the size of the classic CNV component and total area of leakage from CNV. Exploratory outcomes included proportion of patients with visual acuity 20/40 or better, and 20/20 at 12 and 24 months (Snellen equivalent), total area of and change from baseline CNV, area of leakage Safety assessments: IOP measurement 60 minutes after each injection, incidence and severity of ocular and nonocular adverse events, changes and abnormalities in clinical laboratory parameters and vital signs, and immunoreactivity to ranibizumab Reported quality of life indicators: none Intervals at which outcomes assessed: outcomes analyzed at 12 months and 24 months were reported
Notes	Funding: Genentech and Novartis Pharma
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A-Adequate

AMD: age-related macular degeneration

CNV: choroidal neovascularization

IOP: intraocular pressure

PDT: photodynamic therapy

Characteristics of excluded studies [ordered by study ID]

Aggio 2006	Not a randomized trial
Barouch 2004	Review article
Bashshur 2007	Follow-up less than one year
Eyetech Study Group	Phase II uncontrolled study
Fine 2005	Review article
Hahn 2007	Follow-up less than 1 year
Heier 2006	Not randomized
Kim 2004	Review article
Lazic 2007	Follow-up less than one year
Michels 2005a	Uncontrolled, open-label study
Nguyen 2006	Not randomized
Rakic 2005	Letter discussing an included trial
Razavi 2002	Review article
Rosenfeld 2005	Open-label, uncontrolled trial
Scott 2007	Editorial
Siddiqui 2005	Review article

Characteristics of ongoing studies [ordered by study ID]

ABC.

Trial name or title	ABC ISRCTN83325075
Methods
Participants	Men and women aged over 50 years with primary or subfoveal CNV lesions secondary to AMD in the study eye; best corrected visual acuity of 20/40 to 20/320 in the study eye using ETDRS charts, total lesion size less than 12 disc areas, area of fibrosis less than 25% of total lesion area, occult lesions are included if there is evidence of progression as defined for the study
Interventions	Intravitreal bevacizumab with placebo PDT where necessary to maintain masking Verteporfin PDT with sham intravitreal injection Intravitreal pegaptanib Sham intravitreal injection
Outcomes	Primary outcome: Proportion of patients who gain more than 15 letters visual acuity at 12 months Secondary outcomes: Proportion of patients losing fewer than 15 letters vision and proportion gaining five letters or more vision Proportion of patients who meet the above visual acuity criteria at 6 months Safety and tolerability of intravitreal injections of bevacizumab given every 6 weeks Mean change in central OCT thickness
Starting date	11/8/2006
Contact information	Mr Adnan Tufail Moorfields Eye Hospital NHS Foundation Trust 162 City Road London United Kingdom EC1V 2PD
Notes

CATT.

Trial name or title	CATT
Methods
Participants	Men and women aged 50 or older with subfoveal CNV secondary to AMD in at least one eye, visual acuity between 20/40 and 20/320.
Interventions	Ranibizumab on a fixed schedule of every 4 weeks for 1 year, at 1 year, re-randomization to ranibizumab every 4 weeks or to variable dosing Bevacizumab on a fixed schedule of every 4 weeks for 1 year, at 1 year, re-randomization to bevacizumab every 4 weeks or to variable dosing Ranibizumab on variable dosing for 2 years Bevacizumab on variable dosing for 2 years
Outcomes	Primary outcome: Change in visual acuity at 1 year Secondary outcomes: Number of treatments 3-line change in visual acuity (15 letters on ETDRS chart) Change in subretinal and intraretinal fluid on OCT Change in lesion size on fluorescein angiography Incidence of complications of treatment Cost of treatment
Starting date	January 2008
Contact information	Ellen Peskin, University of Pennsylvania peskin@mail.med.upenn.edu
Notes

IVAN.

Trial name or title	IVAN
Methods
Participants	Men and women 50 years and older who are newly referred for treatment of CNV caused AMD in the first or second eye involving the center of the fovea, corrected logMAR visual acuity at least 25 letters read using a standard ETDRS chart
Interventions	Factorial design with patients randomized to either ranibizumab or bevacizumab: continuous treatment with a VEGF inhibitor for 24 months or monthly treatment for the first 3 months or monthly administration for the first 3 months, followed by monthly review to detect disease relapse or progression
Outcomes	Primary outcome for this trial is corrected logMAR visual acuity after two years follow-up. Secondary outcomes at 1 year and 2 years follow-up include: 1. adverse effects, 2. quality of life, 3. cost-effectiveness, 4. treatment satisfaction, 5. clinical measures of vision, 6. CNV morphology, 7. distance logMAR visual acuity, 8.survival free from treatment failure
Starting date
Contact information	http://www.ivan-trial.co.uk/Default.aspx Clinical Trials and Evaluation Unit University of Bristol Level 7, Bristol Royal Infirmary Upper Maudlin Street Bristol BS2 8HW, info@ivan-trial.co.uk
Notes

PIER.

Trial name or title	PIER
Methods
Participants	Men and women older than 50 with “wet” AMD with best corrected visual acuity equal to or worse than 20/40 but not worse than 20/320; classic or occult CNV due to AMD
Interventions	Sham injection Lucentis dose A. Lucentis dose B Three monthly injections are given followed by injections every 3 months thereafter
Outcomes
Starting date	September 2004
Contact information	Genentech (888) 662-6728
Notes

SAILOR.

Trial name or title	SAILOR (ClinicalTrials.gov Identifier: NCT00251459)
Methods
Participants	Two cohorts (2000 in first and 3000 in second) of men and women older than 50 years of age with progressive subfoveal CNV secondary to AMD and best corrected visual acuity (ETDRS charts for cohort 1 and Snellen charts for cohort 2) of 20/40 to 20/320 Snellen equivalent in the study eye. Disease progression is defined as loss of 5 or more letters best corrected visual acuity within 6 months preceding enrollment, increase in lesion area, subretinal hemorrhage associated with CNV within 1 month preceding enrollment, and classic CNV constituting more than 50% of CNV lesion area
Interventions	0.3 mg ranibizumab 0.5 mg ranibizumab
Outcomes	Primary outcomes: Cohort 1 & 2: Incidence of safety adverse effects (ocular and non-ocular)at 12 months follow-up Secondary outcomes: Proportion of subjects losing 15 or more letters in best corrected visual acuity at month 12 Mean time to retreatment following initial three monthly loading doses Mean total number of injections through month 12 of follow-up
Starting date	Cohort 1: November 2005 Cohort 2: March 2006
Contact information	Genentech (888) 662-6728
Notes

SUMMIT.

Trial name or title
Methods
Participants	Men and women aged 50 years or older, with any type of subfoveal CNV secondary to AMD, lesion size less than 9 disc areas, naive to AMD treatment, best corrected visual acuity of 73-24 ETDRS letter score
Interventions	Verteporfin standard fluence at baseline plus intravitreal injection of 0.5 mg ranibizumab at baseline, and months 1 and 2, followed by ranibizumab as needed starting month 3 and verteporfin standard or reduced fluence as needed Verteporfin reduced fluence at baseline plus intravitreal injection of 0.5 mg ranibizumab at baseline, and months 1 and 2, followed by ranibizumab as needed starting month 3 and verteporfin standard or reduced fluence as needed Monthly 0.5 mg ranibizumab monotherapy
Outcomes	Best-corrected visual acuity, fluorescein angiography, OCT measurements, treatment burden of combination therapy including number of retreatments, time to first retreatment, treatment-free interval, health economics outcomes
Starting date
Contact information
Notes

VERITAS.

Trial name or title	VERITAS
Methods
Participants	Patients with subfoveal CNV of any type with CNV more than or equal to 50% of total lesion, aged 50 years or older, best corrected visual acuity ETDRS letter score of 73-24.Exclusion criteria include prior treatment for CNV in study eye and CNV secondary to other causes than AMD
Interventions	Verteporfin PDT with dose 1 of an intravitreal steroid Verteporfin PDT with dose 2 of an intravitreal steroid Verteporfin PDT with an anti-VEGF medication
Outcomes	Outcomes: Specific outcomes not reported in the abstract
Starting date	Unclear
Contact information
Notes

AMD: Age-related macular degeneration

CATT: Comparison of Age-related macular degeneration Treatment Trials

CNV: Choroidal neovascularization

IVAN: A randomised controlled trial of alternative treatments to Inhibit VEGF in Age-related choroidal Neovascularisation

OCT: Optical coherence tomography

PIER: See Table 5

SAILOR: See Table 5

SUMMIT: See Table 5

VERITAS: See Table 5

Data and Analyses

Comparison 1. Pegaptanib vs sham.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Loss of 15 or more letters visual acuity at one year follow-up (patients with measurement at one year)	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
1.1 0.3 mg Pegaptanib versus sham	2	546	Risk Ratio (M-H, Fixed, 95% CI)	0.66 [0.53, 0.83]
1.2 1 mg Pegaptanib versus sham	2	550	Risk Ratio (M-H, Fixed, 95% CI)	0.64 [0.51, 0.80]
1.3 3 mg Pegaptanib versus sham	2	539	Risk Ratio (M-H, Fixed, 95% CI)	0.84 [0.69, 1.04]
1.4 Pegaptanib versus sham	2	1085	Risk Ratio (M-H, Fixed, 95% CI)	0.71 [0.60, 0.84]
2 Loss of 30 or more letters of visual acuity at one year follow-up	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
2.1 0.3 mg Pegaptanib versus sham	2	590	Risk Ratio (M-H, Fixed, 95% CI)	0.43 [0.29, 0.65]
2.2 1 mg Pegaptanib versus sham	2	596	Risk Ratio (M-H, Fixed, 95% CI)	0.36 [0.23, 0.57]
2.3 3 mg Pegaptanib versus sham	2	592	Risk Ratio (M-H, Fixed, 95% CI)	0.62 [0.43, 0.88]
2.4 Pegaptanib versus sham	2	1186	Risk Ratio (M-H, Fixed, 95% CI)	0.47 [0.35, 0.63]
3 Visual acuity worse than 20/200 at one year follow-up	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
3.1 0.3 mg Pegaptanib versus sham	2	590	Risk Ratio (M-H, Fixed, 95% CI)	0.62 [0.49, 0.77]
3.2 1 mg Pegaptanib versus sham	2	596	Risk Ratio (M-H, Fixed, 95% CI)	0.75 [0.61, 0.92]
3.3 3 mg Pegaptanib versus sham	2	592	Risk Ratio (M-H, Fixed, 95% CI)	0.71 [0.58, 0.88]
3.4 Pegaptanib versus sham	2	1186	Risk Ratio (M-H, Fixed, 95% CI)	0.69 [0.59, 0.82]
4 Gain of 15 or more letters of visual acuity at one year follow-up	2		Risk Ratio (M-H, Fixed, 95% CI)	Totals not selected
4.1 0.3 mg Pegaptanib versus sham	2		Risk Ratio (M-H, Fixed, 95% CI)	Not estimable
4.2 1 mg Pegaptanib versus sham	2		Risk Ratio (M-H, Fixed, 95% CI)	Not estimable
4.3 3 mg Pegaptanib versus sham	2		Risk Ratio (M-H, Fixed, 95% CI)	Not estimable
4.4 Pegaptanib versus sham	2		Risk Ratio (M-H, Fixed, 95% CI)	Not estimable
5 Maintenance of visual acuity at one year follow-up	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
5.1 0.3 mg Pegaptanib versus sham	2	590	Risk Ratio (M-H, Fixed, 95% CI)	1.47 [1.13, 1.92]
5.2 1 mg Pegaptanib versus sham	2	596	Risk Ratio (M-H, Fixed, 95% CI)	1.62 [1.25, 2.10]
5.3 3 mg Pegaptanib versus sham	2	592	Risk Ratio (M-H, Fixed, 95% CI)	1.39 [1.06, 1.81]
5.4 Pegaptanib versus sham	2	1186	Risk Ratio (M-H, Fixed, 95% CI)	1.49 [1.19, 1.88]
6 Mean visual acuity at one year follow-up (Mean number of letters on ETDRS chart at one year follow-up)	2		Mean Difference (IV, Fixed, 95% CI)	Subtotals only
6.1 0.3 mg Pegaptanib versus sham	2	590	Mean Difference (IV, Fixed, 95% CI)	7.20 [4.27, 10.13]
6.2 1 mg Pegaptanib versus sham	2	596	Mean Difference (IV, Fixed, 95% CI)	5.76 [2.76, 8.77]
6.3 3 mg Pegaptanib versus sham	2	592	Mean Difference (IV, Fixed, 95% CI)	3.64 [0.78, 6.51]
7 Total CNV size at one year follow-up	2		Mean Difference (IV, Fixed, 95% CI)	Subtotals only
7.1 0.3 mg Pegaptanib versus sham	2	590	Mean Difference (IV, Fixed, 95% CI)	-1.05 [-1.63, -0.47]
7.2 1 mg Pegaptanib versus sham	2	596	Mean Difference (IV, Fixed, 95% CI)	-0.97 [-1.54, -0.40]
7.3 3 mg Pegaptanib versus sham	2	592	Mean Difference (IV, Fixed, 95% CI)	-0.74 [-1.35, -0.13]
8 Mean size of lesion at one year follow-up (number of standard disc areas)	2		Mean Difference (IV, Fixed, 95% CI)	Subtotals only
8.1 0.3 mg Pegaptanib versus sham	2	590	Mean Difference (IV, Fixed, 95% CI)	-1.10 [-1.69, -0.52]
8.2 1 mg Pegaptanib versus sham	2	596	Mean Difference (IV, Fixed, 95% CI)	-0.89 [-1.46, -0.31]
8.3 3 mg Pegaptanib versus sham	2	592	Mean Difference (IV, Fixed, 95% CI)	-0.52 [-1.18, 0.13]

Comparison 2. Sub-group analyses stratified by angiographic subtype: Pegaptanib vs sham.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 0.3 mg Pegaptanib: loss of 15 or more letters of visual acuity at one year follow-up	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
1.1 Predominantly classic	2	148	Risk Ratio (M-H, Fixed, 95% CI)	0.72 [0.47, 1.11]
1.2 Minimally classic	2	213	Risk Ratio (M-H, Fixed, 95% CI)	0.52 [0.35, 0.78]
1.3 Pure occult	2	232	Risk Ratio (M-H, Fixed, 95% CI)	0.77 [0.54, 1.08]
2 1 mg Pegaptanib: loss of 15 or more letters of visual acuity at one year follow-up	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
2.1 Predominantly classic	2	155	Risk Ratio (M-H, Fixed, 95% CI)	0.88 [0.61, 1.26]
2.2 Minimally classic	2	209	Risk Ratio (M-H, Fixed, 95% CI)	0.56 [0.39, 0.81]
2.3 Pure occult	2	235	Risk Ratio (M-H, Fixed, 95% CI)	0.64 [0.45, 0.92]

Comparison 3. Sensitivity analyses for loss of 15 or more letters visual acuity at one year follow-up: Pegaptanib vs sham.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Loss of 15 or more letters visual acuity at one year (Assuming those lost to follow-up lost 15 or more letters	2		Risk Ratio (M-H, Random, 95% CI)	Subtotals only
1.1 0.3 mg Pegaptanib versus sham	2	601	Risk Ratio (M-H, Random, 95% CI)	0.72 [0.59, 0.87]
1.2 1 mg Pegaptanib versus sham	2	609	Risk Ratio (M-H, Random, 95% CI)	0.72 [0.59, 0.86]
1.3 3 mg Pegaptanib versus sham	2	606	Risk Ratio (M-H, Random, 95% CI)	0.92 [0.78, 1.09]
1.4 Pegaptanib versus sham	2	1208	Risk Ratio (M-H, Random, 95% CI)	0.78 [0.68, 0.90]
2 Loss of 15 or more letters at one year (Assuming those lost to follow-up did not lose 15 or more letters)	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
2.1 0.3 mg Pegaptanib versus sham	2	601	Risk Ratio (M-H, Fixed, 95% CI)	0.67 [0.53, 0.84]
2.2 1 mg Pegaptanib versus sham	2	609	Risk Ratio (M-H, Fixed, 95% CI)	0.63 [0.50, 0.80]
2.3 3 mg Pegaptanib versus sham	2	606	Risk Ratio (M-H, Fixed, 95% CI)	0.82 [0.66, 1.01]
2.4 Pegaptanib versus sham	2	1208	Risk Ratio (M-H, Fixed, 95% CI)	0.71 [0.59, 0.84]

Analysis 1.1. Comparison 1 Pegaptanib vs sham, Outcome 1 Loss of 15 or more letters visual acuity at one year follow-up (patients with measurement at one year).

Analysis 1.2. Comparison 1 Pegaptanib vs sham, Outcome 2 Loss of 30 or more letters of visual acuity at one year follow-up.

Analysis 1.3. Comparison 1 Pegaptanib vs sham, Outcome 3 Visual acuity worse than 20/200 at one year follow-up.

Analysis 1.4. Comparison 1 Pegaptanib vs sham, Outcome 4 Gain of 15 or more letters of visual acuity at one year follow-up.

Analysis 1.5. Comparison 1 Pegaptanib vs sham, Outcome 5 Maintenance of visual acuity at one year follow-up.

Analysis 1.6. Comparison 1 Pegaptanib vs sham, Outcome 6 Mean visual acuity at one year follow-up (Mean number of letters on ETDRS chart at one year follow-up).

Analysis 1.7. Comparison 1 Pegaptanib vs sham, Outcome 7 Total CNV size at one year follow-up.

Analysis 1.8. Comparison 1 Pegaptanib vs sham, Outcome 8 Mean size of lesion at one year follow-up (number of standard disc areas).

Analysis 2.1. Comparison 2 Sub-group analyses stratified by angiographic subtype: Pegaptanib vs sham, Outcome 1 0.3 mg Pegaptanib: loss of 15 or more letters of visual acuity at one year follow-up.

Analysis 2.2. Comparison 2 Sub-group analyses stratified by angiographic subtype: Pegaptanib vs sham, Outcome 2 1 mg Pegaptanib: loss of 15 or more letters of visual acuity at one year follow-up.

Analysis 3.1. Comparison 3 Sensitivity analyses for loss of 15 or more letters visual acuity at one year follow-up: Pegaptanib vs sham, Outcome 1 Loss of 15 or more letters visual acuity at one year (Assuming those lost to follow-up lost 15 or more letters.

Analysis 3.2. Comparison 3 Sensitivity analyses for loss of 15 or more letters visual acuity at one year follow-up: Pegaptanib vs sham, Outcome 2 Loss of 15 or more letters at one year (Assuming those lost to follow-up did not lose 15 or more letters).

Appendix 1. CENTRAL search strategy

#1 MeSH descriptor Macular Degeneration

#2 MeSH descriptor Retinal Degeneration

#3 MeSH descriptor Neovascularization, pathologic

#4 ((macul* OR retina* OR choroid*:TI) AND (degener* OR neovasc*:TI))

#5 ((macul* OR retina* OR choroid*:AB) AND (degener* OR neovasc*:AB))

#6 maculopath*

#7 (#1 OR #2 OR #3 OR #4 OR #5 OR #6)

#8 MeSH descriptor Angiogenesis Inhibitors

#9 MeSH descriptor Angiogenesis Inducing

#10 MeSH descriptor Endothelial Growth Factors

#11 macugen or pegapanib or lucentis or rhufab or rhu fab or ranibizumab or bevacizumab

#12 angiogen* or antiangiogen* or neovasculari* or vasculari*

#13 anti-vegf* or anti next vegf

#14 endothelial near growth near factor*

#15 (#8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14)

#16 (#7 AND #15)

Appendix 2. MEDLINE search strategy

1. randomized controlled trial.pt.

2. (randomized or randomised).ab,ti.

3. placebo.ab,ti.

4. dt.fs.

5. randomly.ab,ti.

6. trial.ab,ti.

7. groups.ab,ti.

8. or/1-7

9. exp animals/

10. exp humans/

11. 9 not (9 and 10)

12. 8 not 11

13. exp macular degeneration/

14. exp retinal degeneration/

15. exp retinal neovascularization/

16. exp choroidal neovascularization/

17. exp macula lutea/

18. maculopath$.tw.

19. ((macul$ or retina$ or choroid$) adj3 degener$).tw.)

20. ((macul$ or retina$ or choroid$) adj3 neovasc$).tw.

21. (macula$ adj2 lutea).tw.

22. or/13-21

23. exp angiogenesis inhibitors/

24. exp angiogenesis inducing agents/

25. exp endothelial growth factors/

26. exp vascular endothelial growth factors/

27. (macugen$ or pegaptanib$ or lucentis$ or rhufab$ or ranibizumab$ or bevacizumab$).tw.

28. (anti adj2 VEGF$).tw.

29. (endothelial adj2 growth adj2 factor$).tw.

30. or/23-29

31. 22 and 30

32. 12 and 31

The search filter for trials at the beginning of the MEDLINE strategy is from the published paper by Glanville (Glanville 2006).

Appendix 3. EMBASE search strategy

1. exp randomized controlled trial/

2. exp randomization/

3. exp double blind procedure/

4. exp single blind procedure/

5. random$.tw.

6. or/1-5

7. (animal or animal experiment).sh.

8. human.sh.

9. 7 and 8

10. 7 not 9

11. 6 not 10

12. exp clinical trial/

13. (clin$ adj3 trial$).tw.

14. ((singl$ or doubl$ or trebl$ or tripl$) adj3 (blind$ or mask$)).tw.

15. exp placebo/

16. placebo$.tw.

17. random$.tw.

18. exp experimental design/

19. exp crossover procedure/

20. exp control group/

21. exp latin square design/

22. or/12-21

23. 22 not 10

24. 23 not 11

25. exp comparative study/

26. exp evaluation/

27. exp prospective study/

28. (control$ or prospectiv$ or volunteer$).tw.

29. or/25-28

30. 29 not 10

31. not (11 or 23)

32. 11 or 24 or 31

33. exp retina macula degeneration/

34. exp retinal degeneration/

35. exp subretinal neovascularization/

36. maculopath$.tw.

37. ((macul$ or retina$ or choroid$) adj3 degener$).tw.

38. ((macul$ or retina$ or choroid$) adj3 neovasc$).tw.

39. (macula$ adj2 lutea).tw.

40. or/33-39

41. exp angiogenesis/

42. exp angiogenesis inhibitors/

43. exp angiogenic factor/

44. exp endothelial cell growth factor/

45. exp vasculotropin/

46. (macugen$ or pegaptanib$ or lucentis$ or rhufab$ or ranibizumab$ or bevacizumab$).tw.

47. (anti adj2 VEGF$).tw.

48. (endothelial adj2 growth adj2 factor$).tw.

49. or/41-48

50. 40 and 49

51. 32 and 50

Appendix 4. LILACS search terms

macugen or pegaptanib or lucentis or rhufab or ranibizumab or bevacizumab and macula$ degenerat$ or AMD or ARMD