From Wiki Journal Club
Jump to navigation Jump to search
The CATT Research Group Writers. "Ranibizumab and bevacizumab for neovascular age-related macular degeneration". The New England Journal of Medicine. 2011. 364(20):1897-1908.
PubMedFull textPDF

Clinical Question

Among patients with choroidal neovascularization from age-related macular degeneration, does scheduled or as-needed administration of ranibizumab or bevacizumab result in equivalent changes in visual acuity at one year?

Bottom Line

Among patients with choroidal neovascularization from age-related macular degeneration, administration of ranibizumab or bevacizumab resulted in similar changes in visual acuity at one year, even when administered on a monthly or as-needed basis.

Major Points

Monthly intravitreal injection of the anti-VEGF medication ranibizumab (trade name Lucentis) was shown to be effective in improving visual acuity in choroidal neovascularization from age-related macular degeneration ("wet" AMD) in the 2006 industry-sponsored MARINA study.[1] It received an indication for this from the FDA later that year. The related anti-VEGF medication bevacizumab (trade name Avastin) was being used off-label as a low-cost alternative to the ~$2,000/dose ranibizumab. Both of these medications are manufactured by Genentech and were derived from the same murine monoclonal antibody.[2] Whether bevacizumab was non-inferior to ranibizumab and if they could be used in response to progression of neovascularization rather than on a fixed monthly schedule was unknown.

The NIH-sponsored Comparison of Age-Related Macular Degeneration Treatment Trails (CATT) published its 1 year results in 2011. 1,185 patients with newly-diagnosed, untreated AMD with choroidal neovascularization were randomized to monthly ranibizumab, monthly bevacizumab, ranibizumab as needed (ie, retreatment in response to neovascularization findings on exam), or bevacizumab as needed. It was designed as a noninferiority trial comparing the three other arms against monthly ranibizumab. Treating ophthalmologists were masked to the study medication but not the treatment schedule. At 1 year, the four groups had similar improvement in visual acuity. Groups had similar rates of adverse events, though patients receiving bevacizumab had more occurrences of ≥1 serious adverse event than those in the ranibizumab group. The annual cost of ranibizumab ($23,000 monthly and $13,800 as-needed) was more than 30x that of bevacizumab ($595 monthly and $385 as-needed). Results from the two-year follow-up[3] were similar, though the monthly administration groups outperformed the as-needed groups for gain in visual acuity.

Similar to CATT, the IVAN trial evaluated these two medications on monthly and as-needed administrations in a randomized fashion in 610 patients with this disease. Its two year follow-up data published in 2013[4] reported similar efficacy between the two medications. A pooled analysis of CATT and IVAN outcomes in this publication found no difference in visual acuity between the two medications but a benefit with continuous dosing (weighted difference -2.23; 95% CI -3.93 to -0.53). Safety was similar for the medications though an increased rate of all-cause mortality was found in the as-needed dosing groups.


AAO AMD (2014, adapted)[5]

  • Anti-VEGF therapies (aflibercept, bevacizumab, and ranibizumab) are first-line for neovascular AMD
    • Counsel patients on the off-label nature of bevacizumab and obtain informed consent
  • Follow up for repeat exam in 4 weeks with further follow-up and treatment at the discretion of the ophthalmologist


  • Multicenter, single-blinded, 2x2 factorial, noninferiority, randomized controlled trial
  • N=1,185 patients with a single eye treated with this protocol
    • Ranibizumab (n=301)
    • Bevacizumab (n=286)
    • Ranibizumab as needed (n=298)
    • Bevacizumab as needed (n=300)
  • Setting: 44 US centers
  • Enrollment: 2008-2009
  • Follow-up: 1 year
  • Analysis: Intention-to-treat
  • Primary outcome: Change in visual acuity


Inclusion Criteria

  • Age ≥50 years
  • Females must be on contraceptives, have had a surgical sterilization procedure, or be post-menopausal
  • Presence of neovascularization, fluid, or hemorrhage under the fovea
  • Informed consent
  • Study eye specific:
    • New diagnosis of active choroidal neovascularization from AMD that has not yet been treated, with active disease defined by both of the following:
      • Leakage on fluorescein angiography
      • Subretinal, intraretinal, or sub-retinal pigment epithelium fluid on time-domain optical coherence tomography (OCT)
    • Visual acuity 20/25 to 20/320
    • Choroidal neovascularization must involve the center of the fovea
    • Total fibrotic area <50% of total lesion
    • ≥1 drusen in either eye or late AMD in other eye
    • Clear ocular media and adequate dilitation of the pupil to allow for good fundus imaging
    • Disc/macular steroscopic photographs and fluorescein angiogram <7 days before randomization
    • OCT of macula <7 days before randomization

Exclusion Criteria

  • Prior treatment except for prophylactic
  • Condition preventing 2 years of follow-up
  • Prior treatment with IV bevacizumab
  • In the prior 30 days, use of other investigational therapeutics likely to affect the eye
  • Current systemic anti-VEGF therapies
  • Lesion-specific:
    • Fibrosis or geographic atrophy involving the center of the fovea in the study eye
    • Other causes of choroidal neovascularization
    • Retinal pigment epithelial tear of the macula in the study eye
  • Study-eye specific:
    • Other diseases of the eye (see the supplemental appendix for an extensive list of these)
    • Recent surgery of the eye
  • Pregnancy or lactation
  • Other contraindication to the study drug
  • Active or recurrent systemic infection
  • Other uncontrolled diseases
  • Study compliance issues

Baseline Characteristics

From the ranibizumab group. Groups were similar.

  • Demographics: Age 79 years, 61% female, 99% white
  • PMH: MI 11%, stroke 5%, TIA 4%
  • Health data: BP 134/75
  • Eye and vision data: Foveal thickness 458 um, visual acuity score 60 letters (20/50-80 range)
    • Involvement of the foveal center:
      • Choroidal neovascularization: 58%
      • Fluid: 28%
      • Hemorrhage: 7%
      • Other: 6%
      • No choroidal neovascularization or unable to grade: 1%


  • Only one eye was enrolled in the study in the case of bilateral involvement
  • Patients were randomized to a group with stratification by center with permuted-block method and randomly chosen block size
    • Ranibizumab - 0.5 mg intravitreal injection q28 days
    • Bevacizumab - 1.25 mg intravitreal injection q28 days
    • Ranibizumab as needed - 0.5 mg intravitreal injection only for active neovascularization
    • Bevacizumab as needed - 1.25 mg intravitreal injection only for active neovascularization
  • The two PRN groups had time-domain OCT q28 days and were evaluated for treatment
  • Ophthalmologists at each center were masked to assignments though staff members (who presented the medication to the ophthalmologist) were not
  • Those performing visual acuity testing and grading OCT scans were masked
  • Fluorescein angiography was performed at the discretion of the ophthalmologist
  • Use of antibiotic drops with the injections was also left to the ophthalmologist


Comparisons are ranibizumab vs. bevacizumab vs. ranibizumab as needed vs. bevacizumab as needed. Outcomes are at 1 year except where specified. P-value is for noninferiority among the four study groups.

Primary Outcome

Change in visual acuity
In number of letters on Snellen equivalent.
+8.5 vs. +8.0 vs. +6.8 vs. +8.5 letters (P=0.16)

Secondary Outcomes

Proportion of patients with a change in visual acuity of ≥15 letters
Increase: 34.2% vs. 31.3% vs. 24.9% vs. 28.0%
Decrease: 5.6% vs. 6.0% vs. 4.6% vs. 8.5%
Number of treatments with the study medication
11.7 vs. 11.9 vs. 6.9 vs. 7.7 (P<0.001)
Change in thickness on OCT
Fovea: -196 vs. -164 vs. -168 vs. -152 um (P=0.03)
Fovea plus subfoveal fluid: -100 vs. -79 vs. -81 vs. -79 um (P=0.18)
Change in lesion size at the optic-disc area
0.0 vs. +0.1 vs. +0.2 vs. +0.5 (P=0.047)
Medication cost

Ranibizumab cost $2,000/dose, bevacizumab (used under an investigational new drug application) cost $50/dose.

$23,000 vs. $595 vs. $13,800 vs. $385 through the end of year 1
Study-eye ocular and systemic adverse events
P>0.05 except where specified.
Ocular event
Endophthalmitis: 0.7% vs. 1.4% vs. 0% vs. 0% (among groups P=0.03, between drugs P=0.45)
Pseudoendophthalmitis: 0.3% vs. 0% vs. 0% vs. 0%
≥1 serious: 17.6% vs. 22.4% vs. 20.5% vs. 25.7% (among groups P=0.11, between drugs P=0.04)
All-cause mortality: 1.3% vs. 1.4% vs. 1.7% vs. 3.7%
Arteriothrombotic event: 2.3% vs. 2.1% vs. 2.0% vs. 2.7%
Nonfatal MI: 0.7% vs. 0.7% vs. 1.0% vs. 0.3%
Nonfatal stroke: 1.0% vs. 0.7% vs. 1.0% vs. 0.3%
Fatal MI, fatal stroke, or cardiac arrest: 0.7% vs. 0.7% vs. 0.7% vs. 1.7%
Venous thrombosis: 0% vs. 1.4% vs. 0.7% vs. 0.3%
TIA: 0.3% vs. 0% vs. 0.7% vs. 1.0%
HTN: 0% vs. 0.7% vs. 0 % vs. 0%
Any disorder by organ class
Cardiac: 3.3% vs. 5.6% vs. 4.0% vs. 4.3%
Infection: 2.0% vs. 3.8% vs. 4.0% vs. 6.0%
Nervous system: 2.0% vs. 3.1% vs. 4.0% vs. 3.0%
Injury or procedure complication: 2.3% vs. 3.8% vs. 2.7% vs. 3.0%
Neoplasm, including benign: 2.3% vs. 1.7% vs. 3.4% vs. 3.0%
Surgical or medical procedure: 1.3% vs. 2.1% vs. 1.3% vs. 2.7%
GI: 1.0% vs. 2.1% vs. 0.7% vs. 3.0% (among groups P=0.11, between drugs P=0.02)
Any other system: 6.0% vs. 9.1% vs. 5.4% vs. 9.3% (among groups P=0.14, between drugs P=0.03)

Subgroup Analysis

There was no change in the primary outcome when comparing by clinical center, age, baseline visual acuity, or baseline lesion size.


  • Not powered to detect adverse events due to specific drugs[2]
  • Very little racial diversity, unclear if same outcomes would occur among other groups (specifically Asians)[6]


  • NIH (National Eye Institute)
  • Authors with multiple financial disclosures

Further Reading