LUCAS 2015.
| Methods |
Number randomized (total and per group): 441 participants randomly assigned to study treatment: 220 in bevacizumab group and 221 in ranibizumab group Exclusions after randomization: 10 total participants: 7 in bevacizumab group and 3 in ranibizumab group. "All 9 patients from 1 study center were excluded because of serious protocol violations, and 1 patient was excluded after a serious retinal and vitreous hemorrhage..." Number analyzed (total and per group): 371 total participants: 184 in bevacizumab group and 187 in ranibizumab group Unit of analysis: individuals (1 study eye per participant) Losses to follow‐up: none, but 60 excluded from analysis (29 in bevacizumab group and 31 in ranibizumab group), including 11 total participants who died Compliance: 371/441 participants completed study per protocol Intention‐to‐treat analysis: no; 70 participants enrolled and randomized were not included in analysis Reported power calculation: yes; 181 participants per arm to provide 80% power to detect or rule out a difference in visual acuity outcome, assuming a 10% dropout rate Study design comment: non‐inferiority design using margin of 5 letters on ETDRS chart |
|
| Participants |
Country: 10 clinical centers in Norway Age: mean 78.7 years in bevacizumab group and 78.0 in ranibizumab group Gender (per cent): 140/431 (32.5%) men and 291/431 (67.5%) women Inclusion criteria: age 50 years or older; previously untreated active neovascular AMD in study eye; BCVA in study eye between 20/25 and 20/120, measured at 4 m using an ETDRS "standardized viewer" Exclusion criteria: "pigment epithelial detachments with no associated intraretinal or subretinal edema and lesions comprising more than 50% blood or fibrosis were excluded" Equivalence of baseline characteristics: more participants in ranibizumab group had a history of myocardial infarction Diagnoses in participants: neovascular AMD; 86% had CNV under foveal center |
|
| Interventions |
Intervention 1: 1.25 mg per 0.05 mL intravitreal bevacizumab injections every 4 weeks until no signs of active AMD were found on OCT and biomicroscopic fundus examination, followed by the "treat and extend" protocol Intervention 2: 0.5 mg intravitreal ranibizumab injections every 4 weeks, followed by "treat and extend" protocol The "treat and extend" protocol for each treatment group specified that whenever a new injection was given, the "period" (interval) to the next injection was to be extended by 2 weeks up to a maximum interval of 12 weeks. Whenever recurrent neovascularization was treated, the interval was shortened by 2 weeks until the lesion was inactive. Interval extension was then restarted to a maximum of 2 weeks less than when recurrence was observed Follow‐up Planned length: 24 months Actual length: 12 months Frequency of follow‐up assessments: 4‐week intervals, modified by 2‐week increases or decreases, as described above |
|
| Outcomes |
Primary outcome, as defined: "change in BCVA at 1 year as measured on the ETDRS visual acuity chart" Secondary outcomes, as defined: "number of injections, change in CRT as measured with OCT, and change in lesion size as measured on FA" Safety outcome: occurrence of arteriothrombotic events Intervals at which outcomes were assessed: unclear, but presumably whenever participant was assessed for the need for retreatment |
|
| Notes |
Full study name: Lucentic Compared to Avastin Study Trial registration: NCT01127360 Funding sources: Oslo University Hospital, Oslo, Norway Declarations of interest: "The funding organization had no role in the design of the study but aided in the conduct of the study and data management." One study author had participated in an advisory board meeting for another anti‐VEGF agent for Bayer Study period: random enrollment March 2009 to July 2012 Reported subgroup analyses: none Contacting study investigators: pending |
|
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | "computer generated by a third party at the Norwegian University of Science and Technology, Trondheim, with the use of the block method and stratification by center" |
| Allocation concealment (selection bias) | Low risk | Drugs were allocated by unmasked study nurses, who were also responsible for aseptic filling of a syringe with assigned study drug. "The identical syringes, regardless of which drug was given, were filled by these nurses behind a screen. The syringe was then presented directly to the treating ophthalmologist" |
| Masking of participants (performance bias) | Low risk | "the patient, the treating ophthalmologist, and the assisting nurse were masked to the drug at all times" |
| Masking of study personnel (performance bias) | Low risk | "These study nurses were not involved in any other patient‐related activities in the study" |
| Masking of outcome assessment (detection bias) | Low risk | "Ophthalmic nurses, who also were masked to the drug and patient records, tested the ETDRS visual acuity" |
| Incomplete outcome data (attrition bias) All outcomes | Unclear risk | About 15% of participants were missing 12‐month outcome data, compared with 10% assumed in sample size calculation |
| Selective reporting (reporting bias) | Low risk | All outcomes specified were reported |
| Other bias | Low risk | No other bias was identified |