CATT Trial (Comparison of Age-Related Macular Degeneration Treatments Trials)

CATT 1 (1-year results) - CATT Research Group, NEJM 2011;364:1897-1908

• Primary outcome: mean change in visual acuity (VA) at 1 year.
• Ranibizumab-monthly and bevacizumab-monthly produced nearly identical VA gains (about +8 letters each), meeting non-inferiority criteria.
• Ranibizumab given PRN was equivalent to ranibizumab monthly.
• Bevacizumab-PRN gains were slightly smaller and, by the strictest statistical margin, did not conclusively meet non-inferiority versus ranibizumab-monthly, though the clinical difference was small.
• Anatomic outcomes (retinal thickness on OCT, fluid resolution) showed slightly more residual fluid with bevacizumab, without translating into a meaningful VA difference.
• Ocular serious adverse events (endophthalmitis, uveitis) were low and similar between drugs.
• Bottom line at 1 year: bevacizumab was essentially non-inferior to ranibizumab for VA outcomes, supporting its continued off-label use given the large cost difference - "The Comparison of Age-Related Macular Degeneration Treatments Trial (CATT) study demonstrated the noninferiority of bevacizumab as compared to ranibizumab at 1 year." (The Wills Eye Manual, p. 841)

CATT 2 (2-year results) - Martin DF et al., Ophthalmology 2012

• Patients originally assigned to monthly dosing were re-randomized at year 1 to continue monthly or switch to PRN, allowing assessment of drug effect and dosing-schedule effect independently over 2 years.
• VA outcomes: mean VA gains at 2 years remained similar between ranibizumab and bevacizumab across dosing regimens - drug choice did not significantly affect visual outcome.
• Switching to PRN: eyes switched from monthly to PRN dosing lost a small amount of vision compared with continuing monthly therapy (a few letters), and the "switch" arms did not always meet non-inferiority versus continued monthly dosing, suggesting monthly dosing gives slightly more durable benefit than PRN, at the cost of more injections.
• Geographic atrophy: by year 2, incidence was higher with ranibizumab than bevacizumab (21% vs 17%, p=0.02) and higher with monthly than PRN dosing - raising a hypothesis (not proven causally) that higher VEGF suppression might promote atrophy, though this was not replicated in the similarly designed IVAN trial (28% vs 31%, p=0.46).
• Systemic safety: the rate of systemic serious adverse events (mostly hospitalizations) was higher with bevacizumab than ranibizumab at 2 years (about 24% vs 19%, p=0.04), but there was no consistent pattern implicating any specific organ system, and no significant difference in arteriothrombotic events, myocardial infarction, stroke, or death. This finding has been debated and was not confirmed as causal in later meta-analyses (Cochrane review, Moja et al. 2014, PMID 25220133), which found no significant difference in mortality or arteriothrombotic events between the two drugs across pooled trials, though it flagged a possible signal for increased "any systemic serious adverse event" with bevacizumab, driven mainly by gastrointestinal disorders.

Clinical relevance

• Related trials reached similar conclusions: IVAN (UK) found equivalent visual outcomes between the two drugs, and the geographic atrophy signal was not replicated. Both CATT and IVAN together underpin the widespread off-label use of bevacizumab as a cost-effective alternative to ranibizumab in many health systems (Kanski's Clinical Ophthalmology, p. 613).
• A later CATT Follow-up Study (~5 years) found that the VA gains achieved during the 2-year trial were not sustained long-term once patients returned to routine, less-standardized community care, underscoring that ongoing anti-VEGF treatment (not just initial response) drives long-term visual outcomes.