CATT AND IVAN RESULTS

A special session held during the 2012 annual meeting of the Association for Research in Vision and Ophthalmology featured investigators reporting outcomes from two years of follow-up in the Comparison of Age-related Macular Degeneration Treatments Trials (CATT) and from an interim analysis after one year in the two-year Inhibition of VEGF in Age-related Choroidal Neovascularisation (IVAN) study. Findings from both studies showed visual acuity outcomes in patients with exudative AMD are similar whether they receive intravitreal anti-VEGF treatment with ranibizumab (Lucentis, Genentech) or bevacizumab (Avastin, Genentech). In CATT and in an analysis pooling data from both studies, the rate of serious systemic adverse events was higher for bevacizumab than ranibizumab. However, the interpretation of the latter difference is uncertain for now as is the clinical relevance of small, but statistically significant differences identified between drugs and dosing regimens in various functional and morphological endpoints.

Daniel F Martin MD, study chair for CATT and chairman, Cole Eye Institute, Cleveland Clinic, Cleveland, Ohio, US, noted the new data will help clinicians and their patients make informed treatment decisions. He said that based on the outcomes of CATT, clinicians can proceed with confidence using either anti-VEGF agent. Deciding between the monthly dosing and as needed regimens may involve more of a discussion, he told EuroTimes. Dr Martin explained, “There was more gain in visual acuity with monthly administration, but the difference was small, only 2.4 letters, and required giving 10 more injections over the course of two years. Furthermore, looking at broader metrics of visual acuity, there was not much difference between dosing groups. However, it’s clear from the two-year outcomes that if the decision is made to use the monthly regimen, it needs to be continuous.â€

Dr Martin was referring to the aspect of the study designed to determine the impact of switching to as-needed treatment after one year of regular monthly injections. Investigation of this issue was based on the idea that CNV histology may have been altered after 12 months of continuous VEGF suppression. “However, we found that the lesions are not dormant. Patients switched to a prn regimen after one year of continuous treatment had visual and anatomical results that were similar to the patients receiving prn treatment all along,†Dr Martin said. Usha Chakravarthy MD, PhD, chief investigator, IVAN, and professor of ophthalmology and vision sciences, Queen’s University of Belfast, Ireland, told EuroTimes that, interestingly, in the IVAN trial there was no difference in visual acuity at one year between the continuous and prn dosing regimens. “Furthermore, the equivalence of the prn regimen, which required three monthly injections, and continuous dosing regimens was achieved with a very similar number of injections in the first year as observed in the CATT trial,†she said.

Morphological differences

In analyses of morphological endpoints for CATT, the percentage of patients with no fluid on OCT was higher with monthly dosing and highest with monthly ranibizumab, but monthly dosing was also associated with a higher rate of geographic atrophy development and the rate was highest with monthly ranibizumab. In IVAN, there were no between-drug differences in the morphological measures assessed, but compared with continuously treated eyes, eyes treated as needed for reactivation had a greater total lesion thickness and a higher proportion showed leakage on fluorescein angiogram. “CATT and IVAN have provided a wealth of knowledge on the complex relationships between function and morphology in patients with exudative AMD.

However, the clinical importance of the regimenrelated morphological differences observed in IVAN at one year is unclear given the negligible differences between groups in visual function,†said Dr Chakravarthy. One year earlier, results from the primary endpoint analysis after one year in CATT showed a statistically significant higher risk (1.3-fold) of serious adverse events among patients treated with bevacizumab compared with ranibizumab. The difference persisted at two years, but there was no statistically significant difference between drugs in rates of death or arteriothrombotic events (ATEs). In IVAN at one year, there was no statistically significant difference between drugs in the proportion of patients having serious systemic adverse events, albeit such events were higher among bevacizumabtreated patients. However, there was a statistically significant difference between drugs in the rate of arteriothrombotic events or heart failure (ATEs), with an excess of ATEs among ranibizumab-treated patients. Dr Chakravarthy said the safety findings are generally reassuring, and noted that in an analysis combining the one-year data for IVAN and CATT, the signal observed in IVAN for ATEs disappeared because it was in opposite directions for IVAN and CATT. However, the signal for more SAEs with bevacizumab persisted in the combined analysis because it was in the same direction in both trials (even though not statistically significant in the IVAN trial separately). “Concern remains with respect to the serious systemic adverse events, and we are looking in greater detail into the possible mechanisms that might explain the difference,†she said.

The CATT investigators also undertook further analyses to try to understand if the higher risk of serious systemic adverse events with bevacizumab is true. Curiously, there was a higher risk with prn versus monthly treatment, and dissecting deeper into the data, they found most of the excess events associated with bevacizumab were not previously reported in oncological trials where patients had systemic bevacizumab exposure. “The differences between drugs in CATT are very nonspecific and diffuse. They may be due to chance, perhaps there is some unidentified imbalance between groups at baseline that we did not adjust for, or the difference could be something meaningful. We are still trying to sort that out,†said Dr Martin. Dr Chakravarthy began her presentation by noting that determining the relative effects of the two drugs and two regimens is important in the context that bevacizumab is much cheaper than ranibizumab and because fewer treatments with prn treatment is safer and helps keep costs down, particularly when using ranibizumab. Cost analyses she presented, which included costs for medication, administration, monitoring and the costs of serious expected adverse events (hospitalisations and treatment of these events), showed a range from ￡9656 per patient per year for ranibizumab monthly to ￡1509 for as needed bevacizumab. These estimates were calculated explicitly in the context of the economic evaluation and did not include value-added tax, Dr Chakravarthy noted. Dr Martin also reported cost data, noting cost was a pre-specified secondary outcome for CATT but never drove the study, and he put the per-patient difference into a larger perspective. “Extrapolating to the 220,000 AMD patients treated annually, there is a US$ 5bn difference between ranibizumab monthly and either bevacizumab group and a difference of $3bn comparing prn ranibizumab and the bevacizumab regimens,†he said.

CATT initially randomised 1,185 patients enrolled at 43 US centres into one of four treatment groups to receive ranibizumab 0.5mg or bevacizumab 1.25 monthly or prn. A total of 1,107 patients were followed through year two, but at one year, which was the primary endpoint, patients in the monthly groups were re-randomised to continue monthly treatment or switch to prn. The IVAN study randomised 610 patients at 23 sites in England and Northern Ireland and also compared ranibizumab 0.5mg and bevacizumab 1.25mg monthly or as needed. Patients assigned to prn treatment received three monthly injections initially and whenever re-treatment became indicated based on structured criteria. In CATT, initial treatment and re-treatment for prn patients was with a single injection.