Results from two important studies – the RIVAL study and the FLUID study – were presented at Future Directions in Ophthalmology, a Novartis sponsored conference for ophthalmologists in Sydney during August.
Future Directions in Ophthalmology featured a strong line up of experts in the fields of glaucoma and retinal disease, among them Prof. Christopher Leung (Chinese University of Hong Kong), A/ Prof. Wilson Heriot, A/Prof. Samantha Bell, Prof. Robyn Guymer, and Dr. Paul Beaumont, A/Prof Michael Coote and A/ Prof Tim Roberts. Two highlights were the anticipated primary outcomes from the FLUID study by Dr. Jennifer Arnold and 12 month interim outcomes from the RIVAL study by Prof. Mark Gillies, both of which shed light on treatment for neovascular agerelated macular degeneration (n-AMD).
RIVAL INTERIM STUDY RESULTS
The phase four RIVAL study is the first randomised clinical trial to compare treatment of n-AMD with ranibizumab and aflibercept, head to head, over two years using an identical treat and extend regimen.
Two groups of patients, with an average age of 77 years, were randomly assigned to receive one or the other drug. The primary outpoint was the mean change in geographic atrophy (GA) from baseline to two years.
Prof. Mark Gillies, Head of Research at Save Sight Institute Sydney, who presented on 12 month Interim Rival study results, said the interest in geographic atrophy has arisen due to conclusions from both the CATT, IVAN and SEVEN UP studies, which have suggested that ongoing intravitreal treatment may result in the development of GA over several years. Other studies have looked to determine whether this could be due to the number of injections given, the strength of the particular therapy or other patient related factors.
The prospectively specified secondary outcomes were the number of injections received and visual acuity improvement over 12 months. Prof. Gillies reported that the average number of injections given to patients over 12 months was the same, at 9.7 injections for each drug, and although this was more than anticipated, he stated that it was not much more than is seen in real world practice, as indicated by the Fight Retinal Blindness! Study, of eight injections over 12 months for each drug.
The mean visual acuity (VA) starting score for participants was 65 letters – 6/15. Prof. Gillies said this was the best mean VA starting score of any study in AMD yet described and better than the final VA score in View 1 and II studies, which were the pivotal phase three trials of aflibercept. For this reason a big improvement on 65 letters was not expected.
VA was measured at baseline, one month, two months and 12 months and eyes treated with ranibizumab showed a “fairly robust improvement” of 1.3 letters after just one month. After 12 months, eyes treated with ranibizumab had improved by an average of 7.16 letters while eyes treated with aflibercept had improved by 4.85 letters.
Prof. Gillies said although the improvement in ranibizumab treated eyes was numerically greater than those treated with aflibercept, the difference was not statistically significant. However, for the first time in a head to head study, over 12 months the mean VA improvement with ranibizumab was more than seen with aflibercept. Prof. Gillies concluded by saying “we have two very good agents to treat wet age related macular degeneration” and he emphasised the need to follow a treat and extend protocol for optimum improvement in visual acuity. Findings on GA are awaited.
FLUID STUDY RESULTS
Dr. Jennifer Arnold, Marsden Eye Specialists, presented results from the FLUID Study, an Australian randomised clinical trial sponsored by Novartis.
The frequency of injections to treat patients with n-AMD using a personalised treatment strategy, either PRN or treat and extend, is determined by an assessment of CNV activity using criteria of visual acuity, new haemorrhage, and the presence of intra-retinal and/or sub retinal fluid on an optical coherence tomography (OCT) scan. Although this treat and extend regimen has shown to be efficacious in studies including the CATT and HARBOR, these studies were based on a zero tolerance to fluid on the OCT.
With this in mind, the FLUID study was designed to determine whether some retinal fluid can be tolerated when using treat and extend regimens to treat n-AMD without compromising vision outcomes. Additionally, researchers wanted to determine whether it is necessary or even desirable to try and achieve a dry macula.
Dr. Arnold said previous study results showed a lack of a positive correlation between a dry macula and vision outcomes. “We can get good outcomes even though fluid exists,” she said. “Failure to resolve intra-retinal fluid (IRF) results in worse outcomes; persistence of subretinal fluid (SRF) throughout gives a similar outcome to those people who have a dry macula,” said Dr. Arnold, adding that SRF may even be protective. “It may be possible, therefore, to take a more nuanced and flexible approach to fluid when assessing CNV activity and need for retreatment.”
The FLUID study compared outcomes between two treatment arms – an ‘Intensive arm’ and a ‘Relaxed arm’. After a typical loading phase, both arms were treated with ranibizumab with a treat and extend protocol over two years, with treatment intervals (minimum four and maximum 12 weeks) determined according to changes in vision, any new haemorrhage and presence of fluid as shown on the OCT. IRF was considered a sign of activity for all groups. In the Intensive arm, any SRF was considered a marker of disease activity, however for the Relaxed arm, SRF was not considered a marker of disease activity unless it was greater than 200μm SRF only at foveal centre. The assessment of SRF was carried out by a masked Reading Centre and all other criteria was assessed by the investigator.
The primary outcome found no significant difference in the mean change in vision between the two groups at 12 and 24 months. Importantly, across both groups, the proportion of eyes with driving vision (better than 6/12) increased from 40 per cent at baseline to 55 per cent at 24 months.
The Relaxed arm had less decrease in central retinal thickness than the Intensive arm, as expected because SRF was accepted in the former. Eighty per cent of all patients in the study had some SRF at baseline, which decreased over time in both groups.
There was no significant statistical difference in IRF in either arm at 24 months.
SRF, as the only OCT sign of CNV activity, is a common scenario facing clinicians, occurring in around 70 per cent of patients in both groups on at least one occasion during the study. Dr. Arnold also noted that SRF over 200microns at the foveal centre was detected rarely, at only five visits.
Finally, the study found a statistical difference in the number of ranibizumab injections between the groups – even on a treat and extend regimen.
Dr. Arnold concluded, “The study showed that you could tolerate some SRF when assessing activity and managing patients for n-AMD, and still get comparable vision outcomes, and this is likely to occur with less injections.”