StereoTactic Radiotherapy for Wet Age-Related Macular Degeneration (STAR), 2024

Background: Neovascular age-related macular degeneration (nAMD) is a leading cause of blindness. The first-line therapy is anti-vascular endothelial growth factor (anti-VEGF) agents delivered by intravitreal injection. Ionising radiation mitigates key pathogenic processes underlying nAMD, and therefore has therapeutic potential. STAR aimed to assess whether stereotactic radiotherapy (SRT) reduces the number of anti-VEGF injections required, without sacrificing visual acuity. Methods: This pivotal, randomised, double-masked, sham-controlled trial enrolled participants with pretreated chronic active nAMD from 30 UK hospitals. Participants were randomly allocated in a 2:1 ratio to 16-Gray (Gy) SRT delivered using a robotically controlled device or sham SRT, stratified by treatment centre. Eligible participants were aged 50 years or older and had chronic active nAMD, with at least three previous anti-VEGF injections, including at least one in the last 4 months. Participants and all trial and image reading centre staff were masked to treatment allocation, except one unmasked statistician. The primary outcome was the number of intravitreal ranibizumab injections required over 2 years, tested for superiority (fewer injections). The main secondary outcome was Early Treatment Diabetic Retinopathy Study visual acuity at two years, tested for non-inferiority (five-letter margin). The primary analysis used the intention-to-treat principle, and safety was analysed per-protocol on participants with available data. The study is registered with ClinicalTrials.gov (NCT02243878) and is closed for recruitment. Findings: 411 participants enrolled between Jan 1, 2015, and Dec 27, 2019, and 274 were randomly allocated to the 16-Gy SRT group and 137 to the sham SRT group. 240 (58%) of all participants were female, and 171 (42%) of all participants were male. 241 participants in the 16-Gy SRT group and 118 participants in the sham group were included in the final analysis, and 409 patients were treated and formed the safety population, of whom two patients allocated to sham treatment erroneously received 16-Gy SRT. The SRT group received a mean of 10·7 injections (SD 6·3) over 2 years versus 13·3 injections (5·8) with sham, a reduction of 2·9 injections after adjusting for treatment centre (95% CI -4·2 to -1·6, p<0·0001). The SRT group best-corrected visual acuity change was non-inferior to sham (adjusted mean letter loss difference between groups, -1·7 letters [95% CI -4·2 to 0·8]). Adverse event rates were similar across groups, but reading centre-detected microvascular abnormalities occurred in 77 SRT-treated eyes (35%) and 13 (12%) sham-treated eyes. Overall, eyes with microvascular abnormalities tended to have better best-corrected visual acuity than those without. Fewer ranibizumab injections offset the cost of SRT, saving a mean of £565 per participant (95% CI -332 to 1483). Interpretation: SRT can reduce ranibizumab treatment burden without compromising vision. Funding: Medical Research Council and National Institute for Health and Care Research Efficacy and Mechanism Evaluation Programme. Age-related macular degeneration (AMD) leads more people to go blind than all other eye diseases combined. 'Wet' AMD occurs when abnormal new blood vessels grow into the macula. The macula is an area inside the back of the eye, which is tightly packed with the light-sensitive cells that give fine central vision for driving, reading, facial recognition etc. Most patients are treated with repeated injections of ranibizumab (Lucentis), bevacizumab (Avastin) , or aflibercept (Eylea) into the eye (these drugs suppresses the new vessel growth). Unsurprisingly, the need for hospital visits every 1-2 months, and regular injections into the eye, is not favoured by patients. The proposed study investigates the use of radiation to treat wet AMD. Radiation preferentially damages proliferating cells, such as those forming the abnormal new blood vessels. It has the potential to kill off the abnormal cells, unlike the injections which just suppress the disease. The radiation is delivered using a robotically controlled device that aims three beams of radiation in through the white of eye, to overlap at the macula. Initial studies of stereotactic radiotherapy (SRT) were promising, and established the best dose of radiation. They showed that radiation has the potential to significantly reduce the need for injections. Each injection costs the NHS approximately £600, hence the potential cost savings are substantial. For those who respond well, disease activity is reduced, and injections can be eliminated or greatly reduced. Eye charities have run focus groups of people with wet AMD to help set priorities, and participants identified the burden of treating wet AMD as one of their most pressing concerns, as it greatly restricts their freedom and quality of life. Repeated injections also carry risks of rare but blinding complications, and long term they may possibly lead to other eye diseases such as glaucoma and atrophy of the macula. The preliminary SRT studies were not designed to determine if the visual outcome was the same, better, or worse than standard injection therapy, and follow up was too short to determine long term risks. A larger study is therefore required to confirm that SRT reduces the number of injections and to determine if the visual outcomes are acceptable. There is also a need for longer term safety monitoring, as radiation damage typically has a delayed onset. The aim of the proposed study is to determine if SRT can maintain vision in the context of fewer eye injections. SRT is not designed to replace anti-VEGF therapy in all patients, but to reduce the frequency of anti-VEGF therapy. STAR will use ranibizumab as the anti-VEGF agent in both the treatment and control arms. Ranibizumab was chosen over bevacizumab as it is licensed for use in the eye, and at present bevacizumab is used in only a small minority of NHS hospitals, such that the results with may be less generalisable. Bevacizumab may slow recruitment if prospective participants are anxious about swapping to an off label treatment, and preliminary discussions with prospective sites indicated some investigators would prefer to use ranibizumab. Further, ranibizumab was used in the phase II INTREPID study (detailed below), which helps inform the STAR statistical analysis. Aflibercept's mandated dosing in year 1 means it is not possible to determine if radiation reduces the need for anti-VEGF treatment, the primary outcome measure. Whereas the early studies tested a range of doses, this study will only test the dose shown to have the best balance of safety and effectiveness (16 Gray). Participants will be randomly allocated to dummy SRT (0 Gray dose) or live SRT (16 Gray). Neither the treating doctor nor participant will know if a 0 or 16 Gray dose was given. SRT will be performed in approximately 3 national treatment centres, but 25 NHS recruiting hospitals will refer patients for SRT, and then review them afterwards, treating them with ranibizumab as needed. Participants will attend for examination monthly for two years. They will then revert to standard NHS care, but with two further study visits - one at the end year 3 and one at the end of year 4. These two late safety visits are to exclude delayed radiation damage. At the 24 monthly visits participants will have their vision tested on an eye chart, and a laser scan (OCT) of their macula. If the vision drops or the OCT shows disease activity, then they will receive an injection of ranibizumab. Participants will have imaging of the blood vessels in the macula (angiography) at baseline and yearly thereafter for 4 years. The images will be interpreted by an independent, specialist reading centre to quantify the effect of treatment, to study the interaction of radiation and wet AMD, to look for subtle radiation damage, and to subclassify disease to determine who responds best to SRT. We predict that SRT will produce a similar, or possibly better vision than ranibizumab alone, but with fewer injections. With 411 participants we will have 90% certainty that our results are valid. Participants will provide feedback using established questionnaires about how acceptable they found SRT, their visual function, and their overall quality of life. Experienced health economists will then look at the cost of treatment in relation to any improvement in quality of life, to see if SRT is cost-effective. The main outcomes we will monitor are the number of eye injections and visual function on an eye chart, but several other observations will be made. In particular, we will monitor the safety of treatment, any collateral damage caused by radiation, and the interactions between radiation and diseased eye tissue. If proven to be effective, SRT has the potential to save the NHS money, but more importantly, it may reduce the burden of care faced by more than 250,000 people with wet AMD.