Anti-vascular Endothelial Growth Factor (Anti-VEGF) Monotherapy vs Anti-VEGF Followed by Subthreshold Micropulse Laser for Treating Severe Diabetic Macular Oedema When the Central Retina Goes <400 Microns

Background DMO is a leading cause of sight loss in people living with diabetes (PLWD). Considering the prevalence of DMO (7%) and diabetes (~4.8 million), around 336,000 PLWD have DMO in the UK. With its increasing incidence, DMO will continue to be a burden to society. In DMO, fluid collects at the macula, the area that gives central vision, leading to sight loss. As fluid accumulates, the macula thickens. This is measured in microns (μm) with optical coherence tomography (OCT) scans. A measure of central retinal subfield thickness (CRT) is obtained with OCT scans when DMO is diagnosed, guiding treatment selection. National Institute for Health and Care Excellence (NICE) guidance advises intraocular injections of anti-VEGF for PLWD presenting with severe (CRT ≥400μm) centre involved DMO. For PLWD presenting with milder forms of centre involved DMO (CRT <400μm), NICE advises macular laser because it is as effective as anti-VEGFs but costs less.

Most patients presenting to UK Hospital Eye Services (HES) have DMO with CRT ≥400μm and are treated with anti-VEGFs. Currently, as per standard practice, anti-VEGFs are given monthly initially (loading dose) and then typically every 1-3 months until the macula dries, even if CRT falls to below 400μm. In 72% of people on anti-VEGFs, the fluid remains until at least the second year of treatment. Trials have shown that after 5 years, 38-48% of participants still require anti-VEGFs either because DMO remains or because it recurs after clearing once anti-VEGFs are stopped. Most patients on anti-VEGFs need follow-up for life. Anti-VEGFs are expensive and carry potential harms including increased intraocular pressure, retinal detachment, cataract, and infection (endophthalmitis). The latter, although rare, can lead to total blindness. Furthermore, intraocular injections cause discomfort to many patients during administration and for hours thereafter and elicit anticipatory stress responses. Research has shown that a significant proportion of patients (25%) experience high levels of preprocedural anxiety, and nearly 10% report high levels of pain. Other experiences of treatment burden, not commonly assessed in previous studies, have been found to be important including time and functional disruption associated with intravitreal injections. Moreover, HES are unable to cope with the demand and injections are not being given in a timely manner, which has a negative impact on outcomes and the cost-effectiveness of the treatment. Finding ways to reduce the number of injections to optimise patient experience and maximise adherence is a goal pursued worldwide.

Trials comparing anti-VEGFs with standard macular laser have shown superior efficacy of anti-VEGFs against laser in severe DMO. These trials included predominantly eyes with very thickened retinas (e.g. >460μm in RISE and RIDE and >479μm in VIVID and VISTA. When anti-VEGFs were used in combination with macular laser in some of these trials, combined treatment (anti-VEGFs and macular laser) did not appear to be superior to anti-VEGFs alone. However, macular laser was not necessarily applied when CRT had gone below 400μm following anti-VEGFs. It would be at this stage that macular laser would have more chance to be effective, as its penetration through the neurosensory retina and areas of macular oedema, and subsequently, its uptake by the retinal pigment epithelium (RPE) would be more adequate when compared with its likely reduced penetration and effect on the RPE when the macula is very thickened by marked DMO.

Rationale DAME will be a pragmatic trial comparing clinical- and cost-effectiveness, side effects and participant experience and acceptability of combined treatment with anti-VEGFs + Subthreshold Macular Laser (SML) for participants who present with severe DMO and are treated initially with anti-VEGF, with the SML applied after CRT falls to <400μm (when laser has more chance to succeed), versus the current standard of care of continuing with anti-VEGF monotherapy even when the CRT falls below 400μm if DMO is present.

DAME follows from the DIAMONDS trial, which showed that SML, which does not damage the macula, is as effective to treat DMO of <400μm as standard laser, which produces a burn. Although DIAMONDS participants had poor glycaemic control (mean HbA1c 8.5%), most maintained excellent sight and fulfilled driving standards throughout their 2-year follow-up. Those treated with SML needed, on average, only two sessions of laser, with clinic visits every 3-4 months and only 18% received rescue treatment with anti-VEGFs, with an average total cost of care of £898 per participant (similar to the cost of the drug in a single anti-VEGF injection).

The hypothesis for DAME is that treating people with severe DMO (CRT ≥400μm) initially with anti-VEGFs and then, when CRT goes below 400μm, continuing with SML every 2-3 months until DMO clears will be as effective but more cost-effective, have fewer side effects and be preferred by people with DMO when compared to continuing with anti-VEGF monotherapy. If SML allows people initially treated with anti-VEGFs to maintain the characteristic early vision gains that are observed following the first few anti-VEGF injections this new strategy could become the new standard of care for people with severe DMO and be implemented worldwide. Potential benefits would include fewer injections with subsequent reduction in inconvenience, stress, harms and costs, and fewer clinic visits, which will facilitate patient's compliance with the treatment and reduce costs and inconvenience to people with severe DMO.

A Cochrane network meta-analysis found that anti-VEGFs improve vision in DMO but concluded "evidence from RCTs may not apply to real-world practice where people are often undertreated and under-monitored". In this regard, a large cohort study from Moorfields Eye Hospital found that half of the patients with DMO treated with anti-VEGFs achieved vision of 70 ETDRS letters at 1.9 months of initiating this therapy but, in 50% of these, vision dropped below this by 14.7 months (i.e. visual gain was not maintained). Similarly, a "real world" analysis of 28,658 eyes of participants with DMO treated with anti-VEGFs found that eyes with good vision at baseline (before initiating anti-VEGF therapy), were at risk of visual loss a year following treatment initiation, highlighting that outcomes observed in anti-VEGF trials are not reproduced in clinical practice.

Rationale for the Intervention Macular laser is likely to be effective in combination with anti-VEGFs in people initially presenting with ≥400μm DMO if the macula laser is applied after the CRT has gone below 400μm following anti-VEGFs. A Single Technology Appraisal by NICE found that for people presenting with DMO and CRT of <300μm, there was no statistically significant difference in efficacy between anti-VEGFs and laser, but laser was more cost-effective. When CRT was between 300μm and 400μm, there were gains in vision of 7 ETDRS letters with anti-VEGFs and 4 ETDRS letters with macular laser, a statistically significant difference but of doubtful clinical relevance; and macular laser dominated in cost-effectiveness.

A randomised trial by the Diabetic Retinopathy Clinical Research Network (Protocol V) including people presenting with DMO and good vision (median 85 ETDRS letters), with CRT of <400μm (median 290μm and 299μm in aflibercept and macular laser arms, respectively) and median HbA1c of 7.6% showed comparable efficacy between aflibercept and macular laser, with 16% and 17% of participants experiencing improvement in best corrected visual acuity (BCVA) of >5 ETDRS letters at 2 years. The DIAMONDS trial showed that macular laser is effective and cost-effective for treating people presenting with DMO and CRT <400μm. Participants who had good vision (median 82 ETDRS letters), median CRT of 331μm and median HbA1c of 8.5% (i.e. more severe disease than those included in Protocol V) maintained good sight throughout the 2-year follow-up (mean change in vision of less than 3 ETDRS letters) with 18% of participants experiencing an improvement in BCVA of >5 ETDRS letters at two years.

In clinical practice, macular laser is offered only to people presenting with new DMO with CRT <400μm but not routinely to those who have started anti-VEGFs, even if, at some point, their CRT is <400μm. The proposed strategy (initial anti-VEGF therapy for people presenting with severe DMO with CRT of ≥400μm followed by macular laser after CRT goes below 400μm) would likely allow participants to achieve visual acuity improvement (which often occurs following the first few anti-VEGF injections but is less frequently observed after macular laser monotherapy) but without the need to continue with anti-VEGF injections long-term. The DAME PPI Group felt that participants would be likely to prefer this new proposed strategy (anti-VEGF followed by SML).

The DIAMONDS trial in people presenting with DMO of <400μm showed that SML, which does not produce any deleterious functional or structural changes in the retina, is as effective as standard laser, which produces a burn. A systematic review on SML for DMO identified 5 small (30-56 eyes in total in each trial) randomised trials comparing anti-VEGFs alone with anti-VEGFs + SML. In 4 of these trials, no statistically significant differences in BCVA were found between treatment groups, whereas in one a significant improvement in BCVA was observed only in the combined anti-VEGF+SML group. A statistically significantly reduced number of anti-VEGF injections was required in the anti-VEGF+SML group in 3 of the 4 trials in which this outcome was investigated. None of the trials included other important outcomes such as health-related and visual-related quality of life, participant-reported experience, adverse events, or costs. The CRT in these trials varied (means of 494-513μm, 462-457μm, 458-470μm, 466-451μm, 433-458μm) and the SML was applied after randomisation (i.e. not when CRT had gone below 400μm following anti-VEGFs).

Rationale for the Comparator The comparator in DAME will be the current standard of care for these patients: continuing with anti-VEGF monotherapy until DMO fully clears or if it recurs after having previously dried, based on OCT scans. Sites will use the type of anti-VEGF they routinely use in their standard clinical practice.

Research Hypothesis DAME will test whether, in people presenting with severe DMO (CRT ≥400μm) who are initially treated with anti-VEGFs, treatment with SML (intervention) after their CRT has decreased to <400μm is equivalent (equivalence margin +/- 5 Early Treatment Diabetic Retinopathy Study [ETDRS] letters) to continuing anti-VEGF monotherapy (control, comparator = standard of care) for preserving/improving BCVA in the study eye at 104 weeks (primary outcome).

Aim To conduct a pragmatic randomised equivalence trial to assess clinical- and cost-effectiveness, safety, participant experience and acceptability of SML applied after CRT is <400μm following initial anti-VEGF injections, compared to continued anti-VEGF monotherapy, in people who originally presented with severe DMO (CRT ≥400μm). DAME includes an assessment of service providers and planners of factors that enable sustainable delivery of the service to participants, assuming a positive trial result, after the study.

Objectives

In people initially presenting with severe DMO (CRT ≥400μm) who receive treatment with anti-VEGFs and once their macular CRT, as determined on OCT scans, has decreased to <400μm:

1. To determine if the clinical effectiveness of anti-VEGFs and SML is equivalent to anti-VEGF monotherapy
2. To determine the cost-effectiveness of anti-VEGFs and SML compared to anti-VEGF monotherapy
3. To evaluate the participant experience and acceptability of anti-VEGFs and SML compared to antiVEGF monotherapy
4. To evaluate the post-trial implementation and scalability of anti-VEGFs and SML

Study Design Pragmatic, allocation-concealed, single-masked (outcome assessors), multicentre, randomised, equivalence trial with an internal pilot.

Internal Pilot An internal pilot will be conducted over the first 6 months of recruitment, to assess feasibility of recruitment and determine if the study should continue to a full trial. Progression criteria will be assessed 6 months after the first participant is randomised. The target recruitment will be an average of one participant per month per open site. With staggered opening of sites, it is anticipated that by 6 months after the first participant is randomised, 16% of the required sample size would be met. Therefore, the internal pilot recruitment target is 42.

Criteria for progression to the full trial are GREEN (Average recruitment rate/site/month: 1; number of sites opened: 12; number of participants recruited: 42): Progress to full trial. AMBER (0.5-0.99; 6-11; 21-41): Discuss feasibility with the TSC and NIHR and develop a recovery plan to reach the recruitment target and evaluate options to improve recruitment, including number of eligible participants identified, percentage of these randomised and reasons for non-randomisation, review of site recruitment performance, and a review of recruitment procedures. RED (<0.5; <6; <21): Decision cessation of the trial with the TSC and NIHR.

Setting At least 20 HES across the UK, with catchment areas that cover diverse populations.

Sequence Generation Eligible participants who provide consent will be randomised 1:1 to receive SML or continue with antiVEGF monotherapy. Minimisation will be used to balance allocation of participants across trial arms for centre, duration of DMO (≤1year, >1year), number of doses of anti-VEGFs received before randomisation (1-6; 7-12), type of anti-VEGF used (ranibizumab, ranibizumab-biosimilar, Brolucizumab, aflibercept, or faricimab) before randomisation, which will be continued throughout the trial unless lack of efficacy is observed and rescue treatment is needed, presenting BCVA [BCVA ≥ 69 ETDRS letters (Snellen equivalent ≥ 20/40; logMAR ≥ 0.3), 24-68 ETDRS letters (Snellen equivalent ≤20/50-20/320; logMAR 0.4-1.2) and CI-DMO (Yes, No).

Allocation Concealment and Implementation An automated system with the allocation concealed to the ophthalmologist or designee randomising the participant will be used to generate the random allocation sequence.

After informed consent, participants will be randomised via an automated web-based system. Each participant will be allocated their own unique trial identifier during the randomisation process, which will be used throughout the study for participant identification.

Recruitment People that may become a potential participant in this trial (i.e. people with DMO ≥400μm eligible for anti-VEGFs and being started on this treatment) will be identified by the clinical assessment team through referrals to HES, through electronic databases or logbooks, or whilst in clinic. A member of the care team will introduce the study to them at any time during their first year of anti-VEGF therapy, which might be before their CRT has gone below 400μm. Participants expressing interest in taking part will be given further verbal and written details by a member of the research team, including the DAME Participant Information Leaflet (PIL). After anti-VEGF treatment is initiated and when the CRT is below 400μm the participant, if eligible based on the DAME eligibility criteria, could be enrolled and randomised (once consented appropriately). When someone consents to join DAME, they will be asked if they agree to be approached at a later date to be invited to take part in a focus group discussion, so that they can be approached and consented when these focus groups are organised. Participants will also be asked if following completion of DAME, data collected as part of their standard care, can be reviewed for future follow-up studies. Participants will be randomised only when their DMO has improved and the CRT is <400μm following treatment with anti-VEGFs. Enrolment can happen at any time after the CRT has gone below 400μm provided that it is still within one year of initiating anti-VEGF therapy.

Screening Participants attending clinic for a routine appointment will be screened to check for eligibility. All participants screened for the study will be documented, including reasons for not being enrolled for those not recruited. A minimal dataset will be recorded (including age, sex at birth, sexual orientation, ethnicity and partial postcode, to determine if there are differences with those willing to participate).

Informed Consent The study will be conducted in accordance with the ethical principles that have their origin in the Declaration of Helsinki. Eligible participants may only be included in the trial after written informed consent is obtained.

Sample Size DAME is powered to demonstrate equivalence of treatment strategies for the primary outcome, which will be assessed 104 weeks after randomisation. Based on two one-sided t-tests at the 2.5% significance level, a 10 ETDRS letter standard deviation (SD) and an equivalence margin of +/- 5 ETDRS letters, with a significance level of 2.5% and power of 90%, a total of 210 participants would be required. Allowing for 20% dropout, will require 264 participants. The proposed sample size of 132 per group will also be sufficient to detect a mean difference between groups of 39.5μm in CRT (based on a SD of 86.8) and 6.85 in NEI-VFQ-25 (based on a SD of 15.1) at 104 weeks, which are important secondary outcomes on this study. These differences in CRT and NEI-VFQ scores are both clinically relevant differences.

Nested Process Evaluation to Assess Implementation and Scalability Aspects Post-Trial A process evaluation will collect qualitative data from service providers and managers across sites using i) Focus groups or individual interviews with clinical, nursing and service management leads (or their representatives) across participating sites. Focus groups (analogous to those described above) or 1:1 interviews will be offered to participants, and applied depending on feasibility, preference and timeliness. They will be carried out when sites/participants have had experience with the intervention. The aim is to recruit a minimum of two participants per site for these focus groups/interviews, purposefully selected to have an overview of how the service was introduced and implemented in their hospital and wider knowledge of service setup and funding considerations for eye care.

ii) Implementation strategy analysis will be conducted after the focus groups/interviews. This will be driven by the ERIC framework (Expert Recommendations for Implementing Change). This will lead to a DAME implementation toolkit, assuming a positive result (i.e. new tested SML pathway is shown to be beneficial).

Data from the nested process evaluation will be subjected to a framework analysis, guided by the Consolidated Framework for Implementation Research (CFIR). CFIR allows formal categorisation of emerging barriers or drivers into specific categories - broadly including external and internal context of the service, people involved and the process of implementation and also the actual clinical intervention itself. To ensure representativeness and local relevance, the analysis will create and include new codes for data that do not fit into existing CFIR categories. For the implementation strategy analysis, the CFIR (and any further) categories of barriers to implementation that participants report will be deductively matched to, firstly, the reported strategies they used to implement the new intervention and also further potential strategies as per the ERIC framework. These will formulate the DAME implementation toolkit, which will articulate explicitly barriers to be expected in setting up and delivering the service and what activities and initiatives could be undertaken to mitigate them and sustain delivery in a NHS setting. The toolkit will include how to select what strategies might be relevant for each NHS adopting site (i.e. how to appraise local barriers and match those to implementation strategies as revealed by the trial's process evaluation).

Study Within a Trial (SWAT) A variety of summaries of the results of DAME will be prepared and understanding and potential impact of these will be compared with a variety of stakeholders (including participants in DAME, members of the public, ophthalmologists, policy makers (e.g. those engaged in guideline production) and medical students). They will be required to give consent before taking part in the SWAT. The findings will inform dissemination plans and provide evidence for dissemination plans in other trials. The format of these summaries will be determined when the results of DAME and their potential implications are known. However, they may include a plain language summary, scientific summary, short abstract, infographic, podcast, and its associated script.