Introduction

Background

Diabetes mellitus is a major public health problem that affects 415 million people worldwide.1 Diabetic retinopathy is the most common complication of diabetes mellitus. With the increasing prevalence of diabetes mellitus globally, diabetic retinopathy is emerging as the leading cause of avoidable blindness worldwide. The progression and severity of diabetic retinopathy can be delayed by optimal control of medical risk factors such as hyperglycaemia, hypertension and hyperlipidaemia.2 However, despite better control of these well-established risk factors, good uptake of established national diabetic retinopathy screening programmes in many countries and improved patient awareness, diabetic retinopathy remains a significant morbidity, indicating the need for alternative management options for this condition.3–5 The two vision-threatening complications of diabetic retinopathy are diabetic macular oedema (DMO) and proliferative diabetic retinopathy (PDR). DMO is caused by accumulation of excess extracellular fluid in the macula. PDR is characterised by growth of new blood vessels on the retina and, if left untreated, these blood vessels can bleed and scar the retina, causing severe visual loss because of vitreous haemorrhage, retinal detachment and neovascular glaucoma (NVG). Approximately 110,000 people in the UK have PDR and, of these, 14,000 have severe visual loss in both eyes, highlighting the need to address this prevalent public health problem.5,6

Mechanisms of management of proliferative diabetic retinopathy

Multiple molecular mechanisms are involved in the pathogenesis of diabetic retinopathy. However, all lead to a final common pathway of retinal hypoxia and consequent increased levels of angiogenic growth factors, principally vascular endothelial growth factor (VEGF).7 So the aim of treatment options for PDR is either to increase the oxygen availability to the retina or to decrease the VEGF levels.

Panretinal laser photocoagulation (PRP) has been the mainstay of treatment for DMO and PDR for > 40 years based on a robust evidence base.8,9 In PRP, laser burns are applied to the peripheral retinal tissue to destroy the peripheral retina to reduce retinal oxygen consumption. This reduction in hypoxic drive results in decreased growth factor production, principally VEGF, which in turn causes retinal new vessel regression.10,11 About 60% of patients respond to laser with total regression of new vessels within 3 months.12 However, many need supplemental PRP and 4.5% require vitrectomy following PRP.13 Approximately 13% develop visual loss because of the development or worsening of pre-existing macular oedema.14 In addition, PDR may lead to transient or permanent loss of visual function, including peripheral visual field defects, night vision loss, loss of contrast sensitivity, and progression of visual loss in nearly 5% of individuals despite appropriate treatment.15 Peripheral visual field loss may have an effect on the overall visual field required for safe driving. A proportion of patients treated with repeated PRP fail the driving standards in the long term.16,17 There is therefore an unmet need for an alternative treatment option that could either replace or delay the need for laser treatment for PDR.

Recently, intravitreal treatments targeting VEGF, such as aflibercept (Eylea^®, Regeneron, Tarrytown, NY, USA/Bayer Pharma AG, Berlin, Germany), ranibizumab (Lucentis^®, Genentech, S. San Francisco, CA, USA/Roche, Basel, Switzerland), bevacizumab (Avastin^®, Genentech, S. San Francisco, CA, USA/Roche, Basel, Switzerland) and pegaptanib (Macugen^®, Eyetech, New York, NY, USA/Pfizer, New York, NY, USA), have introduced a paradigm shift in the treatment of a wide array of ocular diseases, including neovascular age-related macular degeneration, DMO and retinal vein occlusions. Anti-VEGF treatment has superseded laser treatment and is now the standard of care in patients with centre-involving DMO. However, our therapeutic options for PDR remain limited to laser despite several clinical and preclinical studies indicating that VEGF is the key causative factor of retinal neovascularisation (NV). Evidence that VEGF is a key stimulus for ocular NV was demonstrated by the injection of VEGF into the eye of a non-human primate, which stimulated growth and permeability of new vessels on the retina simulating PDR and also induced NVG.18 There is also clear evidence that hypoxia from decreased retinal perfusion produces VEGF.19,20 Levels of VEGF messenger ribonucleic acid (mRNA) and protein were shown to be elevated in a manner that is spatially and temporally consistent with the role for VEGF in the growth of new vessels.21 Moreover, it has been hypothesised that high VEGF from ischaemic retina may be a positive feedback for capillary closure from leucostasis induced by VEGF.22 The VEGF levels are highest in ocular fluid in patients with PDR than in those with other retinal diseases.23 Vitreous VEGF levels were also observed with active PDR and decreased in eyes with quiescent PDR or eyes previously treated with PRP. There is therefore a lot of evidence that VEGF plays a critical role in the pathogenesis of PDR. Evidence in support of a direct role of anti-VEGF agents in blocking retinal new vessel growth has also been reported using soluble VEGF receptors, anti-VEGF aptamers and VEGF receptor 1-neutralising antisera.24 Recent evidence also indicates that monthly anti-VEGF treatment can reduce the severity and delay the progression of diabetic retinopathy over 24 months.25 Several case series using different anti-VEGF agents have shown that anti-VEGF therapy is effective in causing transient regression of retinal NV in PDR. We conducted a review of literature on this topic before the start of this study and concluded that, although the current evidence points towards the potential for anti-VEGF treatment for PDR to obviate or delay the need for laser treatment, the efficacy, safety and cost-effectiveness of this treatment relative to PRP were unclear.26

As anti-VEGF has superseded macular laser treatment as the treatment of choice for DMO, it is advantageous for both PDR and DMO to be treated with anti-VEGF agents as doing so will reduce health-care burden, patient burden and potentially improve patient outcomes. At the inception of this study, there were two multicentre trials evaluating the efficacy of ranibizumab in PDR. However, these studies included only treatment-naive eyes with high-risk PDR. This group is less prevalent in the NHS because of prompt referrals and treatment of early PDR thanks to our established screening programmes. The majority of our patients being reviewed in the NHS have eyes that have been previously treated with PRP to evaluate the need for further PRP for persistent or new NV. It was therefore necessary to do a similar study in the UK to assess the benefit of anti-VEGF therapy in our patient cohort with PDR. There are enough pre-clinical and short-term clinical data to support an adequately powered trial to compare the efficacy, safety and cost-effectiveness of anti-VEGF therapy with PRP (standard of care) in PDR.

Choice of anti-vascular endothelial growth factor agent in proliferative diabetic retinopathy

The anti-VEGF agents that are currently available include pegaptanib, ranibizumab, bevacizumab and aflibercept. Although pegaptanib is a selective VEGF-A 165 inhibitor, both ranibizumab and bevacizumab are humanised monoclonal antibodies that inhibit all known isomers of VEGF-A.

Aflibercept (previously VEGF Trap-Eye) is a 115-kD decoy receptor fusion protein.27 Aflibercept is capable of binding both VEGF and placental growth factor. The receptor sequences of the aflibercept provide powerful VEGF binding (140 times that of ranibizumab) and the molecule’s intermediate size of 110 kD (compared with 48 kD for ranibizumab and 148 kD for bevacizumab) creates a 1-month intravitreal binding activity that exceeds both ranibizumab and bevacizumab.28 The pivotal Phase III studies that investigated the efficacy and safety of aflibercept in wet age-related macular degeneration (VIEW 1 and 2 trials) showed that monthly and bimonthly aflibercept were non-inferior to monthly ranibizumab at preventing vision loss (< 15-letter loss) with comparable vision gains and safety.29 Year 2 treatment involved both as-needed treatment and mandatory injections every 3 months and this regimen maintained the vision gains from the first year, with an average of 4.2 injections of aflibercept and 4.7 injections of ranibizumab, suggesting a longer durability of aflibercept over ranibizumab. Aflibercept also has higher binding affinity to VEGF than ranibizumab and bevacizumab. A recently published comparative study of the three intravitreal agents in DMO also showed a more rapid and sustained effect of aflibercept over 2 years.30 In addition, aflibercept is already licensed for use in DMO based on the VIVID (Intravitreal aflibercept injection in vision impairment due to DME) and VISTA (Study of intravitreal aflibercept injection in patients with diabetic macular edema) studies. These studies also proved the beneficial effect of aflibercept in improving the diabetic retinopathy severity scale.31 So there is sufficient evidence that aflibercept is as effective and has a longer duration of action than other anti-VEGF agents. Given that PDR is a progressive disease, an agent with a longer duration of action is preferable so we chose aflibercept as the agent of choice for this study. The NHS-discounted cost of aflibercept is similar to the discounted NHS cost of ranibizumab that is available in the public domain.

In summary, based on the existing research at the time of the grant application, we planned to conduct a robust trial with adequate sample size to evaluate the efficacy, safety and cost-effectiveness of aflibercept in a UK patient cohort with PDR over 12 months. In addition, we compared the ocular and systemic effects of this drug with PRP. In the mechanistic substudy, we evaluated the changes induced by aflibercept on retinal new vessels, capillary non-perfusion and retinal intravascular oxygen saturation.

Objectives

To compare the efficacy, safety and cost-effectiveness of intravitreal aflibercept with standard of care, PRP for PDR for 52 weeks, in a Phase IIb randomised active-controlled clinical trial.

Primary objective

To evaluate whether or not mean change in best corrected visual acuity (BCVA) following intravitreal aflibercept therapy is non-inferior to PRP in eyes with PDR at 52 weeks as measured by Early Treatment Diabetic Retinopathy Study (ETDRS) letters.

Secondary objectives

1. To measure the effect of intravitreal aflibercept therapy, relative to PRP, on additional visual function and quality-of-life outcomes, including:
   1. uniocular and binocular Esterman visual fields defects
   2. binocular visual acuity and low-luminance visual acuity
   3. visual acuity outcomes in terms of visual gain or loss
   4. contrast sensitivity using Pelli–Robson charts
   5. vision-related quality of life measured by the National Eye Institute Visual Function Questionnaire (NEI-VFQ-25) and the Retinopathy-Dependent Quality-of-Life Questionnaire (RetDQoL)
   6. diabetic retinopathy treatment satisfaction outcomes using the retinopathy treatment satisfaction questionnaire (RetTSQ)
   7. generic health-related quality of life (HRQoL) using the EuroQol-5 Dimensions, three-level version (EQ-5D-3L), and capability using the Investigating Choice Experiments Capability Measure for Adults (ICECAP-A)
   8. health and social care service use using a Client Service Receipt Inventory (CSRI).
2. To estimate incremental cost-effectiveness of intravitreal aflibercept versus standard PRP treatment at 52 weeks, where effectiveness is measured in terms of BCVA and quality-adjusted life-year (QALY) gain.
3. To determine the proportions of treatment-naive and post-PRP-treated eyes that do not require PRP through 52 weeks after basic treatment of three loading doses of aflibercept in the aflibercept arm and after initial completion of PRP in the PRP arm.
4. To compare the regression pattern at 12 weeks and the regression and reactivation patterns at 52 weeks between arms.
5. To compare the proportion of patients with one-step and three-step improvement or worsening of diabetic retinopathy between treatment arms at 12 and 52 weeks as per schedule of assessment.
6. To explore the difference in safety profile between intravitreal aflibercept and PRP at 52 weeks, in terms of proportion of patients developing macular oedema [defined as a central subfield thickness (CST) of > 300 µm on spectral-domain optical coherence tomography (SD-OCT) because of clinical evidence of macular oedema], any de novo or increase in existing vitreous haemorrhage, de novo or increasing tractional retinal detachment, NVG, and the requirement for vitrectomy. The indication for vitrectomy will be reported.

Objective for substudy on mechanistic evaluation

1. To explore whether or not intravitreal aflibercept compared with PRP causes measurable regression of retinal NV at 12 and 52 weeks in terms of decimal disc area units in four-field colour photographs and fundus fluorescein angiography (FFA).
2. To explore differences in the mean change in retinal vessel calibre and oxygen saturation in eyes treated with intravitreal aflibercept compared with PRP at 12 and 52 weeks.
3. To explore whether or not intravitreal aflibercept reduces angiographically quantifiable areas of retinal non-perfusion compared with PRP through 52 weeks.