The Individually-Marked Panretinal Laser phoTocoagulation for Proliferative Diabetic Retinopathy Study (TREAT)

Introduction Diabetes mellitus is an epidemic disorder, which in Denmark alone is affecting 320,000 patients. Diabetic retinopathy (DR) is the most frequent long term complication to diabetes mellitus (1) and a feared cause of severe vision loss and blindness (2).

Proliferative diabetic retinopathy (PDR) is the major cause of severe visual loss. Lack of oxygen to the retina (retinal ischemia) results in up-regulation of, in particular, the growth factor vascular endothelial growth factor (VEGF) (3) followed by compensatory retinal proliferations. The neovasculature is fragile and often leads to vitreous hemorrhages or retinal detachment which makes the patient at high risk of irreversible vision loss (4).

In 1976 it was shown that patients with severe PDR can halve the risk of severe vision loss by peripheral retinal laser treatment (photocoagulation panretinal, PRP) (5). This treatment reduces the retina's oxygen demand, which makes the VEGF concentration decrease and the proliferations shrink (5).

PRP has largely been the same for the last 40 years. The standard treatment is basically the same for all patients (4 + 6), which leads to some patients being either over or under treated. If treatment is inadequate, patients are in risk of disease progression and thus difficult vision loss (7). On the other hand, the treatment may cause side effects in the form of loss of visual field (8-9), night vision loss (10) and accumulation of fluid in the eye's macula (diabetic macular edema) (11).

This study is a continuum of the clinical project IMPETUS 2018 - DETECT, which aimed to identify the factors that were important for a successful PRP treatment of PDR. In the study the investigators prospectively followed 65 patients with newly diagnosed PDR. All patients received baseline navigated PRP, as in Scandinavia only offered at Odense University Hospital (OUH). Navigated panretinal laser with a Navilas® laser ensures optimized treatment (12), shorter treatment (13) and increased patient comfort (12-14). Treatment effect was investigated at month three and six, and if necessary, treatment was supplied. All the patients venous retinal oxygen saturation was measured to study whether this had any therapeutic value.

The investigators observed that the retinal oxygen saturation was a strong predictor of treatment response. Compared to patients whose disease was slowed down after treatment, patients with progression three months after PRP had an increase in the venous retinal oxygen saturation (+ 4.1% vs. -1.8%, p = 0.02). Patients with an increase of at least 3.0% in venous retinal oxygen saturation had 4.0 times greater risk of disease progression than patients who were below this threshold (15). This observation is in line with another Danish study, which demonstrated that worsening of DR causes increased venous retinal oxygen saturation (16). By measuring if this increase in venous retinal oxygen saturation has slowed down, one can assess whether PRP treatment is sufficient.

PDR is traditionally perceived as an ischemic disease, which initially affects the entire retina. In our above mentioned study the investigators were able to confirm the results regarding the venous retinal oxygen saturation in the affected segment of the retina, in 24 of the patients in the study, who had only one peripheral proliferation. In these patients the oxygen saturation was increased with disease progression (+ 3.9% vs. -1.5%, p = 0.04). This indicates that the focal hypoxia are more important than previously thought, and thus the local treatment of the diseased area may be a treatment option that reduces the processing volume, thereby minimizing potential side effects.

Retinal proliferations are fragile and often leak contrast fluid. When initiating the study, the investigators expected the leakage of fluorescein over time would be the optimal method to assess disease activity, but had to realize that this method was difficult to objectify (17). As an alternative to this objective evaluation, it is possible to observe the structural conditions at the interface between the retina and vitreous body (18), but technological limitations have so far prevented the possibility of repeated evaluations of the same lesion over time. Optical coherency tomography (OCT)-angiography is, however, a new method that can visualize retinal structures and potential development of these in detail (19).

Purpose In a six-month randomized, prospective study of patients with newly diagnosed PDR the investigators want to investigate 1) whether individualized PRP compared with standard PRP has the same efficacy but less side effects and 2) whether OCT angiography can be used as a marker for disease activity in PDR.

Hypothesis The investigators expect that 1) individualized PDR provides the same effect but fewer side effects and better quality of life than traditional PDR, and 2) OCT angiography has better sensitivity and specificity than wide field fluorescein angiography (FA) for the evaluation of disease activity by PDR.

Methods

Setup:

• Six-months 1: 1 randomized, prospective study.
• 58 consecutively recruited patients with newly diagnosed PDR at the Department of Ophthalmology, University Hospital, included in the period 1 March 2017 to 28 February 2018.
• Patients will be randomized to either 1) standard PRP with Navilas® (n = 29) or individualized PRP with Navilas® (n = 29). To ensure the same degree of ischemic disease, the two groups are balanced in relation to the number of retinal quadrants with proliferations.

Intervention:

• Standard PRP: localized to all four retinal quadrants.
• Individualized PRP: localized to the affected quadrants.
• Both treatments are carried out at baseline (BL) and supplemented if there is increasing disease activity at month three (M3) and / or month six (M6).
• Indications for additional treatment:
• Progression of PDR in the form of subjective growing lesion (assessed by ophthalmoscopy and wide field fundus photo) or increasing leakage wide field FA (M3 or M6).
• Progression of PDR in terms of objectively progressive lesion (≥10% from BL) measured by spectral domain (SD) OCT or OCT angiography (M3 or M6).
• Increase in venous retinal oxygen saturation of at least + 3,0% between BL and M3.

Investigations:

• Demographics: age, sex, type of diabetes, diabetes duration, smoking, drugs (BL).
• Objectively: Blood pressure, height, weight (BL).
• Blood samples: HbA1c, total cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, P creatinine, eGFR (BL, M3, M6).
• Visual acuity (Best Corrected Early Treatment Diabetic Retinopathy Study standard) (BL, M3, M6).
• Intraocular pressure (BL, M3, M6).
• SD-OCT (Topcon 3D OCT 2000): macula and area(s) with PDR (BL, M3, M6).
• OCT angiography (Topcon DRI OCT Triton): region(s) with PDR (BL, M3, M6).
• Wide field fundus photo and FA (Optos) (BL, M3, M6).
• Retinal oximetry (Oxymap T1) (BL, M3, M6).
• Dark-adaptation (Goldmann-weeker adaptometer) (BL, M6).
• Perimetry (Humphrey 30-2) (BL, M6).
• Selected components of quality of life questionnaire (Danish translation of Visual Function Questionnaire-25) (BL, M6).

Endpoints

Primary:

• Need for retreatment between the groups (M3 and M6).
• Loss of visual fields between the groups (from BL to M6).
• Change in dark adaptation between the groups (from BL to M6).
• Sensitivity and specificity of OCT angiography as an expression of disease activity in PDR (BL, M3 and M6).

Secondary:

• Change in visual acuity between the groups (from BL to M6).
• Difference in proportion with the development of vitreous haemorrhage between the groups (from BL to M6).
• Need for surgical removal of the vitreous between the groups (from BL to M6)
• Change in quality of life between the groups (from BL to M6).