OpticAl Coherence TomograpHy Angiography for the dEtection of Neovascular Age-related Macular Degeneration: a Comprehensive Diagnostic Accuracy Study - the ATHENA study

Background:

Age-related Macular Degeneration (AMD) is the most frequent cause of blindness in the UK. Nearly 200 people aged 55-years or above in the UK lose sight daily from the severe form of this condition (neovascular AMD, nAMD). Early detection of nAMD is associated with improved visual outcomes and quality of life.

The diagnosis of this condition is based on the interpretation of multiple imaging tests, including Fluorescein Angiography (FA) which is invasive and not patient friendly. Non-invasive imaging modalities commonly used for nAMD detection, such as Optical Coherence Tomography (OCT), are not always specific requiring further confirmation with other tests such as FA.

Optical Coherence Tomography-Angiography (OCTA) is a more recent technology already available in many NHS Trusts; it is non-invasive, fast and has the potential to replace FA. However, to date there is only a small body of moderate quality evidence to suggest OCTA is sufficiently accurate for the detection of nAMD.

Aim:

To perfom a comprehensive assessment of the diagnostic accuracy of OCTA and FA both alone and in combination with OCT for the detection of nAMD in patients with a positive or suspicious OCT. Primary Objective:

To assess whether the sensitivity and specificity of OCTA in patients with a positive or suspicious OCT is non-inferior to that of FA for the detection of nAMD as interpreted by clinicians. Secondary objectives:

1. To assess the diagnostic accuracy of OCTA alone and FA alone as interpreted by Reading Centre expert graders for the detection of nAMD in patients with a positive or suspicious OCT.

2. To assess the positive predictive value (PPV) of OCT for the detection of nAMD in all patients presenting with suspicion of nAMD.

3. To assess intra- and inter-rater agreement in the detection of nAMD on OCTA and FA as assessed by Reading Centre graders. 4. To develop and validate criteria for the OCTA-based diagnosis of nAMD. 5. To estimate the incremental cost per true positive detected and incremental cost per correct diagnosis for nAMD.

Methods:

Study Design: Non-inferiority, prospective, randomised diagnostic accuracy study with an internal pilot to confirm feasibility of the recruitment plan.

Patient Population: Patients presenting with suspicion of nAMD in secondary care opthalmology units in the first or second eye.

Setting: Recruitment will take place in 20 hospital-based eye units. Of those 6 will be involved in the pilot study and all sites will be involved in the main study.

Analysis Plan: Each patient with a positive or suspicious OCT for nAMD will undergo both OCTA and FA and will be allocated to either the OCTA or the FA arm of the study. The allocation indicates which image will be used by the recruiting clinician to produce a first clinical diagnosis of nAMD. The two arms will be compared for non-inferiority.

A second clinical diagnosis will be produced for each case by the same clinician based on review of both OCTA and FA. A masked review of OCTA alone and FA alone by expert Reading Centre graders will also be conducted.

Reference Standard: To provide the reference standard we will use all available imaging information from multi-modal imaging (except OCTA to minimise incorporation bias) with clinical vignette describing patient symptoms. Timelines: 1) Development phase: 6 months 2) Internal Pilot: 6 months 3) Main study: 18 months

4) Data analysis and reporting of results: 6 months