Adapting for the Future

Just as we might want to look closer at one star from the night sky, sometimes we want to see individual cells of the retina – and adaptive optics (AO) imaging technology lets us do this. Many of you reading this will be familiar with the concept of AO; it is a technology that improves the performance of optical systems by reducing the effects of wavefront distortions (that mostly come from the lens and the cornea), allowing resolution fine enough to visualize single photoreceptor, retinal blood vessels, or retinal surface. AO has three important components: a wavefront sensor, a controller and a wavefront corrector, which is usually a deformable mirror that corrects the distortion. The technology can be coupled to available imaging systems, such as fundus cameras or scanning laser ophthalmoscopes (SLO); for example, we are using an AO-confocal SLO system that was built at our institute. But how can AO help patients?

We have been using the technology to study the retina in normal and in disease states. Until recently, only cones could be imaged with AO systems, but scientists at the imaging laboratory of our institute were able to modify the system to show that rods can also be imaged (1). One interesting feature we have seen is that rod and cone reflectance varies over time; the brightness of rods and cones changes throughout the day, almost like they are twinkling (2). As both rods and cones show similar patterns and durations of “twinkling,” it could be related to circadian rhythm or a metabolic cycle, and we are planning to study this further to elucidate how these changes may indicate photoreceptor function. In retinal diseases, such as cone-rod dystrophy, we have observed bright cones as well as areas of darkness. We have also been studying macular telangiectasia (MacTel), and we also see bright and dark cones. The areas of darkness may indicate absence of cells, while dark cones may indicate abnormal or cones that are not waveguiding. Therefore, the location and the number of dark cones, the extent of dark areas, and the variation of cone density from normal could all be useful biomarkers of disease activity and tracking progression in retinal conditions (3).

I have also found AO useful in examining surgical patients because it can show abnormalities that are not visible with optical coherence tomography (OCT). We have imaged a number of patients following successful surgical closure of macular hole (4). Even though OCT shows excellent anatomic restoration in these eyes, we found significant photoreceptor disruption on AO imaging. We also found that the cone mosaic continues to remodel for over one to two years – which may explain why there can be continued improvement and symptoms of metamorphopsia after surgery in these patients.

Retinal imaging with AO is not limited to the photoreceptors (5). We can go deeper into the eye and see retinal pigment epithelium (RPE) cells and study how disease affects the RPE layer; we can also image the retina superficially at the nerve fiber layer, optic nerve and lamina cribrosa. Due to exquisite resolution, we can image blood flow without the use of a fluorescein dye.

We have found that patients with significantly advanced retinal disease with decreased numbers of photoreceptors compared to normal retina can still have 20/20 visual acuity. Therefore, there may be a “disconnect” between the anatomy (e.g. cone health and number) and function (e.g. visual acuity) at times. Given that accurate assessment of the type and the extent of diseased photoreceptors might be pre-requisite to improving the visual outcome following gene therapy in humans, AO could play an important role in helping identify which patients might be the best candidates for gene therapies. It can be used to monitor non-invasively which patients are anatomically responding to treatment before such information can be obtained by any other method. Therefore, the high image resolution capability of AO to help select ideal candidates may make it an excellent research tool for those who are studying gene therapy.

As wonderful as the AO images are, there are currently issues that prevent wide usage in the clinic. While there is a type of AO fundus camera currently in the market, it is not yet FDA approved. Therefore, in the United States, it is used in a research setting. The type of AO cameras that are being used in the research labs such as ours tend to have higher resolution and can shed more information. However, the imaging time and processing times tend to be quite long at this time. It is hoped that in the future, these deficiencies that prevent use in the clinic can be overcome, allowing the clinicians to be able to fully utilize capabilities of AO imaging.

In summary, functional tests do not tell us everything. AO is a non-invasive imaging technology that can help us detect photoreceptor loss early and may assist us in the future as a biomarker of disease onset and progression. It can help with the selection of ideal candidates for therapies and earlier detection of treatment effect. I believe it will play an important role in the future for assessing the therapeutic potential and outcomes in patients with retinal disorders.