Retinal prosthesis is a general term for implantable devices that partially restore vision by electrically or chemically stimulating the remaining inner retinal cells (bipolar cells and retinal ganglion cells) in patients who have lost photoreceptors due to retinitis pigmentosa (RP) or age-related macular degeneration (AMD).
The concept of a visual assist device dates back to Le Roy in 1755, and Tassicker proposed retinal stimulation in 1956. Since then, advances in electrode array technology have led to the current form. Several devices have reached the clinical stage. 4)
In target patients, photoreceptors are lost, but a considerable number of RGCs in the inner retina survive. 4) The survival of these RGCs is the biological basis for artificial vision.
The main candidates are patients with advanced photoreceptor degeneration due to retinitis pigmentosa, whose visual acuity has declined to light perception or worse. For Argus II, the approved indication is age 25 or older with bare light perception or worse. Survival of the inner retina (RGCs) is a prerequisite and is confirmed by preoperative evaluation.
In retinitis pigmentosa (RP), the main indication for retinal prostheses, the following course is observed.
Typical fundus findings in RP are as follows. 3)
Fundus Findings
Bone-spicule pigmentation: A characteristic finding distributed mainly around the equator.
Optic atrophy and waxy pallor: Pallor of the optic disc, prominent in long-standing cases.
Vascular attenuation: Marked narrowing of retinal arteries.
Electrophysiological and Imaging Findings
Electroretinogram (ERG) non-recordable or severely reduced: Essential for definitive diagnosis.
Loss of ellipsoid zone (EZ) on OCT: Loss of the inner segment/outer segment junction of photoreceptors indicates photoreceptor degeneration.
Ring-shaped hyperautofluorescence on FAF (fundus autofluorescence): Indicates the border of active degeneration.
After implantation of a retinal prosthesis, patients may be able to identify object outlines and motion. However, the vision provided by current devices is limited and does not restore natural sight4).
Functional assessment by electroretinogram is essential for definitive diagnosis. 3) OCT evaluates the status of the ellipsoid zone (EZ), and fundus autofluorescence confirms ring-shaped hyperautofluorescence. Genetic testing is also recommended; the EYS gene accounts for 20–30% of cases in Japanese individuals. 3)
RP is a group of inherited retinal degenerative diseases caused by mutations in over 100 genes. 2) The prevalence is 1 in 4,000–8,000 people, and it is the leading cause of congenital blindness in Japan. 3)
In geographic atrophy (GA), the end stage of AMD, photoreceptors and RPE in the fovea atrophy, leading to loss of central vision. Some retinal prostheses, such as PRIMA, are also being studied for central vision impairment due to GA4).
Assessment of RP progression is essential for evaluating candidacy for retinal prostheses.
The approved indications for Argus II are as follows:
RP is a designated intractable disease under the Intractable Disease Act, and after diagnosis, patients can apply for medical expense subsidies. 3)
Currently, clinically used retinal prosthetic devices are mainly electrical stimulation types. They are classified into epiretinal type (inner retina) and subretinal type (under the retina) based on the placement of the device.
Major electrical stimulation devices are shown below.
| Device | Number of electrodes | Placement |
|---|---|---|
| Argus II | 60 electrodes | Epiretinal |
| Alpha IMS/AMS | 1500 pixels | Subretinal |
| PRIMA | 378 pixels | Subretinal |
Approved by the FDA in 2013, it was one of the earliest commercial retinal prostheses. It combines camera-equipped glasses with an external processor and stimulates the retina with a 60-electrode array. Among all regulatory-approved retinal prostheses, it has been implanted in over 500 people worldwide 4).
A subretinal array of 1500 photodiode-electrode units. It is a self-contained device that does not require external power, converting incoming light directly into electrical signals.
A 378-pixel photoelectric conversion chip placed subretinally. It is primarily indicated for AMD (geographic atrophy). Power is supplied by near-infrared laser light from goggles.
As an alternative to electrical stimulation, an approach that locally releases the neurotransmitter glutamate to stimulate the inner retina is being studied. 1) Compared to electrical stimulation, it has the following characteristics.
Current devices do not restore natural vision. Due to limitations in electrode count and resolution, their main functions are discrimination of brightness, contours, and motion. This technology is still at a stage where it is positioned as an aid for daily life.
In degenerative retinas such as RP, after photoreceptor death, the inner retina undergoes gradual remodeling. 1) This remodeling directly affects the efficacy of artificial vision.
Remodeling progresses in three stages. 1)
| Stage | Main Changes |
|---|---|
| Phase 1 | Shortening and contraction of rod outer segments |
| Phase 2 | Rod death and neural circuit reorganization |
| Phase 3 | Severe neural remodeling and Müller cell hypertrophy |
In Phase 3, Müller cells proliferate and form a fibrous scar spanning the entire retina. 1) This change impedes contact between the electrode and target cells, reducing stimulation efficiency.
In the normal retina, light signals are processed via glutamate into the OFF pathway (inhibiting ON bipolar cells) and the ON pathway (stimulating OFF bipolar cells). 1) It has been shown that the degenerated retina retains similar glutamate responses, 1) providing the biological basis for chemical stimulation.
Glutamate toxicity (excitotoxicity) becomes a problem with excessive exposure. Remaining Müller cells are responsible for glutamate reuptake, but this function declines as degeneration progresses. 1)
RGCs in the inner retina survive for a long time after degeneration. 4) Studies in animal models (S334ter rats, rd1 mice) 1) have confirmed that RGCs remain after photoreceptor degeneration is complete and respond to electrical and chemical stimulation. This RGC survival is the biological basis for retinal prostheses.
Animal implantation experiments of a chemical artificial vision device using glutamate are ongoing. 1) Biocompatibility of SU-8 material, precise flow control by electroosmotic flow, and in vivo experiments in S334ter rats and rd1 mice have been reported. 1) Challenges for clinical application include device miniaturization, long-term stability, and avoidance of glutamate toxicity.
This is a method to reconstruct vision using only light stimulation by introducing light-sensitive proteins (such as channelrhodopsin) into remaining RGCs. 2)
In the GS030 trial by GenSight Biologics, ChrimsonR (a red-shifted channelrhodopsin) was introduced via adeno-associated virus (AAV) vector in patients blinded by RP. In one case, perception of specific visual stimuli was reported. 2)
The biggest difference from electrical stimulation types is that it does not require electrode implantation and is less invasive. However, at this stage, the vision obtained is limited.
Gene replacement therapy using AAV vectors has been approved for retinal degeneration due to RPE65 gene mutations (Luxturna). 2) Research is expanding to other gene mutations (over 100), 2) and future combinations of gene therapy and artificial vision may be considered.
Retinal prostheses require long-term rehabilitation, device updates, and maintenance after implantation. In the development of next-generation devices, not only improvement of visual function but also patient support systems and product sustainability are important evaluation items. 4)
Currently, it is at the early clinical trial stage, 2) and has not yet reached the stage of being offered as general medical care. The ability to apply regardless of the presence of target gene mutations is promising, but verification of the quality, safety, and long-term effects of the obtained vision is necessary.
