The field of retina is evolving at a breakneck pace. It is remarkable to acknowledge that life as a retina specialist exists primarily because of innovations conceptualized within the last 65 years. As opposed to representing a comprehensive review, the technologies described herein were selected to illuminate historical, current, and future biomedical progress. We specifically focus on therapies modulating signaling pathways, gene-based therapies, mitochondrial therapies, and artificial intelligence.
The transformation of our basic science understanding of ocular neovascularization to the advent of antivascular endothelial growth factor (VEGF) therapies represents an important milestone in translational medicine. As early as the 1950s, it was postulated that retinal neovascularization was directly related to ‘relative retinal anoxia’ leading to ‘an unknown factor x that develops in the tissue and stimulates new vessel formation’ . However, our understanding of retinal disease would depend on progress in oncology research. In the 1970s, it was proposed that tumor growth and progression depends on the ability of a tumor to recruit and support the formation of new blood vessels, which led researchers to pursue a tumor-derived angiogenic factor . In 1989, two Science publications reported groundbreaking advancements in our understanding of angiogenesis. One study reported an endothelial mitogen from pituitary follicular cells, which was named VEGF , whereas another study described a tumor-derived factor purified by its ability to induce vascular permeability, termed vascular permeability factor (VPF) . Later research with gene sequencing revealed VEGF and VPF were the same molecules and fundamentally important in angiogenesis. In 1994, researchers identified VEGF as a likely cause of ocular neovascularization .
The first treatment targeting VEGF was bevacizumab (Avastin), a humanized antibody designed to block all VEGF isoforms. In 1997, Genentech (South San Francisco, USA) began trials of bevacizumab for colon cancer, which proved to increase survival time when combined with other chemotherapeutic drugs . In 2004, the US Food and Drug Administration (FDA) approved bevacizumab for the treatment of colon cancer. Although anti-VEGF therapies were being developed in oncology, VEGF was found to play a central role in age-related macular degeneration (AMD), leading to the development of pegaptanib (Macugen), an RNA aptamer that neutralizes the VEGF isomer 165. Pegaptanib was shown to decrease vision loss in AMD leading to FDA approval in 2004, making it the first therapeutic agent approved for ocular neovascularization .
Soon after bevacizumab was approved for cancer treatment, systemic intravenous bevacizumab was used in an off-label fashion for the treatment of AMD and was found to significantly improve visual acuity . Ophthalmologists then began using intravitreal bevacizumab injections for AMD, which was found to decrease retinal fluid and improve vision in patients with AMD . As bevacizumab is a relatively large molecule, it was initially expected that the drug would not diffuse through the retina sufficiently to reach the choroid, leading Genentech to develop a truncated, recombinant monoclonal antibody Fab, known as ranibizumab (Lucentis) . Ranibizumab was subsequently found to improve visual outcomes for all forms of choroidal neovascularization secondary to AMD in two pivotal trials leading to FDA approval in 2004 . In addition to off-label bevacizumab, there are multiple FDA-approved intravitreal anti-VEGF agents including ranibizumab (Lucentis, Genentech), aflibercept (Eylea, Regeneron), and brolocizumab (Beovu, Novartis), and intravitreal injection of anti-VEGF agents is the most commonly performed procedure in ophthalmology and possibly all of medicine .
The Wnt signaling pathway is also highly relevant to the field of retina [13,14]. Wnt signaling guides tissue fetal tissue differentiation, contributes to angiogenesis, helps maintain the blood-brain and blood–retinal barrier, and promotes tissue regeneration . There are two Wnt pathways: the canonical/B-Catenin pathway and the noncanonical pathway. Norrin is a strong activator of the canonical Wnt pathway encoded by the Norrie Disease Protein gene on the X-chromosome. Norrin binding to the Frizzled 4-cell surface receptor (FZD4) along with low-density lipoprotein receptor-related protein-5 (LRP5) and tetraspanin family member-12 (TSPAN12) leads to the accumulation of B-catenin, a transcription factor that guides the expression of genes promoting vascular and neural health .
Mutations in components of the Wnt signaling pathway may result in a myriad of neurovascular diseases including Norrie disease, familial exudative vitreoretinopathy, retinopathy of prematurity, and Coats disease . In addition, acquired retinal vascular diseases result in tissue ischemia, vascular leakage, tissue edema, and pathologic neovascularization [18,19] that may be improved with Norrin . Norrin may also promote the repair and maintenance of retinal neural elements [20–23]. As fundamental Wnt actuators (including Norrin, FZD4, LRP5, TSPAN12) remain expressed in the adult retina , activation of the Wnt pathway with exogenous Norrin protein represents a potential therapeutic avenue to treat both inherited and acquired retinal disease.
Genetic mutations typically cause retinal disease by forming a protein with decreased or absent function, forming a protein that acquires a new detrimental role, or failing to form a protein at all. Genetic testing is the fundamental first step in the diagnosis and treatment of inherited retinal disease (IRD). The field of genetic testing has made considerable progress from when Watson and Crick  first proposed the double helix in 1953. We are now able to perform next-generation sequencing (sequencing huge numbers of samples at once) through multiple service providers  to identify mutations and label their pathogenicity based on reference genomes. There are two rapidly advancing therapeutic approaches to treat a subset of retinal diseases: gene therapy and/or gene editing.
Gene therapy is best thought of as gene supplementation. This typically involves encoding the wild-type DNA sequence of a target gene into a small and circular ‘plasmid’ packaged within a delivery vector. Although there are a wide variety of mechanisms to introduce an engineered plasmid into the cell, the delivery vector is often a recombinant virus (such as adeno-associated virus (AAV) or lentivirus) . Once within the cell, the plasmid expresses the wild-type DNA, thereby generating a ‘normal’ protein to supplement and/or replace the ‘abnormal’ protein (Fig. 1).
Although there a wide variety of gene therapy trials currently underway for both inherited  and acquired retinal diseases (such as age-related macular degeneration ), the only FDA approved gene therapy for the eye is Voretigene neparvovec-rzyl (Luxturna, SPARK Therapeutics) for the treatment of Leber congenital amaurosis (LCA) caused by mutations in the RPE65 gene. Luxturna is an AAV-2 delivery vector encasing a plasmid encoding wildtype (WT) RPE65 that is delivered through a subretinal injection [30▪]. Without treatment, children with LCA due to biallelic RPE65 mutations usually progress to complete vision loss by the third or fourth decade of life , but with treatment (approved for ≥12 months of age in each eye ≥6 days apart), patients experienced improved vision-based navigation as measured by multiluminance mobility testing .
Although the FDA approval of Luxturna in 2017 was a victory for retina and the field of medicine at large, retinal gene therapy has several limitations in its current form. The RPE65 gene is one of the hundreds of (known) genes that lead to IRDs, accounting for an estimated 2% of cases. Furthermore, gene therapy does not treat dominant-negative mutations and is therefore typically limited to addressing autosomal recessive mutations . Although the treatment effect of luxturna appears to last at least 3–5 years based on Phase 1 studies, the exact duration remains unknown . It also appears that retinal degeneration continues to progress in the presence of treatment, albeit much less so in very young patients . Optogenetics overcomes some of these issues, wherein light-sensitive proteins are introduced to cause well defined cellular events in the presence of light . In the field of retina, light-sensitive proteins are introduced into retinal neurons that have no intrinsic light sensitivity thereby imparting light-sensitivity to more downstream retinal elements when photoreceptors and/or other retinal neural elements are damaged . However, optogenetics has several of its own limitations including distortion of the visual experience when downstream retinal elements initiate the signaling cascade and the need for high-intensity light stimulation .
Gene editing comprises two fundamental steps: the creation of double-stranded DNA (dsDNA) breaks at specific locations and dsDNA break correction with gene correction and/or introduction . There are numerous gene-editing technologies available including the widely investigated clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated systems (Cas), zinc-finger nucleases, and transcription activator-like effector nucleases (TALENS) . Gene-editing technologies may be delivered similarly to gene therapy. CRISPRs interface with Cas to form an RNA-guided protein complex that recognizes a component in the target DNA (protospacer adjacent motif) and cleaves at the target nucleic acid sequence  (Fig. 2). CRISPR–Cas can be used to correct a mutated DNA sequence, induce the expression of certain genes, knock down the expressed of others, introduce foreign DNA into the genome, and modify epigenetic DNA changes .
CRISPR–Cas systems have proven efficacy through ex vivo treatment of induced pluripotent stem cells (iPSCs) derived from patients with inherited retinal diseases [43▪▪,44] and animal models of retinal degeneration [33,44–46]. Allergan and Editas Medicine are currently enrolling patients into the Brilliance Phase 1/2 clinical trial of AGN-151587 (EDIT-101), a CRISPR-based gene editing therapy for LCA10 caused by mutations in the CEP290 gene . Of note, this trial is the world's first in-vivo study of CRISPR-based gene editing in medicine. An ongoing concern with CRISPR–Cas is off-targeting effects which can result in unintended deleterious mutations, but significant progress is being made to increase specificity [43▪▪].
Mitochondria are double membrane-bound organelles present in all eukaryotic cells critical for energy metabolism. Although each cell has 2000–10 000 mitochondria, each mitochondrion contains 8–10 copies of maternally inherited mitochondrial DNA encoding for 37 genes, including 13 proteins that are subunits of the electron transport chain (ETC), and all of the machinery (22 tRNAs, 16S and 12S ribosomal RNA) required to produce those proteins (Fig. 3). The neural retina and retinal pigment epithelium (RPE) are among the most metabolically active tissues in the body and are preferentially affected in mitochondrial disease in part because of the resulting high concentrations of reactive oxygen species (ROS) . In addition to the RPE and neural retina being frequently involved in inherited mitochondrial disease , there is a large body of evidence supporting mitochondrial dysfunction as a predominant mechanism of disease in diabetic retinopathy , retinopathy of prematurity , and AMD [51,52].
Mitochondrial diseases are currently untreatable, in part because of difficulties in modeling and understanding mitochondrial dysfunction. The mitochondrial network is highly dynamic with mitochondria undergoing biogenesis, fusion, fission, and mitophagy . In addition, mitochondria do not readily import RNA, making mitochondrial gene editing with CRISPR–Cas difficult (but possible with TALENS that utilize amino acids for their guide sequence). Our group and others have shown that mitochondrial diseases typically manifest only when the % of mutated versus WT mitochondrial DNA (mitochondrial heteroplasmy) exceeds a threshold amount  (Fig. 4), but heteroplasmy is widely variable between patients with the same mutation, between tissue types in the same patient, and even overtime in cell culture from a specific tissue. In spite of this, many compounds have been utilized with variably efficacy to treat mitochondrial dysfunction in retinal cells within in vitro and animal models of disease including rapamycin , metformin , nicotinamide , resveratrol , humanin , coenzyme Q10 , zeaxanthin , and others .
Tetrapeptide SS-31 (Elamipretide, Stealth BioTherapeutics) is a therapeutic candidate for mitochondrial dysfunction currently undergoing an ongoing Phase 2 trial for AMD patients with noncentral geographic atrophy (NCT03891875). Elamipretide stabilizes cardiolipin in the inner mitochondrial membrane, thereby attenuating the damaging effects of ROS . Another potential therapy is ‘photobiomodulation,’ which has shown promising results in randomized clinical trials . Our group and others have shown that light (especially red to infrared light [590–850 nm]) can improve mitochondrial function because cytochrome c-oxidase in the ETC absorbs light and subsequently increases mitochondrial respiration and Adenosine Triphosphate production [65–67].
The field of artificial intelligence has undergone a resurgence in recent years, primarily because of the significant advancements in image recognition through deep learning  (Fig. 5). Artificial intelligence is being increasingly implemented in medical fields that rely heavily on imaging, including the field of retina. Success has been achieved in image segmentation and classification, and computer-aided diagnosis models are being approved by regulatory bodies with FDA approval for the first medical device using artificial intelligence granted in 2018 [69–73]. After segmentation and diagnosis, the next frontier of artificial intelligence in the field of retina is in predictive modeling which is being developed to predict disease progression and treatment response [74–77] including visual acuity prediction after receiving injections for AMD .
Artificial intelligence models have also illuminated clinical correlations that were previously unimaginable. Deep learning algorithms have predicted demographics, cardiovascular risk factors, and anemia from fundus photos alone [79,80], and future algorithms may provide increasing associations between neurodegenerative and cardiovascular disorders [81,82]. Artificial intelligence systems perform at such a high level through integrating large volumes of data to find subtle patterns among millions of pixels in fundus photographs and billions of voxels in three-dimensional optical coherence tomography scans.
However, many of these models have not been validated on large external real-world datasets. As we are currently tied to methods which require large volumes of curated and labeled training data, the setting of ground truth for these input images has a considerable impact on the final performance metrics of the resulting artificial intelligence models. It becomes increasingly important to set strict labeling and adjudication criteria for these image labels . Diagnostic accuracy studies comparing physicians, artificial intelligence models, and artificial intelligence-augmented physicians are necessary to determine the net benefit of this technology. A recent systematic review was the first of its kind in comparing performance between providers and deep learning for detecting disease in medical imaging . It found artificial intelligence model performance to be equivalent to providers; however, few of the articles analyzed reported diagnostic accuracy with externally validated results.
The retina specialist should not be undervalued, as effective retinal medicine will benefit from artificial intelligence enhancement and not by artificial intelligence replacement. Artificial intelligence will lift and homogenize accuracy of retinal diagnosis while enabling personalized care by predicting functional outcomes and treatment response. This integration of data will provide more time with patients, whether to determine a treatment plan or to contextualize the functional impact of a patient's predicted visual changes [85▪]. The new high-value skills will include interpreting and personalizing recommendations made in conjunction with artificial intelligence, and as more time avails, retina specialists will have more ‘time to be human’ with their patients.
Signaling-pathway therapies, genetic therapies, mitochondrial therapies, and artificial intelligence have shaped retina care as we know it and are poised to further impact the future of retina care. Retina specialists have the privilege and responsibility of shaping this future for the visual health of current and future generations.
Figures designed by E.H.W., MD and S.M., BA, MFA and created by S.M., BA, MFA.
Financial support and sponsorship
E.H.W. is supported in part by a grant from the Claire Giannini Foundation
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Wang Z, Liu CH, Huang S, Chen J. Wnt Signaling
in vascular eye diseases. Prog Retin Eye Res 2018.
2. Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971; 285:1182–1186.
3. Leung DW, Cachianes G, Kuang WJ, et al. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989; 246:1306–1309.
4. Keck PJ, Hauser SD, Krivi G, et al. Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science 1989; 246:1309–1312.
5. Miller JW, Adamis AP, Shima DT, et al. Vascular endothelial growth factor/vascular permeability factor is temporally and spatially correlated with ocular angiogenesis in a primate model. Am J Pathol 1994; 145:574–584.
6. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350:2335–2342.
7. Gragoudas ES, Adamis AP, Cunningham ET Jr, et al. Pegaptanib for neovascular age-related macular degeneration. N Engl J Med 2004; 351:2805–2816.
8. Michels S, Rosenfeld PJ, Puliafito CA, et al. Systemic bevacizumab (Avastin) therapy for neovascular age-related macular degeneration twelve-week results of an uncontrolled open-label clinical study. Ophthalmology 2005; 112:1035–1047.
9. Rosenfeld PJ, Moshfeghi AA, Puliafito CA. Optical coherence tomography findings after an intravitreal injection of bevacizumab (avastin) for neovascular age-related macular degeneration. Ophthalmic Surg Lasers Imaging 2005; 36:331–335.
10. Kim LA, D’Amore PA. A brief history of anti-VEGF for the treatment of ocular angiogenesis. Am J Pathol 2012; 181:376–379.
11. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med 2006; 355:1419–1431.
12. Lau PE, Jenkins KS, Layton CJ. Current evidence for the prevention of endophthalmitis in anti-VEGF intravitreal injections. J Ophthalmol 2018; 2018:8567912.
13. Stern JH, Tian Y, Funderburgh J, et al. Regenerating eye tissues to preserve and restore vision. Cell Stem Cell 2018; 22:834–849.
14. Drenser KA. Wnt signaling
pathway in retinal vascularization. Eye Brain 2016; 8:141–146.
15. Zhang C, Lai MB, Khandan L, et al. Norrin-induced Frizzled 4 endocytosis and endo-lysosomal trafficking control retinal angiogenesis and barrier function. Nat Commun 2017; 8:16050.
16. Wise GN. Retinal neovascularization. Trans Am Ophthalmol Soc 1956; 54:729–826.
17. Drenser KA, Fecko A, Dailey W, Trese MT. A characteristic phenotypic retinal appearance in Norrie disease. Retina 2007; 27:243–246.
18. Thanos A, Todorich B, Trese MT. A novel approach to understanding pathogenesis and treatment of capillary dropout in retinal vascular diseases. Ophthalmic Surg Lasers Imaging Retina 2016; 47:288–292.
19. Tokunaga CC, Chen YH, Dailey W, et al. Retinal vascular rescue of oxygen-induced retinopathy in mice by Norrin. Invest Ophthalmol Vis Sci 2013; 54:222–229.
20. Tokunaga CC, Mitton KP, Dailey W, et al. Effects of anti-VEGF treatment on the recovery of the developing retina following oxygen-induced retinopathy. Invest Ophthalmol Vis Sci 2014; 55:1884–1892.
21. Dailey WA, Drenser KA, Wong SC, et al. Norrin treatment improves ganglion cell survival in an oxygen-induced retinopathy model of retinal ischemia. Exp Eye Res 2017; 164:129–138.
22. Del Debbio CB, Balasubramanian S, Parameswaran S, et al. Notch and Wnt signaling
mediated rod photoreceptor regeneration by Müller cells in adult mammalian retina. PLoS One 2010; 5:e12425.
23. Löffler K, Schäfer P, Völkner M, et al. Age-dependent Müller glia neurogenic competence in the mouse retina. Glia 2015; 63:1809–1824.
24. Clevers H. Eyeing up new Wnt pathway players. Cell 2009; 139:227–229.
25. Watson JD, Crick FH. Genetical implications of the structure of deoxyribonucleic acid. Nature 1953; 171:964–967.
26. Wood EH, Hariprasad SM, Drenser KA. Genetic testing for retina specialists. Ophthalmic Surg Lasers Imaging Retina 2018; 49:292–295.
27. Lino CA, Harper JC, Carney JP, Timlin JA. Delivering CRISPR: a review of the challenges and approaches. Drug Deliv 2018; 25:1234–1257.
28. Garafalo AV, Cideciyan AV, Héon E, et al. Progress in treating inherited retinal diseases: early subretinal gene therapy
clinical trials and candidates for future initiatives. Progress Retinal Eye Res 2019; 100827.
29. Reid CA, Nettesheim ER, Connor TB, Lipinski DM. Development of an inducible anti-VEGF rAAV gene therapy
strategy for the treatment of wet AMD. Sci Rep 2018; 8:11763.
30▪. Davis JL, Gregori NZ, MacLaren RE, Lam BL. Surgical technique for subretinal gene therapy
in humans with inherited retinal degeneration. Retina 2019; 39: (Suppl 1): S2–S8.
This article describes the surgical technique for delivery of gene therapy.
31. den Hollander AI, Roepman R, Koenekoop RK, Cremers FP. Leber congenital amaurosis: genes, proteins and disease mechanisms. Prog Retin Eye Res 2008; 27:391–419.
32. Chung DC, McCague S, Yu ZF, et al. Novel mobility test to assess functional vision in patients with inherited retinal dystrophies. Clin Exp Ophthalmol 2018; 46:247–259.
33. DiCarlo JE, Mahajan VB, Tsang SH. Gene therapy
and genome surgery in the retina. J Clin Invest 2018; 128:2177–2188.
34. Darrow JJ. Luxturna: FDA documents reveal the value of a costly gene therapy
. Drug Discov Today 2019; 24:949–954.
35. Ameri H. Prospect of retinal gene therapy
following commercialization of voretigene neparvovec-rzyl for retinal dystrophy mediated by RPE65 mutation. J Curr Ophthalmol 2018; 30:1–2.
36. Deisseroth K. Optogenetics: 10 years of microbial opsins in neuroscience. Nat Neurosci 2015; 18:1213–1225.
37. Stone EM, Andorf JL, Whitmore SS, et al. Clinically focused molecular investigation of 1000 consecutive families with inherited retinal disease. Ophthalmology 2017; 124:1314–1331.
38. Wood EH, Kreymerman A, Sun Y, et al. Considerations for ophthalmic applications of optogenetics. Acta Ophthalmol 2018; 96:e1037.
39. Ceasar SA, Rajan V, Prykhozhij SV, et al. Insert, remove or replace: a highly advanced genome editing system using CRISPR/Cas9. Biochim Biophys Acta 2016; 1863:2333–2344.
40. Khan SH. Genome-editing technologies: concept, pros, and cons of various genome-editing techniques and bioethical concerns for clinical application. Mol Ther Nucleic Acids 2019; 16:326–334.
41. Barrangou R, Doudna JA. Applications of CRISPR technologies in research and beyond. Nat Biotechnol 2016; 34:933–941.
42. Cabral T, DiCarlo JE, Justus S, et al. CRISPR applications in ophthalmologic genome surgery. Curr Opin Ophthalmol 2017; 28:252–259.
43▪▪. Cho GY, Schaefer KA, Bassuk AG, et al. CRISPR genome surgery in the retina in light of off-targeting. Retina 2018; 38:1443–1455.
This article describes the off targeting effects of CRISPR in the retina.
44. Bassuk AG, Zheng A, Li Y, et al. Precision medicine: genetic repair of retinitis pigmentosa in patient-derived stem cells. Sci Rep 2016; 6:19969.
45. Wu WH, Tsai YT, Justus S, et al. CRISPR repair reveals causative mutation in a preclinical model of retinitis pigmentosa. Mol Ther 2016; 24:1388–1394.
46. Sheridan C. Go-ahead for first in-body CRISPR medicine testing. Nat Biotechnol 2018; Available from: https://www.nature.com/articles/d41587-018-00003-2
[cited 2020 Feb 16].
47. Fisher CR, Ferrington DA. Perspective on AMD pathobiology: a bioenergetic crisis in the RPE. Investig Opthalmol Vis Sci 2018; 59:AMD41.
48. Zhu CC, Traboulsi EI, Parikh S. Ophthalmological findings in 74 patients with mitochondrial disease. Ophthalmic Genet 2017; 38:67–69.
49. Barot M, Gokulgandhi MR, Mitra AK. Mitochondrial dysfunction in retinal diseases. Curr Eye Res 2011; 36:1069–1077.
50. Beharry KD, Valencia GB, Lazzaro DR, Aranda JV. Pharmacologic interventions for the prevention and treatment of retinopathy of prematurity. Semin Perinatol 2016; 40:189–202.
51. Brown EE, DeWeerd AJ, Ildefonso CJ, et al. Mitochondrial oxidative stress in the retinal pigment epithelium (RPE) led to metabolic dysfunction in both the RPE and retinal photoreceptors. Redox Biol 2019; 24:101201.
52. Persad PJ, Heid IM, Weeks DE, et al. Joint analysis of nuclear and mitochondrial variants in age-related macular degeneration identifies novel loci TRPM1 and ABHD2/RLBP1. Invest Ophthalmol Vis Sci 2017; 58:4027–4038.
53. Dey S, Sidor A, O’Rourke B. Compartment-specific control of reactive oxygen species scavenging by antioxidant pathway enzymes. J Biol Chem 2016; 291:11185–11197.
54. Stewart JB, Chinnery PF. The dynamics of mitochondrial DNA heteroplasmy: implications for human health and disease. Nat Rev Genet 2015; 16:530–542.
55. Hyttinen JM, Viiri J, Kaarniranta K, Błasiak J. Mitochondrial quality control in AMD: does mitophagy play a pivotal role? Cell Mol Life Sci 2018; 75:2991–3008.
56. Xu L, Kong L, Wang J, Ash JD. Stimulation of AMPK prevents degeneration of photoreceptors and the retinal pigment epithelium. Proc Natl Acad Sci USA 2018; 115:10475–10480.
57. Saini JS, Corneo B, Miller JD, et al. Nicotinamide ameliorates disease phenotypes in a human iPSC model of age-related macular degeneration. Cell Stem Cell 2017; 20:635.e7–647.e7.
58. King RE, Kent KD, Bomser JA. Resveratrol reduces oxidation and proliferation of human retinal pigment epithelial cells via extracellular signal-regulated kinase inhibition. Chemicobiol Interact 2005; 151:143–149.
59. Sreekumar PG, Ishikawa K, Spee C, et al. The mitochondrial-derived peptide humanin protects RPE cells from oxidative stress, senescence, and mitochondrial dysfunction. Invest Ophthalmol Vis Sci 2016; 57:1238–1253.
60. Zhang X, Tohari AM, Marcheggiani F, et al. Therapeutic potential of co-enzyme Q10 in retinal diseases. Curr Med Chem 2017; 24:4329–4339.
61. Zou X, Gao J, Zheng Y, et al. Zeaxanthin induces Nrf2-mediated phase II enzymes in protection of cell death. Cell Death Dis 2014; 5:e1218.
62. Andreux PA, Houtkooper RH, Auwerx J. Pharmacological approaches to restore mitochondrial function. Nat Rev Drug Discov 2013; 12:465–483.
63. El-Hattab AW, Zarante AM, Almannai M, Scaglia F. Therapies for mitochondrial diseases and current clinical trials. Mol Genet Metab 2017; 122:1–9.
64. Markowitz SN, Devenyi RG, Munk MR, et al. A double-masked, randomized, sham-controlled, single-center study with photobiomodulation for the treatment of dry age-related macular degeneration. Retina 2019.
65. Sivapathasuntharam C, Sivaprasad S, Hogg C, Jeffery G. Aging retinal function is improved by near infrared light (670 nm) that is associated with corrected mitochondrial decline. Neurobiol Aging 2017; 52:66–70.
66. Ao J, Wood JP, Chidlow G, et al. Retinal pigment epithelium in the pathogenesis of age-related macular degeneration and photobiomodulation as a potential therapy? Clin Experiment Ophthalmol 2018; 46:670–686.
67. Kirk DK, Gopalakrishnan S. Photobiomodulation reduces photoreceptor death and regulates cytoprotection in early states of P23H retinal dystrophy. SPIE Vol 2013.
68. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015; 521:436–444.
69. De Fauw J, Ledsam JR, Romera-Paredes B, et al. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat Med 2018; 24:1342–1350.
70. Abràmoff MD, Lavin PT, Birch M, et al. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. NPJ Digit Med 2018; 1:39.
71. Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA 2016; 316:2402–2410.
72. Lee CS, Tyring AJ, Deruyter NP, et al. Deep-learning based, automated segmentation of macular edema in optical coherence tomography. Biomed Opt Express 2017; 8:3440–3448.
73. Saha S, Nassisi M, Wang M, et al. Automated detection and classification of early AMD biomarkers using deep learning. Sci Rep 2019; 9:10990.
74. Wen JC, Lee CS, Keane PA, et al. Forecasting future Humphrey Visual Fields using deep learning. PLoS One 2019; 14:e0214875.
75. Arcadu F, Benmansour F, Maunz A, et al. Deep learning algorithm predicts diabetic retinopathy progression in individual patients. NPJ Digit Med 2019; 2:92.
76. Yan Q, Weeks DE, Xin H, et al. Deep-learning-based prediction of late age-related macular degeneration progression. Ophthalmology 2019.
77. Babenko B, Balasubramanian S, Blumer KE, et al. Predicting progression of age-related macular degeneration from fundus images using deep learning. arXiv [csCV] 2019.
78. Schmidt-Erfurth U, Bogunovic H, Sadeghipour A, et al. Machine learning to analyze the prognostic value of current imaging biomarkers in neovascular age-related macular degeneration. Ophthalmol Retina 2018; 2:24–30.
79. Poplin R, Varadarajan AV, Blumer K, et al. Predicting cardiovascular risk factors from retinal fundus photographs using deep learning [Internet]. arXiv [csCV] 2017.
80. Mitani A, Huang A, Venugopalan S, et al. Detection of anaemia from retinal fundus images via deep learning. Nat Biomed Eng [Internet] 2019.
81. Schrijvers EM, Buitendijk GH, Ikram MK, et al. Retinopathy and risk of dementia: the Rotterdam Study. Neurology 2012; 79:365–370.
82. Chan VT, Sun Z, Tang S, et al. Spectral-domain OCT measurements in Alzheimer's disease: a systematic review and meta-analysis. Ophthalmology 2019; 126:497–510.
83. Krause J, Gulshan V, Rahimy E, et al. Grader variability and the importance of reference standards for evaluating machine learning models for diabetic retinopathy. Ophthalmology 2018; 125:1264–1272.
84. Liu X, Faes L, Kale AU, et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. Lancet Digit Health 2019; 1:e271–e297.
85▪. Korot E, Wagner SK, Faes L, et al. Will AI replace ophthalmologists? Transl Vis Sci Technol 2020; 9:2–12.
This article describes skillful applications of artificial intelligence in ophthalmology.