Editorial on the Research TopicIdentification of novel biomarkers for retinopathy of prematurity in preterm infants by use of innovative technologies and artificial intelligence
Advances in our understanding of retinopathy of prematurity (ROP) have improved its management and outcomes (–). However, additional research is needed, including our ability to predict those infants at greatest risk of blindness. This special issue provides insight into new methods for identifying predictive and diagnostic factors, and future research that will refine our approaches.
ROP screening protocols have been implemented worldwide and differ based on the different ROP risk profiles (, ). Despite the differences in ROP progression and risk profiles, there is a recognized need to identify predictive and diagnostic biomarkers. Biomarkers should be interpreted as clinical characteristics, along with ocular signs and disease-related molecular expressions in fluids or tissues. New biomarkers can be used to predict ROP risk, refine ROP risk stratification, and potentially lead to treatment strategies tailored to the individual neonate. Innovative technologies, including machine learning and artificial intelligence (AI) tools and imaging methods using deep learning (DL) algorithms, have been used to better define the fundus features of the International Classification of ROP (ICROP), which has been the gold standard (–). These methods have been proposed to reach the level of human decision-making and to detect retinal characteristics that would otherwise go unnoticed during clinical examination (–).
In this research topic collection, we showcase the development and validation of the most recent DL ROP model by Rao et al. which successfully separated Indian neonates into no ROP and ROP from a large dataset of 227,326 anonymized wide-field fundus images from various camera types. Such innovative initiatives are especially important in countries with limited healthcare resources. For any novel AI tool, it should be kept in mind that model accuracy is the single most important element to ensure before clinical use. Common pitfalls include the application of small, skewed datasets, poor data quality, narrow equipment use, population/selection bias, and ethnic, regulatory, and legal issues. Importantly, AI tools also need to be repeatedly evaluated after use in routine clinical practice.
Recently developed handheld non-invasive optical coherence tomography (OCT) has led to the detection of new retinal and vitreous biomarkers in neonates. In this research topic collection, Mangalesh and Toth and Kubsad et al. summarized recent studies using handheld OCT imaging technology to detect biomarkers in neonates. Mangalesh and Toth described new ROP signs like thinning or splitting of the inner retinal layers, neovascular buds, pre-retinal neovascularization, and hyperreflective spots in both the retina and vitreous. In addition, Kubsad et al. highlighted punctate hyperreflective vitreous opacities and bands, cystoid macular edema, foveal deformation, and a thinner choroid. An interesting next step in the advancement of pediatric imaging technology is handheld, non-invasive wide-field OCT angiography, which may provide interesting ways to quantify areas of choroidal and retinal neovascularization, perhaps as indicators of disease activity.
Several studies have pioneered ROP biomarker research in the blood of preterm infants. Many have focused on the role that vascular endothelial growth factor (VEGF) and insulin-like growth factor (IGF-1) play in neovascular development in the neonatal retina (, , –). More recently, advancements have been made in high-throughput omics techniques that allow the exploration of molecular profiles of metabolites, proteins, lipids, gene variants, and micro-RNA. These OMIC studies aim to identify biological pathways regulated in ROP and upstream regulators of several signaling mechanisms important in the pathogenesis of ROP (, ). Recently, the application of customized panels to target specific proteins in these biological processes has become popular. In this research topic collection, Ling et al. used a prospective panel study to investigate serum levels of 40 inflammatory cytokines associated with ROP in neonates. The authors detected independent and significant regulation of some of these blood biomarkers in zone 1 ROP and stage 3 ROP, along with postmenstrual age, respiratory distress syndrome, necrotizing enterocolitis, and sepsis. Further studies in external large-scale datasets are needed to validate their findings and develop new experimental designs with insight into refined hypotheses.
Another important aspect of biomarker discovery is the acquisition of big data from various data sources such as registries, databases, and patient records, along with data management and bioinformatics. Here, Xie et al. presented an innovative approach to drug discovery from big data sources to unravel safe drug potentials for the treatment of ROP. The authors searched ROP genes and related pharmacologically safe molecules from the major established databases, PubMed and NCBI, and used bioinformatics tools to identify safe choices for future treatments.
The guest editorial team anticipates that advancements in innovative technologies, AI tools, and big data will advance future understanding of the pathophysiology of ROP at the individual infant level, leading to personalized ROP management. The success of this development will largely depend on a multidisciplinary collaborative approach between clinicians, researchers, data scientists, and technology developers. Therefore, we encourage a strong collaborative engagement of experts in this expanding field of research and hope that policymakers and foundations will support these endeavors. Most importantly, the success of these efforts will hopefully lead to lifelong improved vision and quality of life for many children around the world.
Statements
Author contributions
CS: Writing – original draft, Writing – review & editing. AF: Writing – original draft, Writing – review & editing. MH: Writing – original draft, Writing – review & editing.
Acknowledgments
This work was supported by grants from the National Institutes of Health (MEH:PI R01EY015130, R01EY017011).
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1.
PatzAHoeckLEDe La CruzE. Studies on the effect of high oxygen administration in retrolental fibroplasia. I. Nursery observations. Am J Ophthalmol. (1952) 35(9):1248–53. 10.1016/0002-9394(52)91140-9
2.
PalmerEA. Results of U.S. randomized clinical trial of cryotherapy for ROP (CRYO-ROP). Doc Ophthalmol. (1990) 74(3):245–51. 10.1007/BF02482615
3.
FungCCungTNelsonCWangHBretzCRamshekarAet alRetinopathy of prematurity protection conferred by uteroplacental insufficiency through erythropoietin signaling in an experimental murine model. Pediatr Res. (2023) 94(3):950–5. 10.1038/s41390-023-02568-4
4.
SauerLChandlerMHartnettME. Extending peripheral retinal vascularization in retinopathy of prematurity through regulation of VEGF signaling. Am J Ophthalmol. (2023) 260:190–9. 10.1016/j.ajo.2023.12.008
5.
HartnettMEStahlA. Laser versus anti-VEGF: a paradigm shift for treatment-warranted retinopathy of prematurity. Ophthalmol Ther. (2023) 12(5):2241–52. 10.1007/s40123-023-00744-7
6.
HartnettME. Pathophysiology and mechanisms of severe retinopathy of prematurity. Ophthalmology. (2015) 122(1):200–10. 10.1016/j.ophtha.2014.07.050
7.
ArunkumarRLiBAddoEKHartnettMEBernsteinPS. Prenatal carotenoid supplementation with lutein or zeaxanthin ameliorates oxygen-induced retinopathy (OIR) in Bco2-/- macular pigment mice. Invest Ophthalmol Vis Sci. (2023) 64(4):9. 10.1167/iovs.64.4.9
8.
HartnettMEDeAngelisMM. The role of reactive oxygen species and oxidative signaling in retinopathy of prematurity. In: Studies on Retinal and Choroidal Disorders. ResearchGate (2012). p. 559–84.
9.
HartnettMEPennJS. Mechanisms and management of retinopathy of prematurity. N Engl J Med. (2012) 367(26):2515–26. 10.1056/NEJMra1208129
10.
Elizabeth HartnettM. The effects of oxygen stresses on the development of features of severe retinopathy of prematurity: knowledge from the 50/10 OIR model. Doc Ophthalmol. (2010) 120:25–39. 10.1007/s10633-009-9181-x
11.
SimmonsABBretzCAWangHKunzEHajjKKennedyCet alGene therapy knockdown of VEGFR2 in retinal endothelial cells to treat retinopathy. Angiogenesis. (2018) 21(4):751–64. 10.1007/s10456-018-9618-5
12.
SlidsborgCJensenAFormanJLRasmussenSBangsgaardRFledeliusHCet alNeonatal risk factors for treatment-demanding retinopathy of prematurity: a danish national study. Ophthalmology. (2016) 123(4):796–803. 10.1016/j.ophtha.2015.12.019
13.
GilbertCFielderAGordilloLQuinnGSemigliaRVisintinPet alCharacteristics of infants with severe retinopathy of prematurity in countries with low, moderate, and high levels of development: implications for screening programs. Pediatrics. (2005) 115(5):e518–25. 10.1542/peds.2004-1180
14.
KościółekMKisielewskaWĆwiklik-WierzbowskaMWierzbowskiPGilbertC. Systematic review of the guidelines for retinopathy of prematurity. Eur J Ophthalmol. (2023) 33:667–81. 10.1177/11206721221126286
15.
ChiangMFJiangLGelmanRDuYEFlynnJT. Interexpert agreement of plus disease diagnosis in retinopathy of prematurity. Arch Ophthalmol. (2007) 125(7):875–80. 10.1001/archopht.125.7.875
16.
SlidsborgCFormanJLFielderARCrafoordSBaggesenKBangsgaardRet alExperts do not agree when to treat retinopathy of prematurity based on plus disease. Br J Ophthalmol. (2012) 96(4):549–53. 10.1136/bjophthalmol-2011-300573
17.
ChiangMFQuinnGEFielderAROstmoSRPaul ChanRVBerrocalAet alInternational classification of retinopathy of prematurity, third edition. Ophthalmology. (2021) 128(10):e51–68. 10.1016/j.ophtha.2021.05.031
18.
ChiomaRSbordoneAPattiMLPerriAVentoGNobileS. Applications of artificial intelligence in neonatology. Applied Sciences (Switzerland). (2023) 13:3211. 10.3390/app13053211
19.
ReddTKCampbellJPBrownJMKimSJOstmoSChanRVet alEvaluation of a deep learning image assessment system for detecting severe retinopathy of prematurity. Br J Ophthalmol. (2019) 103(5):1–11. 10.1136/bjophthalmol-2018-313156
20.
CoynerASOhMAShahPKSinghPOstmoSValikodathNGet alExternal validation of a retinopathy of prematurity screening model using artificial intelligence in 3 low- and middle-income populations. JAMA Ophthalmol. (2022) 140(8):791–8. 10.1001/jamaophthalmol.2022.2135
21.
HartnettME. Vascular endothelial growth factor antagonist therapy for retinopathy of prematurity. Clin Perinatol. (2014) 41(4):925–43. 10.1016/j.clp.2014.08.011
22.
HartnettME. Role of cytokines and treatment algorithms in retinopathy of prematurity. Curr Opin Ophthalmol. (2017) 28(3):282–8. 10.1097/ICU.0000000000000360
23.
BeckerSWangHSimmonsABSuwanmaneeTStoddardGJKafriTet alTargeted knockdown of overexpressed VEGFA or VEGF164 in muller cells maintains retinal function by triggering different signaling mechanisms. Sci Rep. (2018) 8(1):2003. 10.1038/s41598-018-20278-4
24.
McCloskeyMWangHJiangYSmithGWStrangeJHartnettME. Anti-VEGF antibody leads to later atypical intravitreous neovascularization and activation of angiogenic pathways in a rat model of retinopathy of prematurity. Invest Ophthalmol Vis Sci. (2013) 54(3):2020–6. 10.1167/iovs.13-11625
25.
HartmannJSThompsonHWangHKanekarSHuangWBuddSJet alExpression of vascular endothelial growth factor and pigment epithelial-derived factor in a rat model of retinopathy of prematurity. Mol Vis. (2011) 17:1577–87.
26.
HellgrenGLofqvistCHardALHansen-PuppIGramMLeyDet alSerum concentrations of vascular endothelial growth factor in relation to retinopathy of prematurity. Pediatr Res. (2016) 79(1):70–5. 10.1038/pr.2015.181
27.
LofqvistCHansen-PuppIAnderssonEHolmKSmithLELeyDet alValidation of a new retinopathy of prematurity screening method monitoring longitudinal postnatal weight and insulinlike growth factor I. Arch Ophthalmol. (2009) 127(5):622–7. 10.1001/archophthalmol.2009.69
28.
HellstromAEngstromEHardALAlbertsson-WiklandKCarlssonBNiklassonAet alPostnatal serum insulin-like growth factor I deficiency is associated with retinopathy of prematurity and other complications of premature birth. Pediatrics. (2003) 112(5):1016–20. 10.1542/peds.112.5.1016
29.
DanielssonHTebaniAZhongWFagerbergLBrusselaersNHardALet alBlood protein profiles related to preterm birth and retinopathy of prematurity. Pediatr Res. (2022) 91(4):937–46. 10.1038/s41390-021-01528-0
30.
RamshekarABretzCAHartnettME. RNA-Seq Provides insights into VEGF-induced signaling in human retinal microvascular endothelial cells: implications in retinopathy of prematurity. Int J Mol Sci. (2022) 23(13):1–17. 10.3390/ijms23137354
Summary
Keywords
prematurity, retinopathy of prematurity (ROP), biomarker discovery, innovative technologies, artificial intelligence, risk stratification, predictive factors, diagnosis
Citation
Slidsborg C, Fielder A and Hartnett ME (2024) Editorial: Identification of novel biomarkers for retinopathy of prematurity in preterm infants by use of innovative technologies and artificial intelligence. Front. Pediatr. 12:1382858. doi: 10.3389/fped.2024.1382858
Received
06 February 2024
Accepted
07 February 2024
Published
06 March 2024
Volume
12 - 2024
Edited and reviewed by
Eugene Dempsey, University College Cork, Ireland
Updates
Copyright
© 2024 Slidsborg, Fielder and Hartnett.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Carina Slidsborg carina.slidsborg@regionh.dk Alistair Fielder a.fielder@city.ac.uk M. Elizabeth Hartnett me.hartnett@stanford.edu
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.