Proteomic analysis of diabetic retinopathy identifies potential plasma-protein biomarkers for diagnosis and prognosis

OBJECTIVES

To identify molecular pathways and prognostic- and diagnostic plasma-protein biomarkers for diabetic retinopathy at various stages.

METHODS

This exploratory, cross-sectional proteomics study involved plasma from 68 adults, including 15 healthy controls and 53 diabetes patients for various stages of diabetic retinopathy: non-diabetic retinopathy, non-proliferative diabetic retinopathy, proliferative diabetic retinopathy and diabetic macular edema. Plasma was incubated with peptide library beads and eluted proteins were tryptic digested, analyzed by liquid chromatography-tandem mass-spectrometry followed by bioinformatics.

RESULTS

In the 68 samples, 248 of the 731 identified plasma-proteins were present in all samples. Analysis of variance showed differential expression of 58 proteins across the five disease subgroups. Protein-Protein Interaction network (STRING) showed enrichment of various pathways during the diabetic stages. In addition, stage-specific driver proteins were detected for early and advanced diabetic retinopathy. Hierarchical clustering showed distinct protein profiles according to disease severity and disease type.

CONCLUSIONS

Molecular pathways in the cholesterol metabolism, complement system, and coagulation cascade were enriched in patients at various stages of diabetic retinopathy. The peroxisome proliferator-activated receptor signaling pathway and systemic lupus erythematosus pathways were enriched in early diabetic retinopathy. Stage-specific proteins for early - and advanced diabetic retinopathy as determined herein could be 'key' players in driving disease development and potential 'target' proteins for future therapies. For type 1 and 2 diabetes mellitus, the proteomic profiles were especially distinct during the early disease stage. Validation studies should aim to clarify the role of the detected molecular pathways, potential biomarkers, and potential 'target' proteins for future therapies in diabetic retinopathy.

Whole exome sequencing of patients with varicella-zoster virus and herpes simplex virus induced acute retinal necrosis reveals rare disease-associated genetic variants

Purpose

Herpes simplex virus (HSV) and varicella-zoster virus (VZV) are neurotropic human alphaherpesviruses endemic worldwide. Upon primary infection, both viruses establish lifelong latency in neurons and reactivate intermittently to cause a variety of mild to severe diseases. Acute retinal necrosis (ARN) is a rare, sight-threatening eye disease induced by ocular VZV or HSV infection. The virus and host factors involved in ARN pathogenesis remain incompletely described. We hypothesize an underlying genetic defect in at least part of ARN cases.

Methods

We collected blood from 17 patients with HSV-or VZV-induced ARN, isolated DNA and performed Whole Exome Sequencing by Illumina followed by analysis in Varseq with criteria of CADD score > 15 and frequency in GnomAD < 0.1% combined with biological filters. Gene modifications relative to healthy control genomes were filtered according to high quality and read-depth, low frequency, high deleteriousness predictions and biological relevance.

Results

We identified a total of 50 potentially disease-causing genetic variants, including missense, frameshift and splice site variants and on in-frame deletion in 16 of the 17 patients. The vast majority of these genes are involved in innate immunity, followed by adaptive immunity, autophagy, and apoptosis; in several instances variants within a given gene or pathway was identified in several patients.

Discussion

We propose that the identified variants may contribute to insufficient viral control and increased necrosis ocular disease presentation in the patients and serve as a knowledge base and starting point for the development of improved diagnostic, prophylactic, and therapeutic applications.

Automated artificial intelligence-based system for clinical follow-up of patients with age-related macular degeneration

PURPOSE

In this study, we investigate the potential of a novel artificial intelligence-based system for autonomous follow-up of patients treated for neovascular age-related macular degeneration (AMD).

METHODS

A temporal deep learning model was trained on a data set of 84 489 optical coherence tomography scans from AMD patients to recognize disease activity, and its performance was compared with a published non-temporal model trained on the same data (Acta Ophthalmol, 2021). An autonomous follow-up system was created by augmenting the AI model with deterministic logic to suggest treatment according to the observe-and-plan regimen. To validate the AI-based system, a data set comprising clinical decisions and imaging data from 200 follow-up consultations was collected prospectively. In each case, both the autonomous AI decision and original clinical decision were compared with an expert panel consensus.

RESULTS

The temporal AI model proved superior at detecting disease activity compared with the model without temporal input (area under the curve 0.900 (95% CI 0.894-0.906) and 0.857 (95% CI 0.846-0.867) respectively). The AI-based follow-up system could make an autonomous decision in 73% of the cases, 91.8% of which were in agreement with expert consensus. This was on par with the 87.7% agreement rate between decisions made in the clinic and expert consensus (p = 0.33).

CONCLUSIONS

The proposed autonomous follow-up system was shown to be safe and compliant with expert consensus on par with clinical practice. The system could in the future ease the pressure on public ophthalmology services from an increasing number of AMD patients.

Assessment of Anterior Uveitis Through Anterior-Segment Optical Coherence Tomography and Artificial Intelligence-Based Image Analyses

Purpose

The purpose of this study was to develop an automated artificial intelligence (AI) based method to quantify inflammation in the anterior chamber (AC) using anterior-segment optical coherence tomography (AS-OCT) and to explore the correlation between AI assisted AS-OCT based inflammation analyses and clinical grading of anterior uveitis by Standardization of Uveitis Nomenclature (SUN).

Methods

A prospective double blinded study of AS-OCT images of 32 eyes of 19 patients acquired by Tomey CASIA-II. OCT images were analyzed with proprietary AI-based software. Anatomic boundaries of the AC were segmented automatically by the AI software and Spearman's rank correlation between parameters related to AC cellular inflammation were calculated.

Results

No significant (p = 0.6602) differences were found between the analyzed AC areas between samples of the different SUN grading, suggesting accurate and unbiased border detection/AC segmentation. Segmented AC areas were processed by the AI software and particles within the borders of AC were automatically counted by the software. Statistical analysis found significant (p < 0.001) correlation between clinical SUN grading and AI software detected particle count (Spearman ρ = 0.7077) and particle density (Spearman ρ = 0.7035). Significant (p < 0.001) correlation (Pearson's r = 0.9948) between manually and AI detected particles was found. No significant (p = 0.8080) difference was found between the sizes of the AI detected particles for all studies.

Conclusions

AI-based image analysis of AS-OCT slides show significant and independent correlation with clinical SUN assessment.

Translational Relevance

Automated AI-based AS-OCT image analysis suggests a noninvasive and quantitative assessment of AC inflammation with clear potential application in early detection and management of anterior uveitis.

Detection of oedema on optical coherence tomography images using deep learning model trained on noisy clinical data

PURPOSE

To meet the demands imposed by the continuing growth of the Age-related macular degeneration (AMD) patient population, automation of follow-ups by detecting retinal oedema using deep learning might be a viable approach. However, preparing and labelling data for training is time consuming. In this study, we investigate the feasibility of training a convolutional neural network (CNN) to accurately detect retinal oedema on optical coherence tomography (OCT) images of AMD patients with labels derived directly from clinical treatment decisions, without extensive preprocessing or relabelling.

METHODS

A total of 50 439 OCT images with associated treatment information were retrieved from databases at the Department of Ophthalmology, Rigshospitalet, Copenhagen, Denmark between 01.06.2007 and 01.06.2018. A CNN was trained on the retrieved data with the recorded treatment decisions as labels and validated on a subset of the data relabelled by three ophthalmologists to denote presence of oedema.

RESULTS

Moderate inter-grader agreement on presence of oedema in the relabelled data was found (76.4%). Despite different training and validation labels, the CNN performed on par with inter-grader agreement in detecting oedema on OCT images (AUC 0.97, accuracy 90.9%) and previously published models based on relabelled datasets.

CONCLUSION

The level of performance shown by the current model might make it valuable in detecting disease activity in automated AMD patient follow-up systems. Our approach demonstrates that high accuracy is not necessarily constrained by incongruent training and validation labels. These results might encourage the use of existing clinical databases for development of deep learning based algorithms without labour-intensive preprocessing in the future.

Iron overload and iron chelating agent exposure in anemia-associated outer retinal degeneration: a case report and review of the literature

Background

Deferoxamine retinopathy is the informally designated term used to describe a characteristic pattern of outer retinal degeneration in iron-overloaded chronic anemia patients who are treated with deferoxamine. We hypothesize that insufficiently treated iron overloading and not only deferoxamine is the cause of the retinal degeneration. Our case report is based on exposure histories of two anemia patients and literature review.

Case presentation

Both anemia patients presented with bilateral visual loss secondary to photoreceptor and retinal pigment epithelium degeneration. Chart review showed that visual loss came after a year-long slow, and rather monotonous rise in plasma ferritin concentrations, with no obvious relation to iron chelator exposure. In one patient, the onset of symptomatic visual loss came after a bout of fever followed by two additional febrile episodes, all accompanied by plasma ferritin spikes. Adjustment of iron chelation therapy did not improve visual function. Experimental studies clearly show that both systemic and intraocular exposure to iron ions can induce retinal degeneration.

Conclusion

The available evidence indicates that retinal degeneration in chronic anemia patients treated by deferoxamine is cause by insufficient iron chelation, not by deferoxamine. The actual role of iron chelating agents may be to promote a long enough survival to allow the slow development of retinal siderosis.

A Missense Mutation in RAB28 in a Family with Cone-Rod Dystrophy and Postaxial Polydactyly Prevents Localization of RAB28 to the Primary Cilium

Purpose

Cone-rod dystrophy (CRD) is a rare hereditary eye disorder that causes progressive degeneration of cone and rod photoreceptors. More than 30 genes, including RAB28, have been associated with CRD; however, only a few RAB28 variants have been reported to be associated with CRD. In this study, we describe two brothers with CRD and a homozygous missense variant, c.55G>A (p.Gly19Arg), in RAB28.

Methods

The missense variant was identified as part of a study investigating underlying genetic defects in a large patient cohort (n = 667) using targeted next-generation sequencing of 125 genes associated with retinal dystrophy. Cellular localization of RAB28 and ciliogenesis in patient fibroblasts were investigated by immunofluorescence microscopy. The effect of the missense variant on RAB28 expression level was investigated by quantitative real-time PCR.

Results

Two brothers of a consanguineous couple presented with CRD, postaxial polydactyly (PAP), and myopia. Both brothers had a homozygous missense RAB28 variant located in the G1 box of the guanosine triphosphate/guanosine diphosphate binding domain of RAB28. This missense variant caused a considerable reduction of RAB28 localized to the cilia, whereas ciliogenesis seemed unaffected.

Conclusions

The missense variant in RAB28 is classified as likely pathogenic with functional effect on protein localization. The combination of retinal dystrophy and PAP are well known from ciliopathies; however, more data are needed to finally conclude that the RAB28 variant described here is the cause of PAP in these brothers.

Repeatability of swept-source optical coherence tomography retinal and choroidal thickness measurements in neovascular age-related macular degeneration

BACKGROUND

The aim was to determine the intrasession repeatability of swept-source optical coherence tomography (SS-OCT)-derived retinal and choroidal thickness measurements in eyes with neovascular age-related macular degeneration (nAMD).

METHODS

A prospective study consisting of patients with active nAMD enrolled in the Distance of Choroid Study at Moorfields Eye Hospital, London. Patients underwent three 12×9 mm macular raster scans using the deep range imaging (DRI) OCT-1 SS-OCT (Topcon) device in a single imaging session. Retinal and choroidal thicknesses were calculated for the ETDRS macular subfields. Repeatability was calculated according to methods described by Bland and Altman.

RESULTS

39 eyes of 39 patients with nAMD were included with a mean (±SD) age of 73.9 (±7.2) years. The mean (±SD) retinal thickness of the central macular subfield was 225.7 μm (±12.4 μm). The repeatability this subfield, expressed as a percentage of the mean central macular subfield thickness, was 23.2%. The percentage repeatability of the other macular subfields ranged from 13.2% to 28.7%. The intrasession coefficient of repeatability of choroidal thickness of the central macular subfield was 57.2 μm with a mean choroidal thickness (±SD) of 181 μm (±15.8 μm).

CONCLUSIONS

This study suggests that a change >23.2% of retinal thickness and 57.2 μm choroidal thickness in the central macular subfield is required to distinguish true clinical change from measurement variability when using the DRI OCT-1 device to manage patients with nAMD.

Spectral-domain Optical Coherence Tomography Retinal and Choroidal Thickness Metric Repeatability in Age-related Macular Degeneration

PURPOSE

To determine the intrasession repeatability of spectral-domain OCT (SDOCT)-derived macular retinal and choroidal metrics in patients with neovascular age-related macular degeneration (nAMD) in the Distance of Choroid Study (DOCS).

DESIGN

Validity and reliability analysis.

METHODS

Enrolled patients underwent repeated SDOCT imaging using the Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany). A single technician certified for clinical trials took 3 macular volume scans. Retinal thicknesses were calculated for each of the 9 Early Treatment Diabetic Retinopathy Study (ETDRS) macular subfields. Center point thickness and total macular volume were also included in the analysis. Manual subfoveal choroidal thickness measurements were made by a masked observer.

RESULTS

A total of 40 eyes of 40 patients were included in this analysis (mean [± standard deviation] age: 74.1 [± 7.2] years, 60% male). The coefficient of repeatability (CR) of the central macular subfield was 30.6 μm (95% confidence interval [CI] 29.8-1.4 μm). The CR for the other macular subfields ranged from 7.0 μm to 38.2 μm. The CR for the total macular volume was 0.212 mm(3) (95% CI 0.206-0.217 mm(3)) and the CR for the center point was 47.5 μm (95% CI 46.2-48.7 μm). Images were also reviewed for the presence of segmentation error in the central macular subfield, and after exclusion of these eyes the revised CR for this subfield was 13.7 μm (95% CI 13.3-14.1 μm). The intrasession CR of subfoveal choroidal thickness was 34.7 μm (95% CI 33.7-35.7 μm).

CONCLUSIONS

This study suggests that a change of greater than 31 μm in Spectralis SDOCT-derived retinal thickness measurement of the central macular subfield and 35 μm in subfoveal choroidal thickness is necessary to detect true clinical change associated with disease progression or improvement in nAMD with a revised figure of 14 μm for central macular retinal subfield thickness in the absence of segmentation error.

Ophthalmic imaging

INTRODUCTION OR BACKGROUND

The last two decades have seen a revolution in ophthalmic imaging. In this review we present an overview of the breadth of ophthalmic imaging modalities in use today and describe how the role of ophthalmic imaging has changed from documenting abnormalities visible on clinical examination to the detection of clinically silent abnormalities which can lead to an earlier and more precise diagnosis.

SOURCES OF DATA

This review is based on published literature in the fields of ophthalmic imaging and with focus on most commonly used imaging modalities.

AREAS OF AGREEMENT

New imaging techniques enable non-invasive evaluation of ocular structures at a resolution of a few micrometres. This has led to a re-evaluation of diagnostic criteria for ocular disease, which were previously defined by clinical findings without significant reference to imaging.

AREAS OF CONTROVERSY

Lack of formal training and clinical guidelines regarding use of new imaging techniques in diagnosing and monitoring various ocular conditions. Lack of large normative databases and interchangeability issues between different commercial machines can hinder the detection of disease progression.

GROWING POINTS

Imaging devices are being constantly refined with improved image capture and image analysis tools.

AREAS TIMELY FOR DEVELOPING RESEARCH

Clinical applications of new techniques and devices have yet to be determined using systematic scientific research methods.