RESUMEN

Purpose: To evaluate associations between rare dysfunctional complement factor I (CFI) genetic variant status and progression to advanced age-related macular degeneration (AAMD), geographic atrophy (GA), and neovascular disease (NV). Design: Prospective, longitudinal study. Participants: Patients aged 55 to 80 years at baseline identifying as White with non-AAMD in 1 or both eyes at baseline were included. Follow-up grades were assigned as early, intermediate, or AAMD (GA or NV). CFI variants were categorized using genotyping and sequencing platforms. Methods: Analyses were performed using the Seddon Longitudinal Cohort Study (N = 2116 subjects, 3901 eyes, and mean follow-up of 8.3 years) and the Age-Related Eye Disease Study (N = 2837 subjects, 5200 eyes, and mean follow-up of 9.2 years). CFI rare variants associated with low serum factor I (FI) protein levels and decreased FI function (type 1), other AMD genetic variants, and demographic, behavioral, and ocular factors were evaluated. Generalized estimating equations methods were used to assess the association between CFI rare variants and progression, independent of other genetic variants and covariates. Main Outcome Measures: Progression to AAMD, GA, or NV. Results: In the prospective cohort of 4953 subjects (9101 eyes with non-AAMD at baseline), 1% were type 1 rare CFI carriers. Over 12 years, progression to AAMD was 44% for carriers and 20% for noncarriers (P < 0.001), 30% of carriers versus 10% of noncarriers progressed to GA (P < 0.001), and 18% of carriers compared with 11% of noncarriers progressed to NV (P = 0.049). CFI carriers were more likely to have a family history of AMD (P for trend = 0.035) and a higher baseline AMD grade (P < 0.001). After adjusting for all covariates, CFI carrier status was associated with progression to GA (odds ratio [OR] = 1.91; 95% confidence interval [CI] = 1.03, 3.52) but not NV (OR = 0.96). Higher body mass index was associated with progression among CFI carriers (body mass index ≥ 25 vs. < 25; OR = 5.8; 95% CI 1.5, 22.3) but not for noncarriers (OR = 1.1; 95% CI = 0.9, 1.3), with P_interaction = 0.011. Conclusions: Results suggest that carriers of rare dysfunctional type 1 CFI variants are at higher risk for progression to AAMD with GA. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.

RESUMEN

Purpose: A deep learning model was developed to detect nonexudative macular neovascularization (neMNV) using OCT B-scans. Design: Retrospective review of a prospective, observational study. Participants: Normal control eyes and patients with age-related macular degeneration (AMD) with and without neMNV. Methods: Swept-source OCT angiography (SS-OCTA) imaging (PLEX Elite 9000, Carl Zeiss Meditec, Inc) was performed using the 6 × 6-mm scan pattern. Individual B-scans were annotated to distinguish between drusen and the double-layer sign (DLS) associated with the neMNV. The machine learning model was tested on a dataset graded by humans, and model performance was compared with the human graders. Main Outcome Measures: Intersection over Union (IoU) score was measured to evaluate segmentation network performance. Area under the receiver operating characteristic curve values, sensitivity, specificity, and positive predictive value (PPV) and negative predictive value (NPV) were measured to assess the performance of the final classification performance. Chance-corrected agreement between the algorithm and the human grader determinations was measured with Cohen's kappa. Results: A total of 251 eyes from 210 patients, including 182 eyes with DLS and 115 eyes with drusen, were used for model training. Of 125 500 B-scans, 6879 B-scans were manually annotated. A vision transformer segmentation model was built to extract DLS and drusen from B-scans. The extracted prediction masks from all B-scans in a volume were projected to an en face image, and an eye-level projection map was obtained for each eye. A binary classification algorithm was established to identify eyes with neMNV from the projection map. The algorithm achieved 82%, 90%, 79%, and 91% sensitivity, specificity, PPV, and NPV, respectively, on a separate test set of 100 eyes that were evaluated by human graders in a previous study. The area under the curve value was calculated as 0.91 (95% confidence interval, 0.85-0.98). The results of the algorithm showed excellent agreement with the senior human grader (kappa = 0.83, P < 0.001) and moderate agreement with the junior grader consensus (kappa = 0.54, P < 0.001). Conclusions: Our network (code is available at https://github.com/uw-biomedical-ml/double_layer_vit) was able to detect the presence of neMNV from structural B-scans alone by applying a purely transformer-based model.

RESUMEN

Purpose: To investigate the natural course of pachychoroid pigment epitheliopathy (PPE). Design: A retrospective cohort study. Subjects: From the Kyoto central serous chorioretinopathy (CSC) cohort consisting of 548 patients with CSC as of September 2020, we included consecutive unilateral patients with acute or chronic CSC between January 2013 and December 2016. Methods: All patients underwent complete ophthalmic examination, including multimodal imaging such as fundus autofluorescence, spectral-domain optical coherence tomography, and fluorescein angiography/indocyanine green angiography and/or optimal coherence tomography angiography. The fellow eyes of eyes diagnosed with CSC were screened for PPE, and their natural course was evaluated. We also evaluated the association of ARMS2 rs10490924, CFH rs800292, TNFRSF10A rs13278062, and GATA5 rs6061548 genotypes with the natural course. Main Outcome Measures: Incidence of CSC, pachychoroid neovasculopathy, and pachychoroid geographic atrophy (GA). Results: In total, 165 patients with unilateral CSC (mean age, 55.7 ± 12.6 years; female, 22.4%) were included from the Kyoto CSC cohort. Among them, 148 (89.7%) were diagnosed as having PPE in their non-CSC eye. Survival analysis revealed that 16.8% of PPE eyes developed CSC during the 6-year follow up, whereas non-PPE eyes did not. Although genetic factors did not have significant association with CSC development (P > 0.05, log-rank test), choroidal vascular hyperpermeability (CVH) and subfoveal choroidal thickness (SFCT) were significantly associated with CSC incidence (P = 0.001, log-rank test). Survival analysis showed that eyes without CVH and eyes with SFCT < 300 µm did not develop CSC during the 6-year follow-up. Pachychoroid neovasculopathy developed in only 1 eye with PPE during a follow-up of 46.4 months. Pachychoroid GA did not develop in any of the studied eyes. Conclusions: This study revealed a natural history of PPE in a relatively large Japanese cohort. Choroidal vascular hyperpermeability and SFCT were significant risk factors for the development of CSC in PPE eyes. Although the current results cannot be generalized for all eyes with PPE, these findings present an important clinical implication.

RESUMEN

Age-related macular degeneration (AMD) is the leading cause of blindness for the elderly in high-income countries. Although multivitamin antioxidant nutrients can slow the progression of intermediate "dry" or nonneovascular AMD, no treatment can halt or reverse any stage of dry disease. Multiple biologic pathways have been implicated in AMD pathobiology, including the complement pathway. These pathways have been targeted by various approaches in clinical trials. To date, no treatment has reached their prespecified primary end point in 2 phase III trials, a requirement by the US Food and Drug Administration for a new drug approval. Here, we describe perspectives on the failures and possible successes of various clinical trials that will guide further investigation. These perspectives will also discuss clinical trial design issues to consider in future investigations, and how recent insights into AMD pathobiology might both provide additional explanation for trials not reaching the prespecified primary end points and offer direction for identifying prioritized treatment targets.

RESUMEN

Age-related macular degeneration (AMD) is a leading cause of blindness worldwide. Early to intermediate AMD is characterized by the accumulation of lipid- and protein-rich drusen. Late stages of the disease are characterized by the development of choroidal neovascularization, termed "exudative" or "neovascular AMD," or retinal pigment epithelium (RPE) cell and photoreceptor death, termed "geographic atrophy" (GA) in advanced nonexudative AMD. Although we have effective treatments for exudative AMD in the form of anti-VEGF agents, they have no role for patients with GA. Neuroprotection strategies have emerged as a possible way to slow photoreceptor degeneration and vision loss in patients with GA. These approaches include reduction of oxidative stress, modulation of the visual cycle, reduction of toxic molecules, inhibition of pathologic protein activity, prevention of cellular apoptosis or programmed necrosis (necroptosis), inhibition of inflammation, direct activation of neurotrophic factors, delivery of umbilical tissue-derived cells, and RPE replacement. Despite active investigation in this area and significant promise based on preclinical studies, many clinical studies have not yielded successful results. We discuss selected past and current neuroprotection trials for AMD, highlight the lessons learned from these past studies, and discuss our perspective regarding remaining questions that must be answered before neuroprotection can be successfully applied in the field of AMD research.

RESUMEN

Purpose: To investigate the detection of slower retinal capillary blood flow using commercial OCT angiography (OCTA) with a longer interscan time in diabetic retinopathy (DR). Design: Observational, prospective, cross-sectional study. Participants: A total of 62 eyes from 39 subjects with diabetes mellitus and 10 eyes from 9 healthy subjects. Methods: Commercial spectral domain-OCT was used to obtain 3 × 3-mm fovea-centered OCTA images of all eyes with 3 different interscan times (4.3, 5.7, and 8.6 ms). For each interscan time, OCTA imaging was performed 5 consecutive times, and a ×5 averaged image was obtained. Capillary flow density and visualization of retinal capillaries in the superficial and deep capillary plexuses (SCPs and DCPs, respectively) were compared between the 3 averaged images from the 3 different interscan times. Main Outcome Measures: Capillary flow density and visualization of foveal capillaries in 3 images with different interscan times. Results: Forty-five eyes of 34 patients were analyzed. There was no significant difference in the flow density of the SCP and DCP between the 3 images with different interscan times in all the DR stages. Some capillaries including microaneurysms that could not be observed at 4.3 ms could be observed at 5.7 or 8.6 ms. There were significantly more capillaries with difference points between the 3 images in the group with DR than in the group without DR (P < 0.01). The morphology of some microaneurysms also changed with longer interscan times. Conclusions: OCTA with longer interscan times revealed slower flow points in capillaries and more accurate visualization and morphology of microaneurysms in DR.

RESUMEN

Purpose: The impact of cataract surgery on low luminance visual acuity deficit (LLVAD) measurements was investigated by measuring the LLVAD before and after cataract surgery. Design: Prospective, longitudinal study. Participants: Patients undergoing cataract surgery. Methods: Photopic luminance (PL)-best-corrected visual acuity (BCVA) and low luminance (LL)-BCVA were obtained using the ETDRS chart. Low luminance visual acuity deficit scores were calculated by subtracting the LL-BCVA letter score from the PL-BCVA letter score. To demonstrate the reproducibility of these visual acuity measurements, we used data from drusen-only eyes previously published in the Complement Inhibition with Eculizumab for the Treatment of Nonexudative Age-Related Macular Degeneration (COMPLETE) study. The PL-BCVA, LL-BCVA, and LLVAD measurements obtained at an interval of 3 months in this cohort were used for comparison. In the current study, the impact of cataract surgery on LLVAD measurements was analyzed by comparing the PL-BCVA, LL-BCVA, and LLVAD measurements before and after cataract surgery. Main Outcome Measures: The reproducibility of the visual acuity measurements and the changes in LLVAD measurement after cataract surgery. Results: In the COMPLETE study, no clinically significant differences were found in the PL-BCVA, LL-BCVA, or LLVAD measurements between baseline and the 3-month follow-up visits with a change of -1.1 letters, -1.3 letters, and 0.1 letters, respectively (P = 0.02, P = 0.11, and P = 0.88, respectively). In the current study, significant increases were found in the PL-BCVA and LL-BCVA measurements, with a change of 7.3 letters and 10.2 letters after cataract surgery (P < 0.001 for both), and a statistically significant decrease in LLVAD measurements was found, with a change of -3.0 letters after cataract surgery (P = 0.002). Conclusions: Because of the variable effect of cataracts on LL-BCVA measurements and the significant change in LLVAD measurements after cataract surgery, investigators should be aware that cataract surgery during a trial will have an unpredictable impact on LLVAD measurements, and pseudophakic and phakic patients should be analyzed separately.

RESUMEN

Purpose: To compare the accuracies of the previously proposed square-root-transformed and perimeter-adjusted metrics for estimating length-type geographic atrophy (GA) growth rates. Design: Cross-sectional and simulation-based study. Participants: Thirty-eight eyes with GA from 27 patients. Methods: We used a previously developed atrophy-front growth model to provide analytical and numerical evaluations of the square-root-transformed and perimeter-adjusted growth rate metrics on simulated and semisimulated GA growth data. Main Outcome Measures: Comparison of the accuracies of the square-root-transformed and perimeter-adjusted metrics on simulated and semisimulated GA growth data. Results: Analytical and numerical evaluations showed that the accuracy of the perimeter-adjusted metric is affected minimally by baseline lesion area, focality, and circularity over a wide range of GA growth rates. Average absolute errors of the perimeter-adjusted metric were approximately 20 times lower than those of the square-root-transformed metrics, per evaluation on a semisimulated dataset with growth rate characteristics matching clinically observed data. Conclusions: Length-type growth rates have an intuitive, biophysical interpretation that is independent of lesion geometry, which supports their use in clinical trials of GA therapeutics. Taken in the context of prior studies, our analyses suggest that length-type GA growth rates should be measured using the perimeter-adjusted metric, rather than square-root-transformed metrics.

RESUMEN

Purpose: Assess the safety, tolerability, and feasibility of subcutaneous administration of the mitochondrial-targeted drug elamipretide in patients with dry age-related macular degeneration (AMD) and noncentral geographic atrophy (NCGA) and to perform exploratory analyses of change in visual function. Design: Phase 1, single-center, open-label, 24-week clinical trial with preplanned NCGA cohort. Participants: Adults ≥ 55 years of age with dry AMD and NCGA. Methods: Participants received subcutaneous elamipretide 40-mg daily; safety and tolerability assessed throughout. Ocular assessments included normal-luminance best-corrected visual acuity (BCVA), low-luminance BCVA (LLBCVA), normal-luminance binocular reading acuity (NLBRA), low-luminance binocular reading acuity (LLBRA), spectral-domain OCT, fundus autofluorescence (FAF), and patient self-reported function by low-luminance questionnaire (LLQ). Main Outcome Measures: Primary end point was safety and tolerability. Prespecified exploratory end-points included changes in BCVA, LLBCVA, NLBRA, LLBRA, geographic atrophy (GA) area, and LLQ. Results: Subcutaneous elamipretide was highly feasible. All participants (n = 19) experienced 1 or more nonocular adverse events (AEs), but all AEs were either mild (73.7%) or moderate (26.3%); no serious AEs were noted. Two participants exited the study because of AEs (conversion to neovascular AMD, n = 1; intolerable injection site reaction, n = 1), 1 participant discontinued because of self-perceived lack of efficacy, and 1 participant chose not to continue with study visits. Among participants completing the study (n = 15), mean ± standard deviation (SD) change in BCVA from baseline to week 24 was +4.6 (5.1) letters (P = 0.0032), while mean change (SD) in LLBCVA was +5.4 ± 7.9 letters (P = 0.0245). Although minimal change in NLBRA occurred, mean ± SD change in LLBCVA was -0.52 ± 0.75 logarithm of the minimum angle of resolution units (P = 0.005). Mean ± SD change in GA area (square root transformation) from baseline to week 24 was 0.14 ± 0.08 mm by FAF and 0.13 ± 0.14 mm by OCT. Improvement was observed in LLQ for dim light reading and general dim light vision. Conclusions: Elamipretide seems to be well tolerated without serious AEs in patients with dry AMD and NCGA. Exploratory analyses demonstrated possible positive effect on visual function, particularly under low luminance. A Phase 2b trial is underway to evaluate elamipretide further in dry AMD and NCGA.

RESUMEN

Purpose: To assess safety, tolerability, and feasibility of subcutaneous administration of the mitochondrial-targeted drug elamipretide in patients with intermediate age-related macular degeneration (AMD) and high-risk drusen (HRD) and to perform exploratory analyses of change in visual function. Design: Phase 1, single-center, open-label, 24-week clinical trial with preplanned HRD cohort. Participants: Adult patients ≥55 years of age with intermediate AMD and HRD. Methods: Participants received subcutaneous elamipretide 40 mg daily, with safety and tolerability assessed throughout the study. Ocular assessments included normal-luminance best-corrected visual acuity (BCVA), low-luminance best-corrected visual acuity (LLVA), normal-luminance binocular reading acuity (NLRA), low-luminance binocular reading acuity (LLRA), spectral-domain OCT, fundus autofluorescence (FAF), mesopic microperimetry, dark adaptation, and low-luminance questionnaire (LLQ). Main Outcome Measures: The primary end point was safety and tolerability. Prespecified exploratory end points included changes from baseline in BCVA, LLVA, NLRA, LLRA, retinal pigment epithelium (RPE)-drusen complex (DC) volume by OCT, FAF, mesopic microperimetry, dark adaptation, and LLQ results. Results: Subcutaneous administration of elamipretide was highly feasible. All participants with HRD (n = 21) experienced 1 or more adverse events (AEs), but all were mild (57%) or moderate (43%), with the most common events related to injection site reactions. No serious systemic AEs occurred. One participant discontinued because of injection site reaction, 1 participant withdrew because they did not wish to continue study visits, and 1 participant withdrew after experiencing transient visual impairment. Among the 18 participants who completed the study, mean change in BCVA from baseline to 24 weeks was +3.6 letters (P = 0.014) and LLVA was +5.6 letters (P = 0.004). Compared with baseline, mean NLRA improved by -0.11 logarithm of the minimum angle of resolution (logMAR) units (P = 0.001), and LLRA by -0.28 logMAR units (P < 0.0001). Significant improvements were found in 6 of 7 subscales of the LLQ (P <0.0015). No significant changes were observed for RPE-DC volume, FAF, mesopic microperimetry, or dark adaptation. Conclusions: Elamipretide appeared to be generally safe and well tolerated in treating intermediate AMD and HRD. Exploratory analyses demonstrate a positive effect on visual function, particularly under low-luminance conditions. Further study of elamipretide for treatment of intermediate AMD with HRD is warranted.

RESUMEN

Purpose: To establish the safety, tolerability, pharmacokinetics, and pharmacodynamics of an intravitreal injection of recombinant human complement factor H (CFH), GEM103, in individuals with genetically defined age-related macular degeneration (AMD) and geographic atrophy (GA). Design: Phase I single ascending-dose, open-label clinical trial (ClinicalTrials.gov identifier, NCT04246866). Participants: Twelve individuals 50 years of age or older with a confirmed diagnosis of foveal GA in the study eye. Methods: Participants were assigned to the increasing dose cohorts and received 1 50-µl intravitreal injection of GEM103 at doses of 50 µg/eye, 100 µg/eye, 250 µg/eye, or 500 µg/eye; dose escalation was dependent on the occurrence of dose-limiting toxicities. Main Outcome Measures: Safety assessments included ocular and systemic adverse events (AEs), ocular examinations, clinical laboratory and vital signs, and serum antidrug antibody levels. Biomarkers, measured in the aqueous humor (AH), included CFH and complement activation biomarkers factor Ba and complement component 3a. Results: No dose-limiting toxicities were reported, enabling escalation to the maximum study dose. No anti-GEM103 antidrug antibodies were detected during the study. Four participants experienced AEs; these were nonserious, mild or moderate in severity, and unrelated to GEM103. The AEs in 2 of these participants were related to the intravitreal injection procedure. No clinically significant ophthalmic changes and no ocular inflammation were observed. Visual acuity was maintained and stable throughout the 8-week follow-up period. No choroidal neovascularization occurred. CFH levels increased in a dose-dependent manner after GEM103 administration with supraphysiological levels observed at week 1; levels were more than baseline for 8 weeks or more in all participants receiving single doses of 100 µg or more. Complement activation biomarkers were reduced 7 days after dose administration. Conclusions: A single intravitreal administration of GEM103 (up to 500 µg/eye) was well tolerated in individuals with GA. Of the few mild or moderate AEs reported, none were determined to be related to GEM103. No intraocular inflammation or choroidal neovascularization developed. CFH levels in AH were increased and stable for 8 weeks, with pharmacodynamic data suggesting that GEM103 restored complement regulation. These results support further development in a repeat-dose trial in patients with GA with AMD.

RESUMEN

Purpose: To elucidate the prevalence of soft drusen, pseudodrusen, and pachydrusen and their 5-year changes in a Japanese population. Design: Longitudinal population-based cohort study conducted from 2013 through 2017. Participants: Residents 40 years of age or older. Methods: Nonmydriatic color fundus photographs were used to grade drusen subtypes and retinal pigment epithelium (RPE) abnormalities according to the Three Continent Age-Related Macular Degeneration Consortium. The 5-year changes of each drusen were investigated. Main Outcome Measures: The prevalence of each drusen subtype and the 5-year changes of each drusen. Results: Among 1731 participants, 1660 participants had gradable photographs that were assessed. The age-adjusted prevalence of soft drusen, pachydrusen, and pseudodrusen was 4.3% (95% confidence interval [CI], 3.2%-5.8%), 7.7% (95% CI, 6.2%-9.7%), and 2.8% (95% CI, 1.7%-4.2%), respectively. Pachydrusen accounted for 82.0% (n = 50) of the extramacular drusen (n = 61). Pigment abnormalities were seen in 28.3% and 8.3% of eyes with soft drusen and pachydrusen, respectively (P < 0.0001). Longitudinal changes were investigated in 1444 participants with follow-up examinations, which showed an increase in size in 8.3% and 3.7% and regression in 1.7% and 5.5% for eyes with soft drusen and pachydrusen, respectively. No participants demonstrated RPE atrophy after pachydrusen regression. Conclusions: The prevalence of pachydrusen was higher than that of soft drusen and pseudodrusen combined. Pachydrusen may regress over time and typically is not associated with RPE atrophy as detected using color fundus photographs.

RESUMEN

Purpose: To determine the contribution of common and rare genetic risk variants in families with age-related macular degeneration (AMD). Design: Case-control study. Participants: A family cohort (355 affected and 342 unaffected family members from 144 families with AMD) and an unrelated case-control cohort (1078 patients, 952 controls), recruited from the European Genetic Database. Methods: Genetic data of both cohorts were filtered for carriership of rare genetic variants in the coding and splice-site regions of the complement factor H (CFH) and complement factor I (CFI) genes, and 52 AMD-associated variants were extracted for calculation of genetic risk scores (GRS). To compare GRSs between familial and nonfamilial rare CFH and CFI variant carriers and noncarriers and between AMD disease stages, we performed a 2-way analysis of variance, with Bonferroni correction for multiple testing. Within families with AMD carrying rare CFH and CFI variants, we analyzed segregation patterns by calculating the proportion of affected among carriers. Main Outcome Measures: GRSs and segregation of rare CFH and CFI variants. Results: We observed higher GRSs in familial versus nonfamilial individuals without rare CFH and CFI variants: mean GRS, 1.76 (standard error [SE], 0.08) versus 0.83 (SE, 0.03; P < 0.001). In 51 of 144 families (35.4%), rare CFH and CFI variants were identified. Within the AMD family cohort, carriers of rare CFH and CFI variants showed lower GRSs compared with noncarriers (mean GRS, 1.05 [SE, 0.23] vs. 1.76 [SE, 0.08]; P = 0.02). The proportion of affected family members with a high GRS was 57.3% (176/307). Of the affected family members with a low or intermediate GRS, 40.0% carried rare CFH or CFI variants. Among carriers of 11 rare CFH or CFI variants, the proportion affected by AMD was more than 75%. Conclusions: Genetic risk in families with AMD often is attributed to high GRSs based on common variants. However, in part of the families with a low or intermediate GRS, rare CFH and CFI variants contributed to disease development. We recommend computing GRSs and sequencing the CFH and CFI genes in families with AMD, in particular in the light of ongoing gene-specific clinical trials.

RESUMEN

Purpose: Manually identifying geographic atrophy (GA) presence and location on OCT volume scans can be challenging and time consuming. This study developed a deep learning model simultaneously (1) to perform automated detection of GA presence or absence from OCT volume scans and (2) to provide interpretability by demonstrating which regions of which B-scans show GA. Design: Med-XAI-Net, an interpretable deep learning model was developed to detect GA presence or absence from OCT volume scans using only volume scan labels, as well as to interpret the most relevant B-scans and B-scan regions. Participants: One thousand two hundred eighty-four OCT volume scans (each containing 100 B-scans) from 311 participants, including 321 volumes with GA and 963 volumes without GA. Methods: Med-XAI-Net simulates the human diagnostic process by using a region-attention module to locate the most relevant region in each B-scan, followed by an image-attention module to select the most relevant B-scans for classifying GA presence or absence in each OCT volume scan. Med-XAI-Net was trained and tested (80% and 20% participants, respectively) using gold standard volume scan labels from human expert graders. Main Outcome Measures: Accuracy, area under the receiver operating characteristic (ROC) curve, F1 score, sensitivity, and specificity. Results: In the detection of GA presence or absence, Med-XAI-Net obtained superior performance (91.5%, 93.5%, 82.3%, 82.8%, and 94.6% on accuracy, area under the ROC curve, F1 score, sensitivity, and specificity, respectively) to that of 2 other state-of-the-art deep learning methods. The performance of ophthalmologists grading only the 5 B-scans selected by Med-XAI-Net as most relevant (95.7%, 95.4%, 91.2%, and 100%, respectively) was almost identical to that of ophthalmologists grading all volume scans (96.0%, 95.7%, 91.8%, and 100%, respectively). Even grading only 1 region in 1 B-scan, the ophthalmologists demonstrated moderately high performance (89.0%, 87.4%, 77.6%, and 100%, respectively). Conclusions: Despite using ground truth labels during training at the volume scan level only, Med-XAI-Net was effective in locating GA in B-scans and selecting relevant B-scans within each volume scan for GA diagnosis. These results illustrate the strengths of Med-XAI-Net in interpreting which regions and B-scans contribute to GA detection in the volume scan.

RESUMEN

Age-related macular degeneration (AMD) is a major cause of blindness in the western world. While genetic studies have linked both common and rare variants in genes involved in regulation of the complement system to increased risk of development of AMD, environmental factors, such as smoking and nutrition, can also significantly affect the risk of developing the disease and the rate of disease progression. Since epigenetics has been implicated in mediating, in part, the disease risk associated with some environmental factors, we investigated a possible epigenetic contribution to AMD. We performed genome-wide DNA methylation profiling of blood from AMD patients and controls. No differential methylation site reached genome-wide significance; however, when epigenetic changes in and around known GWAS-defined AMD risk loci were explored, we found small but significant DNA methylation differences in the blood of neovascular AMD patients near age-related maculopathy susceptibility 2 (ARMS2), a top-ranked GWAS locus preferentially associated with neovascular AMD. The methylation level of one of the CpG sites significantly correlated with the genotype of the risk SNP rs10490924, suggesting a possible epigenetic mechanism of risk. Integrating genome-wide DNA methylation analysis of retina samples with and without AMD together with blood samples, we further identified a consistent, replicable change in DNA methylation in the promoter region of protease serine 50 (PRSS50). These methylation changes may identify sites in novel genes that are susceptible to non-genetic factors known to contribute to AMD development and progression.

Asunto(s)

Metilación de ADN/genética , Predisposición Genética a la Enfermedad , Estudio de Asociación del Genoma Completo , Degeneración Macular/genética , Anciano , Anciano de 80 o más Años , Femenino , Genotipo , Humanos , Degeneración Macular/sangre , Degeneración Macular/patología , Masculino , Polimorfismo de Nucleótido Simple , Retina/metabolismo , Retina/patología , Factores de Riesgo

RESUMEN

In recent years, the management of macular disease has undergone radical changes, in part because of new therapeutic approaches, but also due to the introduction of a new imaging modality - optical coherence tomography (OCT). The application of OCT imaging has clarified many aspects of chorioretinal disease pathophysiology and elucidated many hitherto unrecognized disease characteristics. From an early stage in its development, OCT has also been revolutionary in attempting to extract clinically useful measurements from image data in an automated fashion. As a result, OCT-derived measurements of retinal thickness have been rapidly embraced in clinical and research settings. However, as knowledge of OCT image analysis has developed, it has become increasingly clear that even accurate measurements of retinal thickness may fail to predict visual outcomes for many diseases. As a result, the focus of much current clinical imaging research is on the identification of other OCT-derived anatomic biomarkers predictive of visual outcomes - such biomarkers could serve as surrogate endpoints in clinical trials and provide prognostic information in clinical practice. In this review, we begin by highlighting the importance of accurate visual function assessment and describing the fundamentals of OCT image evaluation, before describing the current state-of-the-art with regard to predicting visual outcomes, for a variety of macular diseases, using OCT.