Using Electronic Health Record Data to Determine the Safety of Aqueous Humor Liquid Biopsies for Molecular Analyses.

Purpose: Knowing the surgical safety of anterior chamber liquid biopsies will support the increased use of proteomics and other molecular analyses to better understand disease mechanisms and therapeutic responses in patients and clinical trials. Manual review of operative notes from different surgeons and procedures in electronic health records (EHRs) is cumbersome, but free-text software tools could facilitate efficient searches. Design: Retrospective case series. Participants: A total of 1418 aqueous humor liquid biopsies from patients undergoing intraocular surgery. Methods: Free-text EHR searches were performed using the Stanford Research Repository cohort discovery tool to identify complications associated with anterior chamber paracentesis and subsequent endophthalmitis. Complications of the surgery unrelated to the biopsy were not reviewed. Main Outcome Measures: Biopsy-associated intraoperative complications and endophthalmitis. Results: A total of 1418 aqueous humor liquid biopsies were performed by 17 experienced surgeons. EHR free-text searches were 100% error-free for surgical complications, >99% for endophthalmitis (<1% false positive), and >93.6% for anesthesia type, requiring manual review for only a limited number of cases. More than 85% of cases were performed under local anesthesia without ocular muscle akinesia. Although the most common indication was cataract (50.1%), other diagnoses included glaucoma, diabetic retinopathy, uveitis, age-related macular degeneration, endophthalmitis, retinitis pigmentosa, and uveal melanoma. A 50- to 100-µL sample was collected in all cases using either a 30-gauge needle or a blunt cannula via a paracentesis. The median follow-up was >7 months. There was only one minor complication (0.07%) identified: a case of a small tear in Descemet membrane without long-term sequelae. No other complications occurred, including other corneal injuries, lens or iris trauma, hyphema, or suprachoroidal hemorrhage. There was no case of postoperative endophthalmitis. Conclusions: Anterior chamber liquid biopsy during intraocular surgery is a safe procedure and may be considered for large-scale collection of aqueous humor samples for molecular analyses. Free-text EHR searches are an efficient approach to reviewing intraoperative procedures. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Sensitivity of Magnetic Resonance Imaging in Detection of Choroidal Metastases.

Introduction: The objective of this study was to determine the sensitivity of brain magnetic resonance imaging (MRI) in the detection of choroidal metastasis (CM) from systemic primary cancers. Methods: A retrospective chart review identified patients with clinically confirmed CM seen on the Oncology Service (Byers Eye Institute) between January 2018 and March 2022. Patients had an MRI brain and/or orbits performed within 3 months of CM diagnosis. Evaluation of CM detection by MRI was then divided into two parts: an initial "standard read," where determination of CM detection was based solely on the original radiology report, to reflect real-world performance, and a subsequent "dedicated read," for which a board-certified neuroradiologist, blinded to the laterality and location of the CM, reevaluated the studies to provide an objective "gold standard" interpretation regarding the radiographic detection of CM. Results: The study included 42 eyes of 40 patients with confirmed CM. On standard read, MRI detection of CM occurred in 21 of 42 eyes (50%), with no significant difference between MRI brain and orbit protocols (p = 0.249). Features associated with improved detection were increased tumor basal diameter (p < 0.001) and ultrasonographic tumor thickness (p = 0.003). On dedicated read, MRI detection of CM improved to 26 of 33 eyes (76%; limited to eyes with full complement of pre- and post-gadolinium sequences). Post-gadolinium 3D fluid-attenuated inversion recovery (FLAIR) sequence with fat suppression was the most sensitive (88%) for CM detection. 42% and 58% of lesions were visualized using conventional pre-gadolinium T1- and T2-weighted imaging, respectively. Conclusions: MRI sensitivity to detect CM improved from 50% to 76% with focused reinterpretation. Increased utilization of the post-gadolinium 3D FLAIR sequence and increased ocular scrutiny in cancer patients undergoing brain imaging may facilitate earlier diagnosis of CM.

Progression to Pars Plana Vitrectomy in Patients With Proliferative Diabetic Retinopathy.

Importance: The Diabetic Retinopathy Clinical Research Network Protocol S suggested that vitrectomy for vitreous hemorrhage (VH) or tractional retinal detachment (TRD) was more common among eyes assigned initially to panretinal photocoagulation (PRP) vs anti-vascular endothelial growth factor (anti-VEGF) for proliferative diabetic retinopathy (PDR). These clinical implications warrant further evaluation in the clinical practice setting. Objective: To explore outcomes of PDR treated with PRP monotherapy compared with matched patients treated with anti-VEGF monotherapy. Design, Setting, and Participants: Retrospective cohort study using an aggregated electronic health records research network. Patients with PDR who received PRP or anti-VEGF monotherapy between January and September 2023 were included before propensity score matching. Patients were excluded with 6 or fewer months' follow-up after monotherapy or with a combination of PRP and anti-VEGF. Data were analyzed in September 2023. Exposures: Patients with new PDR diagnoses stratified by monotherapy with PRP or anti-VEGF agents using Current Procedural Terminology code. Main Outcome Measures: Incidence of pars plana vitrectomy (PPV), VH, or TRD. Results: Among 6020 patients (PRP cohort: mean [SD] age, 64.8 [13.4]; 6424 [50.88%] female; 3562 [28.21%] Black, 6180 [48.95%] White, and 2716 [21.51%] unknown race; anti-VEGF cohort: mean [SD] age, 66.1 [13.2]; 5399 [50.52%] male; 2859 [26.75%] Black, 5377 [50.31%] White, and 2382 [22.29%] unknown race) who received treatment, PRP monotherapy was associated with higher rates of PPV when compared with patients treated with anti-VEGF monotherapy at 5 years (RR, 1.18; 95% CI, 1.05-1.36; RD, 1.37%; 95% CI, 0.39%-2.37%; P < .001), with similar associations at 1 and 3 years. PRP monotherapy was associated with higher rates of VH at 5 years (relative risk [RR], 1.72; 95% CI, 1.52-1.95; risk difference [RD], 7.05; 95% CI, 5.41%-8.69%; P < .001) and higher rates of TRD at 5 years (RR, 2.76; 95% CI, 2.26-3.37; RD, 4.25%; 95% CI, 3.45%-5.05%; P < .001), with similar magnitudes of associations at 6 months, 1 year, and 3 years, when compared with patients treated with anti-VEGF monotherapy. Conclusions and Relevance: These findings support the hypothesis that patients with PDR treated with PRP monotherapy are more likely to develop VH, TRD, and undergo PPV when compared with matched patients treated with anti-VEGF monotherapy. However, given the wide range in relative risk, confounding factors may account for some of the association between PRP vs anti-VEGF monotherapy and outcomes evaluated.

Impact of GLP-1 Agonists and SGLT-2 Inhibitors on Diabetic Retinopathy Progression: An Aggregated Electronic Health Record Data Study.

PURPOSE: To examine the effects of glucagon-like peptide-1 receptor (GLP-1) agonists compared to SGLT-2 inhibitors on diabetic retinopathy. DESIGN: Retrospective clinical cohort study using TriNetX, a federated electronic health records network comprising multiple healthcare organizations. METHODS: Patients with an International Classification of Diseases, Tenth Revision (ICD-10) code of nonproliferative diabetic retinopathy (PDR) and monotherapy treatment, excluding insulin, with GLP-1 agonists or SGLT-2 inhibitors. Patients with a history of PDR prior to initiation of treatment were excluded. The rate of progression to PDR and rate of development of diabetic macular edema (DME) were compared between patients on GLP-1 agonists compared to those on SGLT-2 inhibitors. The groups were propensity score matched for age, gender, ethnicity, race, type of diabetes, and severity of PDR. Main outcomes included rate and relative risk (RR) of progression to PDR and risk of DME in the GLP-1 agonist group versus the SGLT-2 inhibitor group. RESULTS: A total of 6481 patients were identified in the GLP-1 cohort and the SGLT-2 inhibitor cohort after propensity score matching. At 1 and 3 years after initiation of therapy, a higher rate of progression of PDR was noted (RR: 1.26, CI 1.04-1.51, P = .017 at 1 year, RR: 1.284, CI 1.1-1.499, P = .002 at 3 years) in the GLP-1 agonist cohort compared to the SGLT-2 inhibitor cohort. There was a higher rate of DME noted at 3 months (RR: 1.192, CI 1.059-1.276, P = .002), 6 months (RR: 1.22, CI 1.13-1.32, P < .001), 1 year (RR: 1.24, CI 1.15-1.33, P < .001), and at 3 years (RR: 1.29, CI 1.21-1.38, P < .001) in the GLP-1 agonist cohort compared to the SGLT-2 inhibitor cohort. CONCLUSIONS: A higher rate of progression of PDR and risk of new-onset DME was observed in patients on monotherapy with GLP-1 agonists compared to those on SGLT-2 inhibitors. It is important for clinicians to be aware of these potential effects and to consider the current retinopathy status when initiating treatment with newer hypoglycemic agents to ensure these patients are appropriately monitored for developing potential vision-threatening complications.

Association of Ocular Manifestations of Marfan Syndrome With Cardiovascular Complications.

PURPOSE: To evaluate associations between ocular manifestations of Marfan syndrome and cardiovascular complications. DESIGN: Retrospective cohort study. METHODS: The TriNetX Analytics platform, a federated health research network of aggregated deidentified electronic health record data of more than 119 million patients, was used to identify patients diagnosed with Marfan syndrome. Univariate logistic regression models were used to evaluate the association of ocular manifestations of Marfan syndrome (such as retinal tears/detachment, lens dislocation, and myopia), with cardiovascular comorbidities. Additional sensitivity analyses were performed using propensity matching. Odds ratios and 95% CIs were calculated for incidence of cardiovascular comorbidities (including aortic dissection, valvular disease, and arrhythmias) following diagnosis of Marfan syndrome. RESULTS: A total of 19,105 patients were identified who were diagnosed with Marfan disease without ocular manifestations, and an additional 3887 Marfan patients with ocular comorbidities. Patients who were diagnosed with ocular disease included 883 with ectopic lens, 417 with retinal tear or detachment, 683 with aphakia, 534 with pseudophakia, and 2465 with myopia. Patients with any ocular manifestations of Marfan were significantly more likely to be diagnosed with all cardiovascular comorbidities modeled including aortic aneurysm and dissection (OR 2.035; P < .0001), mitral valve prolapse (OR 2.725; P < .0001), tricuspid valve disorders (OR 2.142; P < .0001), cardiac arrhythmias (OR 1.836; P < .0001), and all cardiovascular outcomes combined (OR 2.194; P < .0001). CONCLUSIONS: In a large and diverse cohort of patients with Marfan syndrome, ocular manifestations of the disorder appear strongly associated with cardiovascular comorbidities.

Association of Cutaneous Keloids, Hypertrophic Scarring, and Fibrosis with Risk of Postoperative Proliferative Vitreoretinopathy.

PURPOSE: To assess an association between cutaneous keloids, hypertrophic scarring, and fibrosis (KHF) and risk of postoperative proliferative vitreoretinopathy (PVR) after rhegmatogenous retinal detachment (RRD) repair. DESIGN: Retrospective, population-based cohort study. PARTICIPANTS: Patients aged ≥ 18 years who underwent initial retinal detachment (RD) repair with pars plana vitrectomy with or without scleral buckle (SB) (Current Procedural Terminology [CPT] 67108), pneumatic retinopexy (67110), and primary SB (67107) from January 1, 2003, to March 1, 2023. METHODS: A de-identified electronic health record database through TriNetX, a global health research network, was used to analyze patients. Patients were queried for International Classification of Diseases, 10th Revision (ICD-10) codes L91.0 (hypertrophic scar) and L90.5 (scar conditions and fibrosis of skin). Frequency of subsequent diagnosis of PVR (H35.2) and CPT codes for secondary surgery including complex RD repair (67113) were determined. Patients with proliferative diabetic retinopathy (PDR) (ICD-10 H10.35/H11.35) were excluded. Descriptive statistics (Z-test) and propensity score matching (PSM) were used to match for age, sex, and race. MAIN OUTCOME MEASURES: Prevalence of H35.2 and CPT 67113 within 180 days after RRD repair in the KHF cohort versus the non-KHF cohort. RESULTS: Among patients with CPT 67108, 1061 in each cohort (KHF and non-KHF) were analyzed after PSM. The mean (standard deviation) age was 60.7 (15.2) years. Within 180 days, 10.1% of patients in the KHF cohort and 3.4% in the non-KHF cohort had a diagnosis of PVR (H35.2) (P < 0.001, odds ratio [OR], 3.2; 95% confidence interval [CI], 2.13-4.71). A total of 8.3% of patients in the KHF cohort and 5.4% of patients in the non-KHF cohort underwent complex RD repair (CPT 67113) (P = 0.008; OR, 3.2; 95% CI, 1.13-2.25). When including all RD repair types (CPT 67108, 67110, 67107), the rate of PVR diagnosis was still significantly greater in the KHF cohort than in the non-KHF cohort (9.0% vs 4.2%, P < 0.01; OR, 2.28; 95% CI, 1.64-3.16). CONCLUSIONS: A dermatologic history of KHF may be a risk factor for PVR after RD repair. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.

Liquid Biopsy Proteomics in Ophthalmology.

Minimally invasive liquid biopsies from the eye capture locally enriched fluids that contain thousands of proteins from highly specialized ocular cell types, presenting a promising alternative to solid tissue biopsies. The advantages of liquid biopsies include sampling the eye without causing irreversible functional damage, potentially better reflecting tissue heterogeneity, collecting samples in an outpatient setting, monitoring therapeutic response with sequential sampling, and even allowing examination of disease mechanisms at the cell level in living humans, an approach that we refer to as TEMPO (Tracing Expression of Multiple Protein Origins). Liquid biopsy proteomics has the potential to transform molecular diagnostics and prognostics and to assess disease mechanisms and personalized therapeutic strategies in individual patients. This review addresses opportunities, challenges, and future directions of high-resolution liquid biopsy proteomics in ophthalmology, with particular emphasis on the large-scale collection of high-quality samples, cutting edge proteomics technology, and artificial intelligence-supported data analysis.

Risk of Stroke, Myocardial Infarction, Deep Vein Thrombosis, Pulmonary Embolism, and Death After Retinal Vein Occlusion.

PURPOSE: To examine rates of stroke, myocardial infarction (MI), deep vein thrombosis (DVT), pulmonary embolism (PE), and death in patients after retinal vein occlusion (RVO) compared to controls. DESIGN: Retrospective cohort study. METHODS: An aggregated electronic health records research network, TriNetX, was used to identify patients with diagnosis of RVO and a control group of patients with cataract. Patients were excluded if they had history of stroke, MI, DVT, or PE within 2 years of diagnosis of RVO or cataract. Propensity score matching was performed to control for baseline demographics and medical comorbidities. Main outcomes included relative risk (RR) of death, stroke, MI, DVT, and PE after RVO compared to those in matched controls. RESULTS: A total of 45,304 patients were included in each cohort. There was elevated risk of death in the RVO cohort compared to the control cohort at 1 year (RR = 1.30, P < .01), 5 years (RR = 1.22, P < .01), and 10 years (RR = 1.08, P < .01). There was elevated risk of stroke at 1 year (RR = 1.61, P < .01), 5 years (RR = 1.31, P < .01), and 10 years (RR = 1.18, P < .01). There was elevated risk of MI at 1 year (RR = 1.26, P < .01) and 5 years (RR = 1.13, P < .01), but not at 10 years (RR = 1.06, P = .12). There was mildly elevated risk of DVT at 1 year (RR = 1.65, P < .01) but not at 5 years (RR = 0.94, P = .94) or 10 years (RR = 1.05, P = .37). There was no elevated risk of PE at 1 year (RR = 0.98, P = 0.80), 5 years (RR = 0.95, P = .42), or 10 years (RR = 0.85, P =.40). CONCLUSIONS: There is an increased rate of death, stroke, and MI after RVO compared to those in matched controls. We emphasize the need for long-term systemic evaluation after RVO.

Cross-Platform Identification and Validation of Uveal Melanoma Vitreous Protein Biomarkers.

Purpose: The purpose of this study was to profile protein expression liquid vitreous biopsies from patients with uveal melanoma (UM) using mass spectrometry to identify prognostic biomarkers, signaling pathways, and therapeutic targets. Methods: Vitreous biopsies were collected from two cohorts in a pilot study: comparative control eyes with epiretinal membranes (ERM; n = 3) and test eyes with UM (n = 8). Samples were analyzed using liquid chromatography-tandem mass spectrometry (LC-MS/MS). Identified proteins were compared to data from a targeted multiplex ELISA proteomics platform. Results: A total of 69 significantly elevated proteins were detected in the UM vitreous, including LYVE-1. LC-MS/MS identified 62 significantly upregulated proteins in UM vitreous that were not previously identified by ELISA. Analysis of differential protein expression by tumor molecular classification (gene expression profiling [GEP] and preferentially expressed antigen in melanoma [PRAME]) further identified proteins that correlated with these classifications. Patients with high-risk GEP tumors displayed elevated vitreous expression of HGFR (fold-change [FC] = 2.66E + 03, P value = 0.003) and PYGL (FC = 1.02E + 04, P = 1.72E-08). Patients with PRAME positive tumors displayed elevated vitreous expression of ENPP-2 (FC = 3.21, P = 0.04), NEO1 (FC = 2.65E + 03, P = 0.002), and LRP1 (FC = 5.59E + 02, P value = 0.01). IGF regulatory effectors were highly represented (P value = 1.74E-16). Cross-platform analysis validated seven proteins identified by ELISA and LC-MS/MS. Conclusions: Proteomic analysis of liquid biopsies may provide prognostic information supporting gene expression of tumor biopsies. The use of multiple protein detection platforms in the same patient samples increases the sensitivity of candidate biomarker detection and allows for precise characterization of the vitreous proteome.

Artificial Intelligence Improves Patient Follow-Up in a Diabetic Retinopathy Screening Program.

Purpose: We examine the rate of and reasons for follow-up in an Artificial Intelligence (AI)-based workflow for diabetic retinopathy (DR) screening relative to two human-based workflows. Patients and Methods: A DR screening program initiated September 2019 between one institution and its affiliated primary care and endocrinology clinics screened 2243 adult patients with type 1 or 2 diabetes without a diagnosis of DR in the previous year in the San Francisco Bay Area. For patients who screened positive for more-than-mild-DR (MTMDR), rates of follow-up were calculated under a store-and-forward human-based DR workflow ("Human Workflow"), an AI-based workflow involving IDx-DR ("AI Workflow"), and a two-step hybrid workflow ("AI-Human Hybrid Workflow"). The AI Workflow provided results within 48 hours, whereas the other workflows took up to 7 days. Patients were surveyed by phone about follow-up decisions. Results: Under the AI Workflow, 279 patients screened positive for MTMDR. Of these, 69.2% followed up with an ophthalmologist within 90 days. Altogether 70.5% (N=48) of patients who followed up chose their location based on primary care referral. Among the subset of patients that were seen in person at the university eye institute under the Human Workflow and AI-Human Hybrid Workflow, 12.0% (N=14/117) and 11.7% (N=12/103) of patients with a referrable screening result followed up compared to 35.5% of patients under the AI Workflow (N=99/279; χ2df=2 = 36.70, p < 0.00000001). Conclusion: Ophthalmology follow-up after a positive DR screening result is approximately three-fold higher under the AI Workflow than either the Human Workflow or AI-Human Hybrid Workflow. Improved follow-up behavior may be due to the decreased time to screening result.

Liquid-biopsy proteomics combined with AI identifies cellular drivers of eye aging and disease in vivo.

Single-cell analysis in living humans is essential for understanding disease mechanisms, but it is impractical in non-regenerative organs, such as the eye and brain, because tissue biopsies would cause serious damage. We resolve this problem by integrating proteomics of liquid biopsies with single-cell transcriptomics from all known ocular cell types to trace the cellular origin of 5,953 proteins detected in the aqueous humor. We identified hundreds of cell-specific protein markers, including for individual retinal cell types. Surprisingly, our results reveal that retinal degeneration occurs in Parkinson's disease, and the cells driving diabetic retinopathy switch with disease stage. Finally, we developed artificial intelligence (AI) models to assess individual cellular aging and found that many eye diseases not associated with chronological age undergo accelerated molecular aging of disease-specific cell types. Our approach, which can be applied to other organ systems, has the potential to transform molecular diagnostics and prognostics while uncovering new cellular disease and aging mechanisms.

Risk of Stroke, Myocardial Infarction, and Death After Retinal Artery Occlusion.

Importance: Patients with retinal artery occlusions (RAOs) are recommended to have emergent stroke workup, although the true risk of death and subsequent vascular events post-RAO is not clear. Objective: To determine short-term and long-term rates of stroke, myocardial infarction (MI), and death in patients after RAO compared with a control cohort. Design, Setting, and Participants: This retrospective cohort study used aggregated electronic health records from January 1, 2003, through April 14, 2023, from TriNetX, a network with data from more than 111 million patients. Patients with RAO and a cataract control group were identified and matched for age, sex, race, and comorbidities, including hypertension, diabetes, hyperlipidemia, and smoking status. Patients were excluded if they had a stroke or MI within 2 years before the diagnosis of RAO or cataract. Exposure: International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, diagnosis code for RAO or age-related cataract. Main Outcomes and Measures: Rate of death, stroke, and MI at 2 weeks, 30 days, 1 year, 5 years, and 10 years after RAO compared with matched controls. Results: There were a total of 34 874 patients with at least 1 year of follow-up in the RAO cohort. The mean (SD) age at the RAO event was 66 (15.2) years. The rate of death after RAO diagnosis was higher than after cataract diagnosis at 2 weeks (0.14% vs 0.06%; relative risk [RR], 2.45; 95% CI, 1.46-4.12; risk difference [RD], 0.08%; 95% CI, 0.04%-0.13%; P < .001), 30 days (0.29% vs 0.14%; RR, 2.10; 95% CI, 1.49-2.97; RD, 0.15%; 95% CI, 0.08%-0.22%; P < .001), 1 year (3.51% vs 1.99%; RR, 1.78; 95% CI, 1.61-1.94; RD, 1.41%; 95% CI, 1.17%-1.66%; P < .001), 5 years (22.74% vs 17.82%; RR, 1.28; 95% CI, 1.23-1.33; RD, 4.93%; 95% CI, 4.17%-5.68%; P < .001), and 10 years (57.86% vs 55.38%; RR, 1.05; 95% CI, 1.02-1.07; RD, 2.47%; 95% CI, 1.25%-3.69%; P < .001). Risk of stroke after RAO was higher at 2 weeks (1.72% vs 0.08%; RR, 21.43; 95% CI, 14.67-31.29; RD, 1.64%; 95% CI, 1.50%-1.78%; P < .001), 30 days (2.48% vs 0.18%; RR, 14.18; 95% CI, 10.94-18.48; RD, 2.31%; 95% CI, 2.14%-2.47%; P < .001), 1 year (5.89% vs 1.13%; RR, 5.20; 95% CI, 4.67-5.79; RD, 4.64%; 95% CI, 4.37%-4.91%; P < .001), 5 years (10.85% vs 4.86%; RR, 2.24; 95% CI, 2.09-2.40; RD, 6.00%; 95% CI, 5.50%-6.50%; P < .001), and 10 years (14.59% vs 9.18%; RR, 1.59; 95% CI, 1.48-1.70; RD, 5.41%; 95% CI, 4.62%-6.21%; P < .001). Risk of MI after RAO was higher at 2 weeks (0.16% vs 0.06%; RR, 3.00; 95% CI, 1.79-5.04; RD, 0.11%; 95% CI, 0.06%-0.16%; P < .001), 30 days (0.27% vs 0.10%; RR, 2.61; 95% CI, 1.78-3.83; RD, 0.17%; 95% CI, 0.10%-0.23%; P < .001), 1 year (1.66% vs 0.97%; RR, 1.72; 95% CI, 1.51-1.97; RD, 0.59%; 95% CI, 0.42%-0.76%; P < .001), 5 years (6.06% vs 5.00%; RR, 1.21; 95% CI, 1.12-1.31; RD, 1.07%; 95% CI, 0.64%-1.50%; P < .001), and 10 years (10.55% vs 9.43%; RR, 1.12; 95% CI, 1.04-1.21; RD, 1.13%; 95% CI, 0.39%-1.87%; P = .003). Conclusions and Relevance: This study showed an increased risk of death, stroke, and MI in patients with RAO at both short-term and long-term intervals after RAO compared with a matched control population diagnosed with cataract. These findings suggest a potential need for multidisciplinary evaluation and long-term systemic follow-up of patients post-RAO.

AI-Human Hybrid Workflow Enhances Teleophthalmology for the Detection of Diabetic Retinopathy.

Objective: Detection of diabetic retinopathy (DR) outside of specialized eye care settings is an important means of access to vision-preserving health maintenance. Remote interpretation of fundus photographs acquired in a primary care or other nonophthalmic setting in a store-and-forward manner is a predominant paradigm of teleophthalmology screening programs. Artificial intelligence (AI)-based image interpretation offers an alternative means of DR detection. IDx-DR (Digital Diagnostics Inc) is a Food and Drug Administration-authorized autonomous testing device for DR. We evaluated the diagnostic performance of IDx-DR compared with human-based teleophthalmology over 2 and a half years. Additionally, we evaluated an AI-human hybrid workflow that combines AI-system evaluation with human expert-based assessment for referable cases. Design: Prospective cohort study and retrospective analysis. Participants: Diabetic patients ≥ 18 years old without a prior DR diagnosis or DR examination in the past year presenting for routine DR screening in a primary care clinic. Methods: Macula-centered and optic nerve-centered fundus photographs were evaluated by an AI algorithm followed by consensus-based overreading by retina specialists at the Stanford Ophthalmic Reading Center. Detection of more-than-mild diabetic retinopathy (MTMDR) was compared with in-person examination by a retina specialist. Main Outcome Measures: Sensitivity, specificity, accuracy, positive predictive value, and gradability achieved by the AI algorithm and retina specialists. Results: The AI algorithm had higher sensitivity (95.5% sensitivity; 95% confidence interval [CI], 86.7%-100%) but lower specificity (60.3% specificity; 95% CI, 47.7%-72.9%) for detection of MTMDR compared with remote image interpretation by retina specialists (69.5% sensitivity; 95% CI, 50.7%-88.3%; 96.9% specificity; 95% CI, 93.5%-100%). Gradability of encounters was also lower for the AI algorithm (62.5%) compared with retina specialists (93.1%). A 2-step AI-human hybrid workflow in which the AI algorithm initially rendered an assessment followed by overread by a retina specialist of MTMDR-positive encounters resulted in a sensitivity of 95.5% (95% CI, 86.7%-100%) and a specificity of 98.2% (95% CI, 94.6%-100%). Similarly, a 2-step overread by retina specialists of AI-ungradable encounters improved gradability from 63.5% to 95.6% of encounters. Conclusions: Implementation of an AI-human hybrid teleophthalmology workflow may both decrease reliance on human specialist effort and improve diagnostic accuracy. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.

The Biosimilar Paradox: How Anti-Vascular Endothelial Growth Factor Biosimilars Could Increase Patient and Overall Health Care Costs.

PURPOSE: Anti-vascular endothelial growth factor (VEGF) medications for intraocular use are a major and increasing cost, and biosimilars may be a means of reducing the high cost of many biologic medications. However, a bevacizumab biosimilar, which is currently pending Food and Drug Administration (FDA) approval (bevacizumab-vikg), paradoxically may increase the cost burden of intravitreal anti-VEGF agents, because off-label repackaged drugs may no longer be allowed per the Drug Quality and Security Act (DQSA). We aimed to investigate the potential impact of biosimilars on costs in the United States. DESIGN: Cost analysis of anti-VEGF medications. PARTICIPANTS: Medicare data from October 2022 and previously published market share data from 2019. METHODS: Average sales prices (ASPs) of ranibizumab, aflibercept, and bevacizumab were calculated from Medicare allowable payments. The ASPs of biosimilars were calculated from wholesale acquisition costs from a representative distributor. The cost of an intraocular bevacizumab formulation is modeled at $500/1.25-mg dose and $900/1.25-mg dose. MAIN OUTCOME MEASURES: Costs of anti-VEGF drugs to Medicare Part B and patients. RESULTS: If an intraocular bevacizumab biosimilar were to be priced at $500, costs to Medicare would increase by $457 million from $3.01 billion to $3.47 billion (15.2% increase). Patient responsibility would increase by $117 million from $768 million to $884 million. Similarly, if intraocular bevacizumab were priced at $900, Medicare costs would increase by $897 million to $3.91 billion (29.8% increase), and patient responsibility would increase by $229 million to $997 million. If bevacizumab were $500/dose, switching all patients currently receiving ranibizumab or aflibercept to respective biosimilars would compensate for only 28.8% of the increased cost. Current prices of ranibizumab and aflibercept biosimilars would have to decrease by an aggregate of 15.7% to $616.80/injection, $1027.97/injection, and $1436.88/injection for ranibizumab 0.3 mg, ranibizumab 0.5 mg, and aflibercept, respectively. CONCLUSIONS: An FDA-approved bevacizumab biosimilar for ophthalmic use could increase costs to the health care system and patients, raising concerns for access. This increase would not be offset by ranibizumab and aflibercept biosimilar use at current prices. These data support the need for an exemption of section 503B of the DQSA and continued use of repackaged off-label bevacizumab. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Private equity in ophthalmology and optometry: a time series analysis from 2012 to 2021.

Purpose: To identify temporal and geographic trends in private equity (PE)-backed acquisitions of ophthalmology and optometry practices in the United States from 2012 to 2021. Methods: In this cross-sectional time series, acquisition data from 10/21/2019 to 9/1/2021 and previously published data from 1/1/2012 to 10/20/2019 were analyzed. Acquisition data were compiled from 6 financial databases, 5 industry news outlets, and publicly available press releases. Linear regression models were used to compare rates of acquisition. Outcomes included number of total acquisitions, practice type, locations, provider details, and geographic footprint. Results: A total of 245 practices associated with 614 clinical locations and 948 ophthalmologists or optometrists were acquired by 30 PE-backed platform companies between 10/21/2019 and 9/1/2021. Of 30 platform companies, 18 were new vis-à-vis our prior study. Of these acquisitions, 127 were comprehensive practices, 29 were retina practices, and 89 were optometry practices. From 2012 to 2021, monthly acquisitions increased by 0.947 acquisitions per year (P < 0.001*). Texas, Florida, Michigan, and New Jersey were the states with the greatest number of PE acquisitions, with 55, 48, 29, and 28 clinic acquisitions, respectively. Average monthly PE acquisitions were 5.71 per month from 1/1/2019 to 2/29/2020 (pre-COVID), 5.30 per month from 3/1/2020 to 12/31/2020 (COVID pre-vaccine [P = 0.81]), and 8.78 per month from 1/1/2021 to 9/1/2021 (COVID post-vaccine [P = 0.20]). Conclusions: PE acquisitions increased during the period 2012-2021 as companies continue to utilize regionally focused strategies for acquisitions.

Biosimilars for retinal diseases: United States-Europe awareness survey (Bio-USER - survey).

PURPOSE: To assess the awareness of biosimilar intravitreal anti-VEGF agents among retina specialists practicing in the United States (US) and Europe. METHODS: A 16-question online survey was created in English and distributed between Dec 01, 2021 and Jan 31, 2022. A total of 112 respondents (retinal physicians) from the US and Europe participated. RESULTS: The majority of the physicians (56.3%) were familiar with anti-VEGF biosimilars. A significant number of physicians needed more information (18.75%) and real world data (25%) before switching to a biosimilar. About one half of the physicians were concerned about biosimilar safety (50%), efficacy (58.9 %), immunogenicity (50%), and their efficacy with extrapolated indications (67.8 %). Retinal physicians from the US were less inclined to shift from off-label bevacizumab to biosimilar ranibizumab or on-label bevacizumab (if approved) compared to physicians from Europe (p=0.0001). Furthermore, physicians from the US were more concerned about biosimilar safety (p=0.0371) and efficacy compared to Europe (p= 0.0078). CONCLUSIONS: The Bio-USER survey revealed that while the majority of retinal physicians need additional information regarding the safety, efficacy and immunogenicity when making clinical decisions regarding their use. Retinal physicians from US are more comfortable in continuing to use off-label bevacizumab compared to physicians from Europe.