ImportanceOne of the biggest challenges when using anti–vascular endothelial growth factor (VEGF) agents to treat retinopathy of prematurity (ROP) is the need to perform long-term follow-up examinations to identify eyes at risk of ROP reactivation requiring retreatment.ObjectiveTo evaluate whether an artificial intelligence (AI)–based vascular severity score (VSS) can be used to analyze ROP regression and reactivation after anti-VEGF treatment and potentially identify eyes at risk of ROP reactivation requiring retreatment.Design, Setting, and ParticipantsThis prognostic study was a secondary analysis of posterior pole fundus images collected during the multicenter, double-blind, investigator-initiated Comparing Alternative Ranibizumab Dosages for Safety and Efficacy in Retinopathy of Prematurity (CARE-ROP) randomized clinical trial, which compared 2 different doses of ranibizumab (0.12 mg vs 0.20 mg) for the treatment of ROP. The CARE-ROP trial screened and enrolled infants between September 5, 2014, and July 14, 2016. A total of 1046 wide-angle fundus images obtained from 19 infants at predefined study time points were analyzed. The analyses of VSS were performed between January 20, 2021, and November 18, 2022.InterventionsAn AI-based algorithm assigned a VSS between 1 (normal) and 9 (most severe) to fundus images.Main Outcomes and MeasuresAnalysis of VSS in infants with ROP over time and VSS comparisons between the 2 treatment groups (0.12 mg vs 0.20 mg of ranibizumab) and between infants who did and did not receive retreatment for ROP reactivation.ResultsAmong 19 infants with ROP in the CARE-ROP randomized clinical trial, the median (range) postmenstrual age at first treatment was 36.4 (34.7-39.7) weeks; 10 infants (52.6%) were male, and 18 (94.7%) were White. The mean (SD) VSS was 6.7 (1.9) at baseline and significantly decreased to 2.7 (1.9) at week 1 (P < .001) and 2.9 (1.3) at week 4 (P < .001). The mean (SD) VSS of infants with ROP reactivation requiring retreatment was 6.5 (1.9) at the time of retreatment, which was significantly higher than the VSS at week 4 (P < .001). No significant difference was found in VSS between the 2 treatment groups, but the change in VSS between baseline and week 1 was higher for infants who later required retreatment (mean [SD], 7.8 [1.3] at baseline vs 1.7 [0.7] at week 1) vs infants who did not (mean [SD], 6.4 [1.9] at baseline vs 3.0 [2.0] at week 1). In eyes requiring retreatment, higher baseline VSS was correlated with earlier time of retreatment (Pearson r = −0.9997; P < .001).Conclusions and RelevanceIn this study, VSS decreased after ranibizumab treatment, consistent with clinical disease regression. In cases of ROP reactivation requiring retreatment, VSS increased again to values comparable with baseline values. In addition, a greater change in VSS during the first week after initial treatment was found to be associated with a higher risk of later ROP reactivation, and high baseline VSS was correlated with earlier retreatment. These findings may have implications for monitoring ROP regression and reactivation after anti-VEGF treatment.

Purpose Postapproval reports of intraocular inflammation (IOI) and occlusive retinal vasculitis following intravitreal brolucizumab are accumulating. A role of anti‐drug antibodies (ADAs) to brolucizumab is under current scientific discussion. The purpose of the present study was to measure brolucizumab ADAs in a cross‐sectional ophthalmic patient population and to compare the occurrence of brolucizumab ADAs with that of ranibizumab ADAs. Methods One hundred and ninety‐two serum samples and 54 vitreous samples were collected from patients with a range of eye diseases including neovascular age‐related macular degeneration (AMD), diabetic retinopathy, retinal vein occlusion, cataract, glaucoma, dry eye disease, macular hole, epiretinal membranes and intraocular lens (IOL) dislocation. Serum and vitreous samples were analysed for immune globuline (Ig) G ADAs to brolucizumab and ranibizumab using indirect enzyme‐linked immunosorbent assay (ELISA). Optical Density (OD) was read at 450 nm (wavelength correction at 550 nm) for ADA level measurements. Results Presence of brolucizumab ADAs was observed in patients with and without prior brolucizumab exposure. Both the frequency of notable ADA signals (OD > 0.1) and the mean ADA signal in serum samples were higher for brolucizumab than for ranibizumab. Two patients who experienced severe IOI and occlusive retinal vasculitis following intravitreal brolucizumab had high brolucizumab ADA serum levels. In one of these two patients, high brolucizumab ADA levels were also found in vitreous. Another patient developed moderate IOI without retinal vasculitis in the presence of low brolucizumab ADA serum levels. Overall, notable brolucizumab ADA levels were less frequent in vitreous than in the corresponding serum samples but with a tendency for higher prevalence in vitreous from patients with diabetic retinopathy. Conclusion Brolucizumab ADAs occur with significant prevalence in a typical ophthalmic patient population and may represent a risk factor for IOI and occlusive retinal vasculitis following brolucizumab.

Zusammenfassung Hintergrund Seit 2018 ist mit IDx-DR ein Verfahren auf dem Markt, welches den Grad der diabetischen Retinopathie (DR) mittels künstlicher Intelligenz (KI) bestimmt. Methoden Wir haben IDx-DR in die Sprechstunde an einer diabetologischen Schwerpunktklinik integriert und berichten über die Übereinstimmung zwischen IDx-DR (IDx Technologies Inc., Coralville, IA, USA) und Funduskopie sowie IDx-DR und ophthalmologischer Bildbeurteilung sowie über den Einfluss unterschiedlicher Kamerasysteme. Ergebnisse Mit der Topcon-Kamera (n = 456; NW400, Topcon Medical Systems, Oakland, NJ, USA) konnte im Vergleich zur Zeiss-Kamera (n = 47; Zeiss VISUCAM 500, Carl Zeiss Meditec AG, Jena, Deutschland) häufiger eine ausreichende Bildqualität in Miosis erreicht werden. Insgesamt war bei etwa 60 % der Patienten eine IDx-DR-Analyse in Miosis möglich. Alle Patienten, bei denen keine IDx-DR-Analyse in Miosis möglich war, konnten in Mydriasis funduskopiert werden. Innerhalb der Gruppe der auswertbaren Befunde zeigte sich eine Übereinstimmung zwischen IDx-DR und augenärztlicher Funduoskopie in ca. 55 %, ein Überschätzen des Schweregrads durch IDx-DR in ca. 40 % und ein Unterschätzen in ca. 4 %. Die Sensitivität (Spezifität) für das Erkennen einer schweren, behandlungsbedürftigen Retinopathie lag bei 95,7 % (89,1 %) für Fälle mit auswertbaren Fundusaufnahmen und bei 65,2 % (66,7 %), wenn alle Fälle betrachtet werden (inklusive derjeniger ohne verwertbare Aufnahme in Miosis). Der Kappa-Koeffizient zeigt mit 0,334 (p < 0,001) eine ausreichende Übereinstimmung zwischen IDx-DR und ärztlicher Bildauswertung anhand des Fundusfotos unter Berücksichtigung aller Patienten mit auswertbarer IDx-DR-Analyse. Der Vergleich zwischen IDx-DR mit der ärztlichen Funduskopie ergibt unter denselben Voraussetzungen eine geringe Übereinstimmung mit einem Kappa-Wert von 0,168 (p < 0,001). Schlussfolgerung Die vorliegende Studie zeigt Möglichkeiten und Grenzen des KI-gestützten DR-Screenings auf. Eine wesentliche Einschränkung liegt in der Tatsache, dass bei ca. 40 % der Patienten keine ausreichenden Aufnahmen in Miosis gewonnen werden konnten. Wenn ausreichende Aufnahmen vorlagen, stimmten IDx-DR und augenärztliche Diagnose in über 50 % der Fälle überein. Ein Unterschätzen des Schweregrades durch IDx-DR kam selten vor. Für die Integration in augenärztlich unterstützten Sprechstunden erscheint uns das System grundsätzlich geeignet. Die hohe Rate an fehlenden Aufnahmen in Miosis stellt allerdings eine Limitation dar, die einen Einsatz ohne augenärztliche Kontrollmöglichkeit schwierig erscheinen lässt.

No abstract