Cargando…

Automated Retinal Layer Segmentation in CLN2-Associated Disease: Commercially Available Software Characterizing a Progressive Maculopathy

PURPOSE: CLN2-associated disease is a hereditary, fatal lysosomal storage disorder characterized by progressive brain and retinal deterioration. Here, we characterize the inner and outer retinal degeneration using automated segmentation software in optical coherence tomography scans, providing an ob...

Descripción completa

Detalles Bibliográficos
Autores principales: Kovacs, Kyle D., Orlin, Anton, Sondhi, Dolan, Kaminsky, Stephen M., D'Amico, Donald J., Crystal, Ronald G., Kiss, Szilárd
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Association for Research in Vision and Ophthalmology 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8322716/
https://www.ncbi.nlm.nih.gov/pubmed/34313725
http://dx.doi.org/10.1167/tvst.10.8.23
_version_ 1783731113065185280
author Kovacs, Kyle D.
Orlin, Anton
Sondhi, Dolan
Kaminsky, Stephen M.
D'Amico, Donald J.
Crystal, Ronald G.
Kiss, Szilárd
author_facet Kovacs, Kyle D.
Orlin, Anton
Sondhi, Dolan
Kaminsky, Stephen M.
D'Amico, Donald J.
Crystal, Ronald G.
Kiss, Szilárd
author_sort Kovacs, Kyle D.
collection PubMed
description PURPOSE: CLN2-associated disease is a hereditary, fatal lysosomal storage disorder characterized by progressive brain and retinal deterioration. Here, we characterize the inner and outer retinal degeneration using automated segmentation software in optical coherence tomography scans, providing an objective, quantifiable metric for monitoring subtle changes previously identified with a validated disease classification scale (the Weill Cornell Batten Scale). METHODS: This study is a retrospective, single-center cohort review of images from examinations under anesthesia in treatment-naïve patients with CLN2-associated disease. Automated segmentation software was used to delineate retinal nerve fiber, ganglion cell layer (GCL), and outer nuclear layer (ONL) thickness measurements in the fovea, parafovea, and perifovea based on age groups (months): 30 to 38, 39 to 45, 46 to 52, 53 to 59, 60 to 66, and 67 or older. RESULTS: Twenty-seven eyes from 14 patients were included, with 8 serial images yielding 36 interpretable optical coherence tomography scans. There was a significant difference in parafoveal ONL thickness between 39 to 45 and 46 to 52 months of age (P = 0.032) not seen in other regions or retinal layers. Perifoveal ONL demonstrated a difference in thickness between the 60 to 66 and greater than 67 months age cohorts (P = 0.047). There was strong symmetry between eyes, and high segmentation repeatability. CONCLUSIONS: Parafoveal ONL thickness represents a sensitive, early age indicator of CLN2-associated degeneration. Outer retinal degeneration is apparent at younger ages than inner retinal changes though in treatment-naïve patients all retinal layers showed significant differences between 60 to 66 and more than 67 months of age. TRANSLATIONAL RELEVANCE: This study establishes sensitive, quantitative biomarkers for assessing retinal degeneration in a large cohort natural history study in anticipation of future clinical trials.
format Online
Article
Text
id pubmed-8322716
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher The Association for Research in Vision and Ophthalmology
record_format MEDLINE/PubMed
spelling pubmed-83227162021-08-13 Automated Retinal Layer Segmentation in CLN2-Associated Disease: Commercially Available Software Characterizing a Progressive Maculopathy Kovacs, Kyle D. Orlin, Anton Sondhi, Dolan Kaminsky, Stephen M. D'Amico, Donald J. Crystal, Ronald G. Kiss, Szilárd Transl Vis Sci Technol Article PURPOSE: CLN2-associated disease is a hereditary, fatal lysosomal storage disorder characterized by progressive brain and retinal deterioration. Here, we characterize the inner and outer retinal degeneration using automated segmentation software in optical coherence tomography scans, providing an objective, quantifiable metric for monitoring subtle changes previously identified with a validated disease classification scale (the Weill Cornell Batten Scale). METHODS: This study is a retrospective, single-center cohort review of images from examinations under anesthesia in treatment-naïve patients with CLN2-associated disease. Automated segmentation software was used to delineate retinal nerve fiber, ganglion cell layer (GCL), and outer nuclear layer (ONL) thickness measurements in the fovea, parafovea, and perifovea based on age groups (months): 30 to 38, 39 to 45, 46 to 52, 53 to 59, 60 to 66, and 67 or older. RESULTS: Twenty-seven eyes from 14 patients were included, with 8 serial images yielding 36 interpretable optical coherence tomography scans. There was a significant difference in parafoveal ONL thickness between 39 to 45 and 46 to 52 months of age (P = 0.032) not seen in other regions or retinal layers. Perifoveal ONL demonstrated a difference in thickness between the 60 to 66 and greater than 67 months age cohorts (P = 0.047). There was strong symmetry between eyes, and high segmentation repeatability. CONCLUSIONS: Parafoveal ONL thickness represents a sensitive, early age indicator of CLN2-associated degeneration. Outer retinal degeneration is apparent at younger ages than inner retinal changes though in treatment-naïve patients all retinal layers showed significant differences between 60 to 66 and more than 67 months of age. TRANSLATIONAL RELEVANCE: This study establishes sensitive, quantitative biomarkers for assessing retinal degeneration in a large cohort natural history study in anticipation of future clinical trials. The Association for Research in Vision and Ophthalmology 2021-07-27 /pmc/articles/PMC8322716/ /pubmed/34313725 http://dx.doi.org/10.1167/tvst.10.8.23 Text en Copyright 2021 The Authors https://creativecommons.org/licenses/by/4.0/This work is licensed under a Creative Commons Attribution 4.0 International License.
spellingShingle Article
Kovacs, Kyle D.
Orlin, Anton
Sondhi, Dolan
Kaminsky, Stephen M.
D'Amico, Donald J.
Crystal, Ronald G.
Kiss, Szilárd
Automated Retinal Layer Segmentation in CLN2-Associated Disease: Commercially Available Software Characterizing a Progressive Maculopathy
title Automated Retinal Layer Segmentation in CLN2-Associated Disease: Commercially Available Software Characterizing a Progressive Maculopathy
title_full Automated Retinal Layer Segmentation in CLN2-Associated Disease: Commercially Available Software Characterizing a Progressive Maculopathy
title_fullStr Automated Retinal Layer Segmentation in CLN2-Associated Disease: Commercially Available Software Characterizing a Progressive Maculopathy
title_full_unstemmed Automated Retinal Layer Segmentation in CLN2-Associated Disease: Commercially Available Software Characterizing a Progressive Maculopathy
title_short Automated Retinal Layer Segmentation in CLN2-Associated Disease: Commercially Available Software Characterizing a Progressive Maculopathy
title_sort automated retinal layer segmentation in cln2-associated disease: commercially available software characterizing a progressive maculopathy
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8322716/
https://www.ncbi.nlm.nih.gov/pubmed/34313725
http://dx.doi.org/10.1167/tvst.10.8.23
work_keys_str_mv AT kovacskyled automatedretinallayersegmentationincln2associateddiseasecommerciallyavailablesoftwarecharacterizingaprogressivemaculopathy
AT orlinanton automatedretinallayersegmentationincln2associateddiseasecommerciallyavailablesoftwarecharacterizingaprogressivemaculopathy
AT sondhidolan automatedretinallayersegmentationincln2associateddiseasecommerciallyavailablesoftwarecharacterizingaprogressivemaculopathy
AT kaminskystephenm automatedretinallayersegmentationincln2associateddiseasecommerciallyavailablesoftwarecharacterizingaprogressivemaculopathy
AT damicodonaldj automatedretinallayersegmentationincln2associateddiseasecommerciallyavailablesoftwarecharacterizingaprogressivemaculopathy
AT crystalronaldg automatedretinallayersegmentationincln2associateddiseasecommerciallyavailablesoftwarecharacterizingaprogressivemaculopathy
AT kissszilard automatedretinallayersegmentationincln2associateddiseasecommerciallyavailablesoftwarecharacterizingaprogressivemaculopathy