Deep-Learning Assessed Muscular Hypodensity Independently Predicts Mortality in DLBCL Patients Younger Than 60 Years
Background. Muscle depletion (MD) assessed by computed tomography (CT) has been shown to be a predictive marker in solid tumors, but has not been assessed in non-Hodgkin’s lymphomas. Despite software improvements, MD measurement remains highly time-consuming and cannot be used in clinical practice....
Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-09-01
|
Series: | Cancers |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-6694/13/18/4503 |
id |
doaj-af796a44b5d04a559cae7ca0e524e15a |
---|---|
record_format |
Article |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Maxime Jullien Benoit Tessoulin Hervé Ghesquières Lucie Oberic Franck Morschhauser Hervé Tilly Vincent Ribrag Thierry Lamy Catherine Thieblemont Bruno Villemagne Rémy Gressin Kamal Bouabdallah Corinne Haioun Gandhi Damaj Luc-Matthieu Fornecker Jean-Marc Schiano De Colella Pierre Feugier Olivier Hermine Guillaume Cartron Christophe Bonnet Marc André Clément Bailly René-Olivier Casasnovas Steven Le Gouill |
spellingShingle |
Maxime Jullien Benoit Tessoulin Hervé Ghesquières Lucie Oberic Franck Morschhauser Hervé Tilly Vincent Ribrag Thierry Lamy Catherine Thieblemont Bruno Villemagne Rémy Gressin Kamal Bouabdallah Corinne Haioun Gandhi Damaj Luc-Matthieu Fornecker Jean-Marc Schiano De Colella Pierre Feugier Olivier Hermine Guillaume Cartron Christophe Bonnet Marc André Clément Bailly René-Olivier Casasnovas Steven Le Gouill Deep-Learning Assessed Muscular Hypodensity Independently Predicts Mortality in DLBCL Patients Younger Than 60 Years Cancers diffuse large B-cell lymphoma muscle depletion sarcopenia muscle hypodensity U-NET convolutional neural network |
author_facet |
Maxime Jullien Benoit Tessoulin Hervé Ghesquières Lucie Oberic Franck Morschhauser Hervé Tilly Vincent Ribrag Thierry Lamy Catherine Thieblemont Bruno Villemagne Rémy Gressin Kamal Bouabdallah Corinne Haioun Gandhi Damaj Luc-Matthieu Fornecker Jean-Marc Schiano De Colella Pierre Feugier Olivier Hermine Guillaume Cartron Christophe Bonnet Marc André Clément Bailly René-Olivier Casasnovas Steven Le Gouill |
author_sort |
Maxime Jullien |
title |
Deep-Learning Assessed Muscular Hypodensity Independently Predicts Mortality in DLBCL Patients Younger Than 60 Years |
title_short |
Deep-Learning Assessed Muscular Hypodensity Independently Predicts Mortality in DLBCL Patients Younger Than 60 Years |
title_full |
Deep-Learning Assessed Muscular Hypodensity Independently Predicts Mortality in DLBCL Patients Younger Than 60 Years |
title_fullStr |
Deep-Learning Assessed Muscular Hypodensity Independently Predicts Mortality in DLBCL Patients Younger Than 60 Years |
title_full_unstemmed |
Deep-Learning Assessed Muscular Hypodensity Independently Predicts Mortality in DLBCL Patients Younger Than 60 Years |
title_sort |
deep-learning assessed muscular hypodensity independently predicts mortality in dlbcl patients younger than 60 years |
publisher |
MDPI AG |
series |
Cancers |
issn |
2072-6694 |
publishDate |
2021-09-01 |
description |
Background. Muscle depletion (MD) assessed by computed tomography (CT) has been shown to be a predictive marker in solid tumors, but has not been assessed in non-Hodgkin’s lymphomas. Despite software improvements, MD measurement remains highly time-consuming and cannot be used in clinical practice. Methods. This study reports the development of a Deep-Learning automatic segmentation algorithm (DLASA) to measure MD, and investigate its predictive value in a cohort of 656 diffuse large B cell lymphoma (DLBCL) patients included in the GAINED phase III prospective trial (NCT01659099). Results. After training on a series of 190 patients, the DLASA achieved a Dice coefficient of 0.97 ± 0.03. In the cohort, the median skeletal muscle index was 50.2 cm<sup>2</sup>/m<sup>2</sup> and median muscle attenuation (MA) was 36.1 Hounsfield units (HU). No impact of sarcopenia was found on either progression free survival (PFS) or overall survival (OS). Muscular hypodensity, defined as MA below the tenth percentile according to sex, was associated with a lower OS and PFS, respectively (HR = 2.80 (95% CI 1.58–4.95), <i>p</i> < 0.001, and HR = 2.22 (95% CI 1.43–3.45), <i>p</i> < 0.001). Muscular hypodensity appears to be an independent risk factor for mortality in DLBCL and because of DLASA can be estimated in routine practice. |
topic |
diffuse large B-cell lymphoma muscle depletion sarcopenia muscle hypodensity U-NET convolutional neural network |
url |
https://www.mdpi.com/2072-6694/13/18/4503 |
work_keys_str_mv |
AT maximejullien deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT benoittessoulin deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT herveghesquieres deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT lucieoberic deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT franckmorschhauser deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT hervetilly deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT vincentribrag deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT thierrylamy deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT catherinethieblemont deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT brunovillemagne deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT remygressin deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT kamalbouabdallah deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT corinnehaioun deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT gandhidamaj deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT lucmatthieufornecker deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT jeanmarcschianodecolella deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT pierrefeugier deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT olivierhermine deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT guillaumecartron deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT christophebonnet deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT marcandre deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT clementbailly deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT reneoliviercasasnovas deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years AT stevenlegouill deeplearningassessedmuscularhypodensityindependentlypredictsmortalityindlbclpatientsyoungerthan60years |
_version_ |
1717367872453345280 |
spelling |
doaj-af796a44b5d04a559cae7ca0e524e15a2021-09-25T23:49:06ZengMDPI AGCancers2072-66942021-09-01134503450310.3390/cancers13184503Deep-Learning Assessed Muscular Hypodensity Independently Predicts Mortality in DLBCL Patients Younger Than 60 YearsMaxime Jullien0Benoit Tessoulin1Hervé Ghesquières2Lucie Oberic3Franck Morschhauser4Hervé Tilly5Vincent Ribrag6Thierry Lamy7Catherine Thieblemont8Bruno Villemagne9Rémy Gressin10Kamal Bouabdallah11Corinne Haioun12Gandhi Damaj13Luc-Matthieu Fornecker14Jean-Marc Schiano De Colella15Pierre Feugier16Olivier Hermine17Guillaume Cartron18Christophe Bonnet19Marc André20Clément Bailly21René-Olivier Casasnovas22Steven Le Gouill23Department of Hematology, Nantes University Hospital, INSERM CRCINA Nantes-Angers, NeXT Université de Nantes, 44000 Nantes, FranceDepartment of Hematology, Nantes University Hospital, INSERM CRCINA Nantes-Angers, NeXT Université de Nantes, 44000 Nantes, FranceDepartment of Hematology, Hospices Civils de Lyon, Centre Hospitalier Lyon-Sud, Claude Bernard Lyon-1 University, 69310 Pierre Bénite, FranceDepartment of Hematology, IUC Toulouse Oncopole, 31000 Toulouse, FranceDepartment of Hematology, Univ. Lille, CHU Lille, EA 7365-GRITA-Groupe de Recherche sur les Formes Injectables et les Technologies Associées, 59000 Lille, FranceDepartment of Hematology, Centre H. Becquerel, 76000 Rouen, FranceDepartment of Hematology, Gustave Roussy, Université Paris-Saclay, 94800 Villejuif, FranceDepartment of Hematology, University Hospital of Rennes, 35000 Rennes, FranceDepartment of Hematology, APHP, Hopital Saint Louis, Université Paris Diderot, 75011 Paris, FranceDepartment of Hematology, Hopital Departemental de Vendée, 85000 La Roche sur Yon, FranceDepartment of Hematology, CHU Grenoble, 38000 Grenoble, FranceDepartment of Hematology, University Hospital of Bordeaux, F-33000 Bordeaux, FranceLymphoïd Malignancies Unit, Hôpital Henri Mondor, AP-HP, 94000 Créteil, FranceDepartment of Hematology, Institut D’hématologie de Basse Normandie, 14000 Caen, FranceDepartment of Hematology, Institut de Cancérologie Strasbourg Europe (ICANS), University Hospital of Strasbourg, 67000 Strasbourg, FranceDepartment of Hematology, Institut P. Calmette, 13000 Marseille, FranceDepartment of Hematology, University Hospital of Nancy, 54000 Nancy, FranceDepartment of Hematology, Hopital Necker, F-75015 Paris, FranceDepartment of Clinical Hematology, University Hospital of Montpellier, UMR-CNRS 5535, 34000 Montpellier, FranceDepartment of Hematology, CHU Liege, Liege University, 4000 Liege, BelgiumDepartment of Hematology, CHU UCL Namur, Université Catholique de Louvain, 5000 Namur, BelgiumDepartment of Nuclear Medicine, University Hospital of Nantes, 44000 Nantes, FranceDepartment of Hematology, University Hospital F. Mitterrand and Inserm UMR 1231, 21000 Dijon, FranceDepartment of Hematology, Nantes University Hospital, INSERM CRCINA Nantes-Angers, NeXT Université de Nantes, 44000 Nantes, FranceBackground. Muscle depletion (MD) assessed by computed tomography (CT) has been shown to be a predictive marker in solid tumors, but has not been assessed in non-Hodgkin’s lymphomas. Despite software improvements, MD measurement remains highly time-consuming and cannot be used in clinical practice. Methods. This study reports the development of a Deep-Learning automatic segmentation algorithm (DLASA) to measure MD, and investigate its predictive value in a cohort of 656 diffuse large B cell lymphoma (DLBCL) patients included in the GAINED phase III prospective trial (NCT01659099). Results. After training on a series of 190 patients, the DLASA achieved a Dice coefficient of 0.97 ± 0.03. In the cohort, the median skeletal muscle index was 50.2 cm<sup>2</sup>/m<sup>2</sup> and median muscle attenuation (MA) was 36.1 Hounsfield units (HU). No impact of sarcopenia was found on either progression free survival (PFS) or overall survival (OS). Muscular hypodensity, defined as MA below the tenth percentile according to sex, was associated with a lower OS and PFS, respectively (HR = 2.80 (95% CI 1.58–4.95), <i>p</i> < 0.001, and HR = 2.22 (95% CI 1.43–3.45), <i>p</i> < 0.001). Muscular hypodensity appears to be an independent risk factor for mortality in DLBCL and because of DLASA can be estimated in routine practice.https://www.mdpi.com/2072-6694/13/18/4503diffuse large B-cell lymphomamuscle depletionsarcopeniamuscle hypodensityU-NETconvolutional neural network |