Prognostic value of deep learning-derived body composition in advanced pancreatic cancer-a retrospective multicenter study.

body composition computed tomography deep learning pancreatic cancer prognosis

Journal

ESMO open
ISSN: 2059-7029
Titre abrégé: ESMO Open
Pays: England
ID NLM: 101690685

Informations de publication

Date de publication:
08 Jan 2024
Historique:
received: 03 09 2023
revised: 11 12 2023
accepted: 13 12 2023
medline: 10 1 2024
pubmed: 10 1 2024
entrez: 9 1 2024
Statut: aheadofprint

Résumé

Despite the prognostic relevance of cachexia in pancreatic cancer, individual body composition has not been routinely integrated into treatment planning. In this multicenter study, we investigated the prognostic value of sarcopenia and myosteatosis automatically extracted from routine computed tomography (CT) scans of patients with advanced pancreatic ductal adenocarcinoma (PDAC). We retrospectively analyzed clinical imaging data of 601 patients from three German cancer centers. We applied a deep learning approach to assess sarcopenia by the abdominal muscle-to-bone ratio (MBR) and myosteatosis by the ratio of abdominal inter- and intramuscular fat to muscle volume. In the pooled cohort, univariable and multivariable analyses were carried out to analyze the association between body composition markers and overall survival (OS). We analyzed the relationship between body composition markers and laboratory values during the first year of therapy in a subgroup using linear regression analysis adjusted for age, sex, and American Joint Committee on Cancer (AJCC) stage. Deep learning-derived MBR [hazard ratio (HR) 0.60, 95% confidence interval (CI) 0.47-0.77, P < 0.005] and myosteatosis (HR 3.73, 95% CI 1.66-8.39, P < 0.005) were significantly associated with OS in univariable analysis. In multivariable analysis, MBR (P = 0.019) and myosteatosis (P = 0.02) were associated with OS independent of age, sex, and AJCC stage. In a subgroup, MBR and myosteatosis were associated with albumin and C-reactive protein levels after initiation of therapy. Additionally, MBR was also associated with hemoglobin and total protein levels. Our work demonstrates that deep learning can be applied across cancer centers to automatically assess sarcopenia and myosteatosis from routine CT scans. We highlight the prognostic role of our proposed markers and show a strong relationship with protein levels, inflammation, and anemia. In clinical practice, automated body composition analysis holds the potential to further personalize cancer treatment.

Sections du résumé

BACKGROUND BACKGROUND
Despite the prognostic relevance of cachexia in pancreatic cancer, individual body composition has not been routinely integrated into treatment planning. In this multicenter study, we investigated the prognostic value of sarcopenia and myosteatosis automatically extracted from routine computed tomography (CT) scans of patients with advanced pancreatic ductal adenocarcinoma (PDAC).
PATIENTS AND METHODS METHODS
We retrospectively analyzed clinical imaging data of 601 patients from three German cancer centers. We applied a deep learning approach to assess sarcopenia by the abdominal muscle-to-bone ratio (MBR) and myosteatosis by the ratio of abdominal inter- and intramuscular fat to muscle volume. In the pooled cohort, univariable and multivariable analyses were carried out to analyze the association between body composition markers and overall survival (OS). We analyzed the relationship between body composition markers and laboratory values during the first year of therapy in a subgroup using linear regression analysis adjusted for age, sex, and American Joint Committee on Cancer (AJCC) stage.
RESULTS RESULTS
Deep learning-derived MBR [hazard ratio (HR) 0.60, 95% confidence interval (CI) 0.47-0.77, P < 0.005] and myosteatosis (HR 3.73, 95% CI 1.66-8.39, P < 0.005) were significantly associated with OS in univariable analysis. In multivariable analysis, MBR (P = 0.019) and myosteatosis (P = 0.02) were associated with OS independent of age, sex, and AJCC stage. In a subgroup, MBR and myosteatosis were associated with albumin and C-reactive protein levels after initiation of therapy. Additionally, MBR was also associated with hemoglobin and total protein levels.
CONCLUSIONS CONCLUSIONS
Our work demonstrates that deep learning can be applied across cancer centers to automatically assess sarcopenia and myosteatosis from routine CT scans. We highlight the prognostic role of our proposed markers and show a strong relationship with protein levels, inflammation, and anemia. In clinical practice, automated body composition analysis holds the potential to further personalize cancer treatment.

Identifiants

DOI: 10.1016/j.esmoop.2023.102219 PMID: 38194881
pubmed: 38194881
pii: S2059-7029(23)01460-6
doi: 10.1016/j.esmoop.2023.102219
pii:
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Pagination

102219

Informations de copyright

Copyright © 2023 The Author(s). Published by Elsevier Ltd.. All rights reserved.

Auteurs

J Keyl (J)

Institute for Artificial Intelligence in Medicine, University Hospital Essen (AöR), Essen, Germany; Institute of Pathology, University Hospital Essen (AöR), Essen, Germany. Electronic address: Julius.Keyl@uk-essen.de.

A Bucher (A)

Institute for Diagnostic and Interventional Radiology, Goethe University Frankfurt, Frankfurt am Main, Germany; German Cancer Consortium (DKTK), Frankfurt partner site, Heidelberg, Germany.

F Jungmann (F)

Institute of Diagnostic and Interventional Radiology, Technical University of Munich, School of Medicine, Munich, Germany; Artificial Intelligence in Healthcare and Medicine, School of Computation, Information and Technology, Technical University of Munich, Munich, Germany.

R Hosch (R)

Institute for Artificial Intelligence in Medicine, University Hospital Essen (AöR), Essen, Germany.

A Ziller (A)

Institute of Diagnostic and Interventional Radiology, Technical University of Munich, School of Medicine, Munich, Germany; Artificial Intelligence in Healthcare and Medicine, School of Computation, Information and Technology, Technical University of Munich, Munich, Germany.

R Armbruster (R)

Institute for Diagnostic and Interventional Radiology, Goethe University Frankfurt, Frankfurt am Main, Germany.

P Malkomes (P)

Department of General, Visceral and Transplant Surgery, Goethe University Hospital Frankfurt, Frankfurt am Main, Germany.

T M Reissig (TM)

Department of Medical Oncology, University Hospital Essen (AöR), Essen, Germany; West German Cancer Center, University Hospital Essen (AöR), Essen, Germany; German Cancer Consortium (DKTK), Partner site University Hospital Essen (AöR), Essen, Germany; Bridge Institute of Experimental Tumor Therapy, West German Cancer Center, University Hospital Essen, University of Duisburg-Essen, Essen, Germany; Division of Solid Tumor Translational Oncology, German Cancer Consortium (DKTK Partner Site Essen) and German Cancer Research Center, DKFZ, Heidelberg, Germany.

S Koitka (S)

Institute for Artificial Intelligence in Medicine, University Hospital Essen (AöR), Essen, Germany; Institute for Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen (AöR), Essen, Germany.

I Tzianopoulos (I)

Department of Medical Oncology, University Hospital Essen (AöR), Essen, Germany; West German Cancer Center, University Hospital Essen (AöR), Essen, Germany; Bridge Institute of Experimental Tumor Therapy, West German Cancer Center, University Hospital Essen, University of Duisburg-Essen, Essen, Germany; Division of Solid Tumor Translational Oncology, German Cancer Consortium (DKTK Partner Site Essen) and German Cancer Research Center, DKFZ, Heidelberg, Germany.

P Keyl (P)

Institute of Pathology, Ludwig-Maximilians-University Munich, Munich, Germany.

K Kostbade (K)

Department of Medical Oncology, University Hospital Essen (AöR), Essen, Germany; West German Cancer Center, University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

D Albers (D)

Department of Gastroenterology, Elisabeth Hospital Essen, Essen, Germany.

P Markus (P)

Department of General Surgery and Traumatology, Elisabeth Hospital Essen, Essen, Germany.

J Treckmann (J)

West German Cancer Center, University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany; Department of General, Visceral and Transplant Surgery, University Hospital Essen, Essen, Germany.

K Nassenstein (K)

Institute for Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

J Haubold (J)

Institute for Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

M Makowski (M)

Institute of Diagnostic and Interventional Radiology, Technical University of Munich, School of Medicine, Munich, Germany.

M Forsting (M)

German Cancer Consortium (DKTK), Partner site University Hospital Essen (AöR), Essen, Germany; Institute for Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

H A Baba (HA)

Institute of Pathology, University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

S Kasper (S)

Department of Medical Oncology, University Hospital Essen (AöR), Essen, Germany; West German Cancer Center, University Hospital Essen (AöR), Essen, Germany; German Cancer Consortium (DKTK), Partner site University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

J T Siveke (JT)

Department of Medical Oncology, University Hospital Essen (AöR), Essen, Germany; West German Cancer Center, University Hospital Essen (AöR), Essen, Germany; German Cancer Consortium (DKTK), Partner site University Hospital Essen (AöR), Essen, Germany; Bridge Institute of Experimental Tumor Therapy, West German Cancer Center, University Hospital Essen, University of Duisburg-Essen, Essen, Germany; Division of Solid Tumor Translational Oncology, German Cancer Consortium (DKTK Partner Site Essen) and German Cancer Research Center, DKFZ, Heidelberg, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

F Nensa (F)

Institute for Artificial Intelligence in Medicine, University Hospital Essen (AöR), Essen, Germany; German Cancer Consortium (DKTK), Partner site University Hospital Essen (AöR), Essen, Germany; Institute for Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

M Schuler (M)

Department of Medical Oncology, University Hospital Essen (AöR), Essen, Germany; West German Cancer Center, University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany; National Center for Tumor Diseases (NCT), NCT West, Essen, Germany.

G Kaissis (G)

Institute of Diagnostic and Interventional Radiology, Technical University of Munich, School of Medicine, Munich, Germany; Artificial Intelligence in Healthcare and Medicine, School of Computation, Information and Technology, Technical University of Munich, Munich, Germany.

J Kleesiek (J)

Institute for Artificial Intelligence in Medicine, University Hospital Essen (AöR), Essen, Germany; West German Cancer Center, University Hospital Essen (AöR), Essen, Germany; German Cancer Consortium (DKTK), Partner site University Hospital Essen (AöR), Essen, Germany; Medical Faculty, University of Duisburg-Essen, Essen, Germany.

R Braren (R)

Institute of Diagnostic and Interventional Radiology, Technical University of Munich, School of Medicine, Munich, Germany; German Cancer Consortium (DKTK), Munich partner site, Heidelberg, Germany.

Classifications MeSH