Decoding pan-cancer treatment outcomes using multimodal real-world data and explainable artificial intelligence

被引：0

作者：

Julius Keyl ^{[1
]}

Philipp Keyl ^{[2
]}

Grégoire Montavon ^{[3
]}

René Hosch ^{[4
]}

Alexander Brehmer ^{[4
]}

Liliana Mochmann ^{[5
]}

Philipp Jurmeister ^{[6
]}

Gabriel Dernbach ^{[1
]}

Moon Kim ^{[1
]}

Sven Koitka ^{[3
]}

Sebastian Bauer ^{[3
]}

Nikolaos Bechrakis ^{[5
]}

Michael Forsting ^{[1
]}

Dagmar Führer-Sakel ^{[1
]}

Martin Glas ^{[7
]}

Viktor Grünwald ^{[8
]}

Boris Hadaschik ^{[9
]}

Johannes Haubold ^{[10
]}

Ken Herrmann ^{[11
]}

Stefan Kasper ^{[9
]}

Rainer Kimmig ^{[10
]}

Stephan Lang ^{[11
]}

Tienush Rassaf ^{[12
]}

Alexander Roesch ^{[7
]}

Dirk Schadendorf ^{[9
]}

Jens T. Siveke ^{[11
]}

Martin Stuschke ^{[9
]}

Ulrich Sure ^{[10
]}

Matthias Totzeck ^{[13
]}

Anja Welt ^{[9
]}

Marcel Wiesweg ^{[10
]}

Hideo A. Baba ^{[11
]}

Felix Nensa ^{[14
]}

Jan Egger ^{[8
]}

Klaus-Robert Müller ^{[9
]}

Martin Schuler ^{[10
]}

Frederick Klauschen ^{[11
]}

Jens Kleesiek ^{[15
]}

机构：

[1] University Hospital Essen (AöR),Institute for Artificial Intelligence in Medicine

[2] University Hospital Essen (AöR),Institute of Pathology

[3] Ludwig-Maximilians-University Munich,Institute of Pathology

[4] BIFOLD – Berlin Institute for the Foundations of Learning and Data,Machine Learning Group

[5] Technical University of Berlin,Department of Mathematics and Computer Science

[6] Freie Universität Berlin,Institute for Diagnostic and Interventional Radiology and Neuroradiology

[7] University Hospital Essen (AöR),Department of Medical Oncology

[8] University Hospital Essen (AöR),Medical Faculty

[9] University of Duisburg-Essen,West German Cancer Center

[10] University Hospital Essen (AöR),German Cancer Consortium (DKTK)

[11] Partner site University Hospital Essen (AöR),Department of Ophthalmology

[12] University Hospital Essen (AöR),Department of Endocrinology, Diabetes and Metabolism

[13] University Hospital Essen (AöR),Division of Clinical Neurooncology, Department of Neurology and Center for Translational Neuro

[14] University Duisburg-Essen, and Behavioral Sciences (C

[15] University Hospital Essen (AöR),TNBS), University Medicine Essen

[16] University Hospital Essen (AöR),Department of Urology

[17] University Hospital Essen (AöR),Department of Nuclear Medicine

[18] University Hospital Essen (AöR),Department of Gynecology and Obstetrics

[19] University Hospital Essen (AöR),Department of Otorhinolaryngology

[20] University Hospital Essen (AöR),Department of Cardiology and Vascular Medicine, West German Heart and Vascular Center Essen

[21] University of Duisburg-Essen,Department of Dermatology

[22] University of Duisburg-Essen,Research Alliance Ruhr, Research Center One Health

[23] DKFZ,Bridge Institute of Experimental Tumor Therapy, West German Cancer Center, University Hospital Essen (AöR)

[24] University Hospital Essen (AöR),Division of Solid Tumor Translational Oncology, German Cancer Consortium (DKTK Partner Site Essen) and German Cancer Research Center

[25] University Hospital Essen (AöR),Department of Radiotherapy

[26] Korea University,Department of Neurosurgery and Spine Surgery

[27] MPI for Informatics,Department of Artificial Intelligence

[28] Berlin partner site,German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ)

[29] Munich partner site,German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ)

[30] Bavarian Cancer Research Center (BZKF),undefined

来源：

Nature Cancer | 2025年 / 6卷 / 2期

关键词：

D O I：

10.1038/s43018-024-00891-1

中图分类号：

学科分类号：

摘要：

Despite advances in precision oncology, clinical decision-making still relies on limited variables and expert knowledge. To address this limitation, we combined multimodal real-world data and explainable artificial intelligence (xAI) to introduce AI-derived (AID) markers for clinical decision support. We used xAI to decode the outcome of 15,726 patients across 38 solid cancer entities based on 350 markers, including clinical records, image-derived body compositions, and mutational tumor profiles. xAI determined the prognostic contribution of each clinical marker at the patient level and identified 114 key markers that accounted for 90% of the neural network’s decision process. Moreover, xAI enabled us to uncover 1,373 prognostic interactions between markers. Our approach was validated in an independent cohort of 3,288 patients with lung cancer from a US nationwide electronic health record-derived database. These results show the potential of xAI to transform the assessment of clinical variables and enable personalized, data-driven cancer care.

引用

页码：307 / 322

页数：15

共 50 条

[1] Systematic pan-cancer analysis of mutation–treatment interactions using large real-world clinicogenomics data
Ruishan Liu
Shemra Rizzo
Sarah Waliany
Marius Rene Garmhausen
Navdeep Pal
Zhi Huang
Nayan Chaudhary
Lisa Wang
Chris Harbron
Joel Neal
Ryan Copping
James Zou
Nature Medicine, 2022, 28 : 1656 - 1661
[2] An artificial intelligence framework integrating longitudinal electronic health records with real-world data enables continuous pan-cancer prognostication
Olivier Morin
Martin Vallières
Steve Braunstein
Jorge Barrios Ginart
Taman Upadhaya
Henry C. Woodruff
Alex Zwanenburg
Avishek Chatterjee
Javier E. Villanueva-Meyer
Gilmer Valdes
William Chen
Julian C. Hong
Sue S. Yom
Timothy D. Solberg
Steffen Löck
Jan Seuntjens
Catherine Park
Philippe Lambin
Nature Cancer, 2021, 2 : 709 - 722
[3] An artificial intelligence framework integrating longitudinal electronic health records with real-world data enables continuous pan-cancer prognostication
Morin, Olivier
Vallieres, Martin
Braunstein, Steve
Ginart, Jorge Barrios
Upadhaya, Taman
Woodruff, Henry C.
Zwanenburg, Alex
Chatterjee, Avishek
Villanueva-Meyer, Javier E.
Valdes, Gilmer
Chen, William
Hong, Julian C.
Yom, Sue S.
Solberg, Timothy D.
Lock, Steffen
Seuntjens, Jan
Park, Catherine
Lambin, Philippe
NATURE CANCER, 2021, 2 (07) : 709 - +
[4] Systematic pan-cancer analysis of mutation-treatment interactions using large real-world clinicogenomics data
Liu, Ruishan
Rizzo, Shemra
Waliany, Sarah
Garmhausen, Marius Rene
Pal, Navdeep
Huang, Zhi
Chaudhary, Nayan
Wang, Lisa
Harbron, Chris
Neal, Joel
Copping, Ryan
Zou, James
NATURE MEDICINE, 2022, 28 (08) : 1656 - +
[5] Real-time data quality assurance analysis for real-world, pan-cancer data.
Brown, Samantha
Lee, Jasme
Pillai, Anjali
Gandhi, Fenil
Bahadur, Nadia
Barton, Laura
Chan, Kimberly
Niederhausern, Andrew
Nichols, Chelsea
Philip, John
Regazzi, Ashley Marie
Shah, Neil J.
Panageas, Katherine
Lavery, Jessica A.
JOURNAL OF CLINICAL ONCOLOGY, 2021, 39 (15)
[6] Real-world genetic testing outcomes of pan-cancer testing for mismatch repair deficiency
Chai, Teresa S.
Rodgers-Fouche, Linda H.
Walls, Jenna O.
Mattia, Anthony R.
Chung, Daniel C.
CANCER, 2024,
[7] Explainable artificial intelligence models using real-world electronic health record data: a systematic scoping review
Payrovnaziri, Seyedeh Neelufar
Chen, Zhaoyi
Rengifo-Moreno, Pablo
Miller, Tim
Bian, Jiang
Chen, Jonathan H.
Liu, Xiuwen
He, Zhe
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION, 2020, 27 (07) : 1173 - 1185
[8] From real-world electronic health record data to real-world results using artificial intelligence
Knevel, Rachel
Liao, Katherine P.
ANNALS OF THE RHEUMATIC DISEASES, 2023, 82 (03) : 306 - 311
[9] Applications of artificial intelligence in drug development using real-world data
Chen, Zhaoyi
Liu, Xiong
Hogan, William
Shenkman, Elizabeth
Bian, Jiang
DRUG DISCOVERY TODAY, 2021, 26 (05) : 1256 - 1264
[10] Interpretation of Structure-Activity Relationships in Real-World Drug Design Data Sets Using Explainable Artificial Intelligence
Harren, Tobias
Matter, Hans
Hessler, Gerhard
Rarey, Matthias
Grebner, Christoph
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2022, 62 (03) : 447 - 462

← 1 2 3 4 5 →