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

被引：0

作者：

Keyl, Julius ^{[1
,2
]}

Keyl, Philipp ^{[3
,4
]}

Montavon, Gregoire ^{[4
,5
,6
]}

Hosch, Rene ^{[1
]}

Brehmer, Alexander ^{[1
]}

Mochmann, Liliana ^{[3
]}

Jurmeister, Philipp ^{[3
]}

Dernbach, Gabriel ^{[5
]}

Kim, Moon ^{[1
]}

Koitka, Sven ^{[1
,7
]}

Bauer, Sebastian ^{[8
,9
,10
,11
]}

Bechrakis, Nikolaos ^{[9
,10
,11
,12
]}

Forsting, Michael ^{[7
,9
,11
]}

Fuehrer-Sakel, Dagmar ^{[9
,10
,13
]}

Glas, Martin ^{[9
,10
,11
,14
,15
]}

Gruenwald, Viktor ^{[8
,9
,10
,11
,16
]}

Hadaschik, Boris ^{[9
,10
,11
,16
]}

Haubold, Johannes ^{[7
,9
]}

Herrmann, Ken ^{[9
,10
,11
,17
]}

Kasper, Stefan ^{[8
,9
,10
,11
]}

Kimmig, Rainer ^{[9
,10
,18
]}

Lang, Stephan ^{[9
,19
]}

Rassaf, Tienush ^{[9
,20
]}

Roesch, Alexander ^{[9
,10
,11
,21
]}

Schadendorf, Dirk ^{[9
,10
,11
,21
,22
]}

Siveke, Jens T. ^{[8
,9
,10
,11
,23
,24
,25
]}

Stuschke, Martin ^{[9
,10
,11
,26
]}

Sure, Ulrich ^{[9
,10
,11
,27
]}

Totzeck, Matthias ^{[9
,20
]}

Welt, Anja ^{[8
,9
,10
]}

Wiesweg, Marcel ^{[8
,9
,10
,11
]}

Baba, Hideo A. ^{[2
,9
]}

Nensa, Felix ^{[1
,7
,9
,11
]}

Egger, Jan ^{[1
]}

Mueller, Klaus-Robert ^{[4
,5
,28
,29
]}

Schuler, Martin ^{[8
,9
,10
,11
]}

Klauschen, Frederick ^{[3
,4
,30
,31
,32
]}

Kleesiek, Jens ^{[1
,9
,10
,11
]}

机构：

[1] Univ Hosp Essen AoR, Inst Artificial Intelligence Med, Essen, Germany

[2] Univ Hosp Essen AoR, Inst Pathol, Essen, Germany

[3] Ludwig Maximilians Univ Munchen, Inst Pathol, Munich, Germany

[4] BIFOLD Berlin Inst Fdn Learning & Data, Berlin, Germany

[5] Tech Univ Berlin, Machine Learning Grp, Berlin, Germany

[6] Free Univ Berlin, Dept Math & Comp Sci, Berlin, Germany

[7] Univ Hosp Essen AoR, Inst Diagnost & Intervent Radiol & Neuroradiol, Essen, Germany

[8] Univ Hosp Essen AoR, Dept Med Oncol, Essen, Germany

[9] Univ Duisburg Essen, Fac Med, Essen, Germany

[10] Univ Hosp Essen AoR, West German Canc Ctr, Essen, Germany

[11] Univ Hosp Essen AoR, German Canc Consortium DKTK, Partner Site, Essen, Germany

[12] Univ Hosp Essen AoR, Dept Nuclearmedicine, D-45147 Essen, Germany

[13] Univ Hosp Essen AoR, Dept Endocrinol Diabet & Metab, Essen, Germany

[14] Univ Duisburg Essen, Dept Neurol, Div Clin Neurooncol, Essen, Germany

[15] Univ Duisburg Essen, Univ Med Essen, Ctr Translat Neuroand Behav Sci C TNBS, Essen, Germany

[16] Univ Hosp Essen AoR, Dept Urol, Essen, Germany

[17] Univ Hosp Essen AoR, Dept Nucl Med, Essen, Germany

[18] Univ Hosp Essen AoR, Dept Gynecol & Obstet, Essen, Germany

[19] Univ Hosp Essen AoR, Dept Endocrinol & Metab, Essen, Germany

[20] Univ Hosp Essen AoR, West German Heart & Vasc Ctr Essen, Dept Cardiol & Vasc Med, Essen, Germany

[21] Univ Hosp Essen AoR, Dept Dermatol, D-45147 Essen, Germany

[22] Univ Duisburg Essen, Res Ctr One Hlth, Res Alliance Ruhr, Essen, Germany

[23] Univ Duisburg Essen, Univ Hosp Essen AoR, Bridge Inst Expt Tumor Therapy, West German Canc Ctr, Essen, Germany

[24] German Canc Consortium DKTK, Div Solid Tumor Translat Oncol, Partner Site Essen, Heidelberg, Germany

[25] German Canc Res Ctr, DKFZ, Heidelberg, Germany

[26] Univ Hosp Essen AoR, Dept Radiotherapy, Essen, Germany

[27] Univ Hosp Essen AoR, Dept Neurosurg & Spine Surg, Essen, Germany

[28] Korea Univ, Dept Artificial Intelligence, Seoul, South Korea

[29] MPI Informat, Saarbrucken, Germany

[30] German Canc Consortium DKTK, German Canc Res Ctr DKFZ, Berlin Partner Site, Berlin, Germany

[31] German Canc Consortium DKTK, German Canc Res Ctr DKFZ, Munich Partner Site, Munich, Germany

[32] Bavarian Canc Res Ctr BZKF, Erlangen, Germany

来源：

NATURE CANCER | 2025年 / 6卷 / 02期

关键词：

PANCREATIC-CANCER; PROGNOSTIC-FACTOR; NEURAL-NETWORKS; SURVIVAL;

D O I：

10.1038/s43018-024-00891-1

中图分类号：

R73 [肿瘤学];

学科分类号：

100214 ;

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：32

共 64 条

[1] Multimodal biomedical AI
Acosta, Julian N.
Falcone, Guido J.
Rajpurkar, Pranav
Topol, Eric J.
[J]. NATURE MEDICINE, 2022, 28 (09) : 1773 - 1784
[2] LDH correlation with survival in advanced melanoma from two large, randomised trials (Oblimersen GM301 and EORTC 18951)
Agarwala, Sanjiv S.
Keilholz, Ulrich
Gilles, Erard
Bedikian, Agop Y.
Wu, Jane
Kay, Richard
Stein, Cy A.
Itri, Loretta M.
Suciu, Stefan
Eggermont, Alexander M. M.
[J]. EUROPEAN JOURNAL OF CANCER, 2009, 45 (10) : 1807 - 1814
[3] Relationship among circulating tumor cells, CEA and overall survival in patients with metastatic colorectal cancer
Aggarwal, C.
Meropol, N. J.
Punt, C. J.
Iannotti, N.
Saidman, B. H.
Sabbath, K. D.
Gabrail, N. Y.
Picus, J.
Morse, M. A.
Mitchell, E.
Miller, M. C.
Cohen, S. J.
[J]. ANNALS OF ONCOLOGY, 2013, 24 (02) : 420 - 428
[4] On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation
Bach, Sebastian
Binder, Alexander
Montavon, Gregoire
Klauschen, Frederick
Mueller, Klaus-Robert
Samek, Wojciech
[J]. PLOS ONE, 2015, 10 (07):
[5] Fitting Linear Mixed-Effects Models Using lme4
Bates, Douglas
Maechler, Martin
Bolker, Benjamin M.
Walker, Steven C.
[J]. JOURNAL OF STATISTICAL SOFTWARE, 2015, 67 (01): : 1 - 48
[6] Bender D, 2013, COMP MED SY, P326, DOI 10.1109/CBMS.2013.6627810
[7] Survival and New Prognosticators in Metastatic Seminoma: Results From the IGCCCG-Update Consortium
Beyer, Jorg
Collette, Laurence
Sauve, Nicolas
Daugaard, Gedske
Feldman, Darren R.
Tandstad, Torgrim
Tryakin, Alexey
Stahl, Olof
Gonzalez-Billalabeitia, Enrique
De Giorgi, Ugo
Culine, Stephane
de Wit, Ronald
Hansen, Aaron R.
Bebek, Marko
Terbuch, Angelika
Albany, Costantine
Hentrich, Marcus
Gietema, Jourik A.
Negaard, Helene
Huddart, Robert A.
Lorch, Anja
Cafferty, Fay H.
Heng, Daniel Y. C.
Sweeney, Christopher J.
Winquist, Eric
Chovanec, Michal
Fankhauser, Christian
Stark, Daniel
Grimison, Peter
Necchi, Andrea
Tran, Ben
Heidenreich, Axel
Shamash, Jonathan
Sternberg, Cora N.
Vaughn, David J.
Duran, Ignacio
Bokemeyer, Carsten
Patrikidou, Anna
Cathomas, Richard
Assele, Samson
Gillessen, Silke
[J]. JOURNAL OF CLINICAL ONCOLOGY, 2021, 39 (14) : 1553 - +
[8] Birnbaum B, 2020, Arxiv, DOI arXiv:2001.09765
[9] Prognostic factors for progression-free and overall survival in advanced biliary tract cancer
Bridgewater, J.
Lopes, A.
Wasan, H.
Malka, D.
Jensen, L.
Okusaka, T.
Knox, J.
Wagner, D.
Cunningham, D.
Shannon, J.
Goldstein, D.
Moehler, M.
Bekaii-Saab, T.
McNamara, M. G.
Valle, J. W.
[J]. ANNALS OF ONCOLOGY, 2016, 27 (01) : 134 - 140
[10] Preoperative low tri-iodothyronine concentration is associated with worse health status and shorter five year survival of primary brain tumor patients
Bunevicius, Adomas
Deltuva, Vytenis Pranas
Tamasauskas, Sarunas
Smith, Timothy
Laws, Edward R.
Bunevicius, Robertas
Iervasi, Giorgio
Tamasauskas, Arimantas
[J]. ONCOTARGET, 2017, 8 (05) : 8648 - 8656

← 1 2 3 4 5 6 7 →