A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns.

Jiao, Wei; Atwal, Gurnit; Polak, Paz; Karlic, Rosa; Cuppen, Edwin; Danyi, Alexandra; de Ridder, Jeroen; van Herpen, Carla; Lolkema, Martijn P; Steeghs, Neeltje; Getz, Gad; Morris, Quaid; Stein, Lincoln D (2020). A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns. Nature Communications, 11(1), p. 728. Springer Nature 10.1038/s41467-019-13825-8

[img]
Preview
Text
41467_2019_Article_13825.pdf - Published Version
Available under License Creative Commons: Attribution (CC-BY).

Download (827kB) | Preview

In cancer, the primary tumour's organ of origin and histopathology are the strongest determinants of its clinical behaviour, but in 3% of cases a patient presents with a metastatic tumour and no obvious primary. Here, as part of the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium, we train a deep learning classifier to predict cancer type based on patterns of somatic passenger mutations detected in whole genome sequencing (WGS) of 2606 tumours representing 24 common cancer types produced by the PCAWG Consortium. Our classifier achieves an accuracy of 91% on held-out tumor samples and 88% and 83% respectively on independent primary and metastatic samples, roughly double the accuracy of trained pathologists when presented with a metastatic tumour without knowledge of the primary. Surprisingly, adding information on driver mutations reduced accuracy. Our results have clinical applicability, underscore how patterns of somatic passenger mutations encode the state of the cell of origin, and can inform future strategies to detect the source of circulating tumour DNA.

Item Type:

Journal Article (Original Article)

Division/Institute:

04 Faculty of Medicine > Pre-clinic Human Medicine > BioMedical Research (DBMR) > DBMR Forschung Mu35 > Forschungsgruppe Präzisionsonkologie
04 Faculty of Medicine > Pre-clinic Human Medicine > BioMedical Research (DBMR) > DBMR Forschung Mu35 > Forschungsgruppe Präzisionsonkologie

04 Faculty of Medicine > Pre-clinic Human Medicine > BioMedical Research (DBMR) > DBMR Forschung Mu35
04 Faculty of Medicine > Pre-clinic Human Medicine > BioMedical Research (DBMR) > DBMR Forschung Mu35

Subjects:

600 Technology > 610 Medicine & health

ISSN:

2041-1723

Publisher:

Springer Nature

Language:

English

Submitter:

Marla Rittiner

Date Deposited:

24 Dec 2020 11:39

Last Modified:

10 Apr 2021 02:53

Publisher DOI:

10.1038/s41467-019-13825-8

PubMed ID:

32024849

Additional Information:

Collaborator from the DBMR: Mark A Rubin (Director DBMR, Precision Medicine)

BORIS DOI:

10.48350/150188

URI:

https://boris.unibe.ch/id/eprint/150188

Actions (login required)

Edit item Edit item
Provide Feedback