1
|
Ritch EJ, Herberts C, Warner EW, Ng SWS, Kwan EM, Bacon JVW, Bernales CQ, Schönlau E, Fonseca NM, Giri VN, Maurice-Dror C, Vandekerkhove G, Jones SJM, Chi KN, Wyatt AW. A generalizable machine learning framework for classifying DNA repair defects using ctDNA exomes. NPJ Precis Oncol 2023; 7:27. [PMID: 36914848 PMCID: PMC10011564 DOI: 10.1038/s41698-023-00366-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/12/2022] [Accepted: 02/27/2023] [Indexed: 03/16/2023] Open
Abstract
Specific classes of DNA damage repair (DDR) defect can drive sensitivity to emerging therapies for metastatic prostate cancer. However, biomarker approaches based on DDR gene sequencing do not accurately predict DDR deficiency or treatment benefit. Somatic alteration signatures may identify DDR deficiency but historically require whole-genome sequencing of tumour tissue. We assembled whole-exome sequencing data for 155 high ctDNA fraction plasma cell-free DNA and matched leukocyte DNA samples from patients with metastatic prostate or bladder cancer. Labels for DDR gene alterations were established using deep targeted sequencing. Per sample mutation and copy number features were used to train XGBoost ensemble models. Naive somatic features and trinucleotide signatures were associated with specific DDR gene alterations but insufficient to resolve each class. Conversely, XGBoost-derived models showed strong performance including an area under the curve of 0.99, 0.99 and 1.00 for identifying BRCA2, CDK12, and mismatch repair deficiency in metastatic prostate cancer. Our machine learning approach re-classified several samples exhibiting genomic features inconsistent with original labels, identified a metastatic bladder cancer sample with a homozygous BRCA2 copy loss, and outperformed an existing exome-based classifier for BRCA2 deficiency. We present DARC Sign (DnA Repair Classification SIGNatures); a public machine learning tool leveraging clinically-practical liquid biopsy specimens for simultaneously identifying multiple types of metastatic prostate cancer DDR deficiencies. We posit that it will be useful for understanding differential responses to DDR-directed therapies in ongoing clinical trials and may ultimately enable prospective identification of prostate cancers with phenotypic evidence of DDR deficiency.
Collapse
Affiliation(s)
- Elie J Ritch
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Cameron Herberts
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Evan W Warner
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Sarah W S Ng
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Edmond M Kwan
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Jack V W Bacon
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Cecily Q Bernales
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Elena Schönlau
- Department of Medical Oncology, BC Cancer, Vancouver, BC, Canada
| | - Nicolette M Fonseca
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Veda N Giri
- Yale School of Medicine and Yale Cancer Center, New Haven, CT, USA
| | | | - Gillian Vandekerkhove
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada
| | - Steven J M Jones
- Michael Smith Genome Sciences Centre, BC Cancer, Vancouver, BC, Canada
| | - Kim N Chi
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada.,Department of Medical Oncology, BC Cancer, Vancouver, BC, Canada
| | - Alexander W Wyatt
- Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC, Canada. .,Michael Smith Genome Sciences Centre, BC Cancer, Vancouver, BC, Canada.
| |
Collapse
|
2
|
Albuquerque MA, Grande BM, Ritch EJ, Pararajalingam P, Jessa S, Krzywinski M, Grewal JK, Shah SP, Boutros PC, Morin RD. Enhancing knowledge discovery from cancer genomics data with Galaxy. Gigascience 2018; 6:1-13. [PMID: 28327945 PMCID: PMC5437943 DOI: 10.1093/gigascience/gix015] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/04/2016] [Accepted: 03/06/2017] [Indexed: 01/15/2023] Open
Abstract
The field of cancer genomics has demonstrated the power of massively parallel sequencing techniques to inform on the genes and specific alterations that drive tumor onset and progression. Although large comprehensive sequence data sets continue to be made increasingly available, data analysis remains an ongoing challenge, particularly for laboratories lacking dedicated resources and bioinformatics expertise. To address this, we have produced a collection of Galaxy tools that represent many popular algorithms for detecting somatic genetic alterations from cancer genome and exome data. We developed new methods for parallelization of these tools within Galaxy to accelerate runtime and have demonstrated their usability and summarized their runtimes on multiple cloud service providers. Some tools represent extensions or refinement of existing toolkits to yield visualizations suited to cohort-wide cancer genomic analysis. For example, we present Oncocircos and Oncoprintplus, which generate data-rich summaries of exome-derived somatic mutation. Workflows that integrate these to achieve data integration and visualizations are demonstrated on a cohort of 96 diffuse large B-cell lymphomas and enabled the discovery of multiple candidate lymphoma-related genes. Our toolkit is available from our GitHub repository as Galaxy tool and dependency definitions and has been deployed using virtualization on multiple platforms including Docker.
Collapse
Affiliation(s)
- Marco A Albuquerque
- Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, BC, Canada
| | - Bruno M Grande
- Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, BC, Canada
| | - Elie J Ritch
- Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, BC, Canada
| | - Prasath Pararajalingam
- Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, BC, Canada
| | - Selin Jessa
- Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, BC, Canada
| | - Martin Krzywinski
- Canada's Michael Smith Genome Sciences Center, BC Cancer Agency, Vancouver, BC, Canada
| | - Jasleen K Grewal
- Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, BC, Canada
| | - Sohrab P Shah
- Department of Pathology, University of British Columbia, Vancouver, BC, Canada
| | - Paul C Boutros
- Ontario Institute for Cancer Research, Toronto, ON, Canada
| | - Ryan D Morin
- Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, BC, Canada.,Canada's Michael Smith Genome Sciences Center, BC Cancer Agency, Vancouver, BC, Canada
| |
Collapse
|