Abstract 5820
Background
DNA sequencing to identify variants is becoming increasingly valuable in clinical settings; including matching patients to approved targeted therapies, immunotherapies, and/or clinical trials. However, accurate calling of genetic variants from sequencing still remains challenging. With little corroboration between the different tools available, patients are at risk of being treated with therapies that are unsuitable for their cancer.
Methods
Here we present a novel machine learning based method for the accurate identification of somatic variants in cancer patient tumour samples, with a neural network architecture from encoded raw sequencing read information of tumour/normal sample pairings into an image, enabling it to classify whether a variant is germline, somatic, or sequencing error. The model was trained and tested on in-silico spike-in data using bam-surgeon, and then validated on a multi-cancer and multi-center dataset and benchmarked against industry standard variant callers.
Results
The approach, called somaticNET, outperforms existing industry standard tools in sensitivity and specificity, achieving an AUROC of ∼1.00 on the bam-surgeon dataset and an AUROC of ∼0.99 on the multi-cancer multicenter dataset. The model also works faster than other variant callers, in minutes compared to hours.
Conclusions
Using the power of machine learning for accurate somatic variant calling can improve patient matching to approved therapies and clinical trials, thus ensuring patients are given the right therapy at the right time to treat their cancer.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Cambridge Cancer Genomics.
Disclosure
G. Dubourg-Felonneau: Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. D. Rebergen: Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. C. Parsons: Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. H. Thompson: Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. J.W. Cassidy: Leadership role, Shareholder / Stockholder / Stock options, Full / Part-time employment, Officer / Board of Directors: Cambridge Cancer Genomics. N. Patel: Leadership role, Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. H.W. Clifford: Leadership role, Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics.
Resources from the same session
4596 - A Phase 2, Open-Label, Randomized, Multicenter Trial of Encorafenib + Binimetinib Evaluating a Standard-dose and a High-dose Regimen in Patients With BRAFV600-Mutant Melanoma Brain Metastasis (MBM) (POLARIS)
Presenter: Michael Davies
Session: Poster Display session 3
Resources:
Abstract
1891 - Continuation of annual screening mammograms and breast-cancer mortality in women over 70
Presenter: Xabier Garcia De Albeniz
Session: Poster Display session 3
Resources:
Abstract
5587 - Introducing standardized medical procedure and dynamic decision support program in precision oncology for the community of practice
Presenter: Istvan Petak
Session: Poster Display session 3
Resources:
Abstract
4757 - Effectively using primary care givers in oncology care through capacity building, task sharing and techno-mentoring.
Presenter: Dinesh Pendharkar
Session: Poster Display session 3
Resources:
Abstract
4497 - A single institution review of capecitabine related acute admissions and cost analysis
Presenter: Gemma Dart
Session: Poster Display session 3
Resources:
Abstract
2187 - Health status of middle-aged and older cancer survivors in China: results from the China Health and Retirement Longitudinal Study (CHARLS)
Presenter: Jiarui Li
Session: Poster Display session 3
Resources:
Abstract
5101 - Crossed looks on lung cancer perception and knowledge from general public and physicians in France: results of a two-fold survey
Presenter: Céline Mascaux
Session: Poster Display session 3
Resources:
Abstract
4354 - Knowledge and perception of clinical trials (CTs) and attitude towards participation among Polish oncological patients - A pilot survey
Presenter: Artur Kotowski
Session: Poster Display session 3
Resources:
Abstract
3499 - Achieving best possible cancer treatment outcomes in care pathways through benchmarking; ABC-Benchmarking
Presenter: Anke Wind
Session: Poster Display session 3
Resources:
Abstract
2270 - Impact of 10-day Fulbright Specialist Program (FSP) and Project Pink Blue (PPB) Education Sessions on Medical Oncology knowledge among Doctors that treat cancer in Nigeria
Presenter: Mike Martin
Session: Poster Display session 3
Resources:
Abstract