Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research

1707 - Gene embedding: a novel machine learning approach to identify gene candidates related to immunotherapy responsiveness

Date

20 Oct 2018

Session

Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research

Topics

Translational Research

Tumour Site

Presenters

Chi Tung Choy

Citation

Annals of Oncology (2018) 29 (suppl_8): viii14-viii57. 10.1093/annonc/mdy269

Authors

C.T. Choy1, C.H. Wong2, S.L. Chan2

Author affiliations

  • 1 Department Of Clinical Oncology, Faculty Of Medicine, The Chinese University of Hong Kong, N/A - Hong Kong/HK
  • 2 The Cancer Drug Testing Unit, State Key Laboratory Of Oncology In South China, Department Of Clinical Oncology, Faculty Of Medicine, The Chinese University of Hong Kong, Hong Kong/HK
More

Abstract 1707

Background

Apart from PD-L1 and mutational load, there are no genetic predictive biomarkers for checkpoint inhibitors treatment. In this study, gene embedding, a machine learning technique, was used to single out related genes of immune checkpoint proteins (i.e. PD-1, PD-L1, CTLA-4) as new potential predictors for such responders.

Methods

TCGA RNASeqV2 level 3 RSEM normalized read counts (January 2016) were downloaded from the Broad Institute TCGA GDAC Firehose. A shallow neural network, aka embedding layers for samples and genes, were trained using log2 transformed data. Neighbors closeness were evaluated by euclidean distance. The model was kept blind from any additional information, including cancer types, protein-protein interactions and gene ontologies.

Results

Gene expressions of 13045 samples from 36 cancer types were embedded into 50-dimension space, while cancer types were learnt by the model without supervision. Immunotherapy responders and non-responders were stimulated from melanoma (SKMC) and lung squamous cell carcinoma (LUSC) data, and hepatocellular carcinoma and prostate cancer data respectively. 9 genes (TNFRSF8, CLEC10A, FCN1, CD8B, SLA2, IL2RA, CTLA4, GZMH), 3 (CD101, LOC154761, RNF152) genes, and 6 (SH2D1A, MEI1, PDCD1, GFI1, SIT1, SIRPG) genes were found to be closely related neighbors with PD-1, PD-L1, and CTLA-4 respectively in responders but not in non-responders. All neighbors were neither co-expressed in SKMC/LUSC dataset nor indicated as interacting partners on existing databases (BioGRID, MINT, iRefWeb, STRING, HPRD and Reactome). 88.8% genes were evidenced as either directly related to checkpoint proteins and/or T cells activation in literature. Further evaluation of the role of identified targets in immune checkpoint blockade therapy would be warranted.

Conclusions

We identified potential biomarker candidates for immune checkpoint blockade therapy by TCGA data mining and demonstrated the utility of gene embedding learned from big gene expression dataset as a powerful tool to uncover gene relationships that may not be discovered otherwise without prior knowledge on functional interactions.

Clinical trial identification

Legal entity responsible for the study

Department of Clinical Oncology, Faculty of Medicine, The Chinese University of Hong Kong.

Funding

Department of Clinical Oncology, Faculty of Medicine, The Chinese University of Hong Kong.

Editorial Acknowledgement

Disclosure

All authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.