neoDL: a novel neoantigen intrinsic feature-based deep learning model identifies IDH wild-type glioblastomas with the longest survival

Background Neoantigen based personalized immune therapies achieve promising results in melanoma and lung cancer, but few neoantigen based models perform well in IDH wild-type GBM, and the association between neoantigen intrinsic features and prognosis remain unclear in IDH wild-type GBM. We presented a novel neoantigen intrinsic feature-based deep learning model (neoDL) to stratify IDH wild-type GBMs into subgroups with different survivals. Results We first derived intrinsic features for each neoantigen associated with survival, followed by applying neoDL in TCGA data cohort(AUC = 0.988, p value < 0.0001). Leave one out cross validation (LOOCV) in TCGA demonstrated that neoDL successfully classified IDH wild-type GBMs into different prognostic subgroups, which was further validated in an independent data cohort from Asian population. Long-term survival IDH wild-type GBMs identified by neoDL were found characterized by 12 protective neoantigen intrinsic features and enriched in development and cell cycle. Conclusions The model can be therapeutically exploited to identify IDH wild-type GBM with good prognosis who will most likely benefit from neoantigen based personalized immunetherapy. Furthermore, the prognostic intrinsic features of the neoantigens inferred from this study can be used for identifying neoantigens with high potentials of immunogenicity.

[1]  Özlem Türeci,et al.  Systemic RNA delivery to dendritic cells exploits antiviral defence for cancer immunotherapy , 2016, Nature.

[2]  Stephen T. C. Wong,et al.  Prognostic Gene Discovery in Glioblastoma Patients using Deep Learning , 2019, Cancers.

[3]  Yu-Da Lin,et al.  Identifying Risk Stratification Associated With a Cancer for Overall Survival by Deep Learning-Based CoxPH , 2019, IEEE Access.

[4]  Adam A. Margolin,et al.  Population-level distribution and putative immunogenicity of cancer neoepitopes , 2017, BMC Cancer.

[5]  V. Puduvalli,et al.  Treatment of Glioblastoma (GBM) with the Addition of Tumor-Treating Fields (TTF): A Review , 2019, Cancers.

[6]  Levi Garraway,et al.  Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden , 2017, Genome Medicine.

[7]  Pablo Tamayo,et al.  Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[8]  N. T. ten Hacken,et al.  Neoantigens in Chronic Obstructive Pulmonary Disease and Lung Cancer: A Point of View , 2019, Proteomics. Clinical applications.

[9]  P. Stephens,et al.  Tumor Mutational Burden as an Independent Predictor of Response to Immunotherapy in Diverse Cancers , 2017, Molecular Cancer Therapeutics.

[10]  Maxim N. Artyomov,et al.  Tumor neoantigens: building a framework for personalized cancer immunotherapy. , 2015, The Journal of clinical investigation.

[11]  Yang Yang,et al.  Glioma Grading on Conventional MR Images: A Deep Learning Study With Transfer Learning , 2018, Front. Neurosci..

[12]  Jan C Buckner,et al.  Factors influencing survival in high-grade gliomas. , 2003, Seminars in oncology.

[13]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[14]  Sara R. Selitsky,et al.  Machine-Learning Prediction of Tumor Antigen Immunogenicity in the Selection of Therapeutic Epitopes , 2019, Cancer Immunology Research.

[15]  N. McGranahan,et al.  Differential binding affinity of mutated peptides for MHC class I is a predictor of survival in advanced lung cancer and melanoma , 2017, Annals of oncology : official journal of the European Society for Medical Oncology.

[16]  M. Daly,et al.  PGC-1α-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes , 2003, Nature Genetics.

[17]  Gabriel N. Teku,et al.  Pan-cancer analysis of neoepitopes , 2018, Scientific Reports.

[18]  Xinzhong Li,et al.  Large-Scale Analysis Reveals Gene Signature for Survival Prediction in Primary Glioblastoma , 2020, Molecular Neurobiology.

[19]  J. Castle,et al.  Exploiting the mutanome for tumor vaccination. , 2012, Cancer research.

[20]  Erwin G. Van Meir,et al.  Exciting New Advances in Neuro‐Oncology: The Avenue to a Cure for Malignant Glioma , 2010, CA: a cancer journal for clinicians.

[21]  Mark J Daly,et al.  Burden of unique and low prevalence somatic mutations correlates with cancer survival , 2019, Scientific Reports.

[22]  Ellery Wulczyn,et al.  Development and validation of a deep learning algorithm for improving Gleason scoring of prostate cancer , 2018, npj Digital Medicine.

[23]  S. Mohamad R. Soroushmehr,et al.  Deep Learning in Pharmacogenomics: From Gene Regulation to Patient Stratification , 2018, Pharmacogenomics.

[24]  In-Hee Lee,et al.  Clonal evolution of glioblastoma under therapy , 2016, Nature Genetics.

[25]  Steven J. M. Jones,et al.  Molecular Profiling Reveals Biologically Discrete Subsets and Pathways of Progression in Diffuse Glioma , 2016, Cell.

[26]  Jie Hao,et al.  PASNet: pathway-associated sparse deep neural network for prognosis prediction from high-throughput data , 2018, BMC Bioinformatics.

[27]  Nicolai J. Birkbak,et al.  Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade , 2016, Science.

[28]  Yuri Tolkach,et al.  High-accuracy prostate cancer pathology using deep learning , 2020, Nature Machine Intelligence.

[29]  G. Getz,et al.  Inferring tumour purity and stromal and immune cell admixture from expression data , 2013, Nature Communications.

[30]  A. Butte,et al.  Systematic pan-cancer analysis of tumour purity , 2015, Nature Communications.

[31]  Antonello Rizzi,et al.  Cancer Diagnosis Using Deep Learning: A Bibliographic Review , 2019, Cancers.

[32]  J. Devlin,et al.  Structure Based Prediction of Neoantigen Immunogenicity , 2019, Front. Immunol..

[33]  C. Huber,et al.  The response of autologous T cells to a human melanoma is dominated by mutated neoantigens. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[34]  M. Daly,et al.  Gender Disparity and Mutation Burden in Metastatic Melanoma. , 2015, Journal of the National Cancer Institute.

[35]  Christopher A. Miller,et al.  Biological and therapeutic implications of multisector sequencing in newly diagnosed glioblastoma , 2018, Neuro-oncology.

[36]  Nathalie Harder,et al.  Automatic discovery of image-based signatures for ipilimumab response prediction in malignant melanoma , 2019, Scientific Reports.

[37]  A. Bardelli,et al.  Evolving neoantigen profiles in colorectal cancers with DNA repair defects , 2019, Genome Medicine.

[38]  P. Shannon,et al.  Cytoscape: a software environment for integrated models of biomolecular interaction networks. , 2003, Genome research.

[39]  T. Schumacher,et al.  Neoantigens in cancer immunotherapy , 2015, Science.

[40]  Dmitrii Bychkov,et al.  Deep learning based tissue analysis predicts outcome in colorectal cancer , 2018, Scientific Reports.

[41]  M. Ceccarelli,et al.  The combination of neoantigen quality and T lymphocyte infiltrates identifies glioblastomas with the longest survival. , 2019 .

[42]  Zoltan Szallasi,et al.  Tumor Mutation Burden Forecasts Outcome in Ovarian Cancer with BRCA1 or BRCA2 Mutations , 2013, PloS one.

[43]  Shuguang Zuo,et al.  A RNA sequencing-based six-gene signature for survival prediction in patients with glioblastoma , 2019, Scientific Reports.

[44]  Angela E. Leek,et al.  Allele-Specific HLA Loss and Immune Escape in Lung Cancer Evolution , 2017, Cell.

[45]  M. J. van den Bent,et al.  PI3 kinase mutations and mutational load as poor prognostic markers in diffuse glioma patients , 2015, Acta Neuropathologica Communications.

[46]  P. Baldi,et al.  Deep-Learning Convolutional Neural Networks Accurately Classify Genetic Mutations in Gliomas , 2018, American Journal of Neuroradiology.