Driver network as a biomarker: systematic integration and network modeling of multi-omics data to derive driver signaling pathways for drug combination prediction

Lei Huang, David Brunell, Clifford Stephan, James Mancuso, Xiaohui Yu, Bin He, Timothy C Thompson, Ralph Zinner, Jeri Kim, Peter Davies, Stephen T C Wong, Lei Huang, David Brunell, Clifford Stephan, James Mancuso, Xiaohui Yu, Bin He, Timothy C Thompson, Ralph Zinner, Jeri Kim, Peter Davies, Stephen T C Wong

Abstract

Motivation: Drug combinations that simultaneously suppress multiple cancer driver signaling pathways increase therapeutic options and may reduce drug resistance. We have developed a computational systems biology tool, DrugComboExplorer, to identify driver signaling pathways and predict synergistic drug combinations by integrating the knowledge embedded in vast amounts of available pharmacogenomics and omics data.

Results: This tool generates driver signaling networks by processing DNA sequencing, gene copy number, DNA methylation and RNA-seq data from individual cancer patients using an integrated pipeline of algorithms, including bootstrap aggregating-based Markov random field, weighted co-expression network analysis and supervised regulatory network learning. It uses a systems pharmacology approach to infer the combinatorial drug efficacies and synergy mechanisms through drug functional module-induced regulation of target expression analysis. Application of our tool on diffuse large B-cell lymphoma and prostate cancer demonstrated how synergistic drug combinations can be discovered to inhibit multiple driver signaling pathways. Compared with existing computational approaches, DrugComboExplorer had higher prediction accuracy based on in vitro experimental validation and probability concordance index. These results demonstrate that our network-based drug efficacy screening approach can reliably prioritize synergistic drug combinations for cancer and uncover potential mechanisms of drug synergy, warranting further studies in individual cancer patients to derive personalized treatment plans.

Availability and implementation: DrugComboExplorer is available at https://github.com/Roosevelt-PKU/drugcombinationprediction.

Supplementary information: Supplementary data are available at Bioinformatics online.

© The Author(s) 2019. Published by Oxford University Press.

Figures

Fig. 1.
Fig. 1.
The workflow of DrugComboExplorer includes two major components: (A) systematic overview of applying NPBSA on four different types of genomics data to identify driver signaling pathways in cancer. There are five steps in this component, Step 1: to identify the seed (driver) genes from the frequently mutated and copy number variation genes using the DNA-seq and copy number data of specific cancer patients; Step 2: to explore networks from the seed genes by integrating the RNA-seq profiles and pathway knowledge from known pathway databases and generate a gene co-expression network and a gene regulatory network using the RNA-seq profiles; Step 3: to explore networks from the seed genes by integrating the methylation profiles and pathway data; Step 4: to combine the networks generated from the RNA-seq profiles together; Step 5: to combine the networks generated from the RNA-seq data and from the methylation data together. Herein, the color legend shows how the node color represents the fold change. (B) Identification of synergistic drug combinations and the underlying mechanisms based on the pharmacogenomics data and the identified dysregulated driver networks. There are five steps in this component, Step 1: extraction of the drug treated gene expression profiles with genes from the identified driver networks only from the NIH LINCS data; Step 2: to use a Bayesian factor regression approach to factorize the treatment profiles into weight matrices and effect matrices; Step 3: generation of the driver network signatures for each of the drugs; Step 4: evaluation of the synergistic targeting effects of drug combinations on alternative driver signaling networks; Step 5: ranking of the drug combinations according to their quantitative synergistic effects
Fig. 2.
Fig. 2.
(A) Comparison results of different drug combination prediction results in terms of probability concordance index (PCI) on NCI-dream drug combination prediction challenge dataset. (B) The estimated P-value results of different drug combination prediction results for PCI
Fig. 3.
Fig. 3.
Comparison results on the number of top ranked predictions of drug combinations which are synergistic drug combos in vitro on LNCaP and PC3 cells
Fig. 4.
Fig. 4.
(A) Comparison results of the consistency between drug combination prediction results and the in vitro drug combination screening assay results on LNCaP cells; (B) comparison results of the consistency between drug combination prediction results and the in vitro drug combination screening assay results on PC3 cells
Fig. 5.
Fig. 5.
The whole driver PC3-represented PCa signaling pathway. Big nodes are the targets of synergistic drug combo, Elesclomol and Vincristine. The green nodes and red nodes are down and up regulated genes, respectively, and the yellow nodes are genes with non-significant differential expressions

References

    1. Aalinkeel R. (2011) Overexpression of MMP-9 contributes to invasiveness of prostate cancer cell line LNCaP. Immunol. Invest., 40, 447–464.
    1. Al-Lazikani B., et al. (2012) Combinatorial drug therapy for cancer in the post-genomic era. Nat. Biotech., 30, 679–692.
    1. Bansal M., et al. (2014) A community computational challenge to predict the activity of pairs of compounds. Nat. Biotechnol., 32, 1213.
    1. Bayarri M.J., et al. (2003) Bayesian factor regression models in the “Large p, Small n” paradigm, Bayesian Statistics, 7, 733–742.
    1. Beltran H., et al. (2016) Divergent clonal evolution of castration resistant neuroendocrine prostate cancer. Nat. Med., 22, 298–305.
    1. Bernards R. (2012) A missing link in genotype-directed cancer therapy. Cell, 151, 465–468.
    1. Bivona T.G., et al. (2011) FAS and NF-kappaB signalling modulate dependence of lung cancers on mutant EGFR. Nature, 471, 523–526.
    1. Carvalho C.M., et al. (2008) High-dimensional sparse factor modeling: applications in gene expression genomics. J. Am. Stat. Assoc., 103, 1438–1456.
    1. Chan B.A., Hughes B.G.M. (2014) Targeted therapy for non-small cell lung cancer: current standards and the promise of the future. Transl. Lung Cancer Res., 4, 36–54.
    1. Chen X., et al. (2002) TTD: therapeutic target database. Nucleic Acids Res., 30, 412–415.
    1. Choi Y.L., et al. (2010) EML4-ALK mutations in lung cancer that confer resistance to ALK inhibitors. N. Engl. J. Med., 363, 1734–1739.
    1. Engelman J.A., Janne P.A. (2008) Mechanisms of acquired resistance to epidermal growth factor receptor tyrosine kinase inhibitors in non-small cell lung cancer. Clin. Cancer Res., 14, 2895–2899.
    1. Frey B.J., Dueck D. (2007) Clustering by passing messages between data points. Science, 315, 972–976.
    1. Green M.R. (2004) Targeting targeted therapy. N. Engl. J. Med., 350, 2191–2193.
    1. Hamosh A., et al. (2005) Online Mendelian inheritance in man (OMIM), a knowledgebase of human genes and genetic disorders. Nucleic Acids Res., 33, D514–D517.
    1. Harrell Frank E., et al. (1996) Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat. Med., 15, 361–387.
    1. Hirsh V. (2015) Editorial: the treatment of metastatic non-small cell lung cancer (NSCLC) in a new era of personalised medicine. Front. Oncol., 5, 5–6.
    1. Huang H., et al. (2014) Systematic prediction of drug combinations based on clinical side-effects. Sci. Rep., 4, 7160.
    1. Iadevaia S., et al. (2010) Identification of optimal drug combinations targeting cellular networks: integrating phospho-proteomics and computational network analysis. Cancer Res., 70, 6704.
    1. Jia J., et al. (2009) Mechanisms of drug combinations: interaction and network perspectives. Nat. Rev. Drug Discov., 8, 111–128.
    1. Knol M.J., et al. (2011) Estimating measures of interaction on an additive scale for preventive exposures. Eur. J. Epidemiol., 26, 433–438.
    1. Kuhn M., et al. (2012) STITCH 3: zooming in on protein-chemical interactions. Nucleic Acids Res., 40, D876–D880.
    1. Law V., et al. (2014) DrugBank 4.0: shedding new light on drug metabolism. Nucleic Acids Res., 42, D1091–D1097.
    1. Lawrence M.S., et al. (2013) Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature, 499, 214–218.
    1. Lee J.-H., et al. (2012) CDA: combinatorial drug discovery using transcriptional response modules. PLoS One, 7, e42573.
    1. Little A.S., et al. (2011) Amplification of the Driving Oncogene, KRAS or BRAF, Underpins Acquired Resistance to MEK1/2 Inhibitors in Colorectal Cancer Cells.
    1. Liu Y., et al. (2010) DCDB: drug combination database. Bioinformatics, 26, 587–588.
    1. London C.A., et al. (2003) A novel antisense inhibitor of MMP-9 attenuates angiogenesis, human prostate cancer cell invasion and tumorigenicity. Cancer Gene Ther., 10, 823–832.
    1. Mermel C., et al. (2011) GISTIC2.0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancers. Genome Biol., 12, R41.
    1. Mordelet F., Vert J.-P. (2008) SIRENE: supervised inference of regulatory networks. Bioinformatics, 24, i76–i82.
    1. Nelander S., et al. (2008) Models from experiments: combinatorial drug perturbations of cancer cells. Mol. Syst. Biol., 4, 216.
    1. Podolsky S.H., Greene J.A. (2011) Combination drugs—hype, harm, and hope. N. Engl. J. Med., 365, 488–491.
    1. Polyak K., Garber J. (2011) Targeting the missing links for cancer therapy. Nat. Med., 17, 283–284.
    1. Poulikakos P.I., et al. (2011) RAF inhibitor resistance is mediated by dimerization of aberrantly spliced BRAF(V600E). Nature, 480, 387–390.
    1. Preuer K., et al. (2018) DeepSynergy: predicting anti-cancer drug synergy with deep learning. Bioinformatics, 34, 1538–1546.
    1. Shi H., et al. (2012) Melanoma whole-exome sequencing identifies V600EB-RAF amplification-mediated acquired B-RAF inhibitor resistance. Nat. Commun., 3, 724.
    1. Tang J., et al. (2013) Target inhibition networks: predicting selective combinations of druggable targets to block cancer survival pathways. PLoS Comput. Biol., 9, e1003226.
    1. Turke A.B., et al. (2010) Preexistence and clonal selection of MET amplification in EGFR mutant NSCLC. Cancer Cell, 17, 77–88.
    1. Yamaguchi H., et al. (2014) Signaling cross-talk in the resistance to HER family receptor targeted therapy. Oncogene, 33, 1073–1081.
    1. Yang J., et al. (2015) DIGRE: drug-induced genomic residual effect model for successful prediction of multidrug effects. CPT: Pharmacometrics Syst. Pharmacol., 4, e1.
    1. Zhao J., et al. (2014) Predicting cooperative drug effects through the quantitative cellular profiling of response to individual drugs. CPT: Pharmacometrics Syst. Pharmacol., 3, e102.
    1. Zhao X.-M., et al. (2011) Prediction of drug combinations by integrating molecular and pharmacological data. PLoS Comput. Biol., 7, e1002323.
    1. Zhou T., et al. (2010) Solving the apparent diversity-accuracy dilemma of recommender systems. Proc. Natl. Acad. Sci. USA., 107, 4511–4515.

Source: PubMed

Patrocinadores y Colaboradores

Condiciones médicas

Intervenciones de drogas

3
Suscribir