Multi-scale probing, gene network modeling and drug screen 多尺度探测、基因网络建模与药物筛选 熊江辉 2009.12 Click to get the full text paper (PLoS ONE): http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0013937 http://cn.linkedin.com/in/jianghuixiong http://www.researchgate.net/profile/JIANGHUI_XIONG/

How to handle complexity? Probing Components Scales/resolution (zoom in/zoom out) Angles/perspective Integrating Modeling A holistic understanding

Medicine also need “three step paradigm”: probing->integrating->modeling Systems pharmacology and genome medicine: a future perspective. Genome Medicine 2009

-> Oncology drug development : One of most challenging scientific problems What’s wrong with our cancer models? NATURE REVIEWS DRUG DISCOVERY, 2005

Disease models “the current models used for drug testing do not accurately predict how new treatments will fare in clinical trials” Heterogeneity in patient populations Unpredictable physiology “Better predictive power to identify patients that may benefit from specific therapies or that may develop potential drug-induced toxicities Modeling heterogeneous patient populations Modeling physiology behavior of drug in vivo Drug Discov Today Dis Models, 2008

What’s wrong with our cancer models? NATURE REVIEWS DRUG DISCOVERY, 2005

Drug Discov Today Dis Models, 2008

A $1.5 billion, 10-year project Nature 455, 1061-1068 (23 October 2008) Heterogeneity revealed from TCGA project Creating animal models to mirror the natural distribution of mutations ? -- Very hard!

Unpredictable physiology What’s wrong with our cancer models? NATURE REVIEWS DRUG DISCOVERY, 2005

Our solution Pre-clinical in silico <cancer models> for drug prioritization Based on prognosis-guided gene module network Incorporating heterogeneity and in vivo physiology information Hypothesis An intrinsic gene interaction network which affects the clinical outcome in native populations could be established This network could serve as reference for identifying a signature perturbation pattern for drug candidates

What’s benefit of prognosis data? Native population heterogeneity Tumor tissue microenvironment Final point phenotype <survival time> Comprehensive genomic characterization Large data set

Gene modules are robust/reproducible features

How to reflect action pattern of drug candidate?

NCI 60 in vitro screen

How to make the baseline map? “Two genes are synthetic lethal if mutation of either alone is compatible with viability but mutation of both leads to death” NATURE REVIEWS CANCER, 5,2005 “synthetic outcome determination”

How to establish module-module cooperation network Query For each gene included in the query modules, we scanned its synergistic gene partners to determining the prognosis outcome Enrich The over-represented gene modules in above synergy gene list were identified by a Hypergeometric distribution

Hypothesis To disrupt/perturb cancer network, the key to success is to simultaneously perturbs the corresponding gatekeeper modules which cooperatively determine the outcome with the former Gatekeeper module is more important than checkpoint module How to quantify effectness of drug candidate? Hi -- the number of hits by compound c Li -- the active links ( i.e. links in which both source node and target node are matched by compound c) N -- the number of gatekeeper modules Gatekeeper Checkpoint

How to quantify synergistic effect of drug combination? Pool together Calculate the perturbation index

Cancer types

Results

Cisplatin on lung cancer

‘Gatekeeper’ modules for lung cancer

Biological function characterization of inter-module cooperation network

Contribution of various evidence sources for gene module definition

Bootstrapping-based assessment of perturbation index on discriminating successful drugs from the candidate pool

Rank of drugs and agents in clinical development for lung cancer according to their perturbation index

The perturbation index of pair-wise combination of lung cancer agents Validity of Bortezomib-Gemcitabine Notable survival benefits in lung cancer patients using a Bortezomib + gemcitabine/carboplatin combination as first-line treatment (phase II clinical trial reported) Davies, A.M. et al. J Thorac Oncol 4, 87-92 (2009) Validity of Bortezomib-Paclitaxel In an RNA interference (RNAi)-based synthetic lethal screen for seeking paclitaxel chemosensitizer genes in human NSCLC cell line, proteasome is the most enriched gene group Whitehurst, A.W. et al. Nature 446, 815-819 (2007)

Bortezomib-Gemcitabine

Gemcitabine -> baseline perturbation Bortezomib -> add a focused perturbation on key gatekeeper modules

Discussion (1) Unique characteristics as a preclinical in silico modeling tool Mirroring drug behavior on native populations Cost-effectiveness Easy to integrate drug action mechanisms/patterns “Library of lense” to probe the intrinsic gene network Resolution Specificity Multi-faced coverage

Discussion (2) - Strategy against cancer Gatekeeper modules as rate-limiting steps in therapeutic treatment Drug metabolism and accessibility Microenvironment immune system modulation Epigenetic plasticity on gatekeeper modules could exploited by tumor for attaining resistance to treatment drug accessibility <- Multi Drug Resistance Microenvironment <- Inflammatory immune modulation <- Complement activation Battle against cancer know the history of tumorigenesis <etiology> know future survival strategy of tumor under therapeutic interventions Systems biology modeling could provide prediction of the tumor survival strategy Etiology-based strategy Prediction-based strategy The next generation therapeutic strategy

Discussion (3) – Traditional Chinese Medicine 君 - King 臣 - Minister 佐 - Assistant 使 - ambassador The inter-module network - a powerful tool to understanding principle of drug combination

“我们所观测到的不是自然本身，而是自然根据我们探索它的方法的展现”