DIFFERENTIAL PATHWAY DEPENDENCY DISCOVERY ASSOCIATED WITH DRUG RESPONSE ACROSS CANCER CELL LINES

The effort to personalize treatment plans for cancer patients involves the identification of drug treatments that can effectively target the disease while minimizing the likelihood of adverse reactions. In this study, the gene-expression profile of 810 cancer cell lines and their response data to 368 small molecules from the Cancer Therapeutics Research Portal (CTRP) are analyzed to identify pathways with significant rewiring between genes, or differential gene dependency, between sensitive and non-sensitive cell lines. Identified pathways and their corresponding differential dependency networks are further analyzed to discover essentiality and specificity mediators of cell line response to drugs/compounds. For analysis we use the previously published method EDDY (Evaluation of Differential DependencY). EDDY first constructs likelihood distributions of gene-dependency networks, aided by known genegene interaction, for two given conditions, for example, sensitive cell lines vs. non-sensitive cell lines. These sets of networks yield a divergence value between two distributions of network likelihoods that can be assessed for significance using permutation tests. Resulting differential dependency networks are then further analyzed to identify genes, termed mediators, which may play important roles in biological signaling in certain cell lines that are sensitive or non-sensitive to the drugs. Establishing statistical correspondence between compounds and mediators can improve understanding of known gene dependencies associated with drug response while also discovering new dependencies. Millions of compute hours resulted in thousands of these statistical discoveries. EDDY identified 8,811 statistically significant pathways leading to 26,822 compound-pathway-mediator triplets. By incorporating STITCH and STRING databases, we could construct evidence networks for 14,415 compound-pathway-mediator triplets for support. The results of this analysis are presented in a searchable website to aid researchers in studying potential molecular mechanisms underlying cells' drug response as well as in designing experiments for the purpose of personalized treatment regimens.

KNOWLEDGE-ASSISTED APPROACH TO IDENTIFY PATHWAYS WITH DIFFERENTIAL DEPENDENCIES

We have previously developed a statistical method to identify gene sets enriched with condition-specific genetic dependencies. The method constructs gene dependency networks from bootstrapped samples in one condition and computes the divergence between distributions of network likelihood scores from different conditions. It was shown to be capable of sensitive and specific identification of pathways with phenotype-specific dysregulation, i.e., rewiring of dependencies between genes in different conditions. We now present an extension of the method by incorporating prior knowledge into the inference of networks. The degree of prior knowledge incorporation has substantial effect on the sensitivity of the method, as the data is the source of condition specificity while prior knowledge incorporation can provide additional support for dependencies that are only partially supported by the data. Use of prior knowledge also significantly improved the interpretability of the results. Further analysis of topological characteristics of gene differential dependency networks provides a new approach to identify genes that could play important roles in biological signaling in a specific condition, hence, promising targets customized to a specific condition. Through analysis of TCGA glioblastoma multiforme data, we demonstrate the method can identify not only potentially promising targets but also underlying biology for new targets.

Transcriptional profiling of GBM invasion genes identifies effective inhibitors of the LIM kinase-Cofilin pathway

Malignant gliomas are highly proliferative and invasive neoplasms where total surgical resection is often impossible and effective local radiation therapy difficult. Consequently, there is a need to develop a greater understanding of the molecular events driving invasion and to identify novel treatment targets. Using microarray analysis comparing normal brain samples and mesenchymal glioblastoma multiforme (GBM), we identified over 140 significant genes involved in cell migration and invasion. The cofilin (CFL) pathway, which disassembles actin filaments, was highly up-regulated compared to normal brain. Up-regulation of LIM domain kinase 1 and 2 (LIMK1/2), that phosphorylates and inactivates cofilin, was confirmed in an additional independent data set comparing normal brain to GBM. We identified and utilized two small molecule inhibitors BMS-5 and Cucurbitacin I directed against the cofilin regulating kinases, LIMK1 and LIMK2, to target this pathway. Significant decreases in cell viability were observed in glioma cells treated with BMS-5 and Cucurbitacin I, while no cytotoxic effects were seen in normal astrocytes that lack LIMK. BMS-5 and Cucurbitacin I promoted increased adhesion in GBM cells, and decreased migration and invasion. Collectively, these data suggest that use of LIMK inhibitors may provide a novel way to target the invasive machinery in GBM.

Abstract 4671: Identification of novel drugs for glioblastoma using chemical biology fingerprinting

Glioblastoma Multiforme (GBM) is an aggressive brain tumor with very poor prognosis and extremely limited therapeutic options. GBM is the most common malignant brain tumor and the search for novel targets and/ or the repurposing of already extant drugs to treat the disease is therefore of utmost importance. We describe here a comprehensive multidisciplinary approach to identifying said targets and ergo potential therapies. We have applied a novel analytical strategy to The Cancer Genome Atlas (TCGA) GBM expression data to stratify GBM into novel subtypes we call molecular contexts, or mCs. Subsequently, a panel of patient-derived GBM xenografts was ascribed to our novel mCs. Utilizing a technique we term Chemical Biology Fingerprinting, or CBF, short-term cultures derived from these clinically-relevant preclinical models were screened for chemosensitivity with a deeply annotated, yet clinically relevant, chemical library. Agents that were statistically more toxic to one context than another were then re-tested in true drug dose response experiments to confirm sensitivity. Preliminary data demonstrated that mC14, characterized by mutant p53 and transcriptionally similar to the GBM proneural subtype, showed distinct vulnerability to Arsenic Trioxide (ATO) as compared to mC4, enriched for NF1 mutations and with transcriptional patterns similar to the GBM mesenchymal subtype. To validate the ATO vulnerability signature in GBM, we acquired 20 treatment naïve archival patient samples, that were part of a Phase I/II clinical trial to study the efficacy of ATO and Temozolomide in combination with ionizing radiation (NCT00275067). Participants in the trial exhibited varied survival with ATO treatment (91 days to >1000 days) and the clinical samples were subtyped into our molecular contexts using RNAseq data.

In summary, we demonstrate a subclassification of GBM into novel molecular contexts (mCs) and show that these contexts are differentially sensitive to clinically relevant drugs.

Citation Format: Darren Finlay, Harshil Dhruv, Lisa Evers, Sen Peng, Jeff Kiefer, Seungchan Kim, Jeffrey Raizer, Michael Berens, Kristiina Vuori. Identification of novel drugs for glioblastoma using chemical biology fingerprinting. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4671. doi:10.1158/1538-7445.AM2015-4671

Identification of causal genetic drivers of human disease through systems-level analysis of regulatory networks

Identification of driver mutations in human diseases is often limited by cohort size and availability of appropriate statistical models. We propose a framework for the systematic discovery of genetic alterations that are causal determinants of disease, by prioritizing genes upstream of functional disease drivers, within regulatory networks inferred de novo from experimental data. We tested this framework by identifying the genetic determinants of the mesenchymal subtype of glioblastoma. Our analysis uncovered KLHL9 deletions as upstream activators of two previously established master regulators of the subtype, C/EBPβ and C/EBPδ. Rescue of KLHL9 expression induced proteasomal degradation of C/EBP proteins, abrogated the mesenchymal signature, and reduced tumor viability in vitro and in vivo. Deletions of KLHL9 were confirmed in > 50% of mesenchymal cases in an independent cohort, thus representing the most frequent genetic determinant of the subtype. The method generalized to study other human diseases, including breast cancer and Alzheimer's disease.

Abstract 1319: Vulnerability and resistance in glioblastoma to G2-specific drugs

Glioblastoma multiforme (GBM) is the most common and most lethal primary malignant brain tumor, affecting 25,000 patients per year. Approximately 87% of GBMs have abnormalities in the p53 pathway that affects G2 checkpoint. In the absence of p53, Wee1 kinase has been shown to regulate G2 checkpoint upon chemotherapeutic insult. Wee1 inhibits CDC2 function, thereby promoting G2 phase cell cycle arrest, thus preventing damaged cells from entering into premature mitosis without repairing the DNA. We hypothesize that in p53-deficient tumor cells, inhibitors of Wee1 would be effective to heighten DNA damage due to genetic and pharmacologic loss of both the G1 and G2 checkpoints, respectively. MK-1775 (Merck) is a potent and selective small molecule inhibitor of the Wee1 kinase that has demonstrated enhanced cytotoxic effect in combination with several chemotherapeutic agents in preclinical studies in vitro and in vivo. Here, we selected an array of short term cultures of glioma patient derived xenograft (PDX) models (GBM12, GBM43, GBM79, GBM84) and long term established glioma cell lines (U87, SF767, T98G) with aberrant and wild type G1 checkpoint pathways and investigated cytotoxic effects of MK-1775 as a single agent and in combination with Temozolomide (TMZ). Our results demonstrated differential sensitivity to MK-1775 and TMZ as single agents for different glioma cell lines in vitro. However, no correlation between MK-1775, TMZ sensitivity and p53 status of the cell lines was observed. Combination therapy with MK-1775 and TMZ showed a synergistic effect for one PDX model, GBM43, with combination index (CI) of 0.4. Combination therapy of all other glioma cell lines either showed an additive effect or antagonistic effect with CI of >0.75 irrespective of their p53 status. Our results demonstrated that synergistic effects of MK-1775 in combination with chemotherapy (TMZ) are not solely dependent on p53 status of glioma cells and require a more comprehensivce understanding of the molecular mechanism that dictates GBM's dependence on Wee1 for escaping DNA damage therapy. Supported by the Ben & Catherine Ivy Foundation

Citation Format: Harshil D. Dhruv, Andrew D. Nelson, Brock Armstrong, Julie L. Boerner, Jann N. Sarkaria, Nhan L. Tran, Michael Berens. Vulnerability and resistance in glioblastoma to G2-specific drugs. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 1319. doi:10.1158/1538-7445.AM2014-1319

Abstract 1780: Identification and characterization of L524-0366 as a small molecule inhibitor that disrupt TWEAK-Fn14 signaling in glioblastoma

Deregulation of the TWEAK-Fn14 signaling pathway is observed in many diseases including inflammation, autoimmune diseases, and cancer, notably glioblastoma. Activation of Fn14 signaling by TWEAK binding triggers glioma cell invasion and survival and therefore represents an attractive pathway for therapeutic intervention. Based on structural studies of the TWEAK-binding cysteine rich domain of Fn14, several homology models of TWEAK were built to investigate plausible modes of TWEAK-Fn14 interaction. Two promising models, centered on different anchoring residues of TWEAK [Tyrosine 176 (Y176) and Tryptophan 231 (W231)], were prioritized using a data-driven strategy. Site-directed mutagenesis of TWEAK at Y176, but not W231, resulted in the loss of TWEAK binding to Fn14 substantiating Y176 as the anchoring residue. Importantly, mutation of TWEAK at Y176 did not disrupt TWEAK trimerization, but failed to induce Fn14 mediated nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling. The validated structural models were utilized in a virtual screen to design a targeted library of small molecules predicted to disrupt the TWEAK-Fn14 interaction. 129 small molecules were screened iteratively, with identification of molecules producing up to 37% inhibition of TWEAK-Fn14 binding. One specific compound, L524-0366 binds specifically to the Fn14 receptor and disrupt TWEAK-Fn14 interaction. Treatment of L524-0366 suppresses TWEAK-induced cell migration and survival in glioblastoma cells. In summary, we present a data-driven in silico study revealing key structural elements of the TWEAK-Fn14 interaction, followed by experimental validation, serving as a guide for the design of small molecule inhibitors of the TWEAK-Fn14 ligand-receptor interaction. Our results provide the foundation for further exploration utilizing L524-0366 as a small molecule inhibitor to the TWEAK-Fn14 signaling axis.

Citation Format: Harshil D. Dhruv, Joseph C. Loftus, Pooja Narang, Joachim L. Petit, Donald Chow, Holly Yin, Michael Berens, Nathalie Meurice, Nhan L. Tran. Identification and characterization of L524-0366 as a small molecule inhibitor that disrupt TWEAK-Fn14 signaling in glioblastoma. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 1780. doi:10.1158/1538-7445.AM2014-1780

Designing optical free-form surfaces for extended sources

LED lighting has been a strongly growing field for the last decade. The outstanding features of LED, like compactness and low operating temperature take the control of light distributions to a new level. Key for this is the development of sophisticated optical elements that distribute the light as intended. The optics design method known as tailoring relies on the point source assumption. This assumption holds as long as the optical element is large compared to the LED chip. With chip sizes of 1 mm² this is of no concern if each chip is endowed with its own optic. To increase the power of a luminaire, LED chips are arranged to form light engines that reach several cm in diameter. In order to save costs and space it is often desirable to use a single optical element for the light engine. At the same time the scale of the optics must not be increased in order to trivially keep the point source assumption valid. For such design tasks point source algorithms are of limited usefulness. New methods that take into account the extent of the light source have to be developed. We present two such extended source methods. The first method iteratively adapts the target light distribution that is fed into a points source method while the second method employs a full phase space description of the optical system.

Abstract C223: Network-based approach aids in the discovery of context-specific druggable targets for treatment of glioblastoma

Systematic discovery of actionable cancer targets could fill an unmet need for improved approaches to manage Glioblastoma Multiforme (GBM). Critical barriers to the discovery of druggable targets in cancer include: 1) molecular heterogeneity of the disease across patients, 2) implementation of a systematic approach to utilize existing knowledge and molecular data for target discovery, and 3) the lack of a relevant, rapid and systematic pipeline to enable empirical testing of hypotheses. We describe a network-based method to stratify GBM clinical samples into molecularly homogeneous subsets based on gene expression, gene copy number, and miRNAs, which we term “molecular contexts," or mCs. Within these mCs, a knowledge-based topological analysis of pathway elements was used to uncover discrete mC-specific targets. An empiric chemical screen probed the target space across different mCs, mC-4 and mC-14, which were molecularly orthogonal to each other. Specimens in mC-4 were enriched with samples previously classified as Mesenchymal-type GBM, while mC-14 was enriched with Proneural-subtype GBM. The chemical screen was carried out using short-term in vitro cultures derived from patient-derived GBM xenografts, which mapped to mC4 and mC14 based on gene expression. We employed a network-based topological approach to discern targets and pathways as candidate druggable vulnerabilities. The chemical validation screen was carried out with an assembled chemical biology fingerprint (CBF) library comprised of 650 small molecules targeting multiple cancer-associated pathways and processes. Matching specific chemical hits to their respective targets allowed for validation of specific gene hits from the topological analysis. Of particular interest were two molecular context specific lethal compounds in the screen, tamoxifen citrate and arsenic trioxide, specific to the mC4 and mC14, respectively. PKC, among the targets of Tamoxifen citrate, was a predicted node of vulnerability for mC-4 GBM, while PML gene, a target of Arsenic Trioxide, was a predicted target for mC-14. In summary, our results suggest that context analysis coupled with knowledge-based enrichment and topological analysis identifies specific GBM contexts with novel unique drugable targets. Supported by NIH U01 CA168397.

Citation Information: Mol Cancer Ther 2013;12(11 Suppl):C223.

Citation Format: Harshil D. Dhruv, Seungchan Kim, Jeff Kiefer, Dorothea Emig-Agius, Darren Finlay, Sungwon Jung, Kristiina Vuori, Michael Berens. Network-based approach aids in the discovery of context-specific druggable targets for treatment of glioblastoma. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr C223.