Chemosensitivity prediction by transcriptional profiling

Comprehensive anticancer drug response prediction based on a simple cell line-drug complex network model

Background

Accurate prediction of anticancer drug responses in cell lines is a crucial step to accomplish the precision medicine in oncology. Although many popular computational models have been proposed towards this non-trivial issue, there is still room for improving the prediction performance by combining multiple types of genome-wide molecular data.

Results

We first demonstrated an observation on the CCLE and GDSC datasets, i.e., genetically similar cell lines always exhibit higher response correlations to structurally related drugs. Based on this observation we built a cell line-drug complex network model, named CDCN model. It captures different contributions of all available cell line-drug responses through cell line similarities and drug similarities. We executed anticancer drug response prediction on CCLE and GDSC independently. The result is significantly superior to that of some existing studies. More importantly, our model could predict the response of new drug to new cell line with considerable performance. We also divided all possible cell lines into “sensitive” and “resistant” groups by their response values to a given drug, the prediction accuracy, sensitivity, specificity and goodness of fit are also very promising.

Conclusion

CDCN model is a comprehensive tool to predict anticancer drug responses. Compared with existing methods, it is able to provide more satisfactory prediction results with less computational consumption.

Electronic supplementary material

The online version of this article (10.1186/s12859-019-2608-9) contains supplementary material, which is available to authorized users.

A novel gene selection algorithm for cancer classification using microarray datasets

Background

Microarray datasets are an important medical diagnostic tool as they represent the states of a cell at the molecular level. Available microarray datasets for classifying cancer types generally have a fairly small sample size compared to the large number of genes involved. This fact is known as a curse of dimensionality, which is a challenging problem. Gene selection is a promising approach that addresses this problem and plays an important role in the development of efficient cancer classification due to the fact that only a small number of genes are related to the classification problem. Gene selection addresses many problems in microarray datasets such as reducing the number of irrelevant and noisy genes, and selecting the most related genes to improve the classification results.

Methods

An innovative Gene Selection Programming (GSP) method is proposed to select relevant genes for effective and efficient cancer classification. GSP is based on Gene Expression Programming (GEP) method with a new defined population initialization algorithm, a new fitness function definition, and improved mutation and recombination operators. . Support Vector Machine (SVM) with a linear kernel serves as a classifier of the GSP.

Results

Experimental results on ten microarray cancer datasets demonstrate that Gene Selection Programming (GSP) is effective and efficient in eliminating irrelevant and redundant genes/features from microarray datasets. The comprehensive evaluations and comparisons with other methods show that GSP gives a better compromise in terms of all three evaluation criteria, i.e., classification accuracy, number of selected genes, and computational cost. The gene set selected by GSP has shown its superior performances in cancer classification compared to those selected by the up-to-date representative gene selection methods.

Conclusion

Gene subset selected by GSP can achieve a higher classification accuracy with less processing time.

A Review of Microarray Datasets: Where to Find Them and Specific Characteristics

The advent of DNA microarray datasets has stimulated a new line of research both in bioinformatics and in machine learning. This type of data is used to collect information from tissue and cell samples regarding gene expression differences that could be useful for disease diagnosis or for distinguishing specific types of tumor. Microarray data classification is a difficult challenge for machine learning researchers due to its high number of features and the small sample sizes. This chapter is devoted to reviewing the microarray databases most frequently used in the literature. We also make the interested reader aware of the problematic of data characteristics in this domain, such as the imbalance of the data, their complexity, and the so-called dataset shift.

Circulating Tumor and Invasive Cell Gene Expression Profile Predicts Treatment Response and Survival in Pancreatic Adenocarcinoma

Previous studies have shown that pharmacogenomic modeling of circulating tumor and invasive cells (CTICs) can predict response of pancreatic ductal adenocarcinoma (PDAC) to combination chemotherapy, predominantly 5-fluorouracil-based. We hypothesized that a similar approach could be developed to predict treatment response to standard frontline gemcitabine with nab-paclitaxel (G/nab-P) chemotherapy. Gene expression profiles for responsiveness to G/nab-P were determined in cell lines and a test set of patient samples. A prospective clinical trial was conducted, enrolling 37 patients with advanced PDAC who received G/nab-P. Peripheral blood was collected prior to treatment, after two months of treatment, and at progression. The CTICs were isolated based on a phenotype of collagen invasion. The RNA was isolated, cDNA synthesized, and qPCR gene expression analyzed. Patients were most closely matched to one of three chemotherapy response templates. Circulating tumor and invasive cells' SMAD4 expression was measured serially. The CTICs were reliably isolated and profiled from peripheral blood prior to and during chemotherapy treatment. Individual patients could be matched to distinct response templates predicting differential responses to G/nab-P treatment. Progression free survival was significantly correlated to response prediction and ΔSMAD4 was significantly associated with disease progression. These findings support phenotypic profiling and ΔSMAD4 of CTICs as promising clinical tools for choosing effective therapy in advanced PDAC, and for anticipating disease progression.

Anti-cancer Drug Response Prediction Using Neighbor-Based Collaborative Filtering with Global Effect Removal

Patients of the same cancer may differ in their responses to a specific medical therapy. Identification of predictive molecular features for drug sensitivity holds the key in the era of precision medicine. Human cell lines have harbored most of the same genetic changes found in patients’ tumors and thus are widely used in the research of drug response. In this work, we formulated drug-response prediction as a recommender system problem and then adopted a neighbor-based collaborative filtering with global effect removal (NCFGER) method to estimate anti-cancer drug responses of cell lines by integrating cell-line similarity networks and drug similarity networks based on the fact that similar cell lines and similar drugs exhibit similar responses. Specifically, we removed the global effect in the available responses and shrunk the similarity score for each cell line pair as well as each drug pair. We then used the K most similar neighbors (hybrid of cell-line-oriented and drug-oriented) in the available responses to predict the unknown ones. Through 10-fold cross-validation, this approach was shown to reach accurate and reproducible outcomes of drug sensitivity. We also discussed the biological outcomes based on the newly predicted response values.

A Hybrid Interpolation Weighted Collaborative Filtering Method for Anti-cancer Drug Response Prediction

Individualized therapies ask for the most effective regimen for each patient, while the patients' response may differ from each other. However, it is impossible to clinically evaluate each patient's response due to the large population. Human cell lines have harbored most of the same genetic changes found in patients' tumors, thus are widely used to help understand initial responses of drugs. Based on the more credible assumption that similar cell lines and similar drugs exhibit similar responses, we formulated drug response prediction as a recommender system problem, and then adopted a hybrid interpolation weighted collaborative filtering (HIWCF) method to predict anti-cancer drug responses of cell lines by incorporating cell line similarity and drug similarity shown from gene expression profiles, drug chemical structure as well as drug response similarity. Specifically, we estimated the baseline based on the available responses and shrunk the similarity score for each cell line pair as well as each drug pair. The similarity scores were then shrunk and weighted by the correlation coefficients drawn from the know response between each pair. Before used to find the K most similar neighbors for further prediction, they went through the case amplification strategy to emphasize high similarity and neglect low similarity. In the last step for prediction, cell line-oriented and drug-oriented collaborative filtering models were carried out, and the average of predicted values from both models was used as the final predicted sensitivity. Through 10-fold cross validation, this approach was shown to reach accurate and reproducible outcome for those missing drug sensitivities. We also found that the drug response similarity between cell lines or drugs may play important role in the prediction. Finally, we discussed the biological outcomes based on the newly predicted response values in GDSC dataset.

Cancer transcriptome profiling at the juncture of clinical translation

Methodological breakthroughs over the past four decades have repeatedly revolutionized transcriptome profiling. Using RNA sequencing (RNA-seq), it has now become possible to sequence and quantify the transcriptional outputs of individual cells or thousands of samples. These transcriptomes provide a link between cellular phenotypes and their molecular underpinnings, such as mutations. In the context of cancer, this link represents an opportunity to dissect the complexity and heterogeneity of tumours and to discover new biomarkers or therapeutic strategies. Here, we review the rationale, methodology and translational impact of transcriptome profiling in cancer.

A 17-molecule set as a predictor of complete response to neoadjuvant chemotherapy with docetaxel, cisplatin, and 5-fluorouracil in esophageal cancer

Background

Recently, neoadjuvant chemotherapy with docetaxel/cisplatin/5-fluorouracil (NAC-DCF) was identified as a novel strong regimen with a high rate of pathological complete response (pCR) in advanced esophageal cancer in Japan. Predicting pCR will contribute to the therapeutic strategy and the prevention of surgical invasion. However, a predictor of pCR after NAC-DCF has not yet been developed. The aim of this study was to identify a novel predictor of pCR in locally advanced esophageal cancer treated with NAC-DCF.

Patients and methods

A total of 32 patients who received NAC-DCF followed by esophagectomy between June 2013 and March 2016 were enrolled in this study. We divided the patients into the following 2 groups: pCR group (9 cases) and non-pCR group (23 cases), and compared gene expressions between these groups using DNA microarray data and KeyMolnet. Subsequently, a validation study of candidate molecular expression was performed in 7 additional cases.

Results

Seventeen molecules, including transcription factor E2F, T-cell-specific transcription factor, Src (known as “proto-oncogene tyrosine-protein kinase of sarcoma”), interferon regulatory factor 1, thymidylate synthase, cyclin B, cyclin-dependent kinase (CDK) 4, CDK, caspase-1, vitamin D receptor, histone deacetylase, MAPK/ERK kinase, bcl-2-associated X protein, runt-related transcription factor 1, PR domain zinc finger protein 1, platelet-derived growth factor receptor, and interleukin 1, were identified as candidate molecules. The molecules were mainly associated with pathways, such as transcriptional regulation by SMAD, RB/E2F, and STAT. The validation study indicated that 12 of the 17 molecules (71%) matched the trends of molecular expression.

Conclusions

A 17-molecule set that predicts pCR after NAC-DCF for locally advanced esophageal cancer was identified.

Development of the Theta Comparative Cell Scoring Method to Quantify Diverse Phenotypic Responses Between Distinct Cell Types

In this article, we have developed novel data visualization tools and a Theta comparative cell scoring (TCCS) method, which supports high-throughput in vitro pharmacogenomic studies across diverse cellular phenotypes measured by multiparametric high-content analysis. The TCCS method provides a univariate descriptor of divergent compound-induced phenotypic responses between distinct cell types, which can be used for correlation with genetic, epigenetic, and proteomic datasets to support the identification of biomarkers and further elucidate drug mechanism-of-action. Application of these methods to compound profiling across high-content assays incorporating well-characterized cells representing known molecular subtypes of disease supports the development of personalized healthcare strategies without prior knowledge of a drug target. We present proof-of-principle data quantifying distinct phenotypic response between eight breast cancer cells representing four disease subclasses. Application of the TCCS method together with new advances in next-generation sequencing, induced pluripotent stem cell technology, gene editing, and high-content phenotypic screening are well placed to advance the identification of predictive biomarkers and personalized medicine approaches across a broader range of disease types and therapeutic classes.

Over-expression of BAG-1 in head and neck squamous cell carcinomas (HNSCC) is associated with cisplatin-resistance

Background

In order to improve therapy for head and neck squamous cell carcinoma (HNSCC), biomarkers associated with local and/or distant tumor relapses and cancer drug resistance are urgently needed. This study identified a potential biomarker, Bcl-2 associated athanogene-1 (BAG-1), that is implicated in HNSCC insensitive to cisplatin and tumor progression.

Methods

Primary and advanced (relapsed from parental) University of Michigan squamous cell carcinoma cell lines were tested for sensitivity to cisplatin and gene expression profiles were compared between primary (cisplatin sensitive) and the relapsed (cisplatin resistant) cell lines by using Agilent microarrays. Additionally, differentially expressed genes phosphorylated AKT, and BAG-1, and BCL-xL were evaluated for expression using HNSCC tissue arrays.

Results

Advanced HNSCC cells revealed resistant to cisplatin accompanied by increased expression of BAG-1 protein. siRNA knockdown of BAG-1 expression resulted in significant improvement of HNSCC sensitivity to cisplatin. BAG-1 expression enhanced stability of BCL-xL and conferred cisplatin resistant to the HNSCC cells. In addition, high levels of expression of phosphorylated AKT, BAG-1, and BCL-xL were observed in advanced HNSCC compared to in that of primary HNSCC.

Conclusion

Increased expression of BAG-1 was associated with cisplatin resistance and tumor progression in HNSCC patients and warrants further validation in larger independent studies. Over expression of BAG-1 may be a biomarker for cisplatin resistance in patients with primary or recurrent HNSCCs and targeting BAG-1 could be helpful in overcoming cisplatin resistance.

Electronic supplementary material

The online version of this article (doi:10.1186/s12967-017-1289-2) contains supplementary material, which is available to authorized users.

Accelerating glioblastoma drug discovery: Convergence of patient-derived models, genome editing and phenotypic screening

Patients diagnosed with glioblastoma (GBM) continue to face a bleak prognosis. It is critical that new effective therapeutic strategies are developed. GBM stem cells have molecular hallmarks of neural stem and progenitor cells and it is possible to propagate both non-transformed normal neural stem cells and GBM stem cells, in defined, feeder-free, adherent culture. These primary stem cell lines provide an experimental model that is ideally suited to cell-based drug discovery or genetic screens in order to identify tumour-specific vulnerabilities. For many solid tumours, including GBM, the genetic disruptions that drive tumour initiation and growth have now been catalogued. CRISPR/Cas-based genome editing technologies have recently emerged, transforming our ability to functionally annotate the human genome. Genome editing opens prospects for engineering precise genetic changes in normal and GBM-derived neural stem cells, which will provide more defined and reliable genetic models, with critical matched pairs of isogenic cell lines. Generation of more complex alleles such as knock in tags or fluorescent reporters is also now possible. These new cellular models can be deployed in cell-based phenotypic drug discovery (PDD). Here we discuss the convergence of these advanced technologies (iPS cells, neural stem cell culture, genome editing and high content phenotypic screening) and how they herald a new era in human cellular genetics that should have a major impact in accelerating glioblastoma drug discovery.

Drug Repurposing of the Anthelmintic Niclosamide to Treat Multidrug-Resistant Leukemia

Multidrug resistance, a major problem that leads to failure of anticancer chemotherapy, requires the development of new drugs. Repurposing of established drugs is a promising approach for overcoming this problem. An example of such drugs is niclosamide, a known anthelmintic that is now known to be cytotoxic and cytostatic against cancer cells. In this study, niclosamide showed varying activity against different cancer cell lines. It revealed better activity against hematological cancer cell lines CCRF-CEM, CEM/ADR5000, and RPMI-8226 compared to the solid tumor cell lines MDA-MB-231, A549, and HT-29. The multidrug resistant CEM/ADR5000 cells were similar sensitive as their sensitive counterpart CCRF-CEM (resistance ration: 1.24). Furthermore, niclosamide caused elevations in reactive oxygen species and glutathione (GSH) levels in leukemia cells. GSH synthetase (GS) was predicted as a target of niclosamide. Molecular docking showed that niclosamide probably binds to the ATP-binding site of GS with a binding energy of -9.40 kcal/mol. Using microscale thermophoresis, the binding affinity between niclosamide and recombinant human GS was measured (binding constant: 5.64 μM). COMPARE analyses of the NCI microarray database for 60 cell lines showed that several genes, including those involved in lipid metabolism, correlated with cellular responsiveness to niclosamide. Hierarchical cluster analysis showed five major branches with significant differences between sensitive and resistant cell lines (p = 8.66 × 10⁵). Niclosamide significantly decreased nuclear factor of activated T-cells (NFAT) activity as predicted by promoter binding motif analysis. In conclusion, niclosamide was more active against hematological malignancies compared to solid tumors. The drug was particularly active against the multidrug-resistant CEM/ADR5000 leukemia cells. Inhibition of GSH synthesis and NFAT signaling were identified as relevant mechanisms for the anticancer activity of niclosamide. Gene expression profiling predicted the sensitivity or resistance of cancer cells to niclosamide.

Integrating Biological Covariates into Gene Expression-Based Predictors of Radiation Sensitivity

The use of gene expression-based classifiers has resulted in a number of promising potential signatures of patient diagnosis, prognosis, and response to therapy. However, these approaches have also created difficulties in trying to use gene expression alone to predict a complex trait. A practical approach to this problem is to integrate existing biological knowledge with gene expression to build a composite predictor. We studied the problem of predicting radiation sensitivity within human cancer cell lines from gene expression. First, we present evidence for the need to integrate known biological conditions (tissue of origin, RAS, and p53 mutational status) into a gene expression prediction problem involving radiation sensitivity. Next, we demonstrate using linear regression, a technique for incorporating this knowledge. The resulting correlations between gene expression and radiation sensitivity improved through the use of this technique (best-fit adjusted R² increased from 0.3 to 0.84). Overfitting of data was examined through the use of simulation. The results reinforce the concept that radiation sensitivity is not driven solely by gene expression, but rather by a combination of distinct parameters. We show that accounting for biological heterogeneity significantly improves the ability of the model to identify genes that are associated with radiosensitivity.

A systems biology approach to identify microRNAs contributing to cisplatin resistance in human ovarian cancer cells

The dysregulated microRNAs contribute to cisplatin resistance in ovarian cancer cells.

ABCB1 2677G>T/A variant enhances chemosensitivity to anti-cancer agents acting on microtubule dynamics through LAMP1 inhibition

Overexpression of ABCB1 associated with single nucleotide variants in cancers was reported to encode a protein responsible for drug resistance. We studied chemosensitivity-related genes associated with ABCB1 2677G>T/A variant. The associated genes were identified based on the results of the significance analysis of microarray, and then prediction accuracy was evaluated using the prediction analysis of microarray. Functional assay of the selected gene was performed by using siRNA and drug accumulation study. A higher frequency of chemoresistance to microtubule-modulating agents was found in cell lines with wild-type ABCB1 compared to cell lines with 2677G>T/A ABCB1 variant. Based on the pharmacogenetic association study with 2677 variant, we identified seven genes that could predict chemosensitivity to microtubule dynamics modulators. The classification accuracy with these seven genes was 90.0%, and the predicted probability was 0.73. LAMP1 was the only gene that was commonly related to chemosensitivity. LAMP1 expression levels were relatively higher in chemoresistant ABCB1 wild-type compared to chemosensitive polymorphic cells. But, there was no difference in ABCB1 expression levels between the two groups. Following LAMP1 siRNA, chemosensitivity was restored due to increased intracellular drug accumulation in wild type cell line. In conclusion, ABCB1 2677G>T/A variant enhances chemosensitivity on microtubule dynamics through LAMP1 inhibition.

MYC Amplification as a Predictive Factor of Complete Pathologic Response to Docetaxel-based Neoadjuvant Chemotherapy for Breast Cancer

BACKGROUND

Neoadjuvant chemotherapy is a standard treatment for stage II and III breast cancer. The identification of biomarkers that may help in the prediction of response to neoadjuvant therapies is necessary for a more precise definition of the best drug or drug combination to induce a better response.

MATERIAL AND METHODS

We assessed the role of Ki67, hormone receptors expression, HER2, MYC genes and their protein status, and KRAS codon 12 mutations as predictor factors of pathologic response to anthracycline-cyclophosphamide (AC) followed by taxane docetaxel (T) neoadjuvant chemotherapy (AC+T regimen) in 51 patients with invasive ductal breast cancer.

RESULTS

After neoadjuvant chemotherapy, 82.4% of patients showed pathologic partial response, with only 9.8% showing pathologic complete response. In multivariate analysis, MYC immunoreactivity and high MYC gain defined as MYC/nucleus ≥ 5 were significant predictor factors for pathologic partial response. Using the receiver operating characteristic curve analysis, the ratio of 2.5 MYC/CEP8 (sensitivity of 80% and specificity of 89.1%) or 7 MYC/nuclei copies (sensitivity of 80% and specificity of 73.9%) as the best cutoff in predicting a pathologic complete response was identified. Thus, MYC may have a role in chemosensitivity to AC and/or docetaxel drugs. Additionally, MYC amplification may be a predictor factor of pathologic response to the AC+T regimen in patients with breast cancer. Moreover, patients with an increased number of MYC copies showed pathologic complete response to this neoadjuvant treatment more frequently.

CONCLUSION

The analysis of MYC amplification may help in the identification of patients that may have a better response to AC+T treatment.

Systematic drug perturbations on cancer cells reveal diverse exit paths from proliferative state

During a cell state transition, cells travel along trajectories in a gene expression state space. This dynamical systems framework complements the traditional concept of molecular pathways that drive cell phenotype switching. To expose the structure that hinders cancer cells from exiting robust proliferative state, we assessed the perturbation capacity of a drug library and identified 16 non-cytotoxic compounds that stimulate MCF7 breast cancer cells to exit from proliferative state to differentiated state. The transcriptome trajectories triggered by these drugs diverged, then converged. Chemical structures and drug targets of these compounds overlapped minimally. However, a network analysis of targeted pathways identified a core signaling pathway - indicating common stress-response and down-regulation of STAT1 before differentiation. This multi-trajectory analysis explores the cells' state transition with a multitude of perturbations in combination with traditional pathway analysis, leading to an encompassing picture of the dynamics of a therapeutically desired cell-state switching.

A map of mobile DNA insertions in the NCI-60 human cancer cell panel

Background

The National Cancer Institute-60 (NCI-60) cell lines are among the most widely used models of human cancer. They provide a platform to integrate DNA sequence information, epigenetic data, RNA and protein expression, and pharmacologic susceptibilities in studies of cancer cell biology. Genome-wide studies of the complete panel have included exome sequencing, karyotyping, and copy number analyses but have not targeted repetitive sequences. Interspersed repeats derived from mobile DNAs are a significant source of heritable genetic variation, and insertions of active elements can occur somatically in malignancy.

Method

We used Transposon Insertion Profiling by microarray (TIP-chip) to map Long INterspersed Element-1 (LINE-1, L1) and Alu Short INterspersed Element (SINE) insertions in cancer genes in NCI-60 cells. We focused this discovery effort on annotated Cancer Gene Index loci.

Results

We catalogued a total of 749 and 2,100 loci corresponding to candidate LINE-1 and Alu insertion sites, respectively. As expected, these numbers encompass previously known insertions, polymorphisms shared in unrelated tumor cell lines, as well as unique, potentially tumor-specific insertions. We also conducted association analyses relating individual insertions to a variety of cellular phenotypes.

Conclusions

These data provide a resource for investigators with interests in specific cancer gene loci or mobile element insertion effects more broadly. Our data underscore that significant genetic variation in cancer genomes is owed to LINE-1 and Alu retrotransposons. Our findings also indicate that as large numbers of cancer genomes become available, it will be possible to associate individual transposable element insertion variants with molecular and phenotypic features of these malignancies.

Electronic supplementary material

The online version of this article (doi:10.1186/s13100-016-0078-4) contains supplementary material, which is available to authorized users.

Multitask learning improves prediction of cancer drug sensitivity

Precision oncology seeks to predict the best therapeutic option for individual patients based on the molecular characteristics of their tumors. To assess the preclinical feasibility of drug sensitivity prediction, several studies have measured drug responses for cytotoxic and targeted therapies across large collections of genomically and transcriptomically characterized cancer cell lines and trained predictive models using standard methods like elastic net regression. Here we use existing drug response data sets to demonstrate that multitask learning across drugs strongly improves the accuracy and interpretability of drug prediction models. Our method uses trace norm regularization with a highly efficient ADMM (alternating direction method of multipliers) optimization algorithm that readily scales to large data sets. We anticipate that our approach will enhance efforts to exploit growing drug response compendia in order to advance personalized therapy.

Gene selection for cancer classification with the help of bees

BACKGROUND

Development of biologically relevant models from gene expression data notably, microarray data has become a topic of great interest in the field of bioinformatics and clinical genetics and oncology. Only a small number of gene expression data compared to the total number of genes explored possess a significant correlation with a certain phenotype. Gene selection enables researchers to obtain substantial insight into the genetic nature of the disease and the mechanisms responsible for it. Besides improvement of the performance of cancer classification, it can also cut down the time and cost of medical diagnoses.

METHODS

This study presents a modified Artificial Bee Colony Algorithm (ABC) to select minimum number of genes that are deemed to be significant for cancer along with improvement of predictive accuracy. The search equation of ABC is believed to be good at exploration but poor at exploitation. To overcome this limitation we have modified the ABC algorithm by incorporating the concept of pheromones which is one of the major components of Ant Colony Optimization (ACO) algorithm and a new operation in which successive bees communicate to share their findings.

RESULTS

The proposed algorithm is evaluated using a suite of ten publicly available datasets after the parameters are tuned scientifically with one of the datasets. Obtained results are compared to other works that used the same datasets. The performance of the proposed method is proved to be superior.

CONCLUSION

The method presented in this paper can provide subset of genes leading to more accurate classification results while the number of selected genes is smaller. Additionally, the proposed modified Artificial Bee Colony Algorithm could conceivably be applied to problems in other areas as well.

Cytotoxicity of South-African medicinal plants towards sensitive and multidrug-resistant cancer cells

ETHNOPHARMACOLOGICAL RELEVANCE

Traditional medicine plays a major role for primary health care worldwide. Cancer belongs to the leading disease burden in industrialized and developing countries. Successful cancer therapy is hampered by the development of resistance towards established anticancer drugs.

AIM

In the present study, we investigated the cytotoxicity of 29 extracts from 26 medicinal plants of South-Africa against leukemia cell lines, most of which are used traditionally to treat cancer and related symptoms.

MATERIAL AND METHODS

We have investigated the plant extracts for their cytotoxic activity towards drug-sensitive parental CCRF-CEM leukemia cells and their multidrug-resistant P-glycoprotein-overexpressing subline, CEM/ADR5000 by means of the resazurin assay. A panel of 60 NCI tumor cell lines have been investigated for correlations between selected phytochemicals from medicinal plants and the expression of resistance-conferring genes (ABC-transporters, oncogenes, tumor suppressor genes).

RESULTS

Seven extracts inhibited both cell lines (Acokanthera oppositifolia, Hypoestes aristata, Laurus nobilis, Leonotis leonurus, Plectranthus barbatus, Plectranthus ciliates, Salvia apiana). CEM/ADR5000 cells exhibited a low degree of cross-resistance (3.35-fold) towards the L. leonurus extract, while no cross-resistance was observed to other plant extracts, although CEM/ADR5000 cells were highly resistant to clinically established drugs. The log10IC50 values for two out of 14 selected phytochemicals from these plants (acovenoside A and ouabain) of 60 tumor cell lines were correlated to the expression of ABC-transporters (ABCB1, ABCB5, ABCC1, ABCG2), oncogenes (EGFR, RAS) and tumor suppressors (TP53). Sensitivity or resistance of the cell lines were not statistically associated with the expression of these genes, indicating that multidrug-resistant, refractory tumors expressing these genes may still respond to acovenoside A and ouabain.

CONCLUSION

The bioactivity of South African medicinal plants may represent a basis for the development of strategies to treat multidrug-resistant tumors either by phytotherapeutic approaches with whole plant preparations or by classical drug development with isolated compounds such as acovenoside A or ouabain.

Pharmacogenomics of Scopoletin in Tumor Cells

Drug resistance and the severe side effects of chemotherapy necessitate the development of novel anticancer drugs. Natural products are a valuable source for drug development. Scopoletin is a coumarin compound, which can be found in several Artemisia species and other plant genera. Microarray-based RNA expression profiling of the NCI cell line panel showed that cellular response of scopoletin did not correlate to the expression of ATP-binding cassette (ABC) transporters as classical drug resistance mechanisms (ABCB1, ABCB5, ABCC1, ABCG2). This was also true for the expression of the oncogene EGFR and the mutational status of the tumor suppressor gene, TP53. However, mutations in the RAS oncogenes and the slow proliferative activity in terms of cell doubling times significantly correlated with scopoletin resistance. COMPARE and hierarchical cluster analyses of transcriptome-wide mRNA expression resulted in a set of 40 genes, which all harbored binding motifs in their promoter sequences for the transcription factor, NF-κB, which is known to be associated with drug resistance. RAS mutations, slow proliferative activity, and NF-κB may hamper its effectiveness. By in silico molecular docking studies, we found that scopoletin bound to NF-κB and its regulator IκB. Scopoletin activated NF-κB in a SEAP-driven NF-κB reporter cell line, indicating that NF-κB might be a resistance factor for scopoletin. In conclusion, scopoletin might serve as lead compound for drug development because of its favorable activity against tumor cells with ABC-transporter expression, although NF-κB activation may be considered as resistance factor for this compound. Further investigations are warranted to explore the full therapeutic potential of this natural product.

A systematic study on drug-response associated genes using baseline gene expressions of the Cancer Cell Line Encyclopedia

We have studied drug-response associated (DRA) gene expressions by applying a systems biology framework to the Cancer Cell Line Encyclopedia data. More than 4,000 genes are inferred to be DRA for at least one drug, while the number of DRA genes for each drug varies dramatically from almost 0 to 1,226. Functional enrichment analysis shows that the DRA genes are significantly enriched in genes associated with cell cycle and plasma membrane. Moreover, there might be two patterns of DRA genes between genders. There are significantly shared DRA genes between male and female for most drugs, while very little DRA genes tend to be shared between the two genders for a few drugs targeting sex-specific cancers (e.g., PD-0332991 for breast cancer and ovarian cancer). Our analyses also show substantial difference for DRA genes between young and old samples, suggesting the necessity of considering the age effects for personalized medicine in cancers. Lastly, differential module and key driver analyses confirm cell cycle related modules as top differential ones for drug sensitivity. The analyses also reveal the role of TSPO, TP53, and many other immune or cell cycle related genes as important key drivers for DRA network modules. These key drivers provide new drug targets to improve the sensitivity of cancer therapy.

Using activation status of signaling pathways as mechanism-based biomarkers to predict drug sensitivity

Many complex traits, as drug response, are associated with changes in biological pathways rather than being caused by single gene alterations. Here, a predictive framework is presented in which gene expression data are recoded into activity statuses of signal transduction circuits (sub-pathways within signaling pathways that connect receptor proteins to final effector proteins that trigger cell actions). Such activity values are used as features by a prediction algorithm which can efficiently predict a continuous variable such as the IC50 value. The main advantage of this prediction method is that the features selected by the predictor, the signaling circuits, are themselves rich-informative, mechanism-based biomarkers which provide insight into or drug molecular mechanisms of action (MoA).

Identification of cancer-cytotoxic modulators of PDE3A by predictive chemogenomics

High cancer death rates indicate the need for new anticancer therapeutic agents. Approaches to discovering new cancer drugs include target-based drug discovery and phenotypic screening. Here, we identified phosphodiesterase 3A modulators as cell-selective cancer cytotoxic compounds through phenotypic compound library screening and target deconvolution by predictive chemogenomics. We found that sensitivity to 6-(4-(diethylamino)-3-nitrophenyl)-5-methyl-4,5-dihydropyridazin-3(2H)-one, or DNMDP, across 766 cancer cell lines correlates with expression of the gene PDE3A, encoding phosphodiesterase 3A. Like DNMDP, a subset of known PDE3A inhibitors kill selected cancer cells, whereas others do not. Furthermore, PDE3A depletion leads to DNMDP resistance. We demonstrated that DNMDP binding to PDE3A promotes an interaction between PDE3A and Schlafen 12 (SLFN12), suggestive of a neomorphic activity. Coexpression of SLFN12 with PDE3A correlates with DNMDP sensitivity, whereas depletion of SLFN12 results in decreased DNMDP sensitivity. Our results implicate PDE3A modulators as candidate cancer therapeutic agents and demonstrate the power of predictive chemogenomics in small-molecule discovery.

A Copula Based Approach for Design of Multivariate Random Forests for Drug Sensitivity Prediction

Modeling sensitivity to drugs based on genetic characterizations is a significant challenge in the area of systems medicine. Ensemble based approaches such as Random Forests have been shown to perform well in both individual sensitivity prediction studies and team science based prediction challenges. However, Random Forests generate a deterministic predictive model for each drug based on the genetic characterization of the cell lines and ignores the relationship between different drug sensitivities during model generation. This application motivates the need for generation of multivariate ensemble learning techniques that can increase prediction accuracy and improve variable importance ranking by incorporating the relationships between different output responses. In this article, we propose a novel cost criterion that captures the dissimilarity in the output response structure between the training data and node samples as the difference in the two empirical copulas. We illustrate that copulas are suitable for capturing the multivariate structure of output responses independent of the marginal distributions and the copula based multivariate random forest framework can provide higher accuracy prediction and improved variable selection. The proposed framework has been validated on genomics of drug sensitivity for cancer and cancer cell line encyclopedia database.

Cytotoxicity of 35 medicinal plants from Sudan towards sensitive and multidrug-resistant cancer cells

BACKGROUND

Cancer is a complex disease with multiple genetic and epigenetic alterations. Since decades, the hallmark of cancer therapy is chemotherapy. Cytotoxic drugs erase rapidly dividing cells without sufficient differentiation between normal and cancerous cells resulting in severe side effects in normal tissues. Recently, strategies for cancer treatment focused on targeting specific proteins involved in tumor growth and progression. The present study was designed to investigate the cytotoxicity of 65 crude extracts from 35 Sudanese medicinal plants towards various cancer cell lines expressing molecular mechanisms of resistance towards classical chemotherapeutics (two ATP-binding cassette transporters, ABCB1 (P-glycoprotein) and ABCB5, tumor suppressor p53, epidermal growth factors receptor (EGFR). And the aim was to identify plant extracts and isolated compounds thereof with activity towards otherwise drug-resistant tumor cells.

METHODS

Cold maceration was performed to obtain crude extracts from the plants. The resazurin assay was used to determine cytotoxicity of the plant extracts. Microarray-based mRNA expression profiling, COMPARE, and hierarchical cluster analyses were applied to identify, which genes correlate with sensitivity or resistance to ambrosin, the main constituent of the most active extract Ambrosia maritima.

RESULTS

The results of the resazurin assay on different tumors showed that Lawsonia inermis, Trigonella foenum-graecum and Ambrosia maritma were the most active crude extracts. Ambrosin was selected as one active principle of A. maritima for microarray-based expression profiling. Genes from various functional groups (transcriptional regulators, signal transduction, membrane transporters, cytoskeleton organization, chaperones, immune system development and DNA repair) were significantly correlated with response of tumor cell lines to ambrosin.

CONCLUSION

The results revealed cytotoxicity and pharmacogenomics studies of Sudanese medicinal plants provide an attractive strategy for the development of novel cancer therapeutics with activity towards cell lines that resistance to established anticancer agents.

Biological targets and mechanisms of action of natural products from marine cyanobacteria

Marine cyanobacteria are an ancient group of organisms and prolific producers of bioactive secondary metabolites. These compounds are presumably optimized by evolution over billions of years to exert high affinity for their intended biological target in the ecologically relevant organism but likely also possess activity in different biological contexts such as human cells. Screening of marine cyanobacterial extracts for bioactive natural products has largely focused on cancer cell viability; however, diversification of the screening platform led to the characterization of many new bioactive compounds. Targets of compounds have oftentimes been elusive if the compounds were discovered through phenotypic assays. Over the past few years, technology has advanced to determine mechanism of action (MOA) and targets through reverse chemical genetic and proteomic approaches, which has been applied to certain cyanobacterial compounds and will be discussed in this review. Some cyanobacterial molecules are the most-potent-in-class inhibitors and therefore may become valuable tools for chemical biology to probe protein function but also be templates for novel drugs, assuming in vitro potency translates into cellular and in vivo activity. Our review will focus on compounds for which the direct targets have been deciphered or which were found to target a novel pathway, and link them to disease states where target modulation may be beneficial.

Semi-Supervised Projective Non-Negative Matrix Factorization for Cancer Classification

Advances in DNA microarray technologies have made gene expression profiles a significant candidate in identifying different types of cancers. Traditional learning-based cancer identification methods utilize labeled samples to train a classifier, but they are inconvenient for practical application because labels are quite expensive in the clinical cancer research community. This paper proposes a semi-supervised projective non-negative matrix factorization method (Semi-PNMF) to learn an effective classifier from both labeled and unlabeled samples, thus boosting subsequent cancer classification performance. In particular, Semi-PNMF jointly learns a non-negative subspace from concatenated labeled and unlabeled samples and indicates classes by the positions of the maximum entries of their coefficients. Because Semi-PNMF incorporates statistical information from the large volume of unlabeled samples in the learned subspace, it can learn more representative subspaces and boost classification performance. We developed a multiplicative update rule (MUR) to optimize Semi-PNMF and proved its convergence. The experimental results of cancer classification for two multiclass cancer gene expression profile datasets show that Semi-PNMF outperforms the representative methods.

Novel prediction of anticancer drug chemosensitivity in cancer cell lines: evidence of moderation by microRNA expressions

The objectives of this study are (1) to develop a novel "moderation" model of drug chemosensitivity and (2) to investigate if miRNA expression moderates the relationship between gene expression and drug chemosensitivity, specifically for HSP90 inhibitors applied to human cancer cell lines. A moderation model integrating the interaction between miRNA and gene expressions was developed to examine if miRNA expression affects the strength of the relationship between gene expression and chemosensitivity. Comprehensive datasets on miRNA expressions, gene expressions, and drug chemosensitivities were obtained from National Cancer Institute's NCI-60 cell lines including nine different cancer types. A workflow including steps of selecting genes, miRNAs, and compounds, correlating gene expression with chemosensitivity, and performing multivariate analysis was utilized to test the proposed model. The proposed moderation model identified 12 significantly-moderating miRNAs: miR-15b*, miR-16-2*, miR-9, miR-126*, miR-129*, miR-138, miR-519e*, miR-624*, miR-26b, miR-30e*, miR-32, and miR-196a, as well as two genes ERCC2 and SF3B1 which affect chemosensitivities of Tanespimycin and Alvespimycin - both HSP90 inhibitors. A bootstrap resampling of 2,500 times validates the significance of all 12 identified miRNAs. The results confirm that certain miRNA and gene expressions interact to produce an effect on drug response. The lack of correlation between miRNA and gene expression themselves suggests that miRNA transmits its effect through translation inhibition/control rather than mRNA degradation. The results suggest that miRNAs could serve not only as prognostic biomarkers for cancer treatment outcome but also as interventional agents to modulate desired chemosensitivity.

Improved large-scale prediction of growth inhibition patterns using the NCI60 cancer cell line panel

MOTIVATION

Recent large-scale omics initiatives have catalogued the somatic alterations of cancer cell line panels along with their pharmacological response to hundreds of compounds. In this study, we have explored these data to advance computational approaches that enable more effective and targeted use of current and future anticancer therapeutics.

RESULTS

We modelled the 50% growth inhibition bioassay end-point (GI50) of 17,142 compounds screened against 59 cancer cell lines from the NCI60 panel (941,831 data-points, matrix 93.08% complete) by integrating the chemical and biological (cell line) information. We determine that the protein, gene transcript and miRNA abundance provide the highest predictive signal when modelling the GI50 endpoint, which significantly outperformed the DNA copy-number variation or exome sequencing data (Tukey's Honestly Significant Difference, P <0.05). We demonstrate that, within the limits of the data, our approach exhibits the ability to both interpolate and extrapolate compound bioactivities to new cell lines and tissues and, although to a lesser extent, to dissimilar compounds. Moreover, our approach outperforms previous models generated on the GDSC dataset. Finally, we determine that in the cases investigated in more detail, the predicted drug-pathway associations and growth inhibition patterns are mostly consistent with the experimental data, which also suggests the possibility of identifying genomic markers of drug sensitivity for novel compounds on novel cell lines.

CONTACT

terez@pasteur.fr; ab454@ac.cam.uk

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Anticancer drug sensitivity prediction in cell lines from baseline gene expression through recursive feature selection

BACKGROUND

An enduring challenge in personalized medicine is to select right drug for individual patients. Testing drugs on patients in large clinical trials is one way to assess their efficacy and toxicity, but it is impractical to test hundreds of drugs currently under development. Therefore the preclinical prediction model is highly expected as it enables prediction of drug response to hundreds of cell lines in parallel.

METHODS

Recently, two large-scale pharmacogenomic studies screened multiple anticancer drugs on over 1000 cell lines in an effort to elucidate the response mechanism of anticancer drugs. To this aim, we here used gene expression features and drug sensitivity data in Cancer Cell Line Encyclopedia (CCLE) to build a predictor based on Support Vector Machine (SVM) and a recursive feature selection tool. Robustness of our model was validated by cross-validation and an independent dataset, the Cancer Genome Project (CGP).

RESULTS

Our model achieved good cross validation performance for most drugs in the Cancer Cell Line Encyclopedia (≥80% accuracy for 10 drugs, ≥75% accuracy for 19 drugs). Independent tests on eleven common drugs between CCLE and CGP achieved satisfactory performance for three of them, i.e., AZD6244, Erlotinib and PD-0325901, using expression levels of only twelve, six and seven genes, respectively.

CONCLUSIONS

These results suggest that drug response could be effectively predicted from genomic features. Our model could be applied to predict drug response for some certain drugs and potentially play a complementary role in personalized medicine.

A network flow-based method to predict anticancer drug sensitivity

Predicting anticancer drug sensitivity can enhance the ability to individualize patient treatment, thus making development of cancer therapies more effective and safe. In this paper, we present a new network flow-based method, which utilizes the topological structure of pathways, for predicting anticancer drug sensitivities. Mutations and copy number alterations of cancer-related genes are assumed to change the pathway activity, and pathway activity difference before and after drug treatment is used as a measure of drug response. In our model, Contributions from different genetic alterations are considered as free parameters, which are optimized by the drug response data from the Cancer Genome Project (CGP). 10-fold cross validation on CGP data set showed that our model achieved comparable prediction results with existing elastic net model using much less input features.

DISIS: prediction of drug response through an iterative sure independence screening

Prediction of drug response based on genomic alterations is an important task in the research of personalized medicine. Current elastic net model utilized a sure independence screening to select relevant genomic features with drug response, but it may neglect the combination effect of some marginally weak features. In this work, we applied an iterative sure independence screening scheme to select drug response relevant features from the Cancer Cell Line Encyclopedia (CCLE) dataset. For each drug in CCLE, we selected up to 40 features including gene expressions, mutation and copy number alterations of cancer-related genes, and some of them are significantly strong features but showing weak marginal correlation with drug response vector. Lasso regression based on the selected features showed that our prediction accuracies are higher than those by elastic net regression for most drugs.

Improving Drug Sensitivity Prediction Using Different Types of Data

The algorithms and models used to address the two subchallenges that are part of the NCI-DREAM (Dialogue for Reverse Engineering Assessments and Methods) Drug Sensitivity Prediction Challenge (2012) are presented. In subchallenge 1, a bidirectional search algorithm is introduced and optimized using an ensemble scheme and a nonlinear support vector machine (SVM) is then applied to predict the effects of the drug compounds on breast cancer cell lines. In subchallenge 2, a weighted Euclidean distance method is introduced to predict and rank the drug combinations from the most to the least effective in reducing the viability of a diffuse large B-cell lymphoma (DLBCL) cell line.

Pharmacogenomic and molecular docking studies on the cytotoxicity of the natural steroid wortmannin against multidrug-resistant tumor cells

Wortmannin is a cytotoxic compound derived from the endophytic fungi Fusarium oxysporum, Penicillium wortmannii and Penicillium funiculosum that occurs in many plants, including medicinal herbs. The rationale to develop novel anticancer drugs is the frequent development of tumor resistance to the existing antineoplasic agents. Therefore, it is mandatory to analyze resistance mechanisms of novel drug candidates such as wortmannin as well to bring effective drugs into the clinic that have the potential to bypass or overcome resistance to established drugs and to substantially increase life span of cancer patients. In the present project, we found that P-glycoprotein-overexpressing tumor cells displaying the classical multidrug resistance phenotype toward standard anticancer drugs were not cross-resistant to wortmannin. Furthermore, three point-mutated PIK3CA protein structures revealed similar binding energies to wortmannin than wild-type PIK3CA. This protein is the primary target of wortmannin and part of the PI3K/AKT/mTOR signaling pathway. PIK3CA mutations are known to be associated with worse response to therapy and shortened its activity toward wild-type and mutant PIK3CA with similar efficacy.

Measuring circadian clock function in human cells

Circadian clocks are present in most cells and are essential for maintenance of daily rhythms in physiology, mood, and cognition. Thus, not only neurons of the central circadian pacemaker but also many other peripheral tissues possess the same functional and self-sustained circadian clocks. Surprisingly, however, their properties vary widely within the human population. In recent years, this clock variance has been studied extensively both in health and in disease using robust lentivirus-based reporter technologies to probe circadian function in human peripheral cells as proxies for those in neurologically and physiologically relevant but inaccessible tissues. The same procedures can be used to investigate other conserved signal transduction cascades affecting multiple aspects of human physiology, behavior, and disease. Accessing gene expression variation within human populations via these powerful in vitro cell-based technologies could provide important insights into basic phenotypic diversity or to better interpret patterns of gene expression variation in disease.

Integrating phenotypic small-molecule profiling and human genetics: the next phase in drug discovery

Over the past decade, tremendous progress in high-throughput small molecule-screening methods has facilitated the rapid expansion of phenotype-based data. Parallel advances in genomic characterization methods have complemented these efforts by providing a growing list of annotated cell line features. Together, these developments have paved the way for feature-based identification of novel, exploitable cellular dependencies, subsequently expanding our therapeutic toolkit in cancer and other diseases. Here, we provide an overview of the evolution of phenotypic small-molecule profiling and discuss the most significant and recent profiling and analytical efforts, their impact on the field, and their clinical ramifications. We additionally provide a perspective for future developments in phenotypic profiling efforts guided by genomic science.

Prediction of individual response to anticancer therapy: historical and future perspectives

Since the introduction of chemotherapy for cancer treatment in the early 20th century considerable efforts have been made to maximize drug efficiency and at the same time minimize side effects. As there is a great interpatient variability in response to chemotherapy, the development of predictive biomarkers is an ambitious aim for the rapidly growing research area of personalized molecular medicine. The individual prediction of response will improve treatment and thus increase survival and life quality of patients. In the past, cell cultures were used as in vitro models to predict in vivo response to chemotherapy. Several in vitro chemosensitivity assays served as tools to measure miscellaneous endpoints such as DNA damage, apoptosis and cytotoxicity or growth inhibition. Twenty years ago, the development of high-throughput technologies, e.g. cDNA microarrays enabled a more detailed analysis of drug responses. Thousands of genes were screened and expression levels were correlated to drug responses. In addition, mutation analysis became more and more important for the prediction of therapeutic success. Today, as research enters the area of -omics technologies, identification of signaling pathways is a tool to understand molecular mechanism underlying drug resistance. Combining new tissue models, e.g. 3D organoid cultures with modern technologies for biomarker discovery will offer new opportunities to identify new drug targets and in parallel predict individual responses to anticancer therapy. In this review, we present different currently used chemosensitivity assays including 2D and 3D cell culture models and several -omics approaches for the discovery of predictive biomarkers. Furthermore, we discuss the potential of these assays and biomarkers to predict the clinical outcome of individual patients and future perspectives.

A p-Median approach for predicting drug response in tumour cells

Background

The complexity of biological data related to the genetic origins of tumour cells, originates significant challenges to glean valuable knowledge that can be used to predict therapeutic responses. In order to discover a link between gene expression profiles and drug responses, a computational framework based on Consensus p-Median clustering is proposed. The main goal is to simultaneously predict (in silico) anticancer responses by extracting common patterns among tumour cell lines, selecting genes that could potentially explain the therapy outcome and finally learning a probabilistic model able to predict the therapeutic responses.

Results

The experimental investigation performed on the NCI60 dataset highlights three main findings: (1) Consensus p-Median is able to create groups of cell lines that are highly correlated both in terms of gene expression and drug response; (2) from a biological point of view, the proposed approach enables the selection of genes that are strongly involved in several cancer processes; (3) the final prediction of drug responses, built upon Consensus p-Median and the selected genes, represents a promising step for predicting potential useful drugs.

Conclusion

The proposed learning framework represents a promising approach predicting drug response in tumour cells.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-014-0353-7) contains supplementary material, which is available to authorized users.

Cytotoxicity of the bisphenolic honokiol from Magnolia officinalis against multiple drug-resistant tumor cells as determined by pharmacogenomics and molecular docking

A main problem in oncology is the development of drug-resistance. Some plant-derived lignans are established in cancer therapy, e.g. the semisynthetic epipodophyllotoxins etoposide and teniposide. Their activity is, unfortunately, hampered by the ATP-binding cassette (ABC) efflux transporter, P-glycoprotein. Here, we investigated the bisphenolic honokiol derived from Magnolia officinalis. P-glycoprotein-overexpressing CEM/ADR5000 cells were not cross-resistant to honokiol, but MDA-MB-231 BRCP cells transfected with another ABC-transporter, BCRP, revealed 3-fold resistance. Further drug resistance mechanisms analyzed study was the tumor suppressor TP53 and the epidermal growth factor receptor (EGFR). HCT116 p53(-/-) did not reveal resistance to honokiol, and EGFR-transfected U87.MG EGFR cells were collateral sensitive compared to wild-type cells (degree of resistance: 0.34). To gain insight into possible modes of collateral sensitivity, we performed in silico molecular docking studies of honokiol to EGFR and EGFR-related downstream signal proteins. Honokiol bound with comparable binding energies to EGFR (-7.30 ± 0.01 kcal/mol) as the control drugs erlotinib (-7.50 ± 0.30 kcal/mol) and gefitinib (-8.30 ± 0.10 kcal/mol). Similar binding affinities of AKT, MEK1, MEK2, STAT3 and mTOR were calculated for honokiol (range from -9.0 ± 0.01 to 7.40 ± 0.01 kcal/mol) compared to corresponding control inhibitor compounds for these signal transducers. This indicates that collateral sensitivity of EGFR-transfectant cells towards honokiol may be due to binding to EGFR and downstream signal transducers. COMPARE and hierarchical cluster analyses of microarray-based transcriptomic mRNA expression data of 59 tumor cell lines revealed a specific gene expression profile predicting sensitivity or resistance towards honokiol.

Inference of dynamic biological networks based on responses to drug perturbations

Drugs that target specific proteins are a major paradigm in cancer research. In this article, we extend a modeling framework for drug sensitivity prediction and combination therapy design based on drug perturbation experiments. The recently proposed target inhibition map approach can infer stationary pathway models from drug perturbation experiments, but the method is limited to a steady-state snapshot of the underlying dynamical model. We consider the inverse problem of possible dynamic models that can generate the static target inhibition map model. From a deterministic viewpoint, we analyze the inference of Boolean networks that can generate the observed binarized sensitivities under different target inhibition scenarios. From a stochastic perspective, we investigate the generation of Markov chain models that satisfy the observed target inhibition sensitivities.

Conceptualizing cancer drugs as classifiers

Cancer and healthy cells have distinct distributions of molecular properties and thus respond differently to drugs. Cancer drugs ideally kill cancer cells while limiting harm to healthy cells. However, the inherent variance among cells in both cancer and healthy cell populations increases the difficulty of selective drug action. Here we formalize a classification framework based on the idea that an ideal cancer drug should maximally discriminate between cancer and healthy cells. More specifically, this discrimination should be performed on the basis of measurable cell markers. We divide the problem into three parts which we explore with examples. First, molecular markers should discriminate cancer cells from healthy cells at the single-cell level. Second, the effects of drugs should be statistically predicted by these molecular markers. Third, drugs should be optimized for classification performance. We find that expression levels of a handful of genes suffice to discriminate well between individual cells in cancer and healthy tissue. We also find that gene expression predicts the efficacy of some cancer drugs, suggesting that these cancer drugs act as suboptimal classifiers using gene profiles. Finally, we formulate a framework that defines an optimal drug, and predicts drug cocktails that may target cancer more accurately than the individual drugs alone. Conceptualizing cancer drugs as solving a discrimination problem in the high-dimensional space of molecular markers promises to inform the design of new cancer drugs and drug cocktails.

Modeling precision treatment of breast cancer

Background

First-generation molecular profiles for human breast cancers have enabled the identification of features that can predict therapeutic response; however, little is known about how the various data types can best be combined to yield optimal predictors. Collections of breast cancer cell lines mirror many aspects of breast cancer molecular pathobiology, and measurements of their omic and biological therapeutic responses are well-suited for development of strategies to identify the most predictive molecular feature sets.

Results

We used least squares-support vector machines and random forest algorithms to identify molecular features associated with responses of a collection of 70 breast cancer cell lines to 90 experimental or approved therapeutic agents. The datasets analyzed included measurements of copy number aberrations, mutations, gene and isoform expression, promoter methylation and protein expression. Transcriptional subtype contributed strongly to response predictors for 25% of compounds, and adding other molecular data types improved prediction for 65%. No single molecular dataset consistently out-performed the others, suggesting that therapeutic response is mediated at multiple levels in the genome. Response predictors were developed and applied to TCGA data, and were found to be present in subsets of those patient samples.

Conclusions

These results suggest that matching patients to treatments based on transcriptional subtype will improve response rates, and inclusion of additional features from other profiling data types may provide additional benefit. Further, we suggest a systems biology strategy for guiding clinical trials so that patient cohorts most likely to respond to new therapies may be more efficiently identified.

Translational assessment of mitochondrial dysfunction of pancreatic cancer from in vitro gene microarray and animal efficacy studies, to early clinical studies, via the novel tumor-specific anti-mitochondrial agent, CPI-613

STUDY RATIONALE AND OBJECTIVES

Via genetic alterations, malignant transformation and proliferation are associated with extensive alterations of mitochondrial energy metabolism of tumor cells. Thus, inhibition of the altered form of mitochondrial energy metabolism of tumor cells may be an effective therapy for cancers. This study performed translational assessment of mitochondrial dysfunction of pancreatic cancer from in vitro gene microarray and animal efficacy studies, to early clinical studies, via the novel tumor-specific anti-mitochondrial agent, CPI-613.

METHODS

The gene profiles of BxPC-3 human pancreatic tumor cells and non-transformed NIH-3T3 mouse fibroblast cells (negative control), after CPI-613 or sham treatment, were assessed and compared using microarray technique. The anti-cancer efficacies of CPI-613 and Gemcitabine were assessed and compared in mice with xenograft from inoculation of BxPC-3 human pancreatic tumor cells, based on the degree of tumor growth inhibition and prolongation of survival when compared to vehicle treatment. The anti-cancer activities, according to overall survival (OS), of CPI-613 alone and in combination with Gemcitabine were assessed in patients with Stage IV pancreatic cancer.

RESULTS

Microarray studies indicated that CPI-613 down-regulated the expression of Cyclin D3, E1, E2, F, A2, B1 and CDK2 genes of BxPC-3 pancreatic cancer cells but not non-transformed NIH-3T3 mouse fibroblast cells (negative control). In mice with pancreatic carcinoma xenografts, four weekly intraperitoneal injections of either CPI-613 (25 mg/kg/administration) or Gemcitabine (50 mg/kg/administration) inhibited tumor growth and prolonged survival when compared to vehicle treatment. The degree of tumor growth inhibition was ~2×, and prolongation of survival was ~4×, greater with CPI-613 treatment than with Gemcitabine treatment. In patients with Stage IV advanced pancreatic cancer, CPI-613 at 420-1,300 mg/m(2), given twice weekly for three weeks followed by a week of rest (i.e., 3-week-on-1-week-off) as monotherapy, provided median OS of 15 months in three patients. CPI-613 at 150-320 mg/m(2) given twice weekly on the 3-week-on-1-week-off dosing schedule, coinciding with Gemcitabine (1,000 mg/m(2)) given once weekly on the 3-week-on-1-week-off dosing schedule, provided median OS of 17.8 months in four patients. These median OS values from CPI-613 monotherapy and CPI-613 + Gemcitabine treatment tend to be longer than those in patients treated with Abraxane + Gemcitabine combination or FOLFININOX (median OS ~12 months).

CONCLUSIONS

The dysfunctional mitochondria of pancreatic cancer cells was translationable from in vitro gene alteration and animal tumor model studies to patients with advanced Stage IV pancreatic cancer, as reflected by the anti-cancer activities of the tumor-specific anti-mitochondrial agent, CPI-613, in these studies.

Machine learning approaches to investigate the impact of PCBs on the transcriptome of the common bottlenose dolphin (Tursiops truncatus)

As top-level predators, common bottlenose dolphins (Tursiops truncatus) are particularly sensitive to chemical and biological contaminants that accumulate and biomagnify in the marine food chain. This work investigates the potential use of microarray technology and gene expression profile analysis to screen common bottlenose dolphins for exposure to environmental contaminants through the immunological and/or endocrine perturbations associated with these agents. A dolphin microarray representing 24,418 unigene sequences was used to analyze blood samples collected from 47 dolphins during capture-release health assessments from five different US coastal locations (Beaufort, NC, Sarasota Bay, FL, Saint Joseph Bay, FL, Sapelo Island, GA and Brunswick, GA). Organohalogen contaminants including pesticides, polychlorinated biphenyl congeners (PCBs) and polybrominated diphenyl ether congeners were determined in blubber biopsy samples from the same animals. A subset of samples (n = 10, males; n = 8, females) with the highest and the lowest measured values of PCBs in their blubber was used as strata to determine the differential gene expression of the exposure extremes through machine learning classification algorithms. A set of genes associated primarily with nuclear and DNA stability, cell division and apoptosis regulation, intra- and extra-cellular traffic, and immune response activation was selected by the algorithm for identifying the two exposure extremes. In order to test the hypothesis that these gene expression patterns reflect PCB exposure, we next investigated the blood transcriptomes of the remaining dolphin samples using machine-learning approaches, including K-nn and Support Vector Machines classifiers. Using the derived gene sets, the algorithms worked very well (100% success rate) at classifying dolphins according to the contaminant load accumulated in their blubber. These results suggest that gene expression profile analysis may provide a valuable means to screen for indicators of chemical exposure.