Immuno-oncological effects of standard anticancer agents and commonly used concomitant drugs: an in vitro assessment.

BACKGROUND: It has become evident in the field of oncology that the outcome of medical treatment is influenced by the combined effect exerted on both cancer- and immune cells. Therefore, we evaluated potential immunological effects of 46 standard anticancer agents and 22 commonly administered concomitant non-cancer drugs. METHODS: We utilized a miniaturized in vitro model system comprised of fluorescently labeled human colon and lung cancer cell lines grown as monocultures and co-cultured with activated peripheral blood mononuclear cells (PBMCs). The Bliss Independence Model was then applied to detect antagonism and synergy between the drugs and activated immune cells. RESULTS: Among the standard anticancer agents, tyrosine kinase inhibitors (TKIs) stood out as the top inducers of both antagonism and synergy. Ruxolitinib and dasatinib emerged as the most notably antagonistic substances, exhibiting the lowest Bliss scores, whereas sorafenib was shown to synergize with activated PBMCs. Most concomitant drugs did not induce neither antagonism nor synergy. However, the statins mevastatin and simvastatin were uniquely shown to synergize with activated PBMC at all tested drug concentrations in the colon cancer model. CONCLUSION: We utilized a miniaturized tumor-immune model to enable time and cost-effective evaluation of a broad panel of drugs in an immuno-oncology setting in vitro. Using this approach, immunomodulatory effects exerted by TKIs and statins were identified.

Phenotypic screening platform identifies statins as enhancers of immune cell-induced cancer cell death.

BACKGROUND: High-throughput screening (HTS) of small molecule drug libraries has greatly facilitated the discovery of new cancer drugs. However, most phenotypic screening platforms used in the field of oncology are based solely on cancer cell populations and do not allow for the identification of immunomodulatory agents. METHODS: We developed a phenotypic screening platform based on a miniaturized co-culture system with human colorectal cancer- and immune cells, providing a model that recapitulates part of the tumor immune microenvironment (TIME) complexity while simultaneously being compatible with a simple image-based readout. Using this platform, we screened 1,280 small molecule drugs, all approved by the Food and Drug Administration (FDA), and identified statins as enhancers of immune cell-induced cancer cell death. RESULTS: The lipophilic statin pitavastatin had the most potent anti-cancer effect. Further analysis demonstrated that pitavastatin treatment induced a pro-inflammatory cytokine profile as well as an overall pro-inflammatory gene expression profile in our tumor-immune model. CONCLUSION: Our study provides an in vitro phenotypic screening approach for the identification of immunomodulatory agents and thus addresses a critical gap in the field of immuno-oncology. Our pilot screen identified statins, a drug family gaining increasing interest as repurposing candidates for cancer treatment, as enhancers of immune cell-induced cancer cell death. We speculate that the clinical benefits described for cancer patients receiving statins are not simply caused by a direct effect on the cancer cells but rather are dependent on the combined effect exerted on both cancer and immune cells.

Assessment in vitro of interactions between anti-cancer drugs and noncancer drugs commonly used by cancer patients.

Cancer patients often suffer from cancer symptoms, treatment complications and concomitant diseases and are, therefore, often treated with several drugs in addition to anticancer drugs. Whether such drugs, here denoted as 'concomitant drugs', have anticancer effects or interact at the tumor cell level with the anticancer drugs is not very well known. The cytotoxic effects of nine concomitant drugs and their interactions with five anti-cancer drugs commonly used for the treatment of colorectal cancer were screened over broad ranges of drug concentrations in vitro in the human colon cancer cell line HCT116wt. Seven additional tyrosine kinase inhibitors were included to further evaluate key findings as were primary cultures of tumor cells from patients with colorectal cancer. Cytotoxic effects were evaluated using the fluorometric microculture cytotoxicity assay (FMCA) and interaction analysis was based on Bliss independent interaction analysis. Simvastatin and loperamide, included here as an opioid agonists, were found to have cytotoxic effects on their own at reasonably low concentrations whereas betamethasone, enalapril, ibuprofen, metformin, metoclopramide, metoprolol and paracetamol were inactive also at very high concentrations. Drug interactions ranged from antagonistic to synergistic over the concentrations tested with a more homogenous pattern of synergy between simvastatin and protein kinase inhibitors in HCT116wt cells. Commonly used concomitant drugs are mostly neither expected to have anticancer effects nor to interact significantly with anticancer drugs frequently used for the treatment of colorectal cancer.

Single-cell transcriptional pharmacodynamics of trifluridine in a tumor-immune model.

Understanding the immunological effects of chemotherapy is of great importance, especially now that we have entered an era where ever-increasing pre-clinical and clinical efforts are put into combining chemotherapy and immunotherapy to combat cancer. Single-cell RNA sequencing (scRNA-seq) has proved to be a powerful technique with a broad range of applications, studies evaluating drug effects in co-cultures of tumor and immune cells are however scarce. We treated a co-culture comprised of human colorectal cancer (CRC) cells and peripheral blood mononuclear cells (PBMCs) with the nucleoside analogue trifluridine (FTD) and used scRNA-seq to analyze posttreatment gene expression profiles in thousands of individual cancer and immune cells concurrently. ScRNA-seq recapitulated major mechanisms of action previously described for FTD and provided new insight into possible treatment-induced effects on T-cell mediated antitumor responses.

Mebendazole is unique among tubulin-active drugs in activating the MEK-ERK pathway.

We recently showed that the anti-helminthic compound mebendazole (MBZ) has immunomodulating activity in monocyte/macrophage models and induces ERK signalling. In the present study we investigated whether MBZ induced ERK activation is shared by other tubulin binding agents (TBAs) and if it is observable also in other human cell types. Curated gene signatures for a panel of TBAs in the LINCS Connectivity Map (CMap) database showed a unique strong negative correlation of MBZ with MEK/ERK inhibitors indicating ERK activation also in non-haematological cell lines. L1000 gene expression signatures for MBZ treated THP-1 monocytes also connected negatively to MEK inhibitors. MEK/ERK phosphoprotein activity testing of a number of TBAs showed that only MBZ increased the activity in both THP-1 monocytes and PMA differentiated macrophages. Distal effects on ERK phosphorylation of the substrate P90RSK and release of IL1B followed the same pattern. The effect of MBZ on MEK/ERK phosphorylation was inhibited by RAF/MEK/ERK inhibitors in THP-1 models, CD3/IL2 stimulated PBMCs and a MAPK reporter HEK-293 cell line. MBZ was also shown to increase ERK activity in CD4+ T-cells from lupus patients with known defective ERK signalling. Given these mechanistic features MBZ is suggested suitable for treatment of diseases characterized by defective ERK signalling, notably difficult to treat autoimmune diseases.

Predictive Value of Ex Vivo Chemosensitivity Assays for Individualized Cancer Chemotherapy: A Meta-Analysis.

Current treatment strategies for chemotherapy of cancer patients were developed to benefit groups of patients with similar clinical characteristics. In practice, response is very heterogeneous between individual patients within these groups. Precision medicine can be viewed as the development toward a more fine-grained treatment stratification than what is currently in use. Cell-based drug sensitivity testing is one of several options for individualized cancer treatment available today, although it has not yet reached widespread clinical use. We present an up-to-date literature meta-analysis on the predictive value of ex vivo chemosensitivity assays for individualized cancer chemotherapy and discuss their current clinical value and possible future developments.

Ex Vivo Assessment of Drug Activity in Patient Tumor Cells as a Basis for Tailored Cancer Therapy.

Although medical cancer treatment has improved during the past decades, it is difficult to choose between several first-line treatments supposed to be equally active in the diagnostic group. It is even more difficult to select a treatment after the standard protocols have failed. Any guidance for selection of the most effective treatment is valuable at these critical stages. We describe the principles and procedures for ex vivo assessment of drug activity in tumor cells from patients as a basis for tailored cancer treatment. Patient tumor cells are assayed for cytotoxicity with a panel of drugs. Acoustic drug dispensing provides great flexibility in the selection of drugs for testing; currently, up to 80 compounds and/or combinations thereof may be tested for each patient. Drug response predictions are obtained by classification using an empirical model based on historical responses for the diagnosis. The laboratory workflow is supported by an integrated system that enables rapid analysis and automatic generation of the clinical referral response.

A strand specific high resolution normalization method for chip-sequencing data employing multiple experimental control measurements.

BACKGROUND: High-throughput sequencing is becoming the standard tool for investigating protein-DNA interactions or epigenetic modifications. However, the data generated will always contain noise due to e.g. repetitive regions or non-specific antibody interactions. The noise will appear in the form of a background distribution of reads that must be taken into account in the downstream analysis, for example when detecting enriched regions (peak-calling). Several reported peak-callers can take experimental measurements of background tag distribution into account when analysing a data set. Unfortunately, the background is only used to adjust peak calling and not as a pre-processing step that aims at discerning the signal from the background noise. A normalization procedure that extracts the signal of interest would be of universal use when investigating genomic patterns. RESULTS: We formulated such a normalization method based on linear regression and made a proof-of-concept implementation in R and C++. It was tested on simulated as well as on publicly available ChIP-seq data on binding sites for two transcription factors, MAX and FOXA1 and two control samples, Input and IgG. We applied three different peak-callers to (i) raw (un-normalized) data using statistical background models and (ii) raw data with control samples as background and (iii) normalized data without additional control samples as background. The fraction of called regions containing the expected transcription factor binding motif was largest for the normalized data and evaluation with qPCR data for FOXA1 suggested higher sensitivity and specificity using normalized data over raw data with experimental background. CONCLUSIONS: The proposed method can handle several control samples allowing for correction of multiple sources of bias simultaneously. Our evaluation on both synthetic and experimental data suggests that the method is successful in removing background noise.

Quantitative chemogenomics: machine-learning models of protein-ligand interaction.

Chemogenomics is an emerging interdisciplinary field that lies in the interface of biology, chemistry, and informatics. Most of the currently used drugs are small molecules that interact with proteins. Understanding protein-ligand interaction is therefore central to drug discovery and design. In the subfield of chemogenomics known as proteochemometrics, protein-ligand-interaction models are induced from data matrices that consist of both protein and ligand information along with some experimentally measured variable. The two general aims of this quantitative multi-structure-property-relationship modeling (QMSPR) approach are to exploit sparse/incomplete information sources and to obtain more general models covering larger parts of the protein-ligand space, than traditional approaches that focuses mainly on specific targets or ligands. The data matrices, usually obtained from multiple sparse/incomplete sources, typically contain series of proteins and ligands together with quantitative information about their interactions. A useful model should ideally be easy to interpret and generalize well to new unseen protein-ligand combinations. Resolving this requires sophisticated machine-learning methods for model induction, combined with adequate validation. This review is intended to provide a guide to methods and data sources suitable for this kind of protein-ligand-interaction modeling. An overview of the modeling process is presented including data collection, protein and ligand descriptor computation, data preprocessing, machine-learning-model induction and validation. Concerns and issues specific for each step in this kind of data-driven modeling will be discussed.

Improving Bayesian credibility intervals for classifier error rates using maximum entropy empirical priors.

OBJECTIVE: Successful use of classifiers that learn to make decisions from a set of patient examples require robust methods for performance estimation. Recently many promising approaches for determination of an upper bound for the error rate of a single classifier have been reported but the Bayesian credibility interval (CI) obtained from a conventional holdout test still delivers one of the tightest bounds. The conventional Bayesian CI becomes unacceptably large in real world applications where the test set sizes are less than a few hundred. The source of this problem is that fact that the CI is determined exclusively by the result on the test examples. In other words, there is no information at all provided by the uniform prior density distribution employed which reflects complete lack of prior knowledge about the unknown error rate. Therefore, the aim of the study reported here was to study a maximum entropy (ME) based approach to improved prior knowledge and Bayesian CIs, demonstrating its relevance for biomedical research and clinical practice. METHOD AND MATERIAL: It is demonstrated how a refined non-uniform prior density distribution can be obtained by means of the ME principle using empirical results from a few designs and tests using non-overlapping sets of examples. RESULTS: Experimental results show that ME based priors improve the CIs when employed to four quite different simulated and two real world data sets. CONCLUSIONS: An empirically derived ME prior seems promising for improving the Bayesian CI for the unknown error rate of a designed classifier.

Revealing cell cycle control by combining model-based detection of periodic expression with novel cis-regulatory descriptors.

BACKGROUND: We address the issue of explaining the presence or absence of phase-specific transcription in budding yeast cultures under different conditions. To this end we use a model-based detector of gene expression periodicity to divide genes into classes depending on their behavior in experiments using different synchronization methods. While computational inference of gene regulatory circuits typically relies on expression similarity (clustering) in order to find classes of potentially co-regulated genes, this method instead takes advantage of known time profile signatures related to the studied process. RESULTS: We explain the regulatory mechanisms of the inferred periodic classes with cis-regulatory descriptors that combine upstream sequence motifs with experimentally determined binding of transcription factors. By systematic statistical analysis we show that periodic classes are best explained by combinations of descriptors rather than single descriptors, and that different combinations correspond to periodic expression in different classes. We also find evidence for additive regulation in that the combinations of cis-regulatory descriptors associated with genes periodically expressed in fewer conditions are frequently subsets of combinations associated with genes periodically expression in more conditions. Finally, we demonstrate that our approach retrieves combinations that are more specific towards known cell-cycle related regulators than the frequently used clustering approach. CONCLUSION: The results illustrate how a model-based approach to expression analysis may be particularly well suited to detect biologically relevant mechanisms. Our new approach makes it possible to provide more refined hypotheses about regulatory mechanisms of the cell cycle and it can easily be adjusted to reveal regulation of other, non-periodic, cellular processes.

Bayesian detection of periodic mRNA time profiles without use of training examples.

BACKGROUND: Detection of periodically expressed genes from microarray data without use of known periodic and non-periodic training examples is an important problem, e.g. for identifying genes regulated by the cell-cycle in poorly characterised organisms. Commonly the investigator is only interested in genes expressed at a particular frequency that characterizes the process under study but this frequency is seldom exactly known. Previously proposed detector designs require access to labelled training examples and do not allow systematic incorporation of diffuse prior knowledge available about the period time. RESULTS: A learning-free Bayesian detector that does not rely on labelled training examples and allows incorporation of prior knowledge about the period time is introduced. It is shown to outperform two recently proposed alternative learning-free detectors on simulated data generated with models that are different from the one used for detector design. Results from applying the detector to mRNA expression time profiles from S. cerevisiae showsthat the genes detected as periodically expressed only contain a small fraction of the cell-cycle genes inferred from mutant phenotype. For example, when the probability of false alarm was equal to 7%, only 12% of the cell-cycle genes were detected. The genes detected as periodically expressed were found to have a statistically significant overrepresentation of known cell-cycle regulated sequence motifs. One known sequence motif and 18 putative motifs, previously not associated with periodic expression, were also over represented. CONCLUSION: In comparison with recently proposed alternative learning-free detectors for periodic gene expression, Bayesian inference allows systematic incorporation of diffuse a priori knowledge about, e.g. the period time. This results in relative performance improvements due to increased robustness against errors in the underlying assumptions. Results from applying the detector to mRNA expression time profiles from S. cerevisiae include several new findings that deserve further experimental studies.