A Multi-center Study on the Reproducibility of Drug-Response Assays in Mammalian Cell Lines

Optimization of a human induced pluripotent stem cell-derived sensory neuron model for the in vitro evaluation of taxane-induced neurotoxicity

Human induced pluripotent stem cell-derived sensory neuron (iPSC-dSN) models are a valuable resource for the study of neurotoxicity but are affected by poor replicability and reproducibility, often due to a lack of optimization. Here, we identify experimental factors related to culture conditions that substantially impact cellular drug response in vitro and determine optimal conditions for improved replicability and reproducibility. Treatment duration and cell seeding density were both found to be significant factors, while cell line differences also contributed to variation. A replicable dose-response in viability was demonstrated after 48-h exposure to docetaxel or paclitaxel. Additionally, a replicable dose-dependent reduction in neurite outgrowth was demonstrated, demonstrating the applicability of the model for the examination of additional phenotypes. Overall, we have established an optimized iPSC-dSN model for the study of taxane-induced neurotoxicity.

Data Valuation with Gradient Similarity

High-quality data is crucial for accurate machine learning and actionable analytics, however, mislabeled or noisy data is a common problem in many domains. Distinguishing low- from high-quality data can be challenging, often requiring expert knowledge and considerable manual intervention. Data Valuation algorithms are a class of methods that seek to quantify the value of each sample in a dataset based on its contribution or importance to a given predictive task. These data values have shown an impressive ability to identify mislabeled observations, and filtering low-value data can boost machine learning performance. In this work, we present a simple alternative to existing methods, termed Data Valuation with Gradient Similarity (DVGS). This approach can be easily applied to any gradient descent learning algorithm, scales well to large datasets, and performs comparably or better than baseline valuation methods for tasks such as corrupted label discovery and noise quantification. We evaluate the DVGS method on tabular, image and RNA expression datasets to show the effectiveness of the method across domains. Our approach has the ability to rapidly and accurately identify low-quality data, which can reduce the need for expert knowledge and manual intervention in data cleaning tasks.

Selection of clinically relevant drug concentrations for in vitro studies of candidates drugs for cancer repurposing: a proposal

Drug repurposing of widely prescribed patent-off and cheap drugs may provide affordable drugs for cancer treatment. Nevertheless, many preclinical studies of cancer drug repurposing candidates use in vitro drug concentrations too high to have clinical relevance. Hence, preclinical studies must use clinically achievable drug concentrations. In this work, several FDA-approved cancer drugs are analyzed regarding the correlation between the drug inhibitory concentrations 50% (IC50) tested in cancer cell lines and their corresponding peak serum concentration (Cmax) and area under the curve (AUC) reported in clinical studies of these drugs. We found that for most targeted cancer drugs, the AUC and not the Cmax is closest to the IC50; therefore, we suggest that the initial testing of candidate drugs for repurposing could select the AUC pharmacokinetic parameter and not the Cmax as the translated drug concentration for in vitro testing. Nevertheless, this is a suggestion only as experimental evidence does not exist to prove this concept. Studies on this issue are required to advance in cancer drug repurposing.

Improving the power of drug toxicity measurements by quantitative nuclei imaging

Imaging-based anticancer drug screens are becoming more prevalent due to development of automated fluorescent microscopes and imaging stations, as well as rapid advancements in image processing software. Automated cell imaging provides many benefits such as their ability to provide high-content data, modularity, dynamics recording and the fact that imaging is the most direct way to access cell viability and cell proliferation. However, currently most publicly available large-scale anticancer drugs screens, such as GDSC, CTRP and NCI-60, provide cell viability data measured by assays based on colorimetric or luminometric measurements of NADH or ATP levels. Although such datasets provide valuable data, it is unclear how well drug toxicity measurements can be integrated with imaging data. Here we explored the relations between drug toxicity data obtained by XTT assay, two quantitative nuclei imaging methods and trypan blue dye exclusion assay using a set of four cancer cell lines with different morphologies and 30 drugs with different mechanisms of action. We show that imaging-based approaches provide high accuracy and the differences between results obtained by different methods highly depend on drug mechanism of action. Selecting AUC metrics over IC50 or comparing data where significantly drugs reduced cell numbers noticeably improves consistency between methods. Using automated cell segmentation protocols we analyzed mitochondria activity in more than 11 thousand drug-treated cells and showed that XTT assay produces unreliable data for CDK4/6, Aurora A, VEGFR and PARP inhibitors due induced cell size growth and increase in individual mitochondria activity. We also explored several benefits of image-based analysis such as ability to monitor cell number dynamics, dissect changes in total and individual mitochondria activity from cell proliferation, and ability to identify chromatin remodeling drugs. Finally, we provide a web tool that allows comparing results obtained by different methods.

Research rigor and reproducibility in research education: A CTSA institutional survey

We assessed the rigor and reproducibility (R&R) activities of institutions funded by the National Center for Advancing Translational Sciences (NCTSA) through a survey and website search (N = 61). Of 50 institutional responses, 84% reported incorporating some form of R&R training, 68% reported devoted R&R training, 30% monitored R&R practices, and 10% incentivized them. Website searches revealed 9 (15%) freely available training curricula, and 7 (11%) institutional programs specifically created to enhance R&R. NCATS should formally integrate R&R principles into its translational science models and institutional requirements.

Reassessing pharmacogenomic cell sensitivity with multilevel statistical models

Pharmacogenomic experiments allow for the systematic testing of drugs, at varying dosage concentrations, to study how genomic markers correlate with cell sensitivity to treatment. The first step in the analysis is to quantify the response of cell lines to variable dosage concentrations of the drugs being tested. The signal to noise in these measurements can be low due to biological and experimental variability. However, the increasing availability of pharmacogenomic studies provides replicated data sets that can be leveraged to gain power. To do this, we formulate a hierarchical mixture model to estimate the drug-specific mixture distributions for estimating cell sensitivity and for assessing drug effect type as either broad or targeted effect. We use this formulation to propose a unified approach that can yield posterior probability of a cell being susceptible to a drug conditional on being a targeted effect or relative effect sizes conditioned on the cell being broad. We demonstrate the usefulness of our approach via case studies. First, we assess pairwise agreements for cell lines/drugs within the intersection of two data sets and confirm the moderate pairwise agreement between many publicly available pharmacogenomic data sets. We then present an analysis that identifies sensitivity to the drug crizotinib for cells harboring EML4-ALK or NPM1-ALK gene fusions, as well as significantly down-regulated cell-matrix pathways associated with crizotinib sensitivity.

PI3K signaling through a biochemical systems lens

Following 3 decades of extensive research into PI3K signaling, it is now evidently clear that the underlying network does not equate to a simple ON/OFF switch. This is best illustrated by the multifaceted nature of the many diseases associated with aberrant PI3K signaling, including common cancers, metabolic disease, and rare developmental disorders. However, we are still far from a complete understanding of the fundamental control principles that govern the numerous phenotypic outputs that are elicited by activation of this well-characterized biochemical signaling network, downstream of an equally diverse set of extrinsic inputs. At its core, this is a question on the role of PI3K signaling in cellular information processing and decision making. Here, we review the determinants of accurate encoding and decoding of growth factor signals and discuss outstanding questions in the PI3K signal relay network. We emphasize the importance of quantitative biochemistry, in close integration with advances in single-cell time-resolved signaling measurements and mathematical modeling.

Targeting pan-essential pathways in cancer with cytotoxic chemotherapy: challenges and opportunities

Cytotoxic chemotherapy remains a key modality in cancer treatment. These therapies, successfully used for decades, continue to transform the lives of cancer patients daily. With the high attrition rate of current oncology drug development, combined with the knowledge that most new therapies do not displace standard-of-care treatments and that many healthcare systems cannot afford these new therapies; cytotoxic chemotherapies will remain an important component of cancer therapy for many years to come. The clinical value of these therapies is often under-appreciated within the pre-clinical cancer research community, where this diverse class of agents are often grouped together as non-specific cellular poisons killing tumor cells based solely upon proliferation rate; however, this is inaccurate. This review article seeks to reaffirm the importance of focusing research efforts upon improving our basic understanding of how these drugs work, discussing their ability to target pan-essential pathways in cancer cells, the relationship of this to the chemotherapeutic window, and highlighting basic science approaches that can be employed towards refining their use.

Exploiting a living biobank to delineate mechanisms underlying disease-specific chromosome instability

Chromosome instability (CIN) is a cancer hallmark that drives tumour heterogeneity, phenotypic adaptation, drug resistance and poor prognosis. High-grade serous ovarian cancer (HGSOC), one of the most chromosomally unstable tumour types, has a 5-year survival rate of only ~30% - largely due to late diagnosis and rapid development of drug resistance, e.g., via CIN-driven ABCB1 translocations. However, CIN is also a cell cycle vulnerability that can be exploited to specifically target tumour cells, illustrated by the success of PARP inhibitors to target homologous recombination deficiency (HRD). However, a lack of appropriate models with ongoing CIN has been a barrier to fully exploiting disease-specific CIN mechanisms. This barrier is now being overcome with the development of patient-derived cell cultures and organoids. In this review, we describe our progress building a Living Biobank of over 120 patient-derived ovarian cancer models (OCMs), predominantly from HGSOC. OCMs are highly purified tumour fractions with extensive proliferative potential that can be analysed at early passage. OCMs have diverse karyotypes, display intra- and inter-patient heterogeneity and mitotic abnormality rates far higher than established cell lines. OCMs encompass a broad-spectrum of HGSOC hallmarks, including a range of p53 alterations and BRCA1/2 mutations, and display drug resistance mechanisms seen in the clinic, e.g., ABCB1 translocations and BRCA2 reversion. OCMs are amenable to functional analysis, drug-sensitivity profiling, and multi-omics, including single-cell next-generation sequencing, and thus represent a platform for delineating HGSOC-specific CIN mechanisms. In turn, our vision is that this understanding will inform the design of new therapeutic strategies.

Paired evaluation of machine-learning models characterizes effects of confounders and outliers

The true accuracy of a machine-learning model is a population-level statistic that cannot be observed directly. In practice, predictor performance is estimated against one or more test datasets, and the accuracy of this estimate strongly depends on how well the test sets represent all possible unseen datasets. Here we describe paired evaluation as a simple, robust approach for evaluating performance of machine-learning models in small-sample biological and clinical studies. We use the method to evaluate predictors of drug response in breast cancer cell lines and of disease severity in patients with Alzheimer's disease, demonstrating that the choice of test data can cause estimates of performance to vary by as much as 20%. We show that paired evaluation makes it possible to identify outliers, improve the accuracy of performance estimates in the presence of known confounders, and assign statistical significance when comparing machine-learning models.

Targeting metabolism by B-raf inhibitors and diclofenac restrains the viability of BRAF-mutated thyroid carcinomas with Hif-1α-mediated glycolytic phenotype

BACKGROUND

B-raf inhibitors (BRAFi) are effective for BRAF-mutated papillary (PTC) and anaplastic (ATC) thyroid carcinomas, although acquired resistance impairs tumour cells' sensitivity and/or limits drug efficacy. Targeting metabolic vulnerabilities is emerging as powerful approach in cancer.

METHODS

In silico analyses identified metabolic gene signatures and Hif-1α as glycolysis regulator in PTC. BRAF-mutated PTC, ATC and control thyroid cell lines were exposed to HIF1A siRNAs or chemical/drug treatments (CoCl2, EGF, HGF, BRAFi, MEKi and diclofenac). Genes/proteins expression, glucose uptake, lactate quantification and viability assays were used to investigate the metabolic vulnerability of BRAF-mutated cells.

RESULTS

A specific metabolic gene signature was identified as a hallmark of BRAF-mutated tumours, which display a glycolytic phenotype, characterised by enhanced glucose uptake, lactate efflux and increased expression of Hif-1α-modulated glycolytic genes. Indeed, Hif-1α stabilisation counteracts the inhibitory effects of BRAFi on these genes and on cell viability. Interestingly, targeting metabolic routes with BRAFi and diclofenac combination we could restrain the glycolytic phenotype and synergistically reduce tumour cells' viability.

CONCLUSION

The identification of a metabolic vulnerability of BRAF-mutated carcinomas and the capacity BRAFi and diclofenac combination to target metabolism open new therapeutic perspectives in maximising drug efficacy and reducing the onset of secondary resistance and drug-related toxicity.

BID expression determines the apoptotic fate of cancer cells after abrogation of the spindle assembly checkpoint by AURKB or TTK inhibitors

BACKGROUND

Drugs targeting the spindle assembly checkpoint (SAC), such as inhibitors of Aurora kinase B (AURKB) and dual specific protein kinase TTK, are in different stages of clinical development. However, cell response to SAC abrogation is poorly understood and there are no markers for patient selection.

METHODS

A panel of 53 tumor cell lines of different origins was used. The effects of drugs were analyzed by MTT and flow cytometry. Copy number status was determined by FISH and Q-PCR; mRNA expression by nCounter and RT-Q-PCR and protein expression by Western blotting. CRISPR-Cas9 technology was used for gene knock-out (KO) and a doxycycline-inducible pTRIPZ vector for ectopic expression. Finally, in vivo experiments were performed by implanting cultured cells or fragments of tumors into immunodeficient mice.

RESULTS

Tumor cells and patient-derived xenografts (PDXs) sensitive to AURKB and TTK inhibitors consistently showed high expression levels of BH3-interacting domain death agonist (BID), while cell lines and PDXs with low BID were uniformly resistant. Gene silencing rendered BID-overexpressing cells insensitive to SAC abrogation while ectopic BID expression in BID-low cells significantly increased sensitivity. SAC abrogation induced activation of CASP-2, leading to cleavage of CASP-3 and extensive cell death only in presence of high levels of BID. Finally, a prevalence study revealed high BID mRNA in 6% of human solid tumors.

CONCLUSIONS

The fate of tumor cells after SAC abrogation is driven by an AURKB/ CASP-2 signaling mechanism, regulated by BID levels. Our results pave the way to clinically explore SAC-targeting drugs in tumors with high BID expression.

A landscape of response to drug combinations in non-small cell lung cancer

Combination of anti-cancer drugs is broadly seen as way to overcome the often-limited efficacy of single agents. The design and testing of combinations are however very challenging. Here we present a uniquely large dataset screening over 5000 targeted agent combinations across 81 non-small cell lung cancer cell lines. Our analysis reveals a profound heterogeneity of response across the tumor models. Notably, combinations very rarely result in a strong gain in efficacy over the range of response observable with single agents. Importantly, gain of activity over single agents is more often seen when co-targeting functionally proximal genes, offering a strategy for designing more efficient combinations. Because combinatorial effect is strongly context specific, tumor specificity should be achievable. The resource provided, together with an additional validation screen sheds light on major challenges and opportunities in building efficacious combinations against cancer and provides an opportunity for training computational models for synergy prediction.

CellMinerCDB: NCATS Is a Web-Based Portal Integrating Public Cancer Cell Line Databases for Pharmacogenomic Explorations

Major advances have been made in the field of precision medicine for treating cancer. However, many open questions remain that need to be answered to realize the goal of matching every patient with cancer to the most efficacious therapy. To facilitate these efforts, we have developed CellMinerCDB: National Center for Advancing Translational Sciences (NCATS; https://discover.nci.nih.gov/rsconnect/cellminercdb_ncats/), which makes available activity information for 2,675 drugs and compounds, including multiple nononcology drugs and 1,866 drugs and compounds unique to the NCATS. CellMinerCDB: NCATS comprises 183 cancer cell lines, with 72 unique to NCATS, including some from previously understudied tissues of origin. Multiple forms of data from different institutes are integrated, including single and combination drug activity, DNA copy number, methylation and mutation, transcriptome, protein levels, histone acetylation and methylation, metabolites, CRISPR, and miscellaneous signatures. Curation of cell lines and drug names enables cross-database (CDB) analyses. Comparison of the datasets is made possible by the overlap between cell lines and drugs across databases. Multiple univariate and multivariate analysis tools are built-in, including linear regression and LASSO. Examples have been presented here for the clinical topoisomerase I (TOP1) inhibitors topotecan and irinotecan/SN-38. This web application provides both substantial new data and significant pharmacogenomic integration, allowing exploration of interrelationships.

SIGNIFICANCE

CellMinerCDB: NCATS provides activity information for 2,675 drugs in 183 cancer cell lines and analysis tools to facilitate pharmacogenomic research and to identify determinants of response.

A Machine Learning Framework for Diagnosing and Predicting the Severity of Coronary Artery Disease

Background

Although machine learning (ML)-based prediction of coronary artery disease (CAD) has gained increasing attention, assessment of the severity of suspected CAD in symptomatic patients remains challenging.

Methods

The training set for this study consisted of 284 retrospective participants, while the test set included 116 prospectively enrolled participants from whom we collected 53 baseline variables and coronary angiography results. The data was pre-processed with outlier processing and One-Hot coding. In the first stage, we constructed a ML model that used baseline information to predict the presence of CAD with a dichotomous model. In the second stage, baseline information was used to construct ML regression models for predicting the severity of CAD. The non-CAD population was included, and two different scores were used as output variables. Finally, statistical analysis and SHAP plot visualization methods were employed to explore the relationship between baseline information and CAD.

Results

The study included 269 CAD patients and 131 healthy controls. The eXtreme Gradient Boosting (XGBoost) model exhibited the best performance amongst the different models for predicting CAD, with an area under the receiver operating characteristic curve of 0.728 (95% CI 0.623-0.824). The main correlates were left ventricular ejection fraction, homocysteine, and hemoglobin (p < 0.001). The XGBoost model performed best for predicting the SYNTAX score, with the main correlates being brain natriuretic peptide (BNP), left ventricular ejection fraction, and glycated hemoglobin (p < 0.001). The main relevant features in the model predictive for the GENSINI score were BNP, high density lipoprotein, and homocysteine (p < 0.001).

Conclusions

This data-driven approach provides a foundation for the risk stratification and severity assessment of CAD.

Clinical Trial Registration

The study was registered in www.clinicaltrials.gov protocol registration system (number NCT05018715).

Morphodynamical cell state description via live-cell imaging trajectory embedding

Time-lapse imaging is a powerful approach to gain insight into the dynamic responses of cells, but the quantitative analysis of morphological changes over time remains challenging. Here, we exploit the concept of "trajectory embedding" to analyze cellular behavior using morphological feature trajectory histories-that is, multiple time points simultaneously, rather than the more common practice of examining morphological feature time courses in single timepoint (snapshot) morphological features. We apply this approach to analyze live-cell images of MCF10A mammary epithelial cells after treatment with a panel of microenvironmental perturbagens that strongly modulate cell motility, morphology, and cell cycle behavior. Our morphodynamical trajectory embedding analysis constructs a shared cell state landscape revealing ligand-specific regulation of cell state transitions and enables quantitative and descriptive models of single-cell trajectories. Additionally, we show that incorporation of trajectories into single-cell morphological analysis enables (i) systematic characterization of cell state trajectories, (ii) better separation of phenotypes, and (iii) more descriptive models of ligand-induced differences as compared to snapshot-based analysis. This morphodynamical trajectory embedding is broadly applicable to the quantitative analysis of cell responses via live-cell imaging across many biological and biomedical applications.

Response of stem cells derived from human exfoliated deciduous teeth to Bio-C Repair and Mineral Trioxide Aggregate Repair HP: Cytotoxicity and gene expression assessment

Background

The aim of this study was to investigate and compare the cytotoxicity and gene expression of Bio-C Repair, Mineral Trioxide Aggregate (MTA) HP Repair, and Biodentine on stem cells derived from exfoliated deciduous teeth.

Materials and Methods

In this in vitro study MTT assay was used to assess the cellular viability at three different dilutions. The gene expression of Runt-related transcription factor 2 (Runx2), alkaline phosphatase (ALP), osteocalcin [OCN], and dentin matrix protein-1 (DMP-1) was measured with real-time polymerase chain reaction after 7 days, 14 days, and 21 days of incubation. One-way analysis of variance and Bonferroni posttest were used for statistical analysis (p=o.o5).

Results

After 72 h of incubation at dilution 1:4, stem cells derived from human exfoliated deciduous teeth (SHEDs) cultivated in Biodentine, followed by Bio-C Repair and MTA Repair HP reported with highest cellular viability. The highest mRNA expression of Runx2, ALP, OCN, and DMP-1 was reported in SHEDs cultured in Biodentine (after 21 days of incubation).

Conclusion

Bio-C Repair and MTA HP Repair are biocompatible and capable of odontogenic differentiation similar to Biodentine when cultured in stem cells derived from exfoliated primary teeth.

Replacing animal-derived components in in vitro test guidelines OECD 455 and 487

The evaluation of single substances or environmental samples for their genotoxic or estrogenic potential is highly relevant for human- and environment-related risk assessment. To examine the effects on a mechanism-specific level, standardized cell-based in vitro methods are widely applied. However, these methods include animal-derived components like fetal bovine serum (FBS) or rat-derived liver homogenate fractions (S9-mixes), which are a source of variability, reduced assay reproducibility and ethical concerns. In our study, we evaluated the adaptation of the cell-based in vitro OECD test guidelines TG 487 (assessment of genotoxicity) and TG 455 (detection of estrogenic activity) to an animal-component-free methodology. Firstly, the human cell lines A549 (for OECD TG 487), ERα-CALUX® and GeneBLAzer™ ERα-UAS-bla GripTite™ (for OECD TG 455) were investigated for growth in a chemically defined medium without the addition of FBS. Secondly, the biotechnological S9-mix ewoS9R was implemented in comparison to the induced rat liver S9 to simulate in vivo metabolism capacities in both OECD test guidelines. As a model compound, Benzo[a]pyrene was used due to its increased genotoxicity and endocrine activity after metabolization. The metabolization of Benzo[a]Pyrene by S9-mixes was examined via chemical analysis. All cell lines (A549, ERα-CALUX® and GeneBLAzer™ Erα-UAS-bla GripTite™) were successfully cultivated in chemically defined media without FBS. The micronucleus assay could not be conducted in chemically defined medium due to formation of cell clusters. The methods for endocrine activity assessment could be conducted in chemically defined media or reduced FBS content, but with decreased assay sensitivity. The biotechnological ewoS9R showed potential to replace rat liver S9 in the micronucleus in FBS-medium with A549 cells and in the ERα-CALUX® assay in FBS- and chemically defined medium. Our study showed promising steps towards an animal-component free toxicity testing. After further improvements, the new methodology could lead to more reproducible and reliable results for risk assessment.

Moving perfusion culture and live-cell imaging from lab to disc: proof of concept toxicity assay with AI-based image analysis

In vitro, cell-based assays are essential in diagnostics and drug development. There are ongoing efforts to establish new technologies that enable real-time detection of cell-drug interaction during culture under flow conditions. Our compact (10 × 10 × 8.5 cm) cell culture and microscope on disc (CMoD) platform aims to decrease the application barriers of existing lab-on-a-chip (LoC) approaches. For the first time in a centrifugal device, (i) cells were cultured for up to six days while a spindle motor facilitated culture medium perfusion, and (ii) an onboard microscope enabled live bright-field imaging of cells while the data wirelessly transmitted to a computer. The quantification of cells from the acquired images was done using artificial intelligence (AI) software. After optimization, the obtained cell viability data from the AI-based image analysis proved to correlate well with data collected from commonly used image analysis software. The CMoD was also suitable for conducting a proof-of-concept toxicity assay with HeLa cells under continuous flow. The half-maximal inhibitory time (IT50) for various concentrations of doxorubicin (DOX) in the case of HeLa cells in flow, was shown to be lower than the IT50 obtained from a static cytotoxicity assay, indicating a faster onset of cell death in flow. The CMoD proved to be easy to handle, enabled cell culture and monitoring without assistance, and is a promising tool for examining the dynamic processes of cells in real-time assays.

Non-uniformity in in vitro drug-induced cytotoxicity as evidenced by differences in IC50 values - implications and way forward

Cell lines have proven indispensable for in vitro experiments and their utility as experimental models range from understanding the fundamental cell functioning to drug discovery. One of the most common utility of cell lines is for in vitro drug testing. Drug testing involves determining the cytotoxic effects of the drugs and such a measurement is expressed as the IC₅₀ values of drugs. Although determination of IC₅₀ values of drugs on cell lines is one of the most common in vitro experimental approaches, a significant amount of variations can be observed in the results obtained from such studies. Although the variations in the IC₅₀ values of a drug on different cells lines can and should vary, the non-uniformity of such results reported from different studies using a particular drug on a specific cell line is a matter of concern. We present the IC₅₀ values of 5 most commonly used drugs 5-fluorouracil, bleomycin, cisplatin, doxorubicin and methotrexate obtained from several in vitro cell line-based studies. Some of the factors which contribute to the non-uniformity of the IC₅₀ values for a particular drug from different studies are discussed as three types of factors, the biological, non-biological and human factors. Also, ways in which such variations can be reduced to obtain universally common, reliable results are presented.

screenwerk: a modular tool for the design and analysis of drug combination screens

MOTIVATION

There is a rapidly growing interest in high-throughput drug combination screening to identify synergizing drug interactions for treatment of various maladies, such as cancer and infectious disease. This creates the need for pipelines that can be used to design such screens, perform quality control on the data and generate data files that can be analyzed by synergy-finding bioinformatics applications.

RESULTS

screenwerk is an open-source, end-to-end modular tool available as an R-package for the design and analysis of drug combination screens. The tool allows for a customized build of pipelines through its modularity and provides a flexible approach to quality control and data analysis. screenwerk is adaptable to various experimental requirements with an emphasis on precision medicine. It can be coupled to other R packages, such as bayesynergy, to identify synergistic and antagonistic drug interactions in cell lines or patient samples. screenwerk is scalable and provides a complete solution for setting up drug sensitivity screens, read raw measurements and consolidate different datasets, perform various types of quality control and analyze, report and visualize the results of drug sensitivity screens.

AVAILABILITY AND IMPLEMENTATION

The R-package and technical documentation is available at https://github.com/Enserink-lab/screenwerk; the R source code is publicly available at https://github.com/Enserink-lab/screenwerk under GNU General Public License v3.0; bayesynergy is accessible at https://github.com/ocbe-uio/bayesynergy. Selected modules are available through Galaxy, an open-source platform for FAIR data analysis at https://oncotools.elixir.no.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Utilization of Cancer Cell Line Screening to Elucidate the Anticancer Activity and Biological Pathways Related to the Ruthenium-Based Therapeutic BOLD-100

BOLD-100 (sodium trans-[tetrachlorobis(1H indazole)ruthenate(III)]) is a ruthenium-based anticancer compound currently in clinical development. The identification of cancer types that show increased sensitivity towards BOLD-100 can lead to improved developmental strategies. Sensitivity profiling can also identify mechanisms of action that are pertinent for the bioactivity of complex therapeutics. Sensitivity to BOLD-100 was measured in a 319-cancer-cell line panel spanning 24 tissues. BOLD-100's sensitivity profile showed variation across the tissue lineages, including increased response in esophageal, bladder, and hematologic cancers. Multiple cancers, including esophageal, bile duct and colon cancer, had higher relative response to BOLD-100 than to cisplatin. Response to BOLD-100 showed only moderate correlation to anticancer compounds in the Genomics of Drug Sensitivity in Cancer (GDSC) database, as well as no clear theme in bioactivity of correlated hits, suggesting that BOLD-100 may have a differentiated therapeutic profile. The genomic modalities of cancer cell lines were modeled against the BOLD-100 sensitivity profile, which revealed that genes related to ribosomal processes were associated with sensitivity to BOLD-100. Machine learning modeling of the sensitivity profile to BOLD-100 and gene expression data provided moderative predictive value. These findings provide further mechanistic understanding around BOLD-100 and support its development for additional cancer types.

Multiplexed and reproducible high content screening of live and fixed cells using Dye Drop

High-throughput measurement of cells perturbed using libraries of small molecules, gene knockouts, or different microenvironmental factors is a key step in functional genomics and pre-clinical drug discovery. However, it remains difficult to perform accurate single-cell assays in 384-well plates, limiting many studies to well-average measurements (e.g., CellTiter-Glo®). Here we describe a public domain Dye Drop method that uses sequential density displacement and microscopy to perform multi-step assays on living cells. We use Dye Drop cell viability and DNA replication assays followed by immunofluorescence imaging to collect single-cell dose-response data for 67 investigational and clinical-grade small molecules in 58 breast cancer cell lines. By separating the cytostatic and cytotoxic effects of drugs computationally, we uncover unexpected relationships between the two. Dye Drop is rapid, reproducible, customizable, and compatible with manual or automated laboratory equipment. Dye Drop improves the tradeoff between data content and cost, enabling the collection of information-rich perturbagen-response datasets.

Systematic replication enables normalization of high-throughput imaging assays

MOTIVATION

High-throughput fluorescent microscopy is a popular class of techniques for studying tissues and cells through automated imaging and feature extraction of hundreds to thousands of samples. Like other high-throughput assays, these approaches can suffer from unwanted noise and technical artifacts that obscure the biological signal. In this work, we consider how an experimental design incorporating multiple levels of replication enables the removal of technical artifacts from such image-based platforms.

RESULTS

We develop a general approach to remove technical artifacts from high-throughput image data that leverages an experimental design with multiple levels of replication. To illustrate the methods, we consider microenvironment microarrays (MEMAs), a high-throughput platform designed to study cellular responses to microenvironmental perturbations. In application to MEMAs, our approach removes unwanted spatial artifacts and thereby enhances the biological signal. This approach has broad applicability to diverse biological assays.

AVAILABILITY AND IMPLEMENTATION

Raw data are on synapse (syn2862345), analysis code is on github: gjhunt/mema_norm, a reproducible Docker image is available on dockerhub: gjhunt/mema_norm.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

A multi-omic analysis of MCF10A cells provides a resource for integrative assessment of ligand-mediated molecular and phenotypic responses

The phenotype of a cell and its underlying molecular state is strongly influenced by extracellular signals, including growth factors, hormones, and extracellular matrix proteins. While these signals are normally tightly controlled, their dysregulation leads to phenotypic and molecular states associated with diverse diseases. To develop a detailed understanding of the linkage between molecular and phenotypic changes, we generated a comprehensive dataset that catalogs the transcriptional, proteomic, epigenomic and phenotypic responses of MCF10A mammary epithelial cells after exposure to the ligands EGF, HGF, OSM, IFNG, TGFB and BMP2. Systematic assessment of the molecular and cellular phenotypes induced by these ligands comprise the LINCS Microenvironment (ME) perturbation dataset, which has been curated and made publicly available for community-wide analysis and development of novel computational methods ( synapse.org/LINCS_MCF10A ). In illustrative analyses, we demonstrate how this dataset can be used to discover functionally related molecular features linked to specific cellular phenotypes. Beyond these analyses, this dataset will serve as a resource for the broader scientific community to mine for biological insights, to compare signals carried across distinct molecular modalities, and to develop new computational methods for integrative data analysis.

Multiple-low-dose therapy: effective killing of high-grade serous ovarian cancer cells with ATR and CHK1 inhibitors

High-grade serous ovarian cancer (HGSOC) is an aggressive disease that typically develops drug resistance, thus novel biomarker-driven strategies are required. Targeted therapy focuses on synthetic lethality—pioneered by PARP inhibition of BRCA1/2-mutant disease. Subsequently, targeting the DNA replication stress response (RSR) is of clinical interest. However, further mechanistic insight is required for biomarker discovery, requiring sensitive models that closely recapitulate HGSOC. We describe an optimized proliferation assay that we use to screen 16 patient-derived ovarian cancer models (OCMs) for response to RSR inhibitors (CHK1i, WEE1i, ATRi, PARGi). Despite genomic heterogeneity characteristic of HGSOC, measurement of OCM proliferation was reproducible and reflected intrinsic tumour-cell properties. Surprisingly, RSR targeting drugs were not interchangeable, as sensitivity to the four inhibitors was not correlated. Therefore, to overcome RSR redundancy, we screened the OCMs with all two-, three- and four-drug combinations in a multiple-low-dose strategy. We found that low-dose CHK1i-ATRi had a potent anti-proliferative effect on 15 of the 16 OCMs, and was synergistic with potential to minimise treatment resistance and toxicity. Low-dose ATRi-CHK1i induced replication catastrophe followed by mitotic exit and post-mitotic arrest or death. Therefore, this study demonstrates the potential of the living biobank of OCMs as a drug discovery platform for HGSOC.

Design considerations of benchtop fluid flow bioreactors for bio-engineered tissue equivalents in vitro

One of the major aims of bio-engineering tissue equivalents in vitro is to create physiologically relevant culture conditions to accurately recreate the cellular microenvironment. This often includes incorporation of factors such as the extracellular matrix, co-culture of multiple cell types and three-dimensional culture techniques. These advanced techniques can recapitulate some of the properties of tissue in vivo, however fluid flow is a key aspect that is often absent. Fluid flow can be introduced into cell and tissue culture using bioreactors, which are becoming increasingly common as we seek to produce increasingly accurate tissue models. Bespoke technology is continuously being developed to tailor systems for specific applications and to allow compatibility with a range of culture techniques. For effective perfusion of a tissue culture many parameters can be controlled, ranging from impacts of the fluid flow such as increased shear stress and mass transport, to potentially unwanted side effects such as temperature fluctuations. A thorough understanding of these properties and their implications on the culture model can aid with a more accurate interpretation of results. Improved and more complete characterisation of bioreactor properties will also lead to greater accuracy when reporting culture conditions in protocols, aiding experimental reproducibility, and allowing more precise comparison of results between different systems. In this review we provide an analysis of the different factors involved in the development of benchtop flow bioreactors and their potential biological impacts across a range of applications.

An automated do-it-yourself system for dynamic stem cell and organoid culture in standard multi-well plates

We present a low-cost, do-it-yourself system for complex mammalian cell culture under dynamically changing medium formulations by integrating conventional multi-well tissue culture plates with simple microfluidic control and system automation. We demonstrate the generation of complex concentration profiles, enabling the investigation of sophisticated input-response relations. We further apply our automated cell-culturing platform to the dynamic stimulation of two widely employed stem-cell-based in vitro models for early mammalian development: the conversion of naive mouse embryonic stem cells into epiblast-like cells and mouse 3D gastruloids. Performing automated medium-switch experiments, we systematically investigate cell fate commitment along the developmental trajectory toward mouse epiblast fate and examine symmetry-breaking, germ layer formation, and cardiac differentiation in mouse 3D gastruloids as a function of time-varying Wnt pathway activation. With these proof-of-principle examples, we demonstrate a highly versatile and scalable tool that can be adapted to specific research questions, experimental demands, and model systems.

Getting Started with LINCS Datasets and Tools

The Library of Integrated Network-based Cellular Signatures (LINCS) was an NIH Common Fund program that aimed to expand our knowledge about human cellular responses to chemical, genetic, and microenvironment perturbations. Responses to perturbations were measured by transcriptomics, proteomics, cellular imaging, and other high content assays. The second phase of the LINCS program, which lasted 7 years, involved the engagement of six data and signature generation centers (DSGCs) and one data coordination and integration center (DCIC). The DSGCs and the DCIC developed several digital resources, including tools, databases, and workflows that aim to facilitate the use of the LINCS data and integrate this data with other publicly available data. The digital resources developed by the DSGCs and the DCIC can be used to gain new biological and pharmacological insights that can lead to the development of novel therapeutics. This protocol provides step-by-step instructions for processing the LINCS data into signatures, and utilizing the digital resources developed by the LINCS consortia for hypothesis generation and knowledge discovery. © 2022 The Authors. Current Protocols published by Wiley Periodicals LLC. Basic Protocol 1: Navigating L1000 tools and data in CLUE.io Basic Protocol 2: Computing signatures from the L1000 data with the CD method Basic Protocol 3: Analyzing lists of differentially expressed genes and querying them against the L1000 data with BioJupies and the Bulk RNA-seq Appyter Basic Protocol 4: Utilizing the L1000FWD resource for drug discovery Basic Protocol 5: KINOMEscan and the KINOMEscan Appyter Basic Protocol 6: LINCS P100 and GCP Proteomics Assays Basic Protocol 7: The LINCS Joint Project (LJP) Basic Protocol 8: The LINCS Data Portals and SigCom LINCS Basic Protocol 9: Creating and analyzing signatures with iLINCS.

Metabolic flux analysis of 3D spheroids reveals significant differences in glucose metabolism from matched 2D cultures of colorectal cancer and pancreatic ductal adenocarcinoma cell lines

BACKGROUND

Most in vitro cancer cell experiments have been performed using 2D models. However, 3D spheroid cultures are increasingly favored for being more representative of in vivo tumor conditions. To overcome the translational challenges with 2D cell cultures, 3D systems better model more complex cell-to-cell contact and nutrient levels present in a tumor, improving our understanding of cancer complexity. Despite this need, there are few reports on how 3D cultures differ metabolically from 2D cultures.

METHODS

Well-described cell lines from colorectal cancer (HCT116 and SW948) and pancreatic ductal adenocarcinoma (Panc-1 and MIA-Pa-Ca-2) were used to investigate metabolism in 3D spheroid models. The metabolic variation under normal glucose conditions were investigated comparing 2D and 3D cultures by metabolic flux analysis and expression of key metabolic proteins.

RESULTS

We find significant differences in glucose metabolism of 3D cultures compared to 2D cultures, both related to glycolysis and oxidative phosphorylation. Spheroids have higher ATP-linked respiration in standard nutrient conditions and higher non-aerobic ATP production in the absence of supplemented glucose. In addition, ATP-linked respiration is significantly inversely correlated with OCR/ECAR (p = 0.0096). Mitochondrial transport protein, TOMM20, expression decreases in all spheroid models compared to 2D, and monocarboxylate transporter (MCT) expression increases in 3 of the 4 spheroid models.

CONCLUSIONS

In this study of CRC and PDAC cell lines, we demonstrate that glucose metabolism in 3D spheroids differs significantly from 2D cultures, both in terms of glycolytic and oxidative phosphorylation metrics. The metabolic phenotype shift from 2D to 3D culture in one cell line is greater than the phenotypic differences between each cell line and tumor source. The results herein emphasize the need to use 3D cell models for investigating nutrient utilization and metabolic flux for a better understanding of tumor metabolism and potential metabolic therapeutic targets.

In vitro Evaluation of Isoniazid Derivatives as Potential Agents Against Drug-Resistant Tuberculosis

The upsurge of multidrug-resistant tuberculosis has toughened the challenge to put an end to this epidemic by 2030. In 2020 the number of deaths attributed to tuberculosis increased as compared to 2019 and newly identified multidrug-resistant tuberculosis cases have been stably close to 3%. Such a context stimulated the search for new and more efficient antitubercular compounds, which culminated in the QSAR-oriented design and synthesis of a series of isoniazid derivatives active against Mycobacterium tuberculosis. From these, some prospective isonicotinoyl hydrazones and isonicotinoyl hydrazides are studied in this work. To evaluate if the chemical derivatizations are generating compounds with a good performance concerning several in vitro assays, their cytotoxicity against human liver HepG2 cells was determined and their ability to bind human serum albumin was thoroughly investigated. For the two new derivatives presented in this study, we also determined their lipophilicity and activity against both the wild type and an isoniazid-resistant strain of Mycobacterium tuberculosis carrying the most prevalent mutation on the katG gene, S315T. All compounds were less cytotoxic than many drugs in clinical use with IC50 values after a 72 h challenge always higher than 25 µM. Additionally, all isoniazid derivatives studied exhibited stronger binding to human serum albumin than isoniazid itself, with dissociation constants in the order of 10−4–10−5 M as opposed to 10−3 M, respectively. This suggests that their transport and half-life in the blood stream are likely improved when compared to the parent compound. Furthermore, our results are a strong indication that the N′ = C bond of the hydrazone derivatives of INH tested is essential for their enhanced activity against the mutant strain of M. tuberculosis in comparison to both their reduced counterparts and INH.

KDM6A Depletion in Breast Epithelial Cells Leads to Reduced Sensitivity to Anticancer Agents and Increased TGFβ Activity

KDM6A, an X chromosome-linked histone lysine demethylase, was reported to be frequently mutated in many tumor types including breast and bladder cancer. However, the functional role of KDM6A is not fully understood. Using MCF10A as a model of non-tumorigenic epithelial breast cells, we found that silencing KDM6A promoted cell migration and transformation demonstrated by the formation of tumor-like acini in three-dimensional culture. KDM6A loss reduced the sensitivity of MCF10A cells to therapeutic agents commonly used to treat patients with triple-negative breast cancer and also induced TGFβ extracellular secretion leading to suppressed expression of cytotoxic genes in normal human CD8+ T cells in vitro. Interestingly, when cells were treated with TGFβ, de novo synthesis of KDM6A protein was suppressed while TGFB1 transcription was enhanced, indicating a TGFβ/KDM6A-negative regulatory axis. Furthermore, both KDM6A deficiency and TGFβ treatment promoted disorganized acinar structures in three-dimensional culture, as well as transcriptional profiles associated with epithelial-to-mesenchymal transition and metastasis, suggesting KDM6A depletion and TGFβ drive tumor progression.

IMPLICATIONS

Our study provides the preclinical rationale for evaluating KDM6A and TGFβ in breast tumor samples as predictors for response to chemo and immunotherapy, informing personalized therapy based on these findings.

"Cells-to-cDNA on Chip": Phenotypic Assessment and Gene Expression Analysis from Live Cells in Nanoliter Volumes Using Droplet Microarrays

In vitro cell-based experiments are particularly important in fundamental biological research. Microscopy-based readouts to identify cellular changes in response to various stimuli are a popular choice, but gene expression analysis is essential to delineate the underlying molecular dynamics in cells. However, cell-based experiments often suffer from interexperimental variation, especially while using different readout methods. Therefore, establishment of platforms that allow for cell screening, along with parallel investigations of morphological features, as well as gene expression levels, is crucial. The droplet microarray (DMA) platform enables cell screening in hundreds of nanoliter droplets. In this study, a "Cells-to-cDNA on Chip" method is developed enabling on-chip mRNA isolation from live cells and conversion to cDNA in individual droplets of 200 nL. This novel method works efficiently to obtain cDNA from different cell numbers, down to single cell per droplet. This is the first established miniaturized on-chip strategy that enables the entire course of cell screening, phenotypic microscopy-based assessments along with mRNA isolation and its conversion to cDNA for gene expression analysis by real-time PCR on an open DMA platform. The principle demonstrated in this study sets a beginning for myriad of possible applications to obtain detailed information about the molecular dynamics in cultured cells.

A Quantitative Systems Pharmacology Perspective on the Importance of Parameter Identifiability

There is an inherent tension in Quantitative Systems Pharmacology (QSP) between the need to incorporate mathematical descriptions of complex physiology and drug targets with the necessity of developing robust, predictive and well-constrained models. In addition to this, there is no “gold standard” for model development and assessment in QSP. Moreover, there can be confusion over terminology such as model and parameter identifiability; complex and simple models; virtual populations; and other concepts, which leads to potential miscommunication and misapplication of methodologies within modeling communities, both the QSP community and related disciplines. This perspective article highlights the pros and cons of using simple (often identifiable) vs. complex (more physiologically detailed but often non-identifiable) models, as well as aspects of parameter identifiability, sensitivity and inference methodologies for model development and analysis. The paper distills the central themes of the issue of identifiability and optimal model size and discusses open challenges.

A Transient Metabolic State in Melanoma Persister Cells Mediated by Chemotherapeutic Treatments

Persistence is a transient state that poses an important health concern in cancer therapy. The mechanisms associated with persister phenotypes are highly diverse and complex, and many aspects of persister cell physiology remain to be explored. We applied a melanoma cell line and panel of chemotherapeutic agents to show that melanoma persister cells are not necessarily preexisting dormant cells; in fact, they may be induced by cancer chemotherapeutics. Our metabolomics analysis and phenotype microarray assays further demonstrated a transient upregulation in Krebs cycle metabolism in persister cells. We also verified that targeting electron transport chain activity can significantly reduce melanoma persister levels. The reported metabolic remodeling feature seems to be a conserved characteristic of melanoma persistence, as it has been observed in various melanoma persister subpopulations derived from a diverse range of chemotherapeutics. Elucidating a global metabolic mechanism that contributes to persister survival and reversible switching will ultimately foster the development of novel cancer therapeutic strategies.

Matching anticancer compounds and tumor cell lines by neural networks with ranking loss

Computational drug sensitivity models have the potential to improve therapeutic outcomes by identifying targeted drug components that are likely to achieve the highest efficacy for a cancer cell line at hand at a therapeutic dose. State of the art drug sensitivity models use regression techniques to predict the inhibitory concentration of a drug for a tumor cell line. This regression objective is not directly aligned with either of these principal goals of drug sensitivity models: We argue that drug sensitivity modeling should be seen as a ranking problem with an optimization criterion that quantifies a drug's inhibitory capacity for the cancer cell line at hand relative to its toxicity for healthy cells. We derive an extension to the well-established drug sensitivity regression model PaccMann that employs a ranking loss and focuses on the ratio of inhibitory concentration and therapeutic dosage range. We find that the ranking extension significantly enhances the model's capability to identify the most effective anticancer drugs for unseen tumor cell profiles based in on in-vitro data.

Assessing the Fidelity of Translation of Nonclinical Assays: A Pharma Perspective

Advances in nonclinical in vitro models, higher throughput approaches and the promise of human-derived preparations require methods to reliably assess the fidelity of translation of assays compared to in vivo models and clinical studies. This review discusses general principles and parameters useful to evaluate the value of nonclinical assays typically used to guide compound progression. I first consider the biological characteristics (including sensitivity and ability to recapitulate relevant responses) of models that form the foundation of an assay based on the questions posed. I then discuss the quantitative assessment of diagnostic performance and assay utility, including sensitivity and specificity, receiver-operator characteristic curves, positive and negative predictive values, likelihood ratios, along with advantages of combining two independent assays. Understanding the strengths and limitations of the biological model employed along with assay performance and context of use is essential to selecting the best assays supporting the best drug candidates.

Using mathematical modeling to estimate time-independent cancer chemotherapy efficacy parameters

One of the primary cancer treatment modalities is chemotherapy. Unfortunately, traditional anti-cancer drugs are often not selective and cause damage to healthy cells, leading to serious side effects for patients. For this reason more targeted therapeutics and drug delivery methods are being developed. The effectiveness of new treatments is initially determined via in vitro cell viability assays, which determine the IC 50  of the drug. However, these assays are known to result in estimates of IC 50  that depend on the measurement time, possibly resulting in over- or under-estimation of the IC 50 . Here, we test the possibility of using cell growth curves and fitting of mathematical models to determine the IC 50  as well as the maximum efficacy of a drug ( ε max ). We measured cell growth of MCF-7 and HeLa cells in the presence of different concentrations of doxorubicin and fit the data with a logistic growth model that incorporates the effect of the drug. This method leads to measurement time-independent estimates of IC 50  and ε max , but we find that ε max  is not identifiable. Further refinement of this methodology is needed to produce uniquely identifiable parameter estimates.

Preclinical Drug Response Metric Based on Cellular Response Phenotype Provides Better Pharmacogenomic Variables with Phenotype Relevance

High-throughput screening of drug response in cultured cell lines is essential for studying therapeutic mechanisms and identifying molecular variants associated with sensitivity to drugs. Assessment of drug response is typically performed by constructing a dose-response curve of viability and summarizing it to a representative, such as IC₅₀. However, this is limited by its dependency on the assay duration and lack of reflections regarding actual cellular response phenotypes. To address these limitations, we consider how each response-phenotype contributes to the overall growth behavior and propose an alternative method of drug response screening that takes into account the cellular response phenotype. In conventional drug response screening methods, the ranking of sensitivity depends on either the metric used to construct the dose-response curve or the representative factor used to summarize the curve. This ambiguity in conventional assessment methods is due to the fact that assessment methods are not consistent with the underlying principles of population dynamics. Instead, the suggested phenotype metrics provide all phenotypic rates of change that shape overall growth behavior at a given dose and better response classification, including the phenotypic mechanism of overall growth inhibition. This alternative high-throughput drug-response screening would improve preclinical pharmacogenomic analysis and the understanding of a therapeutic mechanism of action.

Robotic high-throughput biomanufacturing and functional differentiation of human pluripotent stem cells

Efficient translation of human induced pluripotent stem cells (hiPSCs) requires scalable cell manufacturing strategies for optimal self-renewal and functional differentiation. Traditional manual cell culture is variable and labor intensive, posing challenges for high-throughput applications. Here, we established a robotic platform and automated all essential steps of hiPSC culture and differentiation under chemically defined conditions. This approach allowed rapid and standardized manufacturing of billions of hiPSCs that can be produced in parallel from up to 90 different patient- and disease-specific cell lines. Moreover, we established automated multi-lineage differentiation and generated functional neurons, cardiomyocytes, and hepatocytes. To validate our approach, we compared robotic and manual cell culture operations and performed comprehensive molecular and cellular characterizations (e.g., single-cell transcriptomics, mass cytometry, metabolism, electrophysiology) to benchmark industrial-scale cell culture operations toward building an integrated platform for efficient cell manufacturing for disease modeling, drug screening, and cell therapy.

Combination treatment optimization using a pan-cancer pathway model

The design of efficient combination therapies is a difficult key challenge in the treatment of complex diseases such as cancers. The large heterogeneity of cancers and the large number of available drugs renders exhaustive in vivo or even in vitro investigation of possible treatments impractical. In recent years, sophisticated mechanistic, ordinary differential equation-based pathways models that can predict treatment responses at a molecular level have been developed. However, surprisingly little effort has been put into leveraging these models to find novel therapies. In this paper we use for the first time, to our knowledge, a large-scale state-of-the-art pan-cancer signaling pathway model to identify candidates for novel combination therapies to treat individual cancer cell lines from various tissues (e.g., minimizing proliferation while keeping dosage low to avoid adverse side effects) and populations of heterogeneous cancer cell lines (e.g., minimizing the maximum or average proliferation across the cell lines while keeping dosage low). We also show how our method can be used to optimize the drug combinations used in sequential treatment plans-that is, optimized sequences of potentially different drug combinations-providing additional benefits. In order to solve the treatment optimization problems, we combine the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) algorithm with a significantly more scalable sampling scheme for truncated Gaussian distributions, based on a Hamiltonian Monte-Carlo method. These optimization techniques are independent of the signaling pathway model, and can thus be adapted to find treatment candidates for other complex diseases than cancers as well, as long as a suitable predictive model is available.

Evaluation and comparison of cytotoxicity and bioactivity of chemomechanical caries removal agents on stem cells from human exfoliated deciduous teeth

AIM

To investigate and compare the cytotoxicity and bioactivity of CMCR agents on stem cells derived from exfoliated deciduous teeth.

METHODOLOGY

MTT assay, flow cytometry, Alizarin Red staining and scratch assay were used to assess the cellular viability, apoptosis, calcium matrix deposits and cell migration, respectively. The gene expression of ALP and BMP-2 was measured with RT-PCR. One-way ANOVA and Bonferroni post-test was used for statistical analysis.

RESULTS

0.5% Carisolv showed highest cell proliferation and calcium matrix formation, whereas 0.5% Papacarie reported the highest% live cells and cell migration. The highest mRNA expression of ALP and BMP-2 was reported in SHEDs cultured in 0.5% Papacarie (after 72 h incubation) and 0.5% Carisolv (after 24 h incubation), respectively.

CONCLUSION

CMCR agents are biocompatible and bioactive when cultured in stem cells derived from exfoliated primary teeth.

Longitudinal stability of molecular alterations and drug response profiles in tumor spheroid cell lines enables reproducible analyses

The utility of patient-derived tumor cell lines as experimental models for glioblastoma has been challenged by limited representation of the in vivo tumor biology and low clinical translatability. Here, we report on longitudinal epigenetic and transcriptional profiling of seven glioblastoma spheroid cell line models cultured over an extended period. Molecular profiles were associated with drug response data obtained for 231 clinically used drugs. We show that the glioblastoma spheroid models remained molecularly stable and displayed reproducible drug responses over prolonged culture times of 30 in vitro passages. Integration of gene expression and drug response data identified predictive gene signatures linked to sensitivity to specific drugs, indicating the potential of gene expression-based prediction of glioblastoma therapy response. Our data thus empowers glioblastoma spheroid disease modeling as a useful preclinical assay that may uncover novel therapeutic vulnerabilities and associated molecular alterations.

Functional drug susceptibility testing using single-cell mass predicts treatment outcome in patient-derived cancer neurosphere models

Functional precision medicine aims to match individual cancer patients to optimal treatment through ex vivo drug susceptibility testing on patient-derived cells. However, few functional diagnostic assays have been validated against patient outcomes at scale because of limitations of such assays. Here, we describe a high-throughput assay that detects subtle changes in the mass of individual drug-treated cancer cells as a surrogate biomarker for patient treatment response. To validate this approach, we determined ex vivo response to temozolomide in a retrospective cohort of 69 glioblastoma patient-derived neurosphere models with matched patient survival and genomics. Temozolomide-induced changes in cell mass distributions predict patient overall survival similarly to O⁶-methylguanine-DNA methyltransferase (MGMT) promoter methylation and may aid in predictions in gliomas with mismatch-repair variants of unknown significance, where MGMT is not predictive. Our findings suggest cell mass is a promising functional biomarker for cancers and drugs that lack genomic biomarkers.

Responsiveness to perturbations is a hallmark of transcription factors that maintain cell identity in vitro

Identifying the particular transcription factors that maintain cell type in vitro is important for manipulating cell type. Identifying such transcription factors by their cell-type-specific expression or their involvement in developmental regulation has had limited success. We hypothesized that because cell type is often resilient to perturbations, the transcriptional response to perturbations would reveal identity-maintaining transcription factors. We developed perturbation panel profiling (P3) as a framework for perturbing cells across many conditions and measuring gene expression responsiveness transcriptome-wide. In human iPSC-derived cardiac myocytes, P3 showed that transcription factors important for cardiac myocyte differentiation and maintenance were among the most frequently upregulated (most responsive). We reasoned that one function of responsive genes may be to maintain cellular identity. We identified responsive transcription factors in fibroblasts using P3 and found that suppressing their expression led to enhanced reprogramming. We propose that responsiveness to perturbations is a property of transcription factors that help maintain cellular identity in vitro. A record of this paper's transparent peer review process is included in the supplemental information.

Evaluation of connectivity map shows limited reproducibility in drug repositioning

The Connectivity Map (CMap) is a popular resource designed for data-driven drug repositioning using a large transcriptomic compendium. However, evaluations of its performance are limited. We used two iterations of CMap (CMap 1 and 2) to assess their comparability and reliability. We queried CMap 2 with CMap 1-derived signatures, expecting CMap 2 would highly prioritize the queried compounds; the success rate was 17%. Analysis of previously published prioritizations yielded similar results. Low recall is caused by low differential expression (DE) reproducibility both between CMaps and within each CMap. DE strength was predictive of reproducibility, and is influenced by compound concentration and cell-line responsiveness. Reproducibility of CMap 2 sample expression levels was also lower than expected. We attempted to identify the "better" CMap by comparison with a third dataset, but they were mutually discordant. Our findings have implications for CMap usage and we suggest steps for investigators to limit false positives.

Drug sensitivity prediction from cell line-based pharmacogenomics data: guidelines for developing machine learning models

The goal of precision oncology is to tailor treatment for patients individually using the genomic profile of their tumors. Pharmacogenomics datasets such as cancer cell lines are among the most valuable resources for drug sensitivity prediction, a crucial task of precision oncology. Machine learning methods have been employed to predict drug sensitivity based on the multiple omics data available for large panels of cancer cell lines. However, there are no comprehensive guidelines on how to properly train and validate such machine learning models for drug sensitivity prediction. In this paper, we introduce a set of guidelines for different aspects of training gene expression-based predictors using cell line datasets. These guidelines provide extensive analysis of the generalization of drug sensitivity predictors and challenge many current practices in the community including the choice of training dataset and measure of drug sensitivity. The application of these guidelines in future studies will enable the development of more robust preclinical biomarkers.

Recent advances in drug repurposing using machine learning

Drug repurposing aims to find new uses for already existing and approved drugs. We now provide a brief overview of recent developments in drug repurposing using machine learning alongside other computational approaches for comparison. We also highlight several applications for cancer using kinase inhibitors, Alzheimer's disease as well as COVID-19.

Human Pluripotent Stem Cells for High-Throughput Drug Screening and Characterization of Small Molecules

Human pluripotent stem cells (hPSCs), such as induced pluripotent stem cells (iPSCs), hold great promise for drug discovery, toxicology studies, and regenerative medicine. Here, we describe standardized protocols and experimental procedures that combine automated cell culture for scalable production of hPSCs with quantitative high-throughput screening (qHTS) in miniaturized 384-well plates. As a proof of principle, we established dose-response assessments and determined optimal concentrations of 12 small molecule compounds that are commonly used in the stem cell field. Multi-parametric analysis of readouts from diverse assays including cell viability, mitochondrial membrane potential, plasma membrane integrity, and ATP production was used to distinguish normal biological responses from cellular stress induced by small molecule treatment. Collectively, the establishment of integrated workflows for cell manufacturing, qHTS, high-content imaging, and data analysis provides an end-to-end platform for industrial-scale projects and should leverage the drug discovery process using hPSC-derived cell types.