Computational models for predicting drug responses in cancer research

MCMVDRP: a multi-channel multi-view deep learning framework for cancer drug response prediction

Drug therapy remains the primary approach to treating tumours. Variability among cancer patients, including variations in genomic profiles, often results in divergent therapeutic responses to analogous anti-cancer drug treatments within the same cohort of cancer patients. Hence, predicting the drug response by analysing the genomic profile characteristics of individual patients holds significant research importance. With the notable progress in machine learning and deep learning, many effective methods have emerged for predicting drug responses utilizing features from both drugs and cell lines. However, these methods are inadequate in capturing a sufficient number of features inherent to drugs. Consequently, we propose a representational approach for drugs that incorporates three distinct types of features: the molecular graph, the SMILE strings, and the molecular fingerprints. In this study, a novel deep learning model, named MCMVDRP, is introduced for the prediction of cancer drug responses. In our proposed model, an amalgamation of these extracted features is performed, followed by the utilization of fully connected layers to predict the drug response based on the IC50 values. Experimental results demonstrate that the presented model outperforms current state-of-the-art models in performance.

Bactericidal, anti-hemolytic, and anticancerous activities of phytofabricated silver nanoparticles of glycine max seeds

Introduction

Soybean is a rich source of bioactive components with good nutritional support and is easily available. In the treatment of cancer, green synthesis of silver nanoparticles (AgNPs) from plant-based samples has gained attentions due to its potency and feasibility. In the present study, using soybean extracts (GM), silver nanoparticles are synthesized and analyzed for their anticancer potency.

Methods

The synthesized GM-AgNPs were characterized via UV-Vis spectroscopy, Fourier transform-infrared (FT-IR), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and energy-dispersive X-ray (EDX) techniques for further analysis. Antibacterial activity was evaluated using the disc method and anti-hemolysis activity using the in vitro method, followed by anticancer property evaluation by cytotoxicity, cell migration, apoptosis, and cell cycle.

Results and discussion

Our results showed that the synthesized GM-AgNPs were spiral-shaped with a size range of 5-50 nm. The antibacterial activity against Staphylococcus aureus and Klebsiella pneumoniae showed the maximum zone of inhibition at 250 μg/mL in comparison with gentamicin. On exploring the anti-hemolysis efficiency, at 200 μg/mL, GM-AgNPs showed no hemolysis in comparison to the extract which showed 40% hemolysis. On analysis of GM-AgNPs against the breast cancer cell line, the nanoparticles displayed the IC50 value of 74.04 μg/mL. Furthermore, at the IC50 concentration, cancer cell migration was reduced. The mechanism of action of GM-AgNPs confirmed the initiation of apoptosis and cell cycle arrest in the sub-G0/G1 (growth phase) phase by 48.19%. In gene expression and protein expression analyses, Bax and Bcl-2 were altered to those of normal physiology.

Computer-aided drug discovery strategies for novel therapeutics for prostate cancer leveraging next-generating sequencing data

INTRODUCTION

Prostate cancer (PC) is the most common malignancy and accounts for a significant proportion of cancer deaths among men. Although initial therapy success can often be observed in patients diagnosed with localized PC, many patients eventually develop disease recurrence and metastasis. Without effective treatments, patients with aggressive PC display very poor survival. To curb the current high mortality rate, many investigations have been carried out to identify efficacious therapeutics. Compared to de novo drug designs, computational methods have been widely employed to offer actionable drug predictions in a fast and cost-efficient way. Particularly, powered by an increasing availability of next-generation sequencing molecular profiles from PC patients, computer-aided approaches can be tailored to screen for candidate drugs.

AREAS COVERED

Herein, the authors review the recent advances in computational methods for drug discovery utilizing molecular profiles from PC patients. Given the uniqueness in PC therapeutic needs, they discuss in detail the drug discovery goals of these studies, highlighting their translational values for clinically impactful drug nomination.

EXPERT OPINION

Evolving molecular profiling techniques may enable new perspectives for computer-aided approaches to offer drug candidates for different tumor microenvironments. With ongoing efforts to incorporate new compounds into large-scale high-throughput screens, the authors envision continued expansion of drug candidate pools.

drGAT: Attention-Guided Gene Assessment of Drug Response Utilizing a Drug-Cell-Gene Heterogeneous Network

Drug development is a lengthy process with a high failure rate. Increasingly, machine learning is utilized to facilitate the drug development processes. These models aim to enhance our understanding of drug characteristics, including their activity in biological contexts. However, a major challenge in drug response (DR) prediction is model interpretability as it aids in the validation of findings. This is important in biomedicine, where models need to be understandable in comparison with established knowledge of drug interactions with proteins. drGAT, a graph deep learning model, leverages a heterogeneous graph composed of relationships between proteins, cell lines, and drugs. drGAT is designed with two objectives: DR prediction as a binary sensitivity prediction and elucidation of drug mechanism from attention coefficients. drGAT has demonstrated superior performance over existing models, achieving 78% accuracy (and precision), and 76% F1 score for 269 DNA-damaging compounds of the NCI60 drug response dataset. To assess the model's interpretability, we conducted a review of drug-gene co-occurrences in Pubmed abstracts in comparison to the top 5 genes with the highest attention coefficients for each drug. We also examined whether known relationships were retained in the model by inspecting the neighborhoods of topoisomerase-related drugs. For example, our model retained TOP1 as a highly weighted predictive feature for irinotecan and topotecan, in addition to other genes that could potentially be regulators of the drugs. Our method can be used to accurately predict sensitivity to drugs and may be useful in the identification of biomarkers relating to the treatment of cancer patients.

An integrated framework for prognosis prediction and drug response modeling in colorectal liver metastasis drug discovery

BACKGROUND

Colorectal cancer (CRC) is the third most prevalent cancer globally, and liver metastasis (CRLM) is the primary cause of death. Hence, it is essential to discover novel prognostic biomarkers and therapeutic drugs for CRLM.

METHODS

This study developed two liver metastasis-associated prognostic signatures based on differentially expressed genes (DEGs) in CRLM. Additionally, we employed an interpretable deep learning model utilizing drug sensitivity databases to identify potential therapeutic drugs for high-risk CRLM patients. Subsequently, in vitro and in vivo experiments were performed to verify the efficacy of these compounds.

RESULTS

These two prognostic models exhibited superior performance compared to previously reported ones. Obatoclax, a BCL-2 inhibitor, showed significant differential responses between high and low risk groups classified by prognostic models, and demonstrated remarkable effectiveness in both Transwell assay and CT26 colorectal liver metastasis mouse model.

CONCLUSIONS

This study highlights the significance of developing specialized prognostication approaches and investigating effective therapeutic drugs for patients with CRLM. The application of a deep learning drug response model provides a new drug discovery strategy for translational medicine in precision oncology.

Cancer drug sensitivity prediction from routine histology images

Drug sensitivity prediction models can aid in personalising cancer therapy, biomarker discovery, and drug design. Such models require survival data from randomised controlled trials which can be time consuming and expensive. In this proof-of-concept study, we demonstrate for the first time that deep learning can link histological patterns in whole slide images (WSIs) of Haematoxylin & Eosin (H&E) stained breast cancer sections with drug sensitivities inferred from cell lines. We employ patient-wise drug sensitivities imputed from gene expression-based mapping of drug effects on cancer cell lines to train a deep learning model that predicts patients' sensitivity to multiple drugs from WSIs. We show that it is possible to use routine WSIs to predict the drug sensitivity profile of a cancer patient for a number of approved and experimental drugs. We also show that the proposed approach can identify cellular and histological patterns associated with drug sensitivity profiles of cancer patients.

Deep learning methods for drug response prediction in cancer: Predominant and emerging trends

Cancer claims millions of lives yearly worldwide. While many therapies have been made available in recent years, by in large cancer remains unsolved. Exploiting computational predictive models to study and treat cancer holds great promise in improving drug development and personalized design of treatment plans, ultimately suppressing tumors, alleviating suffering, and prolonging lives of patients. A wave of recent papers demonstrates promising results in predicting cancer response to drug treatments while utilizing deep learning methods. These papers investigate diverse data representations, neural network architectures, learning methodologies, and evaluations schemes. However, deciphering promising predominant and emerging trends is difficult due to the variety of explored methods and lack of standardized framework for comparing drug response prediction models. To obtain a comprehensive landscape of deep learning methods, we conducted an extensive search and analysis of deep learning models that predict the response to single drug treatments. A total of 61 deep learning-based models have been curated, and summary plots were generated. Based on the analysis, observable patterns and prevalence of methods have been revealed. This review allows to better understand the current state of the field and identify major challenges and promising solution paths.

Evolution of precision oncology-guided treatment paradigms

Cancer treatment is gradually evolving from the classical use of nonspecific cytotoxic drugs targeting generic mechanisms of cell growth and proliferation. Instead, new "patient-specific treatment paradigms" that are based on an individual patient's tumor-specific molecular features are emerging, and these include "druggable" genomic alterations such as oncogenic driver mutations, downstream activities of cancer-signaling pathways, and the expression of specific genes involved in tumorigenesis and cancer progression. This evolving landscape of making evidence-based treatment decisions forms the foundation of precision oncology, which aims to deliver "the right drug, to the right patient and at the right time". The long-term vision for this approach is to maximize the treatment efficacy while minimizing exposure to ineffective therapy and reducing co-morbidity-related side effects. Successful clinical translation and implementation of this vision have the potential to revolutionize treatment paradigms from predominantly reactive, to more evidence-based, proactive and predictive care. In this article, we review the past and current approaches in precision oncology, and describe their remarkable power and limitations. We also speculate on the evolution of newly emerging methodologies of the future that can be used to address some of the key challenges associated with the existing paradigms. This article is categorized under: Cancer > Genetics/Genomics/Epigenetics Cancer > Molecular and Cellular Physiology Cancer > Computational Models.

The Immune Signatures data resource, a compendium of systems vaccinology datasets

Vaccines are among the most cost-effective public health interventions for preventing infection-induced morbidity and mortality, yet much remains to be learned regarding the mechanisms by which vaccines protect. Systems immunology combines traditional immunology with modern 'omic profiling techniques and computational modeling to promote rapid and transformative advances in vaccinology and vaccine discovery. The NIH/NIAID Human Immunology Project Consortium (HIPC) has leveraged systems immunology approaches to identify molecular signatures associated with the immunogenicity of many vaccines. However, comparative analyses have been limited by the distributed nature of some data, potential batch effects across studies, and the absence of multiple relevant studies from non-HIPC groups in ImmPort. To support comparative analyses across different vaccines, we have created the Immune Signatures Data Resource, a compendium of standardized systems vaccinology datasets. This data resource is available through ImmuneSpace, along with code to reproduce the processing and batch normalization starting from the underlying study data in ImmPort and the Gene Expression Omnibus (GEO). The current release comprises 1405 participants from 53 cohorts profiling the response to 24 different vaccines. This novel systems vaccinology data release represents a valuable resource for comparative and meta-analyses that will accelerate our understanding of mechanisms underlying vaccine responses.

NeRD: a multichannel neural network to predict cellular response of drugs by integrating multidimensional data

BACKGROUND

Considering the heterogeneity of tumors, it is a key issue in precision medicine to predict the drug response of each individual. The accumulation of various types of drug informatics and multi-omics data facilitates the development of efficient models for drug response prediction. However, the selection of high-quality data sources and the design of suitable methods remain a challenge.

METHODS

In this paper, we design NeRD, a multidimensional data integration model based on the PRISM drug response database, to predict the cellular response of drugs. Four feature extractors, including drug structure extractor (DSE), molecular fingerprint extractor (MFE), miRNA expression extractor (mEE), and copy number extractor (CNE), are designed for different types and dimensions of data. A fully connected network is used to fuse all features and make predictions.

RESULTS

Experimental results demonstrate the effective integration of the global and local structural features of drugs, as well as the features of cell lines from different omics data. For all metrics tested on the PRISM database, NeRD surpassed previous approaches. We also verified that NeRD has strong reliability in the prediction results of new samples. Moreover, unlike other algorithms, when the amount of training data was reduced, NeRD maintained stable performance.

CONCLUSIONS

NeRD's feature fusion provides a new idea for drug response prediction, which is of great significance for precise cancer treatment.

Molecular pathways enhance drug response prediction using transfer learning from cell lines to tumors and patient-derived xenografts

Computational models have been successful in predicting drug sensitivity in cancer cell line data, creating an opportunity to guide precision medicine. However, translating these models to tumors remains challenging. We propose a new transfer learning workflow that transfers drug sensitivity predicting models from large-scale cancer cell lines to both tumors and patient derived xenografts based on molecular pathways derived from genomic features. We further compute feature importance to identify pathways most important to drug response prediction. We obtained good performance on tumors (AUROC = 0.77) and patient derived xenografts from triple negative breast cancers (RMSE = 0.11). Using feature importance, we highlight the association between ER-Golgi trafficking pathway in everolimus sensitivity within breast cancer patients and the role of class II histone deacetylases and interlukine-12 in response to drugs for triple-negative breast cancer. Pathway information support transfer of drug response prediction models from cell lines to tumors and can provide biological interpretation underlying the predictions, serving as a steppingstone towards usage in clinical setting.

Growth‐rate model predicts in vivo tumor response from in vitro data

A major challenge in oncology drug development is to elucidate why drugs that show promising results in cancer cell lines in vitro fail in mouse studies or human trials. One of the fundamental steps toward solving this problem is to better predict how in vitro potency translates into in vivo efficacy. A common approach to infer whether a model will respond in vivo is based on in vitro half‐maximal inhibitory concentration values (IC₅₀), but yields limited quantitative comparison between cell lines and drugs, potentially because cell division and death rates differ between cell lines and in vivo models. Other methods based either on mechanistic modeling or machine learning require molecular insights or extensive training data, limiting their use for early drug development. To address these challenges, we propose a mathematical model integrating in vitro growth rate inhibition values with pharmacokinetic parameters to estimate in vivo drug response. Upon calibration with a drug‐specific factor, our model yields precise estimates of tumor growth rate inhibition for in vivo studies based on in vitro data. We then demonstrate how our model can be used to study dosing schedules and perform sensitivity analyses. In addition, it provides meaningful metrics to assess association with genotypes and guide clinical trial design. By relying on commonly collected data, our approach shows great promise for optimizing drug development, better characterizing the efficacy of novel molecules targeting proliferation, and identifying more robust biomarkers of sensitivity while limiting the number of in vivo experiments.

Exploring synthetic lethal network for the precision treatment of clear cell renal cell carcinoma

The emerging targeted therapies have revolutionized the treatment of advanced clear cell renal cell carcinoma (ccRCC) over the past 15 years. Nevertheless, lack of personalized treatment limits the development of effective clinical guidelines and improvement of patient prognosis. In this study, large-scale genomic profiles from ccRCC cohorts were explored for integrative analysis. A credible method was developed to identify synthetic lethality (SL) pairs and a list of 72 candidate pairs was determined, which might be utilized to selectively eliminate tumors with genetic aberrations using SL partners of specific mutations. Further analysis identified BRD4 and PRKDC as novel medical targets for patients with BAP1 mutations. After mapping these target genes to the comprehensive drug datasets, two agents (BI-2536 and PI-103) were found to have considerable therapeutic potentials in the BAP1 mutant tumors. Overall, our findings provided insight into the overview of ccRCC mutation patterns and offered novel opportunities for improving individualized cancer treatment.

NSUN2 Promotes Tumor Progression and Regulates Immune Infiltration in Nasopharyngeal Carcinoma

Nasopharyngeal carcinoma (NPC) is one of the most common malignancies in the head and neck with a complex etiology, such as environmental factors, genetic factors, and Epstein-Barr virus infection. The NOP2/Sun domain family, member 2 (NSUN2) is a methyltransferase of m5C methylation modification that has been reported to be involved in the occurrence and progression of various tumors, but its role in NPC remains unclear. In this study, we found that NSUN2 was upregulated in NPC and predicted a poor prognosis for NPC patients in both GEO datasets and our tissue microarrays containing 125 NPC tissues. Next, we demonstrated that NSUN2 promoted the proliferation, migration, and invasion of NPC cells in vitro. Additionally, the differential expression genes between NSUN2-high and low expression patients were mainly enriched in multi-immune cell activation and proliferation. Furthermore, NSUN2 negatively regulates immune cell infiltration in the tumor microenvironment (TME) of NPC, which indicates that the NSUN2 level may be negatively correlated with the sensitivity of immunotherapy and chemotherapy. In conclusion, our findings highlight that NSUN2 might act as an important oncogene involved in NPC progression and serve as a potential biomarker to predict poor prognosis and drug sensitivity of NPC patients.

Pan-Cancer Prediction of Cell-Line Drug Sensitivity Using Network-Based Methods

The development of reliable predictive models for individual cancer cell lines to identify an optimal cancer drug is a crucial step to accelerate personalized medicine, but vast differences in cancer cell lines and drug characteristics make it quite challenging to develop predictive models that result in high predictive power and explain the similarity of cell lines or drugs. Our study proposes a novel network-based methodology that breaks the problem into smaller, more interpretable problems to improve the predictive power of anti-cancer drug responses in cell lines. For the drug-sensitivity study, we used the GDSC database for 915 cell lines and 200 drugs. The theory of optimal mass transport was first used to separately cluster cell lines and drugs, using gene-expression profiles and extensive cheminformatic drug features, represented in a form of data networks. To predict cell-line specific drug responses, random forest regression modeling was separately performed for each cell-line drug cluster pair. Post-modeling biological analysis was further performed to identify potential biological correlates associated with drug responses. The network-based clustering method resulted in 30 distinct cell-line drug cluster pairs. Predictive modeling on each cell-line-drug cluster outperformed alternative computational methods in predicting drug responses. We found that among the four drugs top-ranked with respect to prediction performance, three targeted the PI3K/mTOR signaling pathway. Predictive modeling on clustered subsets of cell lines and drugs improved the prediction accuracy of cell-line specific drug responses. Post-modeling analysis identified plausible biological processes associated with drug responses.

Dissecting the Genome for Drug Response Prediction

The prediction of the cancer cell lines sensitivity to a specific treatment is one of the current challenges in precision medicine. With omics and pharmacogenomics data being available for over 1000 cancer cell lines, several machine learning and deep learning algorithms have been proposed for drug sensitivity prediction. However, deciding which omics data to use and which computational methods can efficiently incorporate data from different sources is the challenge which several research groups are working on. In this review, we summarize recent advances in the representative computational methods that have been developed in the last 2 years on three public datasets: COSMIC, CCLE, NCI-60. These methods aim to improve the prediction of the cancer cell lines sensitivity to a given treatment by incorporating drug's chemical information in the input or using a priori feature selection. Finally, we discuss the latest published method which aims to improve the prediction of clinical drug response of real patients starting from cancer cell line molecular profiles.

Predicting heterogeneity in clone-specific therapeutic vulnerabilities using single-cell transcriptomic signatures

While understanding molecular heterogeneity across patients underpins precision oncology, there is increasing appreciation for taking intra-tumor heterogeneity into account. Based on large-scale analysis of cancer omics datasets, we highlight the importance of intra-tumor transcriptomic heterogeneity (ITTH) for predicting clinical outcomes. Leveraging single-cell RNA-seq (scRNA-seq) with a recommender system (CaDRReS-Sc), we show that heterogeneous gene-expression signatures can predict drug response with high accuracy (80%). Using patient-proximal cell lines, we established the validity of CaDRReS-Sc's monotherapy (Pearson r>0.6) and combinatorial predictions targeting clone-specific vulnerabilities (>10% improvement). Applying CaDRReS-Sc to rapidly expanding scRNA-seq compendiums can serve as in silico screen to accelerate drug-repurposing studies. Availability: https://github.com/CSB5/CaDRReS-Sc .

Predicting Cancer Drug Response In Vivo by Learning an Optimal Feature Selection of Tumour Molecular Profiles

(1) Background: Inter-tumour heterogeneity is one of cancer’s most fundamental features. Patient stratification based on drug response prediction is hence needed for effective anti-cancer therapy. However, single-gene markers of response are rare and/or may fail to achieve a significant impact in the clinic. Machine Learning (ML) is emerging as a particularly promising complementary approach to precision oncology. (2) Methods: Here we leverage comprehensive Patient-Derived Xenograft (PDX) pharmacogenomic data sets with dimensionality-reducing ML algorithms with this purpose. (3) Results: Combining multiple gene alterations via ML leads to better discrimination between sensitive and resistant PDXs in 19 of the 26 analysed cases. Highly predictive ML models employing concise gene lists were found for three cases: paclitaxel (breast cancer), binimetinib (breast cancer) and cetuximab (colorectal cancer). Interestingly, each of these multi-gene ML models identifies some treatment-responsive PDXs not harbouring the best actionable mutation for that case. Thus, ML multi-gene predictors generally have much fewer false negatives than the corresponding single-gene marker. (4) Conclusions: As PDXs often recapitulate clinical outcomes, these results suggest that many more patients could benefit from precision oncology if ML algorithms were also applied to existing clinical pharmacogenomics data, especially those algorithms generating classifiers combining data-selected gene alterations.

Predicting drug sensitivity of cancer cells based on DNA methylation levels

Cancer cell lines, which are cell cultures derived from tumor samples, represent one of the least expensive and most studied preclinical models for drug development. Accurately predicting drug responses for a given cell line based on molecular features may help to optimize drug-development pipelines and explain mechanisms behind treatment responses. In this study, we focus on DNA methylation profiles as one type of molecular feature that is known to drive tumorigenesis and modulate treatment responses. Using genome-wide, DNA methylation profiles from 987 cell lines in the Genomics of Drug Sensitivity in Cancer database, we used machine-learning algorithms to evaluate the potential to predict cytotoxic responses for eight anti-cancer drugs. We compared the performance of five classification algorithms and four regression algorithms representing diverse methodologies, including tree-, probability-, kernel-, ensemble-, and distance-based approaches. We artificially subsampled the data to varying degrees, aiming to understand whether training based on relatively extreme outcomes would yield improved performance. When using classification or regression algorithms to predict discrete or continuous responses, respectively, we consistently observed excellent predictive performance when the training and test sets consisted of cell-line data. Classification algorithms performed best when we trained the models using cell lines with relatively extreme drug-response values, attaining area-under-the-receiver-operating-characteristic-curve values as high as 0.97. The regression algorithms performed best when we trained the models using the full range of drug-response values, although this depended on the performance metrics we used. Finally, we used patient data from The Cancer Genome Atlas to evaluate the feasibility of classifying clinical responses for human tumors based on models derived from cell lines. Generally, the algorithms were unable to identify patterns that predicted patient responses reliably; however, predictions by the Random Forests algorithm were significantly correlated with Temozolomide responses for low-grade gliomas.

Mouse models of glioblastoma for the evaluation of novel therapeutic strategies

Glioblastoma (GBM) is an incurable brain tumor with a median survival of approximately 15 months despite an aggressive standard of care that includes surgery, chemotherapy, and ionizing radiation. Mouse models have advanced our understanding of GBM biology and the development of novel therapeutic strategies for GBM patients. However, model selection is crucial when testing developmental therapeutics, and each mouse model of GBM has unique advantages and disadvantages that can influence the validity and translatability of experimental results. To shed light on this process, we discuss the strengths and limitations of 3 types of mouse GBM models in this review: syngeneic models, genetically engineered mouse models, and xenograft models, including traditional xenograft cell lines and patient-derived xenograft models.

Mapping the landscape of synthetic lethal interactions in liver cancer

Almost all the current therapies against liver cancer are based on the "one size fits all" principle and offer only limited survival benefit. Fortunately, synthetic lethality (SL) may provide an alternate route towards individualized therapy in liver cancer. The concept that simultaneous losses of two genes are lethal to a cell while a single loss is non-lethal can be utilized to selectively eliminate tumors with genetic aberrations. Methods: To infer liver cancer-specific SL interactions, we propose a computational pipeline termed SiLi (statistical inference-based synthetic lethality identification) that incorporates five inference procedures. Based on large-scale sequencing datasets, SiLi analysis was performed to identify SL interactions in liver cancer. Results: By SiLi analysis, a total of 272 SL pairs were discerned, which included 209 unique target candidates. Among these, polo-like kinase 1 (PLK1) was considered to have considerable therapeutic potential. Further computational and experimental validation of the SL pair TP53-PLK1 demonstrated that inhibition of PLK1 could be a novel therapeutic strategy specifically targeting those patients with TP53-mutant liver tumors. Conclusions: In this study, we report a comprehensive analysis of synthetic lethal interactions of liver cancer. Our findings may open new possibilities for patient-tailored therapeutic interventions in liver cancer.

Interpretable deep recommender system model for prediction of kinase inhibitor efficacy across cancer cell lines

Computational models for drug sensitivity prediction have the potential to significantly improve personalized cancer medicine. Drug sensitivity assays, combined with profiling of cancer cell lines and drugs become increasingly available for training such models. Multiple methods were proposed for predicting drug sensitivity from cancer cell line features, some in a multi-task fashion. So far, no such model leveraged drug inhibition profiles. Importantly, multi-task models require a tailored approach to model interpretability. In this work, we develop DEERS, a neural network recommender system for kinase inhibitor sensitivity prediction. The model utilizes molecular features of the cancer cell lines and kinase inhibition profiles of the drugs. DEERS incorporates two autoencoders to project cell line and drug features into 10-dimensional hidden representations and a feed-forward neural network to combine them into response prediction. We propose a novel interpretability approach, which in addition to the set of modeled features considers also the genes and processes outside of this set. Our approach outperforms simpler matrix factorization models, achieving R [Formula: see text] 0.82 correlation between true and predicted response for the unseen cell lines. The interpretability analysis identifies 67 biological processes that drive the cell line sensitivity to particular compounds. Detailed case studies are shown for PHA-793887, XMD14-99 and Dabrafenib.

TUGDA: task uncertainty guided domain adaptation for robust generalization of cancer drug response prediction from in vitro to in vivo settings

MOTIVATION

Large-scale cancer omics studies have highlighted the diversity of patient molecular profiles and the importance of leveraging this information to deliver the right drug to the right patient at the right time. Key challenges in learning predictive models for this include the high-dimensionality of omics data and heterogeneity in biological and clinical factors affecting patient response. The use of multi-task learning techniques has been widely explored to address dataset limitations for in vitro drug response models, while domain adaptation (DA) has been employed to extend them to predict in vivo response. In both of these transfer learning settings, noisy data for some tasks (or domains) can substantially reduce the performance for others compared to single-task (domain) learners, i.e. lead to negative transfer (NT).

RESULTS

We describe a novel multi-task unsupervised DA method (TUGDA) that addresses these limitations in a unified framework by quantifying uncertainty in predictors and weighting their influence on shared feature representations. TUGDA's ability to rely more on predictors with low-uncertainty allowed it to notably reduce cases of NT for in vitro models (94% overall) compared to state-of-the-art methods. For DA to in vivo settings, TUGDA improved over previous methods for patient-derived xenografts (9 out of 14 drugs) as well as patient datasets (significant associations in 9 out of 22 drugs). TUGDA's ability to avoid NT thus provides a key capability as we try to integrate diverse drug-response datasets to build consistent predictive models with in vivo utility.

AVAILABILITYAND IMPLEMENTATION

https://github.com/CSB5/TUGDA.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

A mathematical investigation into the uptake kinetics of nanoparticles in vitro

Nanoparticles have the potential to increase the efficacy of anticancer drugs whilst reducing off-target side effects. However, there remain uncertainties regarding the cellular uptake kinetics of nanoparticles which could have implications for nanoparticle design and delivery. Polymersomes are nanoparticle candidates for cancer therapy which encapsulate chemotherapy drugs. Here we develop a mathematical model to simulate the uptake of polymersomes via endocytosis, a process by which polymersomes bind to the cell surface before becoming internalised by the cell where they then break down, releasing their contents which could include chemotherapy drugs. We focus on two in vitro configurations relevant to the testing and development of cancer therapies: a well-mixed culture model and a tumour spheroid setup. Our mathematical model of the well-mixed culture model comprises a set of coupled ordinary differential equations for the unbound and bound polymersomes and associated binding dynamics. Using a singular perturbation analysis we identify an optimal number of ligands on the polymersome surface which maximises internalised polymersomes and thus intracellular chemotherapy drug concentration. In our mathematical model of the spheroid, a multiphase system of partial differential equations is developed to describe the spatial and temporal distribution of bound and unbound polymersomes via advection and diffusion, alongside oxygen, tumour growth, cell proliferation and viability. Consistent with experimental observations, the model predicts the evolution of oxygen gradients leading to a necrotic core. We investigate the impact of two different internalisation functions on spheroid growth, a constant and a bond dependent function. It was found that the constant function yields faster uptake and therefore chemotherapy delivery. We also show how various parameters, such as spheroid permeability, lead to travelling wave or steady-state solutions.

oncoPredict: an R package for predicting in vivo or cancer patient drug response and biomarkers from cell line screening data

Cell line drug screening datasets can be utilized for a range of different drug discovery applications from drug biomarker discovery to building translational models of drug response. Previously, we described three separate methodologies to (1) correct for general levels of drug sensitivity to enable drug-specific biomarker discovery, (2) predict clinical drug response in patients and (3) associate these predictions with clinical features to perform in vivo drug biomarker discovery. Here, we unite and update these methodologies into one R package (oncoPredict) to facilitate the development and adoption of these tools. This new OncoPredict R package can be applied to various in vitro and in vivo contexts for drug and biomarker discovery.

Predicting breast cancer drug response using a multiple-layer cell line drug response network model

Background

Predicting patient drug response based on a patient’s molecular profile is one of the key goals of precision medicine in breast cancer (BC). Multiple drug response prediction models have been developed to address this problem. However, most of them were developed to make sensitivity predictions for multiple single drugs within cell lines from various cancer types instead of a single cancer type, do not take into account drug properties, and have not been validated in cancer patient-derived data. Among the multi-omics data, gene expression profiles have been shown to be the most informative data for drug response prediction. However, these models were often developed with individual genes. Therefore, this study aimed to develop a drug response prediction model for BC using multiple data types from both cell lines and drugs.

Methods

We first collected the baseline gene expression profiles of 49 BC cell lines along with IC₅₀ values for 220 drugs tested in these cell lines from Genomics of Drug Sensitivity in Cancer (GDSC). Using these data, we developed a multiple-layer cell line-drug response network (ML-CDN2) by integrating a one-layer cell line similarity network based on the pathway activity profiles and a three-layer drug similarity network based on the drug structures, targets, and pan-cancer IC₅₀ profiles. We further used ML-CDN2 to predict the drug response for new BC cell lines or patient-derived samples.

Results

ML-CDN2 demonstrated a good predictive performance, with the Pearson correlation coefficient between the observed and predicted IC₅₀ values for all GDSC cell line-drug pairs of 0.873. Also, ML-CDN2 showed a good performance when used to predict drug response in new BC cell lines from the Cancer Cell Line Encyclopedia (CCLE), with a Pearson correlation coefficient of 0.718. Moreover, we found that the cell line-derived ML-CDN2 model could be applied to predict drug response in the BC patient-derived samples from The Cancer Genome Atlas (TCGA).

Conclusions

The ML-CDN2 model was built to predict BC drug response using comprehensive information from both cell lines and drugs. Compared with existing methods, it has the potential to predict the drug response for BC patient-derived samples.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-08359-6.

kESVR: An Ensemble Model for Drug Response Prediction in Precision Medicine Using Cancer Cell Lines Gene Expression

Background: Cancer cell lines are frequently used in research as in-vitro tumor models. Genomic data and large-scale drug screening have accelerated the right drug selection for cancer patients. Accuracy in drug response prediction is crucial for success. Due to data-type diversity and big data volume, few methods can integrative and efficiently find the principal low-dimensional manifold of the high-dimensional cancer multi-omics data to predict drug response in precision medicine. Method: A novelty k-means Ensemble Support Vector Regression (kESVR) is developed to predict each drug response values for single patient based on cell-line gene expression data. The kESVR is a blend of supervised and unsupervised learning methods and is entirely data driven. It utilizes embedded clustering (Principal Component Analysis and k-means clustering) and local regression (Support Vector Regression) to predict drug response and obtain the global pattern while overcoming missing data and outliers’ noise. Results: We compared the efficiency and accuracy of kESVR to 4 standard machine learning regression models: (1) simple linear regression, (2) support vector regression (3) random forest (quantile regression forest) and (4) back propagation neural network. Our results, which based on drug response across 610 cancer cells from Cancer Cell Line Encyclopedia (CCLE) and Cancer Therapeutics Response Portal (CTRP v2), proved to have the highest accuracy (smallest mean squared error (MSE) measure). We next compared kESVR with existing 17 drug response prediction models based a varied range of methods such as regression, Bayesian inference, matrix factorization and deep learning. After ranking the 18 models based on their accuracy of prediction, kESVR ranks first (best performing) in majority (74%) of the time. As for the remaining (26%) cases, kESVR still ranked in the top five performing models. Conclusion: In this paper we introduce a novel model (kESVR) for drug response prediction using high dimensional cell-line gene expression data. This model outperforms current existing prediction models in terms of prediction accuracy and speed and overcomes overfitting. This can be used in future to develop a robust drug response prediction system for cancer patients using the cancer cell-lines guidance and multi-omics data.

Predicting tumor response to drugs based on gene-expression biomarkers of sensitivity learned from cancer cell lines

Background

Human cancer cell line profiling and drug sensitivity studies provide valuable information about the therapeutic potential of drugs and their possible mechanisms of action. The goal of those studies is to translate the findings from in vitro studies of cancer cell lines into in vivo therapeutic relevance and, eventually, patients’ care. Tremendous progress has been made.

Results

In this work, we built predictive models for 453 drugs using data on gene expression and drug sensitivity (IC₅₀) from cancer cell lines. We identified many known drug-gene interactions and uncovered several potentially novel drug-gene associations. Importantly, we further applied these predictive models to ~ 17,000 bulk RNA-seq samples from The Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx) database to predict drug sensitivity for both normal and tumor tissues. We created a web site for users to visualize and download our predicted data (https://manticore.niehs.nih.gov/cancerRxTissue). Using trametinib as an example, we showed that our approach can faithfully recapitulate the known tumor specificity of the drug.

Conclusions

We demonstrated that our approach can predict drugs that 1) are tumor-type specific; 2) elicit higher sensitivity from tumor compared to corresponding normal tissue; 3) elicit differential sensitivity across breast cancer subtypes. If validated, our prediction could have relevance for preclinical drug testing and in phase I clinical design.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-021-07581-7.

Impact of between-tissue differences on pan-cancer predictions of drug sensitivity

Increased availability of drug response and genomics data for many tumor cell lines has accelerated the development of pan-cancer prediction models of drug response. However, it is unclear how much between-tissue differences in drug response and molecular characteristics may contribute to pan-cancer predictions. Also unknown is whether the performance of pan-cancer models could vary by cancer type. Here, we built a series of pan-cancer models using two datasets containing 346 and 504 cell lines, each with MEK inhibitor (MEKi) response and mRNA expression, point mutation, and copy number variation data, and found that, while the tissue-level drug responses are accurately predicted (between-tissue ρ = 0.88–0.98), only 5 of 10 cancer types showed successful within-tissue prediction performance (within-tissue ρ = 0.11–0.64). Between-tissue differences make substantial contributions to the performance of pan-cancer MEKi response predictions, as exclusion of between-tissue signals leads to a decrease in Spearman’s ρ from a range of 0.43–0.62 to 0.30–0.51. In practice, joint analysis of multiple cancer types usually has a larger sample size, hence greater power, than for one cancer type; and we observe that higher accuracy of pan-cancer prediction of MEKi response is almost entirely due to the sample size advantage. Success of pan-cancer prediction reveals how drug response in different cancers may invoke shared regulatory mechanisms despite tissue-specific routes of oncogenesis, yet predictions in different cancer types require flexible incorporation of between-cancer and within-cancer signals. As most datasets in genome sciences contain multiple levels of heterogeneity, careful parsing of group characteristics and within-group, individual variation is essential when making robust inference.

One of the central goals for precision oncology is to tailor treatment of individual tumors by their molecular characteristics. While drug response predictions have traditionally been sought within each cancer type, it has long been hoped to develop more robust predictions by jointly considering diverse cancer types. While such pan-cancer approaches have improved in recent years, it remains unclear whether between-tissue differences are contributing to the reported pan-cancer prediction performance. This concern stems from the observation that, when cancer types differ in both molecular features and drug response, strong predictive information can come mainly from differences among tissue types. Our study finds that both between- and within-cancer type signals provide substantial contributions to pan-cancer drug response prediction models, and about half of the cancer types examined are poorly predicted despite strong overall performance across all cancer types. We also find that pan-cancer prediction models perform similarly or better than cancer type-specific models, and in many cases the advantage of pan-cancer models is due to the larger number of samples available for pan-cancer analysis. Our results highlight tissue-of-origin as a key consideration for pan-cancer drug response prediction models, and recommend cancer type-specific considerations when translating pan-cancer prediction models for clinical use.

Facilitating Drug Discovery in Breast Cancer by Virtually Screening Patients Using In Vitro Drug Response Modeling

(1) Background: Drug imputation methods often aim to translate in vitro drug response to in vivo drug efficacy predictions. While commonly used in retrospective analyses, our aim is to investigate the use of drug prediction methods for the generation of novel drug discovery hypotheses. Triple-negative breast cancer (TNBC) is a severe clinical challenge in need of new therapies. (2) Methods: We used an established machine learning approach to build models of drug response based on cell line transcriptome data, which we then applied to patient tumor data to obtain predicted sensitivity scores for hundreds of drugs in over 1000 breast cancer patients. We then examined the relationships between predicted drug response and patient clinical features. (3) Results: Our analysis recapitulated several suspected vulnerabilities in TNBC and identified a number of compounds-of-interest. AZD-1775, a Wee1 inhibitor, was predicted to have preferential activity in TNBC (p < 2.2 × 10-16) and its efficacy was highly associated with TP53 mutations (p = 1.2 × 10-46). We validated these findings using independent cell line screening data and pathway analysis. Additionally, co-administration of AZD-1775 with standard-of-care paclitaxel was able to inhibit tumor growth (p < 0.05) and increase survival (p < 0.01) in a xenograft mouse model of TNBC. (4) Conclusions: Overall, this study provides a framework to turn any cancer transcriptomic dataset into a dataset for drug discovery. Using this framework, one can quickly generate meaningful drug discovery hypotheses for a cancer population of interest.

The Aurora kinase/β-catenin axis contributes to dexamethasone resistance in leukemia

Glucocorticoids, such as dexamethasone and prednisolone, are widely used in cancer treatment. Different hematological malignancies respond differently to this treatment which, as could be expected, correlates with treatment outcome. In this study, we have used a glucocorticoid-induced gene signature to develop a deep learning model that can predict dexamethasone sensitivity. By combining gene expression data from cell lines and patients with acute lymphoblastic leukemia, we observed that the model is useful for the classification of patients. Predicted samples have been used to detect deregulated pathways that lead to dexamethasone resistance. Gene set enrichment analysis, peptide substrate-based kinase profiling assay, and western blot analysis identified Aurora kinase, S6K, p38, and β-catenin as key signaling proteins involved in dexamethasone resistance. Deep learning-enabled drug synergy prediction followed by in vitro drug synergy analysis identified kinase inhibitors against Aurora kinase, JAK, S6K, and mTOR that displayed synergy with dexamethasone. Combining pathway enrichment, kinase regulation, and kinase inhibition data, we propose that Aurora kinase or its several direct or indirect downstream kinase effectors such as mTOR, S6K, p38, and JAK may be involved in β-catenin stabilization through phosphorylation-dependent inactivation of GSK-3β. Collectively, our data suggest that activation of the Aurora kinase/β-catenin axis during dexamethasone treatment may contribute to cell survival signaling which is possibly maintained in patients who are resistant to dexamethasone.

Bioinformatics Approaches for Anti-cancer Drug Discovery

Drug discovery is important in cancer therapy and precision medicines. Traditional approaches of drug discovery are mainly based on in vivo animal experiments and in vitro drug screening, but these methods are usually expensive and laborious. In the last decade, omics data explosion provides an opportunity for computational prediction of anti-cancer drugs, improving the efficiency of drug discovery. High-throughput transcriptome data were widely used in biomarkers' identification and drug prediction by integrating with drug-response data. Moreover, biological network theory and methodology were also successfully applied to the anti-cancer drug discovery, such as studies based on protein-protein interaction network, drug-target network and disease-gene network. In this review, we summarized and discussed the bioinformatics approaches for predicting anti-cancer drugs and drug combinations based on the multi-omic data, including transcriptomics, toxicogenomics, functional genomics and biological network. We believe that the general overview of available databases and current computational methods will be helpful for the development of novel cancer therapy strategies.

Personalized logical models to investigate cancer response to BRAF treatments in melanomas and colorectal cancers

The study of response to cancer treatments has benefited greatly from the contribution of different omics data but their interpretation is sometimes difficult. Some mathematical models based on prior biological knowledge of signaling pathways facilitate this interpretation but often require fitting of their parameters using perturbation data. We propose a more qualitative mechanistic approach, based on logical formalism and on the sole mapping and interpretation of omics data, and able to recover differences in sensitivity to gene inhibition without model training. This approach is showcased by the study of BRAF inhibition in patients with melanomas and colorectal cancers who experience significant differences in sensitivity despite similar omics profiles. We first gather information from literature and build a logical model summarizing the regulatory network of the mitogen-activated protein kinase (MAPK) pathway surrounding BRAF, with factors involved in the BRAF inhibition resistance mechanisms. The relevance of this model is verified by automatically assessing that it qualitatively reproduces response or resistance behaviors identified in the literature. Data from over 100 melanoma and colorectal cancer cell lines are then used to validate the model's ability to explain differences in sensitivity. This generic model is transformed into personalized cell line-specific logical models by integrating the omics information of the cell lines as constraints of the model. The use of mutations alone allows personalized models to correlate significantly with experimental sensitivities to BRAF inhibition, both from drug and CRISPR targeting, and even better with the joint use of mutations and RNA, supporting multi-omics mechanistic models. A comparison of these untrained models with learning approaches highlights similarities in interpretation and complementarity depending on the size of the datasets. This parsimonious pipeline, which can easily be extended to other biological questions, makes it possible to explore the mechanistic causes of the response to treatment, on an individualized basis.

Drug sensitivity prediction with normal inverse Gaussian shrinkage informed by external data

In precision medicine, a common problem is drug sensitivity prediction from cancer tissue cell lines. These types of problems entail modelling multivariate drug responses on high-dimensional molecular feature sets in typically >1000 cell lines. The dimensions of the problem require specialised models and estimation methods. In addition, external information on both the drugs and the features is often available. We propose to model the drug responses through a linear regression with shrinkage enforced through a normal inverse Gaussian prior. We let the prior depend on the external information, and estimate the model and external information dependence in an empirical-variational Bayes framework. We demonstrate the usefulness of this model in both a simulated setting and in the publicly available Genomics of Drug Sensitivity in Cancer data.

Q-Rank: Reinforcement Learning for Recommending Algorithms to Predict Drug Sensitivity to Cancer Therapy

In personalized medicine, a challenging task is to identify the most effective treatment for a patient. In oncology, several computational models have been developed to predict the response of drugs to therapy. However, the performance of these models depends on multiple factors. This paper presents a new approach, called Q-Rank, to predict the sensitivity of cell lines to anti-cancer drugs. Q-Rank integrates different prediction algorithms and identifies a suitable algorithm for a given application. Q-Rank is based on reinforcement learning methods to rank prediction algorithms on the basis of relevant features (e.g., omics characterization). The best-ranked algorithm is recommended and used to predict the response of drugs to therapy. Our experimental results indicate that Q-Rank outperforms the integrated models in predicting the sensitivity of cell lines to different drugs.

A Middle-Out Modeling Strategy to Extend a Colon Cancer Logical Model Improves Drug Synergy Predictions in Epithelial-Derived Cancer Cell Lines

Cancer is a heterogeneous and complex disease and one of the leading causes of death worldwide. The high tumor heterogeneity between individuals affected by the same cancer type is accompanied by distinct molecular and phenotypic tumor profiles and variation in drug treatment response. In silico modeling of cancer as an aberrantly regulated system of interacting signaling molecules provides a basis to enhance our biological understanding of disease progression, and it offers the means to use computer simulations to test and optimize drug therapy designs on particular cancer types and subtypes. This sets the stage for precision medicine: the design of treatments tailored to individuals or groups of patients based on their tumor-specific molecular cancer profiles. Here, we show how a relatively large manually curated logical model can be efficiently enhanced further by including components highlighted by a multi-omics data analysis of data from Consensus Molecular Subtypes covering colorectal cancer. The model expansion was performed in a pathway-centric manner, following a partitioning of the model into functional subsystems, named modules. The resulting approach constitutes a middle-out modeling strategy enabling a data-driven expansion of a model from a generic and intermediate level of molecular detail to a model better covering relevant processes that are affected in specific cancer subtypes, comprising 183 biological entities and 603 interactions between them, partitioned in 25 functional modules of varying size and structure. We tested this model for its ability to correctly predict drug combination synergies, against a dataset of experimentally determined cell growth responses with 18 drugs in all combinations, on eight cancer cell lines. The results indicate that the extended model had an improved accuracy for drug synergy prediction for the majority of the experimentally tested cancer cell lines, although significant improvements of the model's predictive performance are still needed. Our study demonstrates how a tumor-data driven middle-out approach toward refining a logical model of a biological system can further customize a computer model to represent specific cancer cell lines and provide a basis for identifying synergistic effects of drugs targeting specific regulatory proteins. This approach bridges between preclinical cancer model data and clinical patient data and may thereby ultimately be of help to develop patient-specific in silico models that can steer treatment decisions in the clinic.

Leveraging TCGA gene expression data to build predictive models for cancer drug response

BACKGROUND

Machine learning has been utilized to predict cancer drug response from multi-omics data generated from sensitivities of cancer cell lines to different therapeutic compounds. Here, we build machine learning models using gene expression data from patients' primary tumor tissues to predict whether a patient will respond positively or negatively to two chemotherapeutics: 5-Fluorouracil and Gemcitabine.

RESULTS

We focused on 5-Fluorouracil and Gemcitabine because based on our exclusion criteria, they provide the largest numbers of patients within TCGA. Normalized gene expression data were clustered and used as the input features for the study. We used matching clinical trial data to ascertain the response of these patients via multiple classification methods. Multiple clustering and classification methods were compared for prediction accuracy of drug response. Clara and random forest were found to be the best clustering and classification methods, respectively. The results show our models predict with up to 86% accuracy; despite the study's limitation of sample size. We also found the genes most informative for predicting drug response were enriched in well-known cancer signaling pathways and highlighted their potential significance in chemotherapy prognosis.

CONCLUSIONS

Primary tumor gene expression is a good predictor of cancer drug response. Investment in larger datasets containing both patient gene expression and drug response is needed to support future work of machine learning models. Ultimately, such predictive models may aid oncologists with making critical treatment decisions.

An Improved Anticancer Drug-Response Prediction Based on an Ensemble Method Integrating Matrix Completion and Ridge Regression

In this study, we proposed an ensemble learning method, simultaneously integrating a low-rank matrix completion model and a ridge regression model to predict anticancer drug response on cancer cell lines. The model was applied to two benchmark datasets, including the Cancer Cell Line Encyclopedia (CCLE) and the Genomics of Drug Sensitivity in Cancer (GDSC). As previous studies suggest, the dual-layer integrated cell line-drug network model was one of the best models by far and outperformed most state-of-the-art models. Thus, we performed a head-to-head comparison between the dual-layer integrated cell line-drug network model and our model by a 10-fold crossvalidation study. For the CCLE dataset, our model has a higher Pearson correlation coefficient between predicted and observed drug responses than that of the dual-layer integrated cell line-drug network model in 18 out of 23 drugs. For the GDSC dataset, our model is better in 26 out of 28 drugs in the phosphatidylinositol 3-kinase (PI3K) pathway and 26 out of 30 drugs in the extracellular signal-regulated kinase (ERK) signaling pathway, respectively. Based on the prediction results, we carried out two types of case studies, which further verified the effectiveness of the proposed model on the drug-response prediction. In addition, our model is more biologically interpretable than the compared method, since it explicitly outputs the genes involved in the prediction, which are enriched in functions, like transcription, Src homology 2/3 (SH2/3) domain, cell cycle, ATP binding, and zinc finger.

The E2F Pathway Score as a Predictive Biomarker of Response to Neoadjuvant Therapy in ER+/HER2- Breast Cancer

E2F transcription factors play critical roles in the cell cycle. Therefore, their activity is expected to reflect tumor aggressiveness and responsiveness to therapy. We scored 3905 tumors of nine breast cancer cohorts for this activity based on their gene expression for the Hallmark E2F targets gene set. As expected, tumors with a high score had an increased expression of cell proliferation-related genes. A high score was significantly associated with shorter patient survival, greater MKI67 expression, histological grade, stage, and genomic aberrations. Furthermore, metastatic tumors had higher E2F scores than the primary tumors from which they arose. Although tumors with a high score had greater infiltration by both pro- and anti-cancerous immune cells, they had an increased expression of immune checkpoint genes. Estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative cancer with a high E2F score achieved a significantly higher pathological complete response (pCR) rate to neoadjuvant chemotherapy. The E2F score was significantly associated with the expression of cyclin-dependent kinase (CDK)-related genes and strongly correlated with sensitivity to CDK inhibition in cell lines. In conclusion, the E2F score is a marker of breast cancer aggressiveness and predicts the responsiveness of ER-positive/HER2-negative patients to neoadjuvant chemotherapy and possibly to CDK and immune checkpoint inhibitors.

Concise Polygenic Models for Cancer-Specific Identification of Drug-Sensitive Tumors from Their Multi-Omics Profiles

In silico models to predict which tumors will respond to a given drug are necessary for Precision Oncology. However, predictive models are only available for a handful of cases (each case being a given drug acting on tumors of a specific cancer type). A way to generate predictive models for the remaining cases is with suitable machine learning algorithms that are yet to be applied to existing in vitro pharmacogenomics datasets. Here, we apply XGBoost integrated with a stringent feature selection approach, which is an algorithm that is advantageous for these high-dimensional problems. Thus, we identified and validated 118 predictive models for 62 drugs across five cancer types by exploiting four molecular profiles (sequence mutations, copy-number alterations, gene expression, and DNA methylation). Predictive models were found in each cancer type and with every molecular profile. On average, no omics profile or cancer type obtained models with higher predictive accuracy than the rest. However, within a given cancer type, some molecular profiles were overrepresented among predictive models. For instance, CNA profiles were predictive in breast invasive carcinoma (BRCA) cell lines, but not in small cell lung cancer (SCLC) cell lines where gene expression (GEX) and DNA methylation profiles were the most predictive. Lastly, we identified the best XGBoost model per cancer type and analyzed their selected features. For each model, some of the genes in the selected list had already been found to be individually linked to the response to that drug, providing additional evidence of the usefulness of these models and the merits of the feature selection scheme.

Identification of four hub genes as promising biomarkers to evaluate the prognosis of ovarian cancer in silico

BACKGROUND

Ovarian cancer (OvCa) is one of the most fatal cancers among females in the world. With growing numbers of individuals diagnosed with OvCa ending in deaths, it is urgent to further explore the potential mechanisms of OvCa oncogenesis and development and related biomarkers.

METHODS

The gene expression profiles of GSE49997 were downloaded from the Gene Expression Omnibus (GEO) database. Weighted gene co-expression network analysis (WGCNA) was applied to explore the most potent gene modules associated with the overall survival (OS) and progression-free survival (PFS) events of OvCa patients, and the prognostic values of these genes were exhibited and validated based on data from training and validation sets. Next, protein-protein interaction (PPI) networks were built by GeneMANIA. Besides, enrichment analysis was conducted using DAVID website.

RESULTS

According to the WGCNA analysis, a total of eight modules were identified and four hub genes (MM > 0.90) in the blue module were reserved for next analysis. Kaplan-Meier analysis exhibited that these four hub genes were significantly associated with worse OS and PFS in the patient cohort from GSE49997. Moreover, we validated the short-term (4-years) and long-term prognostic values based on the GSE9891 data, respectively. Last, PPI networks analysis, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis revealed several potential mechanisms of four hub genes and their co-operators participating in OvCa progression.

CONCLUSION

Four hub genes (COL6A3, CRISPLD2, FBN1 and SERPINF1) were identified to be associated with the prognosis in OvCa, which might be used as monitoring biomarkers to evaluate survival time of OvCa patients.

Machine learning approaches to drug response prediction: challenges and recent progress

Cancer is a leading cause of death worldwide. Identifying the best treatment using computational models to personalize drug response prediction holds great promise to improve patient's chances of successful recovery. Unfortunately, the computational task of predicting drug response is very challenging, partially due to the limitations of the available data and partially due to algorithmic shortcomings. The recent advances in deep learning may open a new chapter in the search for computational drug response prediction models and ultimately result in more accurate tools for therapy response. This review provides an overview of the computational challenges and advances in drug response prediction, and focuses on comparing the machine learning techniques to be of utmost practical use for clinicians and machine learning non-experts. The incorporation of new data modalities such as single-cell profiling, along with techniques that rapidly find effective drug combinations will likely be instrumental in improving cancer care.

Dr.VAE: improving drug response prediction via modeling of drug perturbation effects

Motivation

Individualized drug response prediction is a fundamental part of personalized medicine for cancer. Great effort has been made to discover biomarkers or to develop machine learning methods for accurate drug response prediction in cancers. Incorporating prior knowledge of biological systems into these methods is a promising avenue to improve prediction performance. High-throughput cell line assays of drug-induced transcriptomic perturbation effects are a prior knowledge that has not been fully incorporated into a drug response prediction model yet.

Results

We introduce a unified probabilistic approach, Drug Response Variational Autoencoder (Dr.VAE), that simultaneously models both drug response in terms of viability and transcriptomic perturbations. Dr.VAE is a deep generative model based on variational autoencoders. Our experimental results showed Dr.VAE to do as well or outperform standard classification methods for 23 out of 26 tested Food and Drug Administration-approved drugs. In a series of ablation experiments we showed that the observed improvement of Dr.VAE can be credited to the incorporation of drug-induced perturbation effects with joint modeling of treatment sensitivity.

Availability and implementation

Processed data and software implementation using PyTorch (Paszke et al., 2017) are available at: https://github.com/rampasek/DrVAE.

Supplementary information

Supplementary data are available at Bioinformatics online.

G2M Cell Cycle Pathway Score as a Prognostic Biomarker of Metastasis in Estrogen Receptor (ER)-Positive Breast Cancer

The vast majority of breast cancer death is a result of metastasis. Thus, accurate identification of patients who are likely to have metastasis is expected to improve survival. The G2M checkpoint plays a critical role in cell cycle. We hypothesized that breast cancer tumors with high activity of G2M pathway genes are more aggressive and likely to metastasize. To test this, we used the single-sample gene set variation analysis method to calculate the score for the Hallmark G2M checkpoint pathway using gene expression data of a total of 4626 samples from 12 human breast cancer cohorts. As expected, a high G2M pathway score correlated with enriched tumor expression of other cell proliferation-related gene sets. The score was significantly associated with clinical aggressive features of tumors and patient survival in estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer. Interestingly, a high G2M score of metastasis tumors was also significantly associated with worse survival. In primary as well as metastasis tumors with high scores, the infiltration of both pro- and anti-cancerous immune cells increased. Tumor G2M score was also associated with treatment response to systemic chemotherapy in ER-positive/HER2-negative cancer, and was predictive of response to cyclin-dependent kinase inhibition therapy.

Tissue-guided LASSO for prediction of clinical drug response using preclinical samples

Prediction of clinical drug response (CDR) of cancer patients, based on their clinical and molecular profiles obtained prior to administration of the drug, can play a significant role in individualized medicine. Machine learning models have the potential to address this issue but training them requires data from a large number of patients treated with each drug, limiting their feasibility. While large databases of drug response and molecular profiles of preclinical in-vitro cancer cell lines (CCLs) exist for many drugs, it is unclear whether preclinical samples can be used to predict CDR of real patients. We designed a systematic approach to evaluate how well different algorithms, trained on gene expression and drug response of CCLs, can predict CDR of patients. Using data from two large databases, we evaluated various linear and non-linear algorithms, some of which utilized information on gene interactions. Then, we developed a new algorithm called TG-LASSO that explicitly integrates information on samples' tissue of origin with gene expression profiles to improve prediction performance. Our results showed that regularized regression methods provide better prediction performance. However, including the network information or common methods of including information on the tissue of origin did not improve the results. On the other hand, TG-LASSO improved the predictions and distinguished resistant and sensitive patients for 7 out of 13 drugs. Additionally, TG-LASSO identified genes associated with the drug response, including known targets and pathways involved in the drugs' mechanism of action. Moreover, genes identified by TG-LASSO for multiple drugs in a tissue were associated with patient survival. In summary, our analysis suggests that preclinical samples can be used to predict CDR of patients and identify biomarkers of drug sensitivity and survival.

Improving drug response prediction by integrating multiple data sources: matrix factorization, kernel and network-based approaches

Predicting the response of cancer cell lines to specific drugs is one of the central problems in personalized medicine, where the cell lines show diverse characteristics. Researchers have developed a variety of computational methods to discover associations between drugs and cell lines, and improved drug sensitivity analyses by integrating heterogeneous biological data. However, choosing informative data sources and methods that can incorporate multiple sources efficiently is the challenging part of successful analysis in personalized medicine. The reason is that finding decisive factors of cancer and developing methods that can overcome the problems of integrating data, such as differences in data structures and data complexities, are difficult. In this review, we summarize recent advances in data integration-based machine learning for drug response prediction, by categorizing methods as matrix factorization-based, kernel-based and network-based methods. We also present a short description of relevant databases used as a benchmark in drug response prediction analyses, followed by providing a brief discussion of challenges faced in integrating and interpreting data from multiple sources. Finally, we address the advantages of combining multiple heterogeneous data sources on drug sensitivity analysis by showing an experimental comparison. Contact:  betul.guvenc@aalto.fi

A Deep Learning Model for Cell Growth Inhibition IC50 Prediction and Its Application for Gastric Cancer Patients

Heterogeneity in intratumoral cancers leads to discrepancies in drug responsiveness, due to diverse genomics profiles. Thus, prediction of drug responsiveness is critical in precision medicine. So far, in drug responsiveness prediction, drugs’ molecular “fingerprints”, along with mutation statuses, have not been considered. Here, we constructed a 1-dimensional convolution neural network model, DeepIC50, to predict three drug responsiveness classes, based on 27,756 features including mutation statuses and various drug molecular fingerprints. As a result, DeepIC50 showed better cell viability IC50 prediction accuracy in pan-cancer cell lines over two independent cancer cell line datasets. Gastric cancer (GC) is not only one of the lethal cancer types in East Asia, but also a heterogeneous cancer type. Currently approved targeted therapies in GC are only trastuzumab and ramucirumab. Responsive GC patients for the drugs are limited, and more drugs should be developed in GC. Due to the importance of GC, we applied DeepIC50 to a real GC patient dataset. Drug responsiveness prediction in the patient dataset by DeepIC50, when compared to the other models, were comparable to responsiveness observed in GC cell lines. DeepIC50 could possibly accurately predict drug responsiveness, to new compounds, in diverse cancer cell lines, in the drug discovery process.

Opportunities for improving cancer treatment using systems biology

Current cancer therapies target a limited set of tumor features, rather than considering the tumor as a whole. Systems biology aims to reveal therapeutic targets associated with a variety of facets in an individual's tumor, such as genetic heterogeneity and its evolution, cancer cell-autonomous phenotypes, and microenvironmental signaling. These disparate characteristics can be reconciled using mathematical modeling that incorporates concepts from ecology and evolution. This provides an opportunity to predict tumor growth and response to therapy, to tailor patient-specific approaches in real time or even prospectively. Importantly, as data regarding patient tumors is often available from only limited time points during treatment, systems-based approaches can address this limitation by interpolating longitudinal events within a principled framework. This review outlines areas in medicine that could benefit from systems biology approaches to deconvolve the complexity of cancer.

Modeling cancer drug response through drug-specific informative genes

Recent advances in pharmacogenomics have generated a wealth of data of different types whose analysis have helped in the identification of signatures of different cellular sensitivity/resistance responses to hundreds of chemical compounds. Among the different data types, gene expression has proven to be the more successful for the inference of drug response in cancer cell lines. Although effective, the whole transcriptome can introduce noise in the predictive models, since specific mechanisms are required for different drugs and these realistically involve only part of the proteins encoded in the genome. We analyzed the pharmacogenomics data of 961 cell lines tested with 265 anti-cancer drugs and developed different machine learning approaches for dissecting the genome systematically and predict drug responses using both drug-unspecific and drug-specific genes. These methodologies reach better response predictions for the vast majority of the screened drugs using tens to few hundreds genes specific to each drug instead of the whole genome, thus allowing a better understanding and interpretation of drug-specific response mechanisms which are not necessarily restricted to the drug known targets.