Search

A major challenge of precision oncology is the identification and prioritization of suitable treatment options based on molecular biomarkers of the considered tumor. In pursuit of this goal, large cancer cell line panels have successfully been studied to elucidate the relationship between cellular features and treatment response. Due to the high dimensionality of these datasets, machine learning (ML) is commonly used for their analysis. However, choosing a suitable algorithm and set of input features can be challenging. We performed a comprehensive benchmarking of ML methods and dimension reduction (DR) techniques for predicting drug response metrics. Using the Genomics of Drug Sensitivity in Cancer cell line panel, we trained random forests, neural networks, boosting trees and elastic nets for 179 anti-cancer compounds with feature sets derived from nine DR approaches. We compare the results regarding statistical performance, runtime and interpretability. Additionally, we provide strategies for assessing model performance compared with a simple baseline model and measuring the trade-off between models of different complexity. Lastly, we show that complex ML models benefit from using an optimized DR strategy, and that standard models—even when using considerably fewer features—can still be superior in performance. [ABSTRACT FROM AUTHOR]

As key oncogenic drivers in non-small-cell lung cancer (NSCLC), various mutations in the epidermal growth factor receptor (EGFR) with variable drug sensitivities have been a major obstacle for precision medicine. To achieve clinical-level drug recommendations, a platform for clinical patient case retrieval and reliable drug sensitivity prediction is highly expected. Therefore, we built a database, D3EGFRdb, with the clinicopathologic characteristics and drug responses of 1339 patients with EGFR mutations via literature mining. On the basis of D3EGFRdb, we developed a deep learning-based prediction model, D3EGFRAI, for drug sensitivity prediction of new EGFR mutation-driven NSCLC. Model validations of D3EGFRAI showed a prediction accuracy of 0.81 and 0.85 for patients from D3EGFRdb and our hospitals, respectively. Furthermore, mutation scanning of the crucial residues inside drug-binding pockets, which may occur in the future, was performed to explore their drug sensitivity changes. D3EGFR is the first platform to achieve clinical-level drug response prediction of all approved small molecule drugs for EGFR mutation-driven lung cancer and is freely accessible at https://www.d3pharma.com/D3EGFR/index.php. [ABSTRACT FROM AUTHOR]

Members of the interleukin (IL) family are closely linked to cancer development and progression. However, research on the prognosis of colorectal cancer (CRC) related to IL is still lacking. This study investigated new CRC prognostic markers and offered new insights for CRC prognosis and treatment. CRC-related data and IL gene data were collected from public databases. Sample clustering was done with the NMF package to divide samples into different subtypes. Differential, enrichment, survival, and immune analyses were conducted on subtypes. A prognostic model was constructed using regression analysis. Drug sensitivity analysis was performed using GDSC database. Western blot analysis was performed to assess the effect of IL-7 on the JAK/STAT signaling pathway. Flow cytometry was used to examine the impact of IL-7 on CD8+ T cell apoptosis. Two CRC subtypes based on IL-associated genes were obtained. Cluster 1 had a higher survival rate than cluster 2, and they showed differences in some immune levels. The two clusters were mainly enriched in the JAK-STAT signaling pathway, T helper 17 cell differentiation, and the IL-17 signaling pathway. An 11-gene signature was built, and risk score was an independent prognosticator for CRC. The low-risk group showed a higher sensitivity to nine common targeted anticancer drugs. Western blot and flow cytometry results demonstrated that IL-7 could phosphorylate STAT5 and promote survival of CD8+ T cells. In conclusion, this study divided CRC samples into two IL-associated subtypes and obtained an 11-gene signature. In addition, targeted drugs that may improve the prognosis of patients with CRC were identified. These findings are of paramount importance for patient prognosis and CRC treatment. [ABSTRACT FROM AUTHOR]

Simple Summary: This study aimed to evaluate the potential of circular RNA (circRNA) expression profiles for the early detection of chronic lymphocytic leukemia (CLL) using bioinformatic algorithms on verified CLL patient datasets. We analyzed and validated the diagnostic performance of circRNAs as potential biomarkers in different CLL sample sets, which reveal better prognostic value than existing clinical risk scales for the prediction of 5-year overall survival. The identification of specific circRNAs from our biomarker panel are also involved in cancer-related pathways, which possess druggable targets for therapeutic interests. Our findings therefore suggest that circRNA signatures represent significant biomarkers for the early detection and surveillance of CLL in addition to providing pharmacogenomic value to personalized medicine. Chronic lymphocytic leukemia (CLL) is a known hematologic malignancy associated with a growing incidence and post-treatment relapse. Hence, finding a reliable diagnostic biomarker for CLL is crucial. Circular RNAs (circRNAs) represent a new class of RNA involved in many biological processes and diseases. This study aimed to define a circRNA-based panel for the early diagnosis of CLL. To this point, the list of the most deregulated circRNAs in CLL cell models was retrieved using bioinformatic algorithms and applied to the verified CLL patients' online datasets as the training cohort (n = 100). The diagnostic performance of potential biomarkers represented in individual and discriminating panels, was then analyzed between CLL Binet stages and validated in individual sample sets I (n = 220) and II (n = 251). We also estimated the 5-year overall survival (OS), introduced the cancer-related signaling pathways regulated by the announced circRNAs, and provided a list of possible therapeutic compounds to control the CLL. These findings show that the detected circRNA biomarkers exhibit better predictive performance compared to current validated clinical risk scales, and are applicable for the early detection and treatment of CLL. [ABSTRACT FROM AUTHOR]

Despite incredible progress in cancer treatment, therapy resistance remains the leading limiting factor for long-term survival. During drug treatment, several genes are transcriptionally upregulated to mediate drug tolerance. Using highly variable genes and pharmacogenomic data for acute myeloid leukemia (AML), we developed a drug sensitivity prediction model for the receptor tyrosine kinase inhibitor sorafenib and achieved more than 80% prediction accuracy. Furthermore, by using Shapley additive explanations for determining leading features, we identified AXL as an important feature for drug resistance. Drug-resistant patient samples displayed enrichment of protein kinase C (PKC) signaling, which was also identified in sorafenib-treated FLT3-ITD-dependent AML cell lines by a peptide-based kinase profiling assay. Finally, we show that pharmacological inhibition of tyrosine kinase activity enhances AXL expression, phosphorylation of the PKC-substrate cyclic AMP response element binding (CREB) protein, and displays synergy with AXL and PKC inhibitors. Collectively, our data suggest an involvement of AXL in tyrosine kinase inhibitor resistance and link PKC activation as a possible signaling mediator. [ABSTRACT FROM AUTHOR]

It is infeasible to test many different chemotherapy drugs on actual patients in large clinical trials, which motivates computational methods with the ability to learn and exploit associations between drug effectiveness and patient characteristics. This work proposes a machine learning approach to infer robust predictors of drug responses from patient genomic information. Rather than predicting the exact drug response on a given cell line, we introduce an elastic-net regression methodology to compare a drug-cell line pair against an alternative pair. Using predicted pairwise comparisons we rank the effectiveness of different drugs on the same cell line. A total of 173 cell lines and 100 drug responses were used in various settings for training and testing the proposed models. By comparing our approach against twelve baseline methods, we demonstrate that it outperforms the state-of-the-art methods in the literature. In contrast to most other methods, the algorithm is able to maintain its high performance even when we use a large number of drugs and few cell lines. [ABSTRACT FROM AUTHOR]

Background: Gastric cancer (GC) is a highly molecular heterogeneous tumor with poor prognosis. Epithelial-mesenchymal transition (EMT) process and cancer stem cells (CSCs) are reported to share common signaling pathways and cause poor prognosis in GC. Considering about the close relationship between these two processes, we aimed to establish a gene signature based on both processes to achieve better prognostic prediction in GC. Methods: The gene signature was constructed by univariate Cox and the least absolute shrinkage and selection operator (LASSO) Cox regression analyses by using The Cancer Genome Atlas (TCGA) GC cohort. We performed enrichment analyses to explore the potential mechanisms of the gene signature. Kaplan-Meier analysis and time-dependent receiver operating characteristic (ROC) curves were implemented to assess its prognostic value in TCGA cohort. The prognostic value of gene signature on overall survival (OS), disease-free survival (DFS), and drug sensitivity was validated in different cohorts. Quantitative reverse transcription polymerase chain reaction (RT-qPCR) validation of the prognostic value of gene signature for OS and DFS prediction was performed in the Fudan cohort. Results: A prognostic signature including SERPINE1, EDIL3, RGS4, and MATN3 (SERM signature) was constructed to predict OS, DFS, and drug sensitivity in GC. Enrichment analyses illustrated that the gene signature has tight connection with the CSC and EMT processes in GC. Patients were divided into two groups based on the risk score obtained from the formula. The Kaplan-Meier analyses indicated high-risk group yielded significantly poor prognosis compared with low-risk group. Pearson's correlation analysis indicated that the risk score was positively correlated with carboplatin and 5-fluorouracil IC50 of GC cell lines. Multivariate Cox regression analyses showed that the gene signature was an independent prognostic factor for predicting GC patients' OS, DFS, and susceptibility to adjuvant chemotherapy. Conclusions: Our SERM prognostic signature is of great value for OS, DFS, and drug sensitivity prediction in GC, which may give guidance to the development of targeted therapy for CSC- and EMT-related gene in the future. [ABSTRACT FROM AUTHOR]

Background: Drug sensitivity prediction and drug responsive biomarker selection on high-throughput genomic data is a critical step in drug discovery. Many computational methods have been developed to serve this purpose including several deep neural network models. However, the modular relations among genomic features have been largely ignored in these methods. To overcome this limitation, the role of the gene co-expression network on drug sensitivity prediction is investigated in this study. Methods: In this paper, we first introduce a network-based method to identify representative features for drug response prediction by using the gene co-expression network. Then, two graph-based neural network models are proposed and both models integrate gene network information directly into neural network for outcome prediction. Next, we present a large-scale comparative study among the proposed network-based methods, canonical prediction algorithms (i.e., Elastic Net, Random Forest, Partial Least Squares Regression, and Support Vector Regression), and deep neural network models for drug sensitivity prediction. All the source code and processed datasets in this study are available at https://github.com/compbiolabucf/drug-sensitivity-prediction. Results: In the comparison of different feature selection methods and prediction methods on a non-small cell lung cancer (NSCLC) cell line RNA-seq gene expression dataset with 50 different drug treatments, we found that (1) the network-based feature selection method improves the prediction performance compared to Pearson correlation coefficients; (2) Random Forest outperforms all the other canonical prediction algorithms and deep neural network models; (3) the proposed graph-based neural network models show better prediction performance compared to deep neural network model; (4) the prediction performance is drug dependent and it may relate to the drug's mechanism of action. Conclusions: Network-based feature selection method and prediction models improve the performance of the drug response prediction. The relations between the genomic features are more robust and stable compared to the correlation between each individual genomic feature and the drug response in high dimension and low sample size genomic datasets. [ABSTRACT FROM AUTHOR]

Background: Clinical studies often track dose-response curves of subjects over time. One can easily model the dose-response curve at each time point with Hill equation, but such a model fails to capture the temporal evolution of the curves. On the other hand, one can use Gompertz equation to model the temporal behaviors at each dose without capturing the evolution of time curves across dosage. Results: In this article, we propose a parametric model for dose-time responses that follows Gompertz law in time and Hill equation across dose approximately. We derive a recursion relation for dose-response curves over time capturing the temporal evolution and then specify a regression model connecting the parameters controlling the dose-time responses with individual level proteomic data. The resultant joint model allows us to predict the dose-response curves over time for new individuals. Conclusion: We have compared the efficacy of our proposed Recursive Hybrid model with individual dose-response predictive models at desired time points. We note that our proposed model exhibits a superior performance compared to the individual ones for both synthetic data and actual pharmacological data. For the desired dose-time varying genetic characterization and drug response values, we have used the HMS-LINCS database and demonstrated the effectiveness of our model for all available anticancer compounds. [ABSTRACT FROM AUTHOR]

Background: In precision medicine, scarcity of suitable biological data often hinders the design of an appropriate predictive model. In this regard, large scale pharmacogenomics studies, like CCLE and GDSC hold the promise to mitigate the issue. However, one cannot directly employ data from multiple sources together due to the existing distribution shift in data. One way to solve this problem is to utilize the transfer learning methodologies tailored to fit in this specific context. Results: In this paper, we present two novel approaches for incorporating information from a secondary database for improving the prediction in a target database. The first approach is based on latent variable cost optimization and the second approach considers polynomial mapping between the two databases. Utilizing CCLE and GDSC databases, we illustrate that the proposed approaches accomplish a better prediction of drug sensitivities for different scenarios as compared to the existing approaches. Conclusion: We have compared the performance of the proposed predictive models with database-specific individual models as well as existing transfer learning approaches. We note that our proposed approaches exhibit superior performance compared to the abovementioned alternative techniques for predicting sensitivity for different anti-cancer compounds, particularly the nonlinear mapping model shows the best overall performance. [ABSTRACT FROM AUTHOR]

Precision medicine entails the design of therapies that are matched for each individual patient. Thus, predictive modeling of drug responses for specific patients constitutes a significant challenge for personalized therapy. In this article, we consider a review of approaches that have been proposed to tackle the drug sensitivity prediction problem especially with respect to personalized cancer therapy. We first discuss modeling approaches that are based on genomic characterizations alone and further the discussion by including modeling techniques that integrate both genomic and functional information. A comparative analysis of the prediction performance of four representative algorithms, elastic net, random forest, kernelized Bayesian multi-task learning and deep learning, reflecting the broad classes of regularized linear, ensemble, kernelized and neural network-based models, respectively, has been included in the paper. The review also considers the challenges that need to be addressed for successful implementation of the algorithms in clinical practice. [ABSTRACT FROM AUTHOR]

Background: An enduring challenge in personalized medicine is to select right drug for individual patients. Testing drugs on patients in large clinical trials is one way to assess their efficacy and toxicity, but it is impractical to test hundreds of drugs currently under development. Therefore the preclinical prediction model is highly expected as it enables prediction of drug response to hundreds of cell lines in parallel. Methods: Recently, two large-scale pharmacogenomic studies screened multiple anticancer drugs on over 1000 cell lines in an effort to elucidate the response mechanism of anticancer drugs. To this aim, we here used gene expression features and drug sensitivity data in Cancer Cell Line Encyclopedia (CCLE) to build a predictor based on Support Vector Machine (SVM) and a recursive feature selection tool. Robustness of our model was validated by cross-validation and an independent dataset, the Cancer Genome Project (CGP). Results: Our model achieved good cross validation performance for most drugs in the Cancer Cell Line Encyclopedia (=80 % accuracy for 10 drugs, = 75 % accuracy for 19 drugs). Independent tests on eleven common drugs between CCLE and CGP achieved satisfactory performance for three of them, i.e., AZD6244, Erlotinib and PD-0325901, using expression levels of only twelve, six and seven genes, respectively. Conclusions: These results suggest that drug response could be effectively predicted from genomic features. Our model could be applied to predict drug response for some certain drugs and potentially play a complementary role in personalized medicine. [ABSTRACT FROM AUTHOR]

The article focuses on the role of epigenetic modifications in cancer physiopathology including haematological malignancies. Topics discussed include analysis of DNA hypomethylation during myelomagenesis in multiple myeloma, role of DNA methyltransferases in the regulation of DNA methylation, and aspects of epigenetic modifiers 5-azacytidine and decitabine which results in DNA damage induction, and DNA hypomethylation.

Affinity fingerprints report the activity of small molecules across a set of assays, and thus permit to gather information about the bioactivities of structurally dissimilar compounds, where models based on chemical structure alone are often limited, and model complex biological endpoints, such as human toxicity and in vitro cancer cell line sensitivity. Here, we propose to model in vitro compound activity using computationally predicted bioactivity profiles as compound descriptors. To this aim, we apply and validate a framework for the calculation of QSAR-derived affinity fingerprints (QAFFP) using a set of 1360 QSAR models generated using Ki, Kd, IC50 and EC50 data from ChEMBL database. QAFFP thus represent a method to encode and relate compounds on the basis of their similarity in bioactivity space. To benchmark the predictive power of QAFFP we assembled IC50 data from ChEMBL database for 18 diverse cancer cell lines widely used in preclinical drug discovery, and 25 diverse protein target data sets. This study complements part 1 where the performance of QAFFP in similarity searching, scaffold hopping, and bioactivity classification is evaluated. Despite being inherently noisy, we show that using QAFFP as descriptors leads to errors in prediction on the test set in the ~ 0.65–0.95 pIC50 units range, which are comparable to the estimated uncertainty of bioactivity data in ChEMBL (0.76–1.00 pIC50 units). We find that the predictive power of QAFFP is slightly worse than that of Morgan2 fingerprints and 1D and 2D physicochemical descriptors, with an effect size in the 0.02–0.08 pIC50 units range. Including QSAR models with low predictive power in the generation of QAFFP does not lead to improved predictive power. Given that the QSAR models we used to compute the QAFFP were selected on the basis of data availability alone, we anticipate better modeling results for QAFFP generated using more diverse and biologically meaningful targets. Data sets and Python code are publicly available at https://github.com/isidroc/QAFFP_regression. [ABSTRACT FROM AUTHOR]

Background: A significant problem in precision medicine is the prediction of drug sensitivity for individual cancer cell lines. Predictive models such as Random Forests have shown promising performance while predicting from individual genomic features such as gene expressions. However, accessibility of various other forms of data types including information on multiple tested drugs necessitates the examination of designing predictive models incorporating the various data types. Results: We explore the predictive performance of model stacking and the effect of stacking on the predictive bias and squared error. In addition we discuss the analytical underpinnings supporting the advantages of stacking in reducing squared error and inherent bias of random forests in prediction of outliers. The framework is tested on a setup including gene expression, drug target, physical properties and drug response information for a set of drugs and cell lines. Conclusion: The performance of individual and stacked models are compared. We note that stacking models built on two heterogeneous datasets provide superior performance to stacking different models built on the same dataset. It is also noted that stacking provides a noticeable reduction in the bias of our predictors when the dominant eigenvalue of the principle axis of variation in the residuals is significantly higher than the remaining eigenvalues. [ABSTRACT FROM AUTHOR]