Models Of Drug Response Research Articles

Methodological challenges in translational drug response modeling in cancer: A systematic analysis with FORESEE.

Translational models directly relating drug response specific processes that can be observed in vitro to their in vivo role in cancer patients constitute a crucial part of the development of personalized medication. Unfortunately, current studies often focus on the optimization of isolated model characteristics instead of examining the overall modeling workflow and the interplay of the individual model components. Moreover, they are often limited to specific data sets only. Therefore, they are often confined by the irreproducibility of the results and the non-transferability of the approaches into other contexts. In this study, we present a thorough investigation of translational models and their ability to predict the drug responses of cancer patients originating from diverse data sets using the R-package FORESEE. By systematically scanning the modeling space for optimal combinations of different model settings, we can determine models of extremely high predictivity and work out a few modeling guidelines that promote simplicity. Yet, we identify noise within the data, sample size effects, and drug unspecificity as factors that deteriorate the models’ robustness. Moreover, we show that cell line models of high accuracy do not necessarily excel in predicting drug response processes in patients. We therefore hope to motivate future research to consider in vivo aspects more carefully to ultimately generate deeper insights into applicable precision medicine.

Integration of Drug Perturbed Dose Dependent Response Data in Predictive Computational Models of Chemotherapeutic Response Using Multiple‐Kernel‐Learning

Recently there has been a great deal of work to understand the dynamics of genome expression under small molecule perturbation. This has been demonstrated by the recent expansion of the Connectivity Map (CMAP) ( www.broadinstitute.org/connectivity‐map‐cmap) with roughly 5000 small molecule signatures. While this has proven to be invaluable resource it is still limited due to many drugs being constrained to a handful of cancer cell genomes, because of this, it is difficult to leverage these datasets to make an accurate and robust predictive model. Outside CMAP there is limited data that systematically characterize small molecule induced genomic changes, however, recently Monks et al. (Cancer Res 2018 DEC 15) has made available data in the NCI60 cell line panel for treated and untreated cell lines at 2, 6, and 24hr and two different dosing concentrations for 15 anti‐cancer agents. In our work we explore how each dataset can be optimally leveraged to build a predictive model of cell IC50 within each of the 15 drugs using a multiple kernel learning approach. With this method we can reach correlations as high as 0.88 (p ~1e−6) between predicted and measured IC50 with an average correlation of 0.57 on an independent test set. Additionally, perturbation with higher drug dose yields better individual models with average correlation around 0.451 compared to non‐treated data of 0.391 suggesting drug induced genomic changes are an important predictor of drug performance. Furthermore, we present a novel method that derives signatures present in perturbed data that can be captured in basal expression data. This approach to selecting features yields model performance that is equal or better than a signature derived basally alone. Our work demonstrates the importance of both basal and dynamic gene expression in modeling and understanding of drug response. We hope this work motivates further experiments to characterize drug induced gene dynamics for both modeling applications and understanding the underlying genomic influences in drug response.Support or Funding InformationShipley Chair in Comparative Oncology

IGMDR: Integrated Pharmacogenetic Resource Guide to Cancer Therapy and Research

Current pharmacogenetic studies have obtained many genetic models that can predict the therapeutic efficacy of anticancer drugs. Although some of these models are of crucial importance and have been used in clinical practice, these very valuable models have not been well adopted into cancer research to promote the development of cancer therapies due to the lack of integration and standards for the existing data of the pharmacogenetic studies. For this purpose, we built a resource investigating genetic model of drug response (iGMDR), which integrates the models from in vitro and in vivo pharmacogenetic studies with different omics data from a variety of technical systems. In this study, we introduced a standardized process for all integrations, and described how users can utilize these models to gain insights into cancer. iGMDR is freely accessible at https://igmdr.modellab.cn.

A pharmacodynamic model of clinical synergy in multiple myeloma.

BackgroundMultiagent therapies, due to their ability to delay or overcome resistance, are a hallmark of treatment in multiple myeloma (MM). The growing number of therapeutic options in MM requires high-throughput combination screening tools to better allocate treatment, and facilitate personalized therapy.MethodsA second-order drug response model was employed to fit patient-specific ex vivo responses of 203 MM patients to single-agent models. A novel pharmacodynamic model, developed to account for two-way combination effects, was tested with 130 two-drug combinations. We have demonstrated that this model is sufficiently parameterized by single-agent and fixed-ratio combination responses, by validating model estimates with ex vivo combination responses for different concentration ratios, using a checkerboard assay. This new model reconciles ex vivo observations from both Loewe and BLISS synergy models, by accounting for the dimension of time, as opposed to focusing on arbitrary time-points or drug effect. Clinical outcomes of patients were simulated by coupling patient-specific drug combination models with pharmacokinetic data.FindingsCombination screening showed 1 in 5 combinations (21.43% by LD50, 18.42% by AUC) were synergistic ex vivo with statistical significance (P < 0.05), but clinical synergy was predicted for only 1 in 10 combinations (8.69%), which was attributed to the role of pharmacokinetics and dosing schedules.InterpretationThe proposed framework can inform clinical decisions from ex vivo observations, thus providing a path toward personalized therapy using combination regimens.FundingThis research was funded by the H. Lee Moffitt Cancer Center Physical Sciences in Oncology (PSOC) Grant (1U54CA193489-01A1) and by H. Lee Moffitt Cancer Center's Team Science Grant. This work has been supported in part by the PSOC Pilot Project Award (5U54CA193489-04), the Translational Research Core Facility at the H. Lee Moffitt Cancer Center & Research Institute, an NCI-designated Comprehensive Cancer Center (P30-CA076292), the Pentecost Family Foundation, and Miles for Moffitt Foundation.

An Organotypic Microcosm for the Pancreatic Tumor Microenvironment.

Pancreatic duct adenocarcinoma (PDAC) is projected to become the second leading cause of cancer-related deaths in the next few years. Unfortunately, the development of novel therapies for PDAC has been challenged by a uniquely complex tumor microenvironment. The development of in vitro cancer organoids in recent years has demonstrated potential to increase therapies for patients with PDAC. Organoids have been established from PDAC murine and human tissues and they are representative of the primary tumor. Further, organoids have been shown beneficial in studies of molecular mechanisms and drug sensitivity testing. This review will cover the use of organoids to study PDAC development, invasiveness, and therapeutic resistance in the context of the tumor microenvironment, which is characterized by a dense desmoplastic reaction, hindered immune activity, and pro-tumor metabolic signaling. We describe investigations utilizing organoids to characterize the tumor microenvironment and also describe their limitations. Overall, organoids have great potential to serve as a versatile model of drug response and may be used to increase available therapies and improve survival for patients with PDAC.

Luminal B subtype as a predictive biomarker of docetaxel benefit for newly diagnosed metastatic hormone sensitive prostate cancer (mHSPC): A correlative study of E3805 CHAARTED.

162 Background: Through gene expression profiling (GEP), the PAM50 classifier demonstrates prognostic value in localized prostate cancer (PCa). Pre-clinical drug response models predict increased taxane sensitivity in luminal subtypes compared to basal subtype. Men with mHSPC and high-risk features have greatest benefit from androgen deprivation therapy (ADT) plus docetaxel (D) vs ADT alone. We therefore sought to test the prognostic and predictive value of PAM50 in pre-ADT specimens from E3805 CHAARTED. Methods: Whole transcriptomic profiling of formalin-fixed, paraffin-embedded primary PCa biopsies from pts enrolled in the E3805 CHAARTED trial of ADT vs ADT+D was performed using the Human Exon 1.0 ST microarray platform (Decipher Biosciences). Normalized gene expression was used to classify subjects as luminal A, luminal B or basal subtype. Multivariable analyses (MVA) adjusted for ECOG status, de novo metastasis vs prior local therapy and volume of disease. The primary endpoint was overall survival (OS). Secondary endpoint was time to castration resistant PCa (TTCRPC). Results: Successful GEP was completed in 160 of 198 pts with available specimens. Eighty (50%), 77 (48%) and 3 (2%) pts were classified as luminal B, basal and luminal A, respectively. High volume disease was similarly present in luminal B (79%) and basal (78%) subtypes. In the ADT arm, luminal B subtype was associated with shorter OS vs basal (HR 1.75, p=0.05); consistent in MVA. Pts with luminal B subtype treated with ADT+D showed significant improvement in TTCRPC and OS (Table). By contrast, basal subtype showed no OS benefit from ADT+D even in pts with high volume disease. Conclusions: We demonstrate that GEP identifies tumor subtypes associated with differential benefit from chemohormonal therapy for mHSPC. Luminal B subtype is associated with poorer OS with ADT alone and benefits from addition of D. Basal subtype shows a lack of OS benefit from upfront ADT+D. We plan to validate these findings in independent trial cohorts.[Table: see text]

Experimentally-driven mathematical modeling to improve combination targeted and cytotoxic therapy for HER2+ breast cancer

The goal of this study is to experimentally and computationally investigate combination trastuzumab-paclitaxel therapies and identify potential synergistic effects due to sequencing of the therapies with in vitro imaging and mathematical modeling. Longitudinal alterations in cell confluence are reported for an in vitro model of BT474 HER2+ breast cancer cells following various dosages and timings of paclitaxel and trastuzumab combination regimens. Results of combination drug regimens are evaluated for drug interaction relationships based on order, timing, and quantity of dose of the drugs. Altering the order of treatments, with the same total therapeutic dose, provided significant changes in overall cell confluence (p < 0.001). Two mathematical models are introduced that are constrained by the in vitro data to simulate the tumor cell response to the individual therapies. A collective model merging the two individual drug response models was designed to investigate the potential mechanisms of synergy for paclitaxel-trastuzumab combinations. This collective model shows increased synergy for regimens where trastuzumab is administered prior to paclitaxel and suggests trastuzumab accelerates the cytotoxic effects of paclitaxel. The synergy derived from the model is found to be in agreement with the combination index, where both indicate a spectrum of additive and synergistic interactions between the two drugs dependent on their dose order. The combined in vitro results and development of a mathematical model of drug synergy has potential to evaluate and improve standard-of-care combination therapies in cancer.

Machine learning and data mining frameworks for predicting drug response in cancer: An overview and a novel in silico screening process based on association rule mining.

A major challenge in cancer treatment is predicting the clinical response to anti-cancer drugs on a personalized basis. The success of such a task largely depends on the ability to develop computational resources that integrate big "omic" data into effective drug-response models. Machine learning is both an expanding and an evolving computational field that holds promise to cover such needs. Here we provide a focused overview of: 1) the various supervised and unsupervised algorithms used specifically in drug response prediction applications, 2) the strategies employed to develop these algorithms into applicable models, 3) data resources that are fed into these frameworks and 4) pitfalls and challenges to maximize model performance. In this context we also describe a novel in silico screening process, based on Association Rule Mining, for identifying genes as candidate drivers of drug response and compare it with relevant data mining frameworks, for which we generated a web application freely available at: https://compbio.nyumc.org/drugs/. This pipeline explores with high efficiency large sample-spaces, while is able to detect low frequency events and evaluate statistical significance even in the multidimensional space, presenting the results in the form of easily interpretable rules. We conclude with future prospects and challenges of applying machine learning based drug response prediction in precision medicine.

Abstract 676: Leveraging TCGA gene expression data to build predictive models for cancer drug response

Abstract Personalized oncology promises to increase the success rate of cancer drug therapy by using molecular tumor profiles to determine the optimal therapeutic for an individual. 5-Fluorouracil and Gemcitabine are two Food and Drug Administration approved chemotherapeutics commonly prescribed to cancer patients. Here, we build machine learning models using gene expression data from patients’ primary tumors to predict their response, positive or negative, to these drugs. We compared several clustering and classification methods for predicting response. Of those tested, Clara was found to be the best clustering algorithm while random forest was the best classification method, with prediction accuracy up to 86%. We determined that models trained across several cancers outperform single cancer models. We also found that genes most informative for predicting drug response were enriched in well-known cancer signaling pathways, highlighting their potential significance in chemotherapy prognosis. Ultimately, such predictive models may aid oncologists with making critical treatment decisions. Citation Format: Evan Clayton. Leveraging TCGA gene expression data to build predictive models for cancer drug response [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 676.

A systematic analysis of genomics-based modeling approaches for prediction of drug response to cytotoxic chemotherapies

BackgroundThe availability and generation of large amounts of genomic data has led to the development of a new paradigm in cancer treatment emphasizing a precision approach at the molecular and genomic level. Statistical modeling techniques aimed at leveraging broad scale in vitro, in vivo, and clinical data for precision drug treatment has become an active area of research. As a rapidly developing discipline at the crossroads of medicine, computer science, and mathematics, techniques ranging from accepted to those on the cutting edge of artificial intelligence have been utilized. Given the diversity and complexity of these techniques a systematic understanding of fundamental modeling principles is essential to contextualize influential factors to better understand results and develop new approaches.MethodsUsing data available from the Genomics of Drug Sensitivity in Cancer (GDSC) and the NCI60 we explore principle components regression, linear and non-linear support vector regression, and artificial neural networks in combination with different implementations of correlation based feature selection (CBF) on the prediction of drug response for several cytotoxic chemotherapeutic agents.ResultsOur results indicate that the regression method and features used have marginal effects on Spearman correlation between the predicted and measured values as well as prediction error. Detailed analysis of these results reveal that the bulk relationship between tissue of origin and drug response is a major driving factor in model performance.ConclusionThese results display one of the challenges in building predictive models for drug response in pan-cancer models. Mainly, that bulk genotypic traits where the signal to noise ratio is high is the dominant behavior captured in these models. This suggests that improved techniques of feature selection that can discriminate individual cell response from histotype response will yield more successful pan-cancer models.

On-target Resistance to the Mutant-Selective EGFR Inhibitor Osimertinib Can Develop in an Allele-Specific Manner Dependent on the Original EGFR-Activating Mutation.

The third-generation EGFR inhibitor, osimertinib, is the first mutant-selective inhibitor that has received regulatory approval for the treatment of patients with EGFR-mutant lung cancer. Despite the development of highly selective third-generation inhibitors, acquired resistance remains a significant clinical challenge. Recently, we and others have identified a novel osimertinib resistance mutation, G724S, which was not predicted in in vitro screens. Here, we investigate how G724S confers resistance to osimertinib.Experimental Design: We combine structure-based predictive modeling of G724S in combination with the 2 most common EGFR-activating mutations, exon 19 deletion (Ex19Del) and L858R, with in vitro drug-response models and patient genomic profiling. Our simulations suggest that the G724S mutation selectively reduces osimertinib-binding affinity in the context of Ex19Del. Consistent with our simulations, cell lines transduced with Ex19Del/G724S demonstrate resistance to osimertinib, whereas cells transduced with L858R/G724S are sensitive to osimertinib. Subsequent clinical genomic profiling data further suggest G724S occurs with Ex19Del but not L858R. Furthermore, we demonstrate that Ex19Del/G724S retains sensitivity to afatinib, but not to erlotinib, suggesting a possible therapy for patients at the time of disease relapse. Altogether, these data suggest that G724S is an allele-specific resistance mutation emerging in the context of Ex19Del but not L858R. Our results fundamentally reframe the problem of targeted therapy resistance from one focused on the "drug-resistance mutation" pair to one focused on the "activating mutation-drug-resistance mutation" trio. This has broad implications across clinical oncology.

Modeling of Hepatic Drug Metabolism and Responses in CYP2C19 Poor Metabolizer Using Genetically Manipulated Human iPS cells.

Cytochrome P450 family 2 subfamily C member 19 (CYP2C19), in liver, plays important roles in terms of drug metabolism. It is known that CYP2C19 poor metabolizers (PMs) lack CYP2C19 metabolic capacity. Thus, unexpected drug-induced liver injury or decrease of drug efficacy would be caused in CYP2C19 substrate-treated CYP2C19 PMs. However, it is difficult to evaluate the safety and effectiveness of drugs and candidate compounds for CYP2C19 PMs because there is currently no model for this phenotype. Here, using human induced pluripotent stem cells (human iPS cells) and our highly efficient genome-editing and hepatocyte differentiation technologies, we generated CYP2C19-knockout human iPS cell-derived hepatocyte-like cells (CYP2C19-KO HLCs) as a novel CYP2C19 PM model for drug development and research. The gene expression levels of hepatocyte markers were similar between wild-type iPS cell-derived hepatocyte-like cells (WT HLCs) and CYP2C19-KO HLCs, suggesting that CYP2C19 deficiency did not affect the hepatic differentiation potency. We also examined CYP2C19 metabolic activity by measuring S-mephenytoin metabolites using ultra-performance liquid chromatography-tandem mass spectrometry. The CYP2C19 metabolic activity was almost eliminated by CYP2C19 knockout. Additionally, we evaluated whether clopidogrel (CYP2C19 substrate)-induced liver toxicity could be predicted using our model. Unexpectedly, there was no significant difference in cell viability between clopidogrel-treated WT HLCs and CYP2C19-KO HLCs. However, the cell viability in clopidogrel- and ketoconazole (CYP3A4 inhibitor)-treated CYP2C19-KO HLCs was significantly enhanced as compared with that in clopidogrel- and DMSO-treated CYP2C19-KO HLCs. This result suggests that CYP2C19-KO HLCs can predict clopidogrel-induced liver toxicity. We succeeded in generating CYP2C19 PM model cells using human iPS cells and genome-editing technologies for pharmaceutical research. SIGNIFICANCE STATEMENT: Although unexpected drug-induced liver injury or decrease of drug efficacy would be caused in CYP2C19 substrate-treated CYP2C19 poor metabolizers, it is difficult to evaluate the safety and effectiveness of drugs and candidate compounds for CYP2C19 poor metabolizers because there is currently no model for this phenotype. Using human iPS cells and our highly efficient genome editing and hepatocyte differentiation technologies, we generated CYP2C19-knockout human iPS cell-derived hepatocyte-like cells as a novel CYP2C19 poor metabolizer model for drug development and research.

Meta-GDBP: a high-level stacked regression model to improve anticancer drug response prediction.

Anticancer drug response prediction plays an important role in personalized medicine. In particular, precisely predicting drug response in specific cancer types and patients is still a challenge problem. Here we propose Meta-GDBP, a novel anticancer drug-response model, which involves two levels. At the first level of Meta-GDBP, we build four optimized base models (BMs) using genetic information, chemical properties and biological context with an ensemble optimization strategy, while at the second level, we construct a weighted model to integrate the four BMs. Notably, the weights of the models are learned upstream, thus the parameter cost is significantly reduced compared to previous methods. We evaluate the Meta-GDBP on Genomics of Drug Sensitivity in Cancer (GDSC) and the Cancer Cell Line Encyclopedia (CCLE) data sets. Benchmarking results demonstrate that compared to other methods, the Meta-GDBP achieves a much higher correlation between the predicted and the observed responses for almost all the drugs. Moreover, we apply the Meta-GDBP to predict the GDSC-missing drug response and use the CCLE-known data to validate the performance. The results show quite a similar tendency between these two response sets. Particularly, we here for the first time introduce a biological context-based frequency matrix (BCFM) to associate the biological context with the drug response. It is encouraging that the proposed BCFM is biologically meaningful and consistent with the reported biological mechanism, further demonstrating its efficacy for predicting drug response. The R implementation for the proposed Meta-GDBP is available at https://github.com/RanSuLab/Meta-GDBP.

FORESEE: a tool for the systematic comparison of translational drug response modeling pipelines.

SummaryTranslational models that utilize omics data generated in in vitro studies to predict the drug efficacy of anti-cancer compounds in patients are highly distinct, which complicates the benchmarking process for new computational approaches. In reaction to this, we introduce the uniFied translatiOnal dRug rESponsE prEdiction platform FORESEE, an open-source R-package. FORESEE not only provides a uniform data format for public cell line and patient datasets, but also establishes a standardized environment for drug response prediction pipelines, incorporating various state-of-the-art pre-processing methods, model training algorithms and validation techniques. The modular implementation of individual elements of the pipeline facilitates a straightforward development of combinatorial models, which can be used to re-evaluate and improve already existing pipelines as well as to develop new ones.Availability and implementationFORESEE is licensed under GNU General Public License v3.0 and available at https://github.com/JRC-COMBINE/FORESEE and https://doi.org/10.17605/OSF.IO/RF6QK, and provides vignettes for documentation and application both online and in the Supplementary Files 2 and 3.Supplementary information Supplementary data are available at Bioinformatics online.

Statistical assessment and visualization of synergies for large-scale sparse drug combination datasets

BackgroundDrug combinations have the potential to improve efficacy while limiting toxicity. To robustly identify synergistic combinations, high-throughput screens using full dose-response surface are desirable but require an impractical number of data points. Screening of a sparse number of doses per drug allows to screen large numbers of drug pairs, but complicates statistical assessment of synergy. Furthermore, since the number of pairwise combinations grows with the square of the number of drugs, exploration of large screens necessitates advanced visualization tools.ResultsWe describe a statistical and visualization framework for the analysis of large-scale drug combination screens. We developed an approach suitable for datasets with large number of drugs pairs even if small number of data points are available per drug pair. We demonstrate our approach using a systematic screen of all possible pairs among 108 cancer drugs applied to melanoma cell lines. In this dataset only two dose-response data points per drug pair and two data points per single drug test were available. We used a Bliss-based linear model, effectively borrowing data from the drug pairs to obtain robust estimations of the singlet viabilities, consequently yielding better estimates of drug synergy. Our method improves data consistency across dosing thus likely reducing the number of false positives. The approach allows to compute p values accounting for standard errors of the modeled singlets and combination viabilities. We further develop a synergy specificity score that distinguishes specific synergies from those arising with promiscuous drugs. Finally, we developed a summarized interactive visualization in a web application, providing efficient access to any of the 439,000 data points in the combination matrix (http://www.cmtlab.org:3000/combo_app.html). The code of the analysis and the web application is available at https://github.com/arnaudmgh/synergy-screen.ConclusionsWe show that statistical modeling of single drug response from drug combination data can help determine significance of synergy and antagonism in drug combination screens with few data point per drug pair. We provide a web application for the rapid exploration of large combinatorial drug screen. All codes are available to the community, as a resource for further analysis of published data and for analysis of other drug screens.

Predictive modeling of therapeutic response to chondroitin sulfate/glucosamine hydrochloride in knee osteoarthritis.

Background:In the present study, we explored potential protein biomarkers useful to predict the therapeutic response of knee osteoarthritis (KOA) patients treated with pharmaceutical grade Chondroitin sulfate/Glucosamine hydrochloride (CS+GH; Droglican, Bioiberica), in order to optimize therapeutic outcomes.Methods:A shotgun proteomic analysis by iTRAQ labelling and liquid chromatography–mass spectrometry (LC-MS/MS) was performed using sera from 40 patients enrolled in the Multicentre Osteoarthritis interVEntion trial with Sysadoa (MOVES). The panel of proteins potentially useful to predict KOA patient’s response was clinically validated in the whole MOVES cohort at baseline (n = 506) using commercially available enzyme-linked immunosorbent assays kits. Logistic regression models and receiver-operating-characteristics (ROC) curves were used to analyze the contribution of these proteins to our prediction models of symptomatic drug response in KOA.Results:In the discovery phase of the study, a panel of six putative predictive biomarkers of response to CS+GH (APOA2, APOA4, APOH, ITIH1, C4BPa and ORM2) were identified by shotgun proteomics. Data are available via ProteomeXchange with identifier PXD012444. In the verification phase, the panel was verified in a larger set of KOA patients (n = 262). Finally, ITIH1 and ORM2 were qualified by a blind test in the whole MOVES cohort at baseline. The combination of these biomarkers with clinical variables predict the patients’ response to CS+GH with a specificity of 79.5% and a sensitivity of 77.1%.Conclusions:Combining clinical and analytical parameters, we identified one biomarker that could accurately predict KOA patients’ response to CS+GH treatment. Its use would allow an increase in response rates and safety for the patients suffering KOA.

Integrated mapping of pharmacokinetics and pharmacodynamics in a patient-derived xenograft model of glioblastoma

Therapeutic options for the treatment of glioblastoma remain inadequate despite concerted research efforts in drug development. Therapeutic failure can result from poor permeability of the blood-brain barrier, heterogeneous drug distribution, and development of resistance. Elucidation of relationships among such parameters could enable the development of predictive models of drug response in patients and inform drug development. Complementary analyses were applied to a glioblastoma patient-derived xenograft model in order to quantitatively map distribution and resulting cellular response to the EGFR inhibitor erlotinib. Mass spectrometry images of erlotinib were registered to histology and magnetic resonance images in order to correlate drug distribution with tumor characteristics. Phosphoproteomics and immunohistochemistry were used to assess protein signaling in response to drug, and integrated with transcriptional response using mRNA sequencing. This comprehensive dataset provides simultaneous insight into pharmacokinetics and pharmacodynamics and indicates that erlotinib delivery to intracranial tumors is insufficient to inhibit EGFR tyrosine kinase signaling.

Machine learning predicts individual cancer patient responses to therapeutic drugs with high accuracy

Precision or personalized cancer medicine is a clinical approach that strives to customize therapies based upon the genomic profiles of individual patient tumors. Machine learning (ML) is a computational method particularly suited to the establishment of predictive models of drug response based on genomic profiles of targeted cells. We report here on the application of our previously established open-source support vector machine (SVM)-based algorithm to predict the responses of 175 individual cancer patients to a variety of standard-of-care chemotherapeutic drugs from the gene-expression profiles (RNA-seq or microarray) of individual patient tumors. The models were found to predict patient responses with >80% accuracy. The high PPV of our algorithms across multiple drugs suggests a potential clinical utility of our approach, particularly with respect to the identification of promising second-line treatments for patients failing standard-of-care first-line therapies.

Abstract 455: Cellular response to oxaliplatin treatment shows altered DNA damage, nucleolar and ribosomal stresses as main altered pathways

Abstract Oxaliplatin is a platinum-based drug playing a critical role in the treatment of colorectal carcinoma and various Hodgkin and non-Hodgkin lymphomas. The primary mechanism of action is reported to be DNA damage and double strand breaks response. However, platinum-based drugs are usually very reactive and covalently binds not only to DNA but also to RNA, proteins or glutathione. Concerning very high reactivity of platinum compounds, we have performed whole-cell proteomic profiling on highly chemosensitive CCRF-CEM cell line to identify all possible cellular response mechanisms. For modeling of actual drug response, CCRF-CEM cell line was treated with 5 x IC50 (29.3 µM) for half-time to caspase activation (240 min). Such cells were analyzed using current high-resolution mass spectrometry. Totally, 4049 proteins were identified over three biological replicates. Among these proteins, 76 were significantly downregulated and 31 significantly upregulated in the whole experiment. Bioinformatic analysis showed nucleolar and ribosomal stress, changes in secretory, mitotic and DNA damage related proteins. These response pathways were confirmed by functional analyses and targeted experiments aimed to expression changes of selected proteins. Results observed in this experiment suggests nucleolar and ribosomal stress as the most abundant stress responses to oxaliplatin treatment at a halftime to apoptosis. Down-regulation of ribosomal proteins indicates stop of ribosome synthesis, which is one kind of response to stress. In such a case, energy saved by stopping ribosome biogenesis could be invested in solving stress conditions. Dysregulated ribosomal proteins are reported activators of p53 as well. In conclusion, oxaliplatin-induced cellular stress caused an acute response at an early stage of the experiment. Acute response stopped ribosomal biosynthesis and reduced the nucleolar size by the currently unknown mechanism. Nucleolar and ribosomal stress then triggers apoptosis in both p53 dependent and independent pathways. This work was supported by Internal Grant of Palacky University in Olomouc (LF_2016_019), Ministry of School, Education and Youth of the Czech Republic (LO1304 and LM2015064) and Technological Agency of the Czech Republic (TE02000058). Citation Format: Tomáš Oždian, Dušan Holub, Zuzana Macečková, Lakshman Varanasi, Gabriela Rylova, Jiří Rehulka, Jana Vaclavkova, Hanuš Slavík, Pavel Moudry, Pawel Znojek, Jarmila Stanková, Juan Bautista de Sanctis, Marian Hajduch, Petr Džubák. Cellular response to oxaliplatin treatment shows altered DNA damage, nucleolar and ribosomal stresses as main altered pathways [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 455.

Abstract 3883: Gene expression predicts ex vivo drug sensitivity in acute myeloid leukemia

Abstract Introduction: Ex vivo drug sensitivity studies of samples derived from acute myeloid leukemia (AML) patients have been shown to be predictive of in vivo response. These findings are based on a limited number of well-characterized agents for which in vivo patient response data and ex vivo drug sensitivity data—on that same patient—are available. To show the feasibility of scaling such ex vivo studies to large drug screens, we characterized the reproducibility of expression-based models of drug response across two independent data sets—one generated at the Oregon Health and Science University (OHSU) and the second at the Institute for Molecular Medicine Finland (FIMM). Methods: We harmonized two large-scale AML ex vivo studies screened for drug response and profiled transcriptomically—OHSU (303 AML patient samples and 160 drugs) and FIMM (48 AML samples and 480 drugs). The two panels have 94 drugs in common. Log-logistic curves were fit to the dose-response data and area under the dose-response curves (AUCs) were calculated. Predictive modeling using Ridge regression or an integrative Bayesian approach was performed for each drug AUC independently using 202 highly-variable and/or cancer-associated genes as features. Results: For each of the 94 drugs in common between the two data sets, we trained a Ridge regression model on the OHSU data set, used the model to predict response in the FIMM data set, and calculated the Pearson correlation between the predicted and observed FIMM responses. 41 of the 94 drug models had a positive and statistically significant correlation [false discovery rate (FDR) &amp;lt; 20%; mean ρ = 0.43; 95% CI = 0.29 – 0.77]. Drugs corresponding to the top decile of these significant models (mean ρ = 0.54; 95% CI = 0.48 – 0.77) clustered into four primary classes: MEK inhibitors (PD184352, Selumetinib, and Trametinib), EGFR/VEGFR inhibitors (Cabozantinib, Erlotinib, Foretinib, and Sorafenib), and singletons Venetoclax and Sirolimus. To confirm these results, we applied a second modeling approach—an integrative Bayesian machine learning method—that allows systematic combination of both data sets. Training and evaluation of this approach using 10-fold cross validation yielded 82 positive and statistically significant correlations (FDR &amp;lt; 20%; mean ρ = 0.35; 95% CI = 0.13 – 0.58). Five of nine drugs (Cabozantinib, Selumetinib, Sirolimus, Sorafenib, and Trametinib) corresponding to the top decile of these significant models (mean ρ = 0.54; 95% CI = 0.49 – 0.60) overlapped with drugs from the top decile of Ridge results (one-sided Fisher p = 2.5 x 10-4) Conclusions: Our results using independent data sets and two statistical approaches suggest that certain drugs (including MEK and EGFR/VEGFR inhibitors) are amenable to expression-based predictive modeling in AML. Future work will focus on inferring individual biomarkers of response. Citation Format: Brian S. White, Suleiman A. Khan, Muhammad Ammad-ud-din, Swapnil Potdar, Mike J. Mason, Cristina E. Tognon, Brian J. Druker, Caroline A. Heckman, Olli P. Kallioniemi, Stephen E. Kurtz, Kimmo Porkka, Jeffrey W. Tyner, Tero Aittokallio, Krister Wennerberg, Justin Guinney. Gene expression predicts ex vivo drug sensitivity in acute myeloid leukemia [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3883.