Dose-response modeling in high-throughput cancer drug screenings: A case study with recommendations for practitioners

Personalized cancer treatments based on the molecular profile of a patient's tumor are becoming a standard of care in oncology. Experimentalists and pharmacologists rely on high-throughput, \textit{in vitro} screenings of many compounds against many different cell lines to build models of drug response. These models help them discover new potential therapeutics that may apply to broad classes of tumors matching some molecular pattern. We propose a hierarchical Bayesian model of how cancer cell lines respond to drugs in these experiments and develop a method for fitting the model to real-world data. Through a case study, the model is shown both quantitatively and qualitatively to capture nontrivial associations between molecular features and drug response. Finally, we draw five conclusions and recommendations that may benefit experimentalists, analysts, and clinicians working in the field of personalized medicine for cancer therapeutics.

[1]  Julio Saez-Rodriguez,et al.  Machine Learning Prediction of Cancer Cell Sensitivity to Drugs Based on Genomic and Chemical Properties , 2012, PloS one.

[2]  S. Gabriel,et al.  Pan-cancer patterns of somatic copy-number alteration , 2013, Nature Genetics.

[3]  David M. Simcha,et al.  Tackling the widespread and critical impact of batch effects in high-throughput data , 2010, Nature Reviews Genetics.

[4]  Scott E. Martin,et al.  Reproducible pharmacogenomic profiling of cancer cell line panels , 2016, Nature.

[5]  Ryan P. Adams,et al.  Elliptical slice sampling , 2009, AISTATS.

[6]  Alexander J. Smola,et al.  The Falling Factorial Basis and Its Statistical Applications , 2014, ICML.

[7]  Andrea Califano,et al.  Detection and removal of spatial bias in multiwell assays , 2016, Bioinform..

[8]  Francesco Iorio,et al.  Multilevel models improve precision and speed of IC50 estimates. , 2016, Pharmacogenomics.

[9]  H. Zou,et al.  Regularization and variable selection via the elastic net , 2005 .

[10]  Adam A. Margolin,et al.  The Cancer Cell Line Encyclopedia enables predictive modeling of anticancer drug sensitivity , 2012, Nature.

[11]  Jing Zhang,et al.  The real cost of sequencing: scaling computation to keep pace with data generation , 2016, Genome biology.

[12]  James G. Scott,et al.  Bayesian Inference for Logistic Models Using Pólya–Gamma Latent Variables , 2012, 1205.0310.

[13]  S. Ramaswamy,et al.  Systematic identification of genomic markers of drug sensitivity in cancer cells , 2012, Nature.

[14]  Cheng Li,et al.  Adjusting batch effects in microarray expression data using empirical Bayes methods. , 2007, Biostatistics.

[15]  B. Efron.  Large-Scale Simultaneous Hypothesis Testing , 2004 .

[16]  A. Califano,et al.  Inhibition of the autocrine IL-6-JAK2-STAT3-calprotectin axis as targeted therapy for HR-/HER2+ breast cancers. , 2015, Genes & development.

[17]  Petr Smirnov,et al.  Dr.VAE: Drug Response Variational Autoencoder , 2017, 1706.08203.

[18]  David B. Dunson,et al.  Bayesian monotone regression using Gaussian process projection , 2013, 1306.4041.

[19]  Robert Nadon,et al.  Identification and correction of spatial bias are essential for obtaining quality data in high-throughput screening technologies , 2017, Scientific Reports.

[20]  Andrea Califano,et al.  Direct reversal of glucocorticoid resistance by AKT inhibition in acute lymphoblastic leukemia. , 2013, Cancer cell.

[21]  Sridhar Ramaswamy,et al.  Genomics of Drug Sensitivity in Cancer (GDSC): a resource for therapeutic biomarker discovery in cancer cells , 2012, Nucleic Acids Res..

[22]  Tero Aittokallio,et al.  Drug response prediction by inferring pathway-response associations with kernelized Bayesian matrix factorization , 2016, Bioinform..

[23]  K. Tomczak,et al.  The Cancer Genome Atlas (TCGA): an immeasurable source of knowledge , 2015, Contemporary oncology.

[24]  Donavan T. Cheng,et al.  Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT): A Hybridization Capture-Based Next-Generation Sequencing Clinical Assay for Solid Tumor Molecular Oncology. , 2015, The Journal of molecular diagnostics : JMD.

[25]  R. Tibshirani.  Adaptive piecewise polynomial estimation via trend filtering , 2013, 1304.2986.

[26]  Laura M. Heiser,et al.  A community effort to assess and improve drug sensitivity prediction algorithms , 2014, Nature Biotechnology.