Clinical drug response can be predicted using baseline gene expression levels and in vitro drug sensitivity in cell lines

Geeleher, Paul; Cox, Nancy J.; Huang, R. Stephanie

doi:10.1186/gb-2014-15-3-r47

Cited by 732 publications

(712 citation statements)

References 53 publications

Supporting

Mentioning

708

Contrasting

Order By: Relevance

“…This information can be used to systematically train computational models of the molecular signaling pathways contributing to drug sensitivity and resistance in various cancer settings, and to propose novel drug targets and combination approaches. Cell line screens have provided some success in explaining or predicting drug responses by driver gene mutations (2)(3)(4); however, in many cases the true mechanism of resistance remains elusive or more complex. Most predictive methods routinely used today use correlative statistics or feature-based learning techniques such as machine learning, while network methods remain scarce despite their potential for extracting mechanistic insights and actionable biomarkers.…”

Section: Introductionmentioning

confidence: 99%

Cell-Specific Computational Modeling of the PIM Pathway in Acute Myeloid Leukemia

et al. 2017

View full text Add to dashboard Cite

Personalized therapy is a major goal of modern oncology, as patient responses vary greatly even within a histologically defined cancer subtype. This is especially true in acute myeloid leukemia (AML), which exhibits striking heterogeneity in molecular segmentation. When calibrated to cell-specific data, executable network models can reveal subtle differences in signaling that help explain differences in drug response. Furthermore, they can suggest drug combinations to increase efficacy and combat acquired resistance. Here, we experimentally tested dynamic proteomic changes and phenotypic responses in diverse AML cell lines treated with pan-PIM kinase inhibitor and fms-related tyrosine kinase 3 (FLT3) inhibitor as single agents and in combination. We constructed cell-specific executable models of the signaling axis, connecting genetic aberrations in FLT3, tyrosine kinase 2 (TYK2), platelet-derived growth factor receptor alpha (PDGFRA), and fibroblast growth factor receptor 1 (FGFR1) to cell proliferation and apoptosis via the PIM and PI3K kinases. The models capture key differences in signaling that later enabled them to accurately predict the unique proteomic changes and phenotypic responses of each cell line. Furthermore, using cell-specific models, we tailored combination therapies to individual cell lines and successfully validated their efficacy experimentally. Specifically, we showed that cells mildly responsive to PIM inhibition exhibited increased sensitivity in combination with PIK3CA inhibition. We also used the model to infer the origin of PIM resistance engineered through prolonged drug treatment of MOLM16 cell lines and successfully validated experimentally our prediction that this resistance can be overcome with AKT1/2 inhibition.

show abstract

Section: Introductionmentioning

confidence: 99%

Cell-Specific Computational Modeling of the PIM Pathway in Acute Myeloid Leukemia

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Ridge regression has been used in [57] to predict chemotherapeutic response in patients using only before-treatment baseline tumor gene expression data. The authors compared the performance with other approaches like random forests [58], nearest shrunken centroids [59], principal component regression [60,61], LASSO [42] and elastic net [44] regression and observed that ridge regression was the best performer.…”

Section: Linear Regression Modelsmentioning

confidence: 99%

“…As mentioned earlier, both the drug type that is being predicted or the different molecular features being used to design the predictive model can play a significant role. For instance, in the study [57], they used chemotherapeutic, whereas [25] used primarily targeted drugs. Furthermore, [57] used gene expression data alone, whereas [25] used multiple types of genomic data, including gene expression, RNA-Seq, RPPA, methylation, CNV and SNP.…”

Section: Linear Regression Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Algorithms for Drug Sensitivity Prediction

et al. 2016

View full text Add to dashboard Cite

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.

show abstract

“…Geeleher et al . (2014) fitted a ridge regression model of whole‐genome gene expression in cell lines against in vitro drug half maximal inhibitory concentration (IC 50 ) values and then used the model with tumor expression data from a clinical trial to estimate drug response. Chang et al .…”

Section: Introductionmentioning

confidence: 99%

Inferences of individual drug responses across diverse cancer types using a novel competing endogenous RNA network

Zhang

Zhou

et al. 2018

Molecular Oncology

View full text Add to dashboard Cite

Differences in individual drug responses are an obstacle to progression in cancer treatment, and predicting responses would help to plan treatment. The accumulation of cancer molecular profiling and drug response data provides opportunities and challenges to identify novel molecular signatures and mechanisms of tumor responsiveness to drugs. This study evaluated drug responses with a competing endogenous RNA (ceRNA) system that depended on competition between diverse RNA species. We identified drug response‐related ceRNA (DRCEs) by combining the sequence and expression data of long noncoding RNA (lncRNA), microRNA (miRNA), and messenger RNA (mRNA), and the survival data of cancer patients treated with drugs. We constructed a patient–drug two‐layer integrated network and used a linear weighting method to predict individual drug responses. DRCEs were found to be significantly enriched in known cancer and drug‐associated data resources, involved in biological processes known to mediate drug responses, and correlated to drug activity in cancer cell lines. The dysregulation of DRCE expression influenced drug response‐associated functions and pathways, suggesting DRCEs as potential therapeutic targets affecting drug responses. A further case study in breast invasive carcinoma (BRCA) found that DRCE expression was consistent with the drug response pattern and the aberrant expression of the two NEAT1‐related DRCEs may lead to poor response to tamoxifen therapy for patients with TP53 mutations. In summary, this study provides a framework for ceRNA‐based evaluation of clinical drug responses across multiple cancer types. Understanding the underlying molecular mechanisms of drug responses will allow improved response to chemotherapy and outcomes of cancer treatment.

show abstract

Clinical drug response can be predicted using baseline gene expression levels and in vitro drug sensitivity in cell lines

Cited by 732 publications

References 53 publications

Cell-Specific Computational Modeling of the PIM Pathway in Acute Myeloid Leukemia

Cell-Specific Computational Modeling of the PIM Pathway in Acute Myeloid Leukemia

Algorithms for Drug Sensitivity Prediction

Inferences of individual drug responses across diverse cancer types using a novel competing endogenous RNA network

Contact Info

Product

Resources

About