Modeling cancer drug response through drug-specific informative genes

Parca, Luca; Pepe, Gerardo; Pietrosanto, Marco; Galvan, Giulio; Galli, Leonardo; Palmeri, Antonio; Sciandrone, Marco; Ferré, Fabrizio; Ausiello, Gabriele; Helmer‐Citterich, Manuela

doi:10.1038/s41598-019-50720-0

Cited by 54 publications

(48 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…8 and 9 ). These results supported recent observations that the expression profiles of drug targets alone as well as the network neighbors of drug targets were not predictive of drug response 6 , 7 , 30 , 31 , suggesting that utilization of network proximity allowed robust signals to be captured at the pathway level, but not at the gene level. Moreover, we also applied the feature selection procedure by Bolis et al, which leverages 10 iterations of leave-half-out cross-validation to select genes for ML training (see “Methods”).…”

Section: Resultssupporting

confidence: 87%

Network-based machine learning in colorectal and bladder organoid models predicts anti-cancer drug efficacy in patients

et al. 2020

View full text Add to dashboard Cite

Cancer patient classification using predictive biomarkers for anti-cancer drug responses is essential for improving therapeutic outcomes. However, current machine-learning-based predictions of drug response often fail to identify robust translational biomarkers from preclinical models. Here, we present a machine-learning framework to identify robust drug biomarkers by taking advantage of network-based analyses using pharmacogenomic data derived from three-dimensional organoid culture models. The biomarkers identified by our approach accurately predict the drug responses of 114 colorectal cancer patients treated with 5-fluorouracil and 77 bladder cancer patients treated with cisplatin. We further confirm our biomarkers using external transcriptomic datasets of drug-sensitive and -resistant isogenic cancer cell lines. Finally, concordance analysis between the transcriptomic biomarkers and independent somatic mutation-based biomarkers further validate our method. This work presents a method to predict cancer patient drug responses using pharmacogenomic data derived from organoid models by combining the application of gene modules and network-based approaches.

show abstract

Section: Resultssupporting

confidence: 87%

Network-based machine learning in colorectal and bladder organoid models predicts anti-cancer drug efficacy in patients

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Consistently throughout this study, we observed that the higher expression of CD36 in HT29 LuM3 cells ( 36 ) as compared to parental HT29 cells, makes these cells more sensitive to CD36 inhibition via CD36 shRNA and inhibits xenograft tumor growth to a higher extent as compared to HT29 xenografts. Furthermore, the mutational and metabolic profiles of tumors determine tumor cell response to multiple therapies including metabolic inhibitors ( 48 , 49 ). Different genetic profiles and metabolic features can explain the varying levels of response of cell lines to FASN and CD36 inhibition.…”

Section: Discussionmentioning

confidence: 99%

Inhibition of Fatty Acid Synthase Upregulates Expression of CD36 to Sustain Proliferation of Colorectal Cancer Cells

Drury

Rychahou

et al. 2020

Front. Oncol.

View full text Add to dashboard Cite

Fatty acid synthase, a key enzyme of de novo lipogenesis, is an attractive therapeutic target in cancer. The novel fatty acid synthase inhibitor, TVB-3664, shows anti-cancer activity in multiple cancers including colorectal cancer; however, it is unclear whether uptake of exogeneous fatty acids can compensate for the effect of fatty acid synthase inhibition. This study demonstrates that inhibition of fatty acid synthase selectively upregulates fatty acid translocase (CD36), a fatty acid transporter, in multiple colorectal cancer models including colorectal cancer cells with shRNA mediated knockdown of fatty acid synthase and genetically modified mouse tissues with heterozygous and homozygous deletion of fatty acid synthase. Furthermore, human colorectal cancer tissues treated with TVB-3664 show a significant and selective upregulation of CD36 mRNA. shRNA-mediated knockdown of CD36 and inhibition of CD36 via sulfosuccinimidyl oleate, a chemical inhibitor of CD36, decreased cell proliferation in vitro and reduced tumor growth in subcutaneous xenograft models. Isogenic cell populations established from patient derived xenografts and expressing high levels of CD36 show a significantly increased ability to grow tumors in vivo. The tumor-promoting effect of CD36 is associated with an increase in the levels of pAkt and survivin. Importantly, combinatorial treatment of primary and established colorectal cancer cells with TVB-3664 and sulfosuccinimidyl oleate shows a synergistic effect on cell proliferation. In summary, our study demonstrates that upregulation of CD36 expression is a potential compensatory mechanism for fatty acid synthase inhibition and that inhibition of CD36 can improve the efficacy of fatty acid synthase-targeted therapy.

show abstract

“…A variety of computational methods have been proposed to predict drug responses for cell lines based on molecular data. Classical algorithms like decision trees and support vector machines have been used to predict the clinical efficiency of anti-cancer drugs and classify drug responses (Stetson et al, 2015;Borisov et al, 2018;Oskooei et al, 2018;Webber et al, 2018;Parca et al, 2019;Su et al, 2019). Neural networks (Menden et al, 2013) and deep neural networks (Chiu et al, 2019) have been used to predict drug response based on genomic profiles from cell lines.…”

Section: Discussionmentioning

confidence: 99%

“…Other techniques have included elastic net regression (Basu et al, 2013;Webber et al, 2018;Parca et al, 2019), linear ridge regression (Geeleher et al, 2017), and LASSO regression (Huang et al, 2020). Alternative approaches based on computational linear algebra or network structures have also been applied to infer drug response in cell lines; these include matrix factorization (Guan et al, 2019), matrix completion (Nguyen and Le, 2018), and link prediction (Stanfield et al, 2017) methods.…”

Section: Discussionmentioning

confidence: 99%

“…Recent cell-line studies have emphasized the potential to predict drug responses based on geneexpression profiles (Costello et al, 2014;Yuan et al, 2014;Zhao et al, 2015;Chiu et al, 2019;Parca et al, 2019). Technologies for profiling gene-expression levels are widely available and reflect the downstream effects of genomic and epigenomic aberrations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting drug sensitivity of cancer cells based on DNA methylation levels

Fleck

Miranda

Baião

et al. 2020

Preprint

View full text Add to dashboard Cite

Cancer cell lines, which are cell cultures developed from tumor samples, represent one of the least expensive and most studied preclinical models for drug development. Accurately predicting drug response 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 from the Genomics of Drug Sensitivity in Cancer database, we applied 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 that use diverse methodologies, including tree-, probability-, kernel-, ensemble-, and distance-based approaches. For both types of algorithm, we artificially subsampled the data to varying degrees, aiming to understand whether training models based on relatively extreme outcomes would yield improved performance. We also performed an information-gain analysis to examine which genes were most predictive of drug responses. Finally, we used tumor data from The Cancer Genome Atlas to evaluate the feasibility of predicting clinical responses in humans based on models derived from cell lines. 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 both consisted of cell-line data. However, classification models derived from cell-line data failed to generalize effectively for tumors.

show abstract

Modeling cancer drug response through drug-specific informative genes

Cited by 54 publications

References 45 publications

Network-based machine learning in colorectal and bladder organoid models predicts anti-cancer drug efficacy in patients

Network-based machine learning in colorectal and bladder organoid models predicts anti-cancer drug efficacy in patients

Inhibition of Fatty Acid Synthase Upregulates Expression of CD36 to Sustain Proliferation of Colorectal Cancer Cells

Predicting drug sensitivity of cancer cells based on DNA methylation levels

Contact Info

Product

Resources

About