2019
DOI: 10.1158/1535-7163.mct-19-0273
|View full text |Cite
|
Sign up to set email alerts
|

Expression Levels of Therapeutic Targets as Indicators of Sensitivity to Targeted Therapeutics

Abstract: Cancer precision medicine aims to predict the drug likely to yield the best response for a patient. Genomic sequencing of tumors is currently being used to better inform treatment options; however, this approach has had a limited clinical impact due to the paucity of actionable mutations. An alternative to mutation status is the use of gene expression signatures to predict response. Using data from two largescale studies, The Genomics of Drug Sensitivity of Cancer (GDSC) and The Cancer Therapeutics Response Po… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
7
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
8
1
1

Relationship

0
10

Authors

Journals

citations
Cited by 17 publications
(7 citation statements)
references
References 43 publications
(40 reference statements)
0
7
0
Order By: Relevance
“…The drug sensitivity data in the GDSC dataset were quantified using two continuous variables, IC50 and AUC, representing half maximal inhibitory concentration and the area under the fitted dose response curve, respectively. We converted IC50 to -log 10 IC50 and use 1-AUC to define drug sensitive as the AUC values ranges from 0 to 1 45 . We eventually used only the model that predicts 1-AUC as the final pre-trained model because it had better performance in terms of Root Mean Squared Error (RMSE) on the GDSC data, with an average RMSE of 0.096 across five test folds, comparing to mean RMSE of 0.42 achieved by the model that used − log 10 IC50 as the response variable.…”
Section: Methodsmentioning
confidence: 99%
“…The drug sensitivity data in the GDSC dataset were quantified using two continuous variables, IC50 and AUC, representing half maximal inhibitory concentration and the area under the fitted dose response curve, respectively. We converted IC50 to -log 10 IC50 and use 1-AUC to define drug sensitive as the AUC values ranges from 0 to 1 45 . We eventually used only the model that predicts 1-AUC as the final pre-trained model because it had better performance in terms of Root Mean Squared Error (RMSE) on the GDSC data, with an average RMSE of 0.096 across five test folds, comparing to mean RMSE of 0.42 achieved by the model that used − log 10 IC50 as the response variable.…”
Section: Methodsmentioning
confidence: 99%
“…Related to our findings in this context, a recent large-scale study examined the sensitivity of hundreds of cancer cell lines to hundreds of drugs, which was correlated with expression of drug targets within the cells. Here, Roy et al described inverse correlations between target expression and drug sensitivity for 8% of targets, suggesting drug efficacy may not only be determined by expression levels of the drug target, but may also depend on other factors such as genetic background and other molecules that could affect drug-target interactions, including the expression of other gene family members or interacting proteins ( 40 ).…”
Section: Discussionmentioning
confidence: 99%
“…K nearest neighbor imputation was applied to impute the missing AUC values. We used the normalization method to modify the drug sensitivity data matrix of CCLs ( Roy et al, 2019 ). The drugs with >0.1 were considered to have differential sensitivity in different clusters ( Yang et al, 2021 ).…”
Section: Methodsmentioning
confidence: 99%