Modeling precision treatment of breast cancer.

Anneleen Daemen, Obi L. Griffith, Laura M. Heiser, Nicholas J. Wang, Oana M. Enache, Zachary Sanborn, Francois Pepin, Steffen Durinck, James E. Korkola, Malachi Griffith, Joe S. Hur, Nam Huh, Jongsuk Chung, Leslie Cope, Mary Jo Fackler, Christopher B Umbricht, Saraswati Sukumar, Pankaj Seth, Vikas P. Sukhatme, Lakshmi R. Jakkula & 7 others Yiling Lu, Gordon B. Mills, Raymond J. Cho, Eric A. Collisson, Laura J. van't Veer, Paul T. Spellman, Joe W. Gray

Research output: Contribution to journalArticle

Abstract

First-generation molecular profiles for human breast cancers have enabled the identification of features that can predict therapeutic response; however, little is known about how the various data types can best be combined to yield optimal predictors. Collections of breast cancer cell lines mirror many aspects of breast cancer molecular pathobiology, and measurements of their omic and biological therapeutic responses are well-suited for development of strategies to identify the most predictive molecular feature sets. We used least squares-support vector machines and random forest algorithms to identify molecular features associated with responses of a collection of 70 breast cancer cell lines to 90 experimental or approved therapeutic agents. The datasets analyzed included measurements of copy number aberrations, mutations, gene and isoform expression, promoter methylation and protein expression. Transcriptional subtype contributed strongly to response predictors for 25% of compounds, and adding other molecular data types improved prediction for 65%. No single molecular dataset consistently out-performed the others, suggesting that therapeutic response is mediated at multiple levels in the genome. Response predictors were developed and applied to TCGA data, and were found to be present in subsets of those patient samples. These results suggest that matching patients to treatments based on transcriptional subtype will improve response rates, and inclusion of additional features from other profiling data types may provide additional benefit. Further, we suggest a systems biology strategy for guiding clinical trials so that patient cohorts most likely to respond to new therapies may be more efficiently identified.

Original languageEnglish (US)
JournalGenome Biology
Volume14
Issue number10
StatePublished - 2013

Fingerprint

breast neoplasms
cancer
Breast Neoplasms
therapeutics
modeling
cell lines
methylation
Therapeutics
Cell Line
mutation
Systems Biology
genome
least squares
clinical trials
Least-Squares Analysis
Methylation
protein synthesis
protein
promoter regions
gene

ASJC Scopus subject areas

  • Genetics
  • Cell Biology
  • Ecology, Evolution, Behavior and Systematics

Cite this

Daemen, A., Griffith, O. L., Heiser, L. M., Wang, N. J., Enache, O. M., Sanborn, Z., ... Gray, J. W. (2013). Modeling precision treatment of breast cancer. Genome Biology, 14(10).

Modeling precision treatment of breast cancer. / Daemen, Anneleen; Griffith, Obi L.; Heiser, Laura M.; Wang, Nicholas J.; Enache, Oana M.; Sanborn, Zachary; Pepin, Francois; Durinck, Steffen; Korkola, James E.; Griffith, Malachi; Hur, Joe S.; Huh, Nam; Chung, Jongsuk; Cope, Leslie; Fackler, Mary Jo; Umbricht, Christopher B; Sukumar, Saraswati; Seth, Pankaj; Sukhatme, Vikas P.; Jakkula, Lakshmi R.; Lu, Yiling; Mills, Gordon B.; Cho, Raymond J.; Collisson, Eric A.; van't Veer, Laura J.; Spellman, Paul T.; Gray, Joe W.

In: Genome Biology, Vol. 14, No. 10, 2013.

Research output: Contribution to journalArticle

Daemen, A, Griffith, OL, Heiser, LM, Wang, NJ, Enache, OM, Sanborn, Z, Pepin, F, Durinck, S, Korkola, JE, Griffith, M, Hur, JS, Huh, N, Chung, J, Cope, L, Fackler, MJ, Umbricht, CB, Sukumar, S, Seth, P, Sukhatme, VP, Jakkula, LR, Lu, Y, Mills, GB, Cho, RJ, Collisson, EA, van't Veer, LJ, Spellman, PT & Gray, JW 2013, 'Modeling precision treatment of breast cancer.', Genome Biology, vol. 14, no. 10.
Daemen A, Griffith OL, Heiser LM, Wang NJ, Enache OM, Sanborn Z et al. Modeling precision treatment of breast cancer. Genome Biology. 2013;14(10).
Daemen, Anneleen ; Griffith, Obi L. ; Heiser, Laura M. ; Wang, Nicholas J. ; Enache, Oana M. ; Sanborn, Zachary ; Pepin, Francois ; Durinck, Steffen ; Korkola, James E. ; Griffith, Malachi ; Hur, Joe S. ; Huh, Nam ; Chung, Jongsuk ; Cope, Leslie ; Fackler, Mary Jo ; Umbricht, Christopher B ; Sukumar, Saraswati ; Seth, Pankaj ; Sukhatme, Vikas P. ; Jakkula, Lakshmi R. ; Lu, Yiling ; Mills, Gordon B. ; Cho, Raymond J. ; Collisson, Eric A. ; van't Veer, Laura J. ; Spellman, Paul T. ; Gray, Joe W. / Modeling precision treatment of breast cancer. In: Genome Biology. 2013 ; Vol. 14, No. 10.
@article{98ed33b01d154fd28a07cc41e6e37d7d,
title = "Modeling precision treatment of breast cancer.",
abstract = "First-generation molecular profiles for human breast cancers have enabled the identification of features that can predict therapeutic response; however, little is known about how the various data types can best be combined to yield optimal predictors. Collections of breast cancer cell lines mirror many aspects of breast cancer molecular pathobiology, and measurements of their omic and biological therapeutic responses are well-suited for development of strategies to identify the most predictive molecular feature sets. We used least squares-support vector machines and random forest algorithms to identify molecular features associated with responses of a collection of 70 breast cancer cell lines to 90 experimental or approved therapeutic agents. The datasets analyzed included measurements of copy number aberrations, mutations, gene and isoform expression, promoter methylation and protein expression. Transcriptional subtype contributed strongly to response predictors for 25{\%} of compounds, and adding other molecular data types improved prediction for 65{\%}. No single molecular dataset consistently out-performed the others, suggesting that therapeutic response is mediated at multiple levels in the genome. Response predictors were developed and applied to TCGA data, and were found to be present in subsets of those patient samples. These results suggest that matching patients to treatments based on transcriptional subtype will improve response rates, and inclusion of additional features from other profiling data types may provide additional benefit. Further, we suggest a systems biology strategy for guiding clinical trials so that patient cohorts most likely to respond to new therapies may be more efficiently identified.",
author = "Anneleen Daemen and Griffith, {Obi L.} and Heiser, {Laura M.} and Wang, {Nicholas J.} and Enache, {Oana M.} and Zachary Sanborn and Francois Pepin and Steffen Durinck and Korkola, {James E.} and Malachi Griffith and Hur, {Joe S.} and Nam Huh and Jongsuk Chung and Leslie Cope and Fackler, {Mary Jo} and Umbricht, {Christopher B} and Saraswati Sukumar and Pankaj Seth and Sukhatme, {Vikas P.} and Jakkula, {Lakshmi R.} and Yiling Lu and Mills, {Gordon B.} and Cho, {Raymond J.} and Collisson, {Eric A.} and {van't Veer}, {Laura J.} and Spellman, {Paul T.} and Gray, {Joe W.}",
year = "2013",
language = "English (US)",
volume = "14",
journal = "Genome Biology",
issn = "1474-7596",
publisher = "BioMed Central",
number = "10",

}

TY - JOUR

T1 - Modeling precision treatment of breast cancer.

AU - Daemen, Anneleen

AU - Griffith, Obi L.

AU - Heiser, Laura M.

AU - Wang, Nicholas J.

AU - Enache, Oana M.

AU - Sanborn, Zachary

AU - Pepin, Francois

AU - Durinck, Steffen

AU - Korkola, James E.

AU - Griffith, Malachi

AU - Hur, Joe S.

AU - Huh, Nam

AU - Chung, Jongsuk

AU - Cope, Leslie

AU - Fackler, Mary Jo

AU - Umbricht, Christopher B

AU - Sukumar, Saraswati

AU - Seth, Pankaj

AU - Sukhatme, Vikas P.

AU - Jakkula, Lakshmi R.

AU - Lu, Yiling

AU - Mills, Gordon B.

AU - Cho, Raymond J.

AU - Collisson, Eric A.

AU - van't Veer, Laura J.

AU - Spellman, Paul T.

AU - Gray, Joe W.

PY - 2013

Y1 - 2013

N2 - First-generation molecular profiles for human breast cancers have enabled the identification of features that can predict therapeutic response; however, little is known about how the various data types can best be combined to yield optimal predictors. Collections of breast cancer cell lines mirror many aspects of breast cancer molecular pathobiology, and measurements of their omic and biological therapeutic responses are well-suited for development of strategies to identify the most predictive molecular feature sets. We used least squares-support vector machines and random forest algorithms to identify molecular features associated with responses of a collection of 70 breast cancer cell lines to 90 experimental or approved therapeutic agents. The datasets analyzed included measurements of copy number aberrations, mutations, gene and isoform expression, promoter methylation and protein expression. Transcriptional subtype contributed strongly to response predictors for 25% of compounds, and adding other molecular data types improved prediction for 65%. No single molecular dataset consistently out-performed the others, suggesting that therapeutic response is mediated at multiple levels in the genome. Response predictors were developed and applied to TCGA data, and were found to be present in subsets of those patient samples. These results suggest that matching patients to treatments based on transcriptional subtype will improve response rates, and inclusion of additional features from other profiling data types may provide additional benefit. Further, we suggest a systems biology strategy for guiding clinical trials so that patient cohorts most likely to respond to new therapies may be more efficiently identified.

AB - First-generation molecular profiles for human breast cancers have enabled the identification of features that can predict therapeutic response; however, little is known about how the various data types can best be combined to yield optimal predictors. Collections of breast cancer cell lines mirror many aspects of breast cancer molecular pathobiology, and measurements of their omic and biological therapeutic responses are well-suited for development of strategies to identify the most predictive molecular feature sets. We used least squares-support vector machines and random forest algorithms to identify molecular features associated with responses of a collection of 70 breast cancer cell lines to 90 experimental or approved therapeutic agents. The datasets analyzed included measurements of copy number aberrations, mutations, gene and isoform expression, promoter methylation and protein expression. Transcriptional subtype contributed strongly to response predictors for 25% of compounds, and adding other molecular data types improved prediction for 65%. No single molecular dataset consistently out-performed the others, suggesting that therapeutic response is mediated at multiple levels in the genome. Response predictors were developed and applied to TCGA data, and were found to be present in subsets of those patient samples. These results suggest that matching patients to treatments based on transcriptional subtype will improve response rates, and inclusion of additional features from other profiling data types may provide additional benefit. Further, we suggest a systems biology strategy for guiding clinical trials so that patient cohorts most likely to respond to new therapies may be more efficiently identified.

UR - http://www.scopus.com/inward/record.url?scp=84906213716&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906213716&partnerID=8YFLogxK

M3 - Article

VL - 14

JO - Genome Biology

JF - Genome Biology

SN - 1474-7596

IS - 10

ER -