Rationalizing Drug Response in Cancer Cell Lines

J Mol Biol. 2018 Sep 14;430(18 Pt A):3016-3027. doi: 10.1016/j.jmb.2018.03.021. Epub 2018 Apr 4.


Cancer cell lines (CCLs) play an important role in the initial stages of drug discovery allowing, among others, for the screening of drug candidates. As CCL panels continue to grow in size and diversity, many polymorphisms in genes encoding drug-metabolizing enzymes, transporters and drug targets, as well as disease-related genes have been linked to altered drug sensitivity. However, identifying the correlation between this variability and pharmacological responses remains challenging due to the heterogeneity of cancer biology and the intricate interplay between cell lines and drug molecules. Here, we propose a network-based strategy that exploits information on gene expression and somatic mutations of CCLs to group cells according to their molecular similarity. We then identify genes that are characteristic of each cluster and correlate their status with drug response. We find that CCLs with similar characteristic active network regions present specific responses to certain drugs, and identify a limited set of genes that might be directly involved in drug sensitivity or resistance.

Keywords: antineoplastic drugs; cancer cell lines; drug response; molecular signatures; network-based stratification.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Antineoplastic Agents / pharmacology*
  • Bayes Theorem
  • Cell Line, Tumor
  • Drug Discovery
  • Drug Screening Assays, Antitumor* / methods
  • Gene Expression Profiling
  • Humans
  • Mutation
  • Protein Interaction Mapping
  • ROC Curve
  • Therapeutic Index, Drug*


  • Antineoplastic Agents