Targeted cancer therapies - drugs that kill cancer cells with certain "driver oncogenes" - shrink tumors and extend patient survival. Ultimately though, the cancers become resistant to the targeted therapies.

"Unfortunately, virtually all patients with metastatic cancer develop disease progression, limiting the effectiveness of these agents," said William Pao, M.D., Ph.D., director of the Division of Hematology and Oncology at Vanderbilt-Ingram Cancer Center.

Now, Pao and colleagues including Zhongming Zhao, Ph.D., associate professor of Biomedical Informatics, and Peilin Jia, Ph.D., assistant professor of Biomedical Informatics, have combined next-generation sequencing technologies and bioinformatics analyses to screen for genome-wide genetic mutations associated with drug resistance in a series of lung cancer cell lines.

Their report in the journal Genome Research demonstrates the power of their approach and highlights these diverse changes in the context of different tumor cell backgrounds.

"The recent advances in next-generation sequencing offer us the opportunity to explore the full range of DNA changes acquired in the course of cancer cell drug resistance," Jia said.

The investigators focused on lung cancer cell lines that harbor mutations in the EGFR (epidermal growth factor receptor) gene.

They developed genetically matched pairs of drug-sensitive and drug-resistant cell lines by culturing the "parental" drug-sensitive cells in escalating doses of EGFR-targeted inhibitors, such as erlotinib (Tarceva), until the cells developed resistance to the drug.

Using next-generation sequencing technologies, they sequenced the whole genome or whole exome (the "expressed" part of the genome) and compared mutational changes in resistant lines and their parental counterparts. They found that resistant cells acquired between 18 and 91 new single nucleotide variations, insertions and deletions, and lost between one and 27 mutations.

The researchers were surprised to find more copy number variations (CNVs) - a type of mutation where large repeated regions of DNA are inserted or deleted - across all resistant cell lines.

"We found a large burden of CNV changes across all resistant cell lines," Zhao said. "Among these changes, a surprising finding was that one of the cell lines displaying an EMT (epithelial mesenchymal transition) phenotype had far more CNV alterations than the others. This study in general reflects the strong mutational heterogeneity of cancer."

The results, Pao said, provide a framework for studying how drug-related genetic variations evolve over time. In future studies, the investigators will use this framework to examine the effects of different types and doses of targeted therapies on the evolution of drug resistance, and they will extend their analyses of resistance mechanisms to non-targeted chemotherapies and radiation.