Impact of EGFR gene polymorphisms on anticancer drug cytotoxicity in vitro.

BACKGROUND AND
OBJECTIVE: The epidermal growth factor receptor (EGFR) plays a major role in cell proliferation of epithelial tissues, and its alterations frequently contribute to oncogenesis. Several common polymorphisms of the EGFR gene have been described, at the level of both coding and regulatory sequences. Some of these polymorphisms are associated with alterations of EGFR expression and/or activity and may have an impact on the activity of anticancer agents. This study aims to analyze the relationships between specific EGFR functional polymorphisms and anticancer drug activity.
METHOD: We investigated, in the panel of 60 human tumor cell lines established by the National Cancer Institute (NCI-60), whether the EGFR polymorphisms -216G>T, -191C>A, Arg521Lys (R521K), Val592Ala (V592A), and Cys624Phe (C624F), and the intron 1 (CA)n repeat were associated with EGFR gene expression and the in vitro cytotoxicity of anticancer agents using data extracted from the NCI database. We also looked for mutations of exons 18-21, known to enhance the activity of tyrosine kinase inhibitors, and the deletion of exons 2-7, associated to the oncogenesis of glioblastomas.
RESULTS: In the NCI-60 cell lines, only two mutations were observed, both in exon 19, in a leukemia and melanoma cell line. These mutations have not been described previously in clinical samples and their functional role is uncertain. The allele frequencies of the -216G>T, -191C>A, and R521K single nucleotide polymorphisms (SNPs) in the NCI-60 panel were 33%, 8.5%, and 27%, respectively; the V592A and C624F SNPs were not found in any NCI-60 cell line. The intron 1 CA repeat was highly variable in the cell lines; 32 cell lines having a total number of repeats below 35, and 27 having a total number of repeats above 35. The heterozygous and variant homozygous cell lines for the -216G>T SNP presented a significantly higher expression of the EGFR gene than the homozygous wild-type lines. In contrast, there was no association between the -191C>A or R521K SNPs and EGFR gene expression. No association could be detected between the number of CA repeats in intron 1 and the expression of EGFR. The cell lines having at least one variant T allele at the -216G>T SNP were more sensitive to erlotinib and less sensitive to geldanamycin, topoisomerase I and II inhibitors, and alkylating agents than those without a variant allele. No relationship was detected between anticancer drug sensitivity and the -191C>A SNP. The R521K SNP was associated to lower sensitivity to alkylating agents. The number of CA repeats was associated with significant differences in anticancer drug activity: a high total number of CA repeats (>35 per diploid genome) was associated to increased sensitivity to alkylating agents and topoisomerase I and II inhibitors. DISCUSSION: We provide evidence in this work that EGFR polymorphisms are associated with significant differences in the in vitro cytotoxicity of several types of DNA-interfering agents. Studies attempting a clinical validation of these clues are warranted.