The estrogen receptor-alpha A908G (K303R) mutation occurs at a low frequency in invasive breast tumors: results from a population-based study

Abstract

Introduction: Evidence suggests that alterations in estrogen signaling pathways, including estrogen receptor-alpha (ER-alpha), occur during breast cancer development. A point mutation in ER-alpha (nucleotide A908G), producing an amino acid change from lysine to arginine at codon 303 (K303R) results in receptor hypersensitivity to estrogen. This mutation was initially reported in one-third of hyperplastic benign breast lesions, although several recent studies failed to detect it in benign or malignant breast tissues.

Methods: We screened 653 microdissected, newly diagnosed invasive breast tumors from patients in the Carolina Breast Cancer Study, a population-based case-control study of breast cancer in African American and white women in North Carolina, for the presence of the ER-alpha A908G mutation by using single-strand conformational polymorphism (SSCP) analysis and 33P-cycle sequencing.

Results: We detected the ER-alpha A908G mutation in 37 of 653 (5.7%) breast tumors. The absence of this mutation in germline DNA confirmed it to be somatic. Three tumors exhibited only the mutant G base at nucleotide 908 on sequencing, indicating that the wild-type ER-alpha allele had been lost. The ER-alpha A908G mutation was found more frequently in higher-grade breast tumors (odds ratio (OR) 2.83; 95% confidence interval (CI) 1.09 to 7.34, grade II compared with grade I), and in mixed lobular/ductal tumors (OR 2.10; 95% CI 0.86 to 5.12) compared with ductal carcinomas, although the latter finding was not statistically significant.

Conclusion: This population-based study, the largest so far to screen for the ER-alpha A908G mutation in breast cancer, confirms the presence of the mutant in invasive breast tumors. The mutation was associated with higher tumor grade and mixed lobular/ductal breast tumor histology.

Figure 1

SSCP and sequencing analysis of the ER-α A908G mutation in invasive breast tumors. (a) Single-strand conformational polymorphism (SSCP) analysis showing the band shift characteristic of the estrogen receptor-α (ER-α) A908G mutation in ductal breast tumor 52 and in the positive control tissue MN31 (Pos). Tumors 51 and 53 to 55 show the wild-type SSCP pattern, as does the MCF-7 negative control cell line (Wt). Und is the undenatured wild-type control. (b) ³³P-Cycle sequencing analysis confirming the presence of the ER-α A908G mutation in breast tumors 52, 15, 28, 313 and 341, as indicated by the upper arrow (A→G). Tumor 87 shows only wild-type ER-α. Note that tumor 52 also carried a second mutation, a deletion of A in codon 302, indicated by the lower arrow (del A), which seems to be in the allele exhibiting the A908G mutation in codon 303.

Figure 2

Comparison of screening methods in one ER-α A908G mutation-positive breast tumor from the CBCS. (a) Single-strand conformational polymorphism analysis of three Carolina Breast Cancer Study (CBCS) breast tumors, with the mutant band in tumor 622 indicated by an arrow. Tumors 615 and 617 contain only wild-type estrogen receptor (ER). (b) ³³P-cycle sequencing showing the A908G point mutation in CBCS breast tumor 622 (indicated by the arrow). (c) SNaPshot dideoxy primer extension analysis showing the wild-type (wt) and mutant (mut) ER peaks in both the 5' and 3' directions in CBCS breast tumor 622. Note that in the 5' direction the wild-type peak (A) is green and the mutant (G) is blue, whereas in the 3' direction, the wild-type peak (T) is red and the mutant (C) is black. (d) Big Dye terminator cycle sequencing analysis in the exon 4 region of ER-α surrounding codon 303 in CBCS breast tumor 622. Codon 303 is boxed; note that only the wild-type sequence (AAG) is observed.