Tamoxifen-induced hot flashes are associated with estrogen receptor polymorphisms

Abstract

501 Background: Hot flashes are the most common side effect of tamoxifen. The pharmacogenetic predictors of hot flashes are not known. We hypothesized that estrogen receptor (ER) genotypes that have been associated with clinical phenotypes may predict the frequency and composite score of tamoxifen-associated hot flashes. Methods: We conducted a prospective observational study in 3 academic centers to evaluate the associations between ER and hot flash composite scores in women with breast cancer during the first year of adjuvant tamoxifen treatment. Menopausal status and prior chemotherapy history were collected at baseline. Medication records and validated hot flash diaries were prospectively collected before and following 1, 4, 8 and 12 months of starting tamoxifen. ERa (ESR1) PvuII (rs2234693) and XbaI (rs9340799) genotypes were determined using a Restriction Fragment Length Polymorphism (RFLP) assay; and ERβ (ESR2–02; rs4986938) genotype was determined using a Taqman™ assay. The haplotype of two ESR1 SNPs and their association with the baseline hot flash score were analyzed with haplo.stat function in R. Individual SNPs and their interactions in baseline hot flash score and month 4 hot flash score change were analyzed with SAS PROC GLM. Results: Of the 298 women recruited to the trial, 286 returned their baseline diaries and 251, 239, 214, and 213 women returned their diaries 1, 4, 8, and 12 months after tamoxifen treatment. At baseline, premenopausal women who were homozygous for the ESR1 CG haplotype reported higher hot flash scores (17.2±3.7) than carriers with one or less CG allele (5.2±3.9), p=0.01. After tamoxifen treatment, postmenopausal women with ESR1 PvuII CC and ESR2–02 GG genotype had the greatest increase (p=0.0004) in hot flash score (53.9±16.4) as compared with women with other genotypes (12.5±3.6). Women who were homozygous for the ESR2–02 AA allele were significantly less likely to experience tamoxifen-induced hot flashes than women with at least one G allele (OR=4.0, P=0.0006) regardless of menopausal status and chemotherapy history. CYP2D6 analysis in underway and results will be available for the meeting. Conclusions: Menopausal status, history of prior chemotherapy, and ER genotype may be used to predict who may suffer hot flashes during tamoxifen treatment. [Table: see text]