EFFECTS OF CYP2D6, CYP3A5 AND SULT1A1 POLYMORPHISMS ON TAMOXIFEN AND ITS METABOLITES AND ESTROGEN CONCENTRATIONS IN THAI BREAST CANCER PATIENTS

Abstract

Tamoxifen (TAM) is a member of selective estrogen receptor modulators which has been used as antiestrogen therapy in estrogen receptor positive breast cancer patients. TAM needs several CYP450 enzymes to convert its pro-drug form to the two active metabolites (endoxifen (END) and 4-hydroxytamoxifen (4OHT)) which are the competitive inhibitors with estrogen for acting on estrogen receptor. The polymorphisms of TAM-metabolizing enzymes encoded genes; CYP2D6 or CYP3A5, showed potentially effects on TAM and its metabolites concentrations which might subsequently leaded to treatment failure including recurrent or death. The high prevalence of CYP2D6 and CYP3A5 incomplete functional alleles were reported in Thai but the associations of those alleles with TAM and its metabolites in large numbers of patients including estrogen concentration and estrogen-metabolizing enzyme polymorphisms (SULT1A1 copy numbers variations (CNVs)) in Thai breast cancer patients have never been investigated before. This research was conducted to determine the effects of CYP2D6*2, CYP2D6*10 and CYP3A5*3 on TAM, NDMT, END and 4OHT concentrations including estrogen concentrations, SULT1A1 CNVs and TAM-associated ADRs. The prevalence of SULT1A1 CNVs including the association of those genes with estrogen concentrations and ADRs in Thai were also firstly reported from this study. The 134 Thai breast cancer patients were recruited to the study. The blood samples were analyzed by real-time PCR with TaqMan® assay, HPLC-FLU and ELISA. Most patients were in stage I and stage II breast cancer ranging in age from 27.0-82.0 years. Their BMI were ranged from 15.4-40.0. Mean of TAM used was 21.4 (SD 16.1) months. They were ER+/PR+(71.6%), ER+/PR-(26.9%), ER-/PR+(0.8%) and ER-/PR-(0.8%). CYP2D6*10/*10, CYP3A5*1/*3 and SULT1A1x2 copies were the most common genotypes while CYP2D6-IM and CYP3A5-EM were the most common phenotypes. CYP2D6 phenotypes were statistically affected NDMT (P=0.013), END (P=0.014) and 4OHT (P=0.017) concentrations while those effects from CYP3A5 polymorphisms were not suggested. SULT1A1 CNVs were not associated with estrogen concentrations or ADRs in premenopausal sub-group analysis; however, patients with ADRs had statistically higher estrogen concentrations than those patients without ADRs (P=0.014).