Electronic searches were performed by databases of PubMed from its inception to June 2010. To achieve the maximum sensitivity of the search strategy and identify all studies about breast cancer and endocrine therapy resistance, we used appropriate free text and thesaurus terms including “breast cancer”, “breast carcinoma”, “breast tumour”, “mammary cancer”, “endocrine therapy”, “drug resistance”, “drug tolerance” and all other relative information about breast. The MeSH table was searched by “breast neoplasms” [MeSH Terms] AND (“endocrine system”[MeSH Terms] AND “therapy” [Subheading] OR “therapeutics” [MeSH Terms]) AND resistance [All Fields]. The reference lists of all retrieved articles were reviewed for further identification of potentially relevant studies.

All studies assessing breast cancer, endocrine therapy and drug resistance published were included. No restrictions were placed on abstracts and conference proceedings. We excluded studies that were not directly relevant to drug resistance to breast cancer endocrine therapy, such as resistance to chemotherapy combined with endocrine therapy. Review about drug resistance to breast cancer endocrine therapy was excluded because we were about to explore the original mechanisms and countermeasures to drug resistance to breast cancer endocrine therapy, actually there is no reviews concentrating on drug resistance to breast cancer endocrine therapy.

Several breast cancer cell lines have been applied in research of mechanism and countermeasures of drug resistance in breast cancer endocrine therapy, for example, MCF-7 cell line (6,15-17), MDA-MB-231cell line (6), LTLTCa cell line (4) and T47D cell line (15). Under a certain process, those cell lines could evolve various cell lines that have different characteristics but all of them have resistance to breast cancer endocrine therapy. For example, if cultured in medium rich in OH-TAM for 6 months, MCF-7 breast cancer cell line would evolve into CL6.8 cell strain, which has resistance to TAM (6); if cultured in medium rich in OH-TAM and Fulvestrant, cell strain resistant to OH-TAM and Fulvestrant would be evolved (6). And MCF-7/HER2-18 is ER positive cell lines with HER2 over-expression, which was built by stable transfecting MCF-7 with over-expressed HER2; while MCF-7 wild type (MCFwt) is positive ER cell lines without HER2 over-expression (19). Through assays of cell growth and cell migration&invasion, changes in different experimental group were evaluated (15); through western blotting, relevant proteins’ level in cell lines with or without resistance to endocrine drugs could be compared, hence the mechanism of endocrine resistance could be further analysed (15). After inoculated with endocrineresponsive cell lines (for example MCF-7Ca cells), mice could be assigned into different groups when the tumors reached a measurable size. The mice could be killed and tumor tissues were collected with addition of endocrine drugs when: 1) tumor volume began to shrink; 2) tumor volume stopped shrinking and began to accrete; 3) tumor accreted to several times of origin volume. Immuno-blotting and other methods could be used to analyse relevant proteins’ level of these tumor tissues and then the mechanism of endocrine resistance could be further analysed (4).

As to the problem of endocrine resistance, common research of animal models in exploring mechanisms and countermeasures is as follows: establishing animal models with endocrine resistance through subcutaneous inoculation of tumor cells with resistance to endocrine drugs, then in vivo observation of tumor volume which would reveal the difference of various drugs in inducing and reversing endocrine resistance to tumors was performed. Usually when tumors reached a sufficient size, for example, 150-200mm³, the animals were randomly assigned to various treatment groups (19). Mice were frequently chosen to build animal models (18-19), for example, subcutaneous inoculation tumor cells in ovariectomized & immunosuppressed mice so as to simulate the postmenopausal breast cancer patients because the source of estrogen after menopause is from nonovarian tissue and is not under regulation by gonadotropins, which could be used to explore mechanisms of endocrine resistance and countermeasures to AIs (4). To explore mechanisms of endocrine resistance and countermeasures to TAM, premenopausal animal models could be built by subcutaneous heeling-in slow-release estradiol pellets in ovariectomized mice which could simulate the in vivo estrogen release (20). Ovariectomized athymic nude mice in the presence of estrogen could also be used in establishing xenografts (19). In vivo observation of resistancerelevant proteins expression levels would promote thorough analysis the mechanism of endocrine resistance.

Phosphorylation is one of the post-translational modification (PTMs) in cells including phosphorylation, glycosylation and acetylization, and ER is the main target of phosphorylation (21). Phosphorylation of ERs plays a pivotal role in acquiring endocrine resistance to TAM (22). In TAM-resistant MCF-7 Her-2/neu breast cancer cell lines and TAM-resistant breast cancer tumor tissues, it has been demonstrated by western blotting that specific ligand-dependent endogenous phosphorylation of ER occurs at S118 and S167 (22-25). Phosphorylated ER would lose ligand-dependance. While in v itro research revealed that different mechanisms of phosphor ylation would lead to different biological characteristics of ER: ER phosphorylation induced by steroid receptor coactivator (SRC) and protein kinase A (PKA) would increase ER’s affinity to estrogen (E2); ER phosphorylation induced by mitogen-activated protein kinase (MAPK) would reduce ER’s affinity to trans-hydroxytamoxifen (TOT) (22). Phosphorylated-ER’s affinity to estrogen response elements (ERE) would be significantly reduced with stimulation of exogenous TOT no matter by which mechanisms ERs were phosphorylated (22). With presence of E2, ER phosphorylation induced by AKT, MAPK and PKA would all increase the mutual interaction between DNA combining receptors and SRC3 receptors (22). Compared with TAM, although it takes a much longer time for AIs to induce breast cancer endocrine resistance, phosphorylation of ERs still plays an important role in acquiring endocrine resistance to AIs (3,21). It has been demonstrated that expression level of phosphorylated-ERα in tumor tissues resistant to letrozole is much higher than that in control group, which is sensitive to letrozole; while expression level of ERα(not phosphorylated) is much lower than that of control group (4). Meanwhile, expression level of phosphorylated-ERα will be even higher in tumor tissues resistant to letrozole if tumors continue growing under a certain concentration of letrozole, and expression level of ERα (not phosphorylated) will be even lower, while expression level of MAPK will be higher and expression level of PR will be constant (4). There are many phosphorylation sites of ERα, and different sites were phosphorylated by different mechanisms (21).

It has been stated above that SRC plays an important role in the process of ER phosphorylation. SRC acts as a non-receptor tyrosine kinase, and its overexpression has a close correlation with various malignant tumor genesis including breast cancer (26-27). In vitro research has revealed that SRC has correlation with development of breast cancer endocrine resistance (17). Overexpression of SRC will weaken tumor cells’ sensitivity to TAM in MCF-7 breast cancer cells which are with positive ERα and sensitive to TAM; while using inhibitor of SRC could reduce its expression and restore tumor cells’ sensitivity to TAM in breast cancer endocrine resistant cells developed from TAM-sensitive MCF-7 breast cancer cells. The expression level of SRC in cytoplasma of breast cancer tumor cells is much higher than that in cytoplasma of breast cells besides tumor in human (p＜0.01); while in nucleolus the expression level of SRC is just the other way. This implies that the correlation between the reactivity of breast cancer cells to endocrine drugs and SRC expression levels in cytoplasma or nucleolus might be poles apart. AZD0530 could inhibit SRC kinase activity dosedependently, as shown by a decrease in phosphorylaiton of SRC at Y419 (15). In addition, AZD0530 together with TAM significantly suppress expression of the Ki-67 antigen, and has correlation with expression of both cyclin-D1 (necessary for the progression of cells from G1 to S phase) and C-Myc (a positive regulator of cellular proliferation) (15).

The ErbB family has 4 members, including ErbB1, ErbB2, ErbB3 and ErbB4, and all of them are tyrosine kinase receptors. Research conducted by Ghayad et al (6) has demonstrated that ErbB1, ErbB2 and ErbB3 are activated and ErbB4 is highly expressed in endocrine resistant breast cancer cells, while ErbB heterodimers and various ligands relevant to ErbB are also highly expressed. AKT and MAPK are the main biological targets in the downstream of ErbB family associated signal transduction pathway, and activated AKT and MAPK have a close correlation with breast cancer endocrine resistance (16). The activated AKT has correlation not only with endocrine resistance but also with poor prognosis (28). In breast cancer cell lines with resistance to TAM and Fulvestrant, the MAPK pathway is highly activated (29-30), which will make ERαfurther phosphorylated and make cells even more resistant to endocrine drugs through various ways (31,32). It has been demonstrated by western blotting that the biological target MAPK and PI3K/AKT are activated in breast cancer cells resistant to endocrine drugs, while highly expressed MAPK has close correlation with phosphorylation of serine at site 118 in region AF1 of ERα, and highly expressed PI3K/ AKT has close correlation with phosphorylation of serine at site 167 in region AF1 of ERα (6). In clinical settings, patients usually benefit less from TAM if the tumors are with positive ERαand MAPK highly phosphorylated (33).