Breast cancer is the most common female malignancy in the Western world. Two-thirds of all breast cancers are estrogen receptor (ER)-positive, a phenotypic characteristic that is prognostic of disease-free survival and predictive of response to endocrine therapy.
Through its binding to either ER-alpha or -beta, estrogen is known to regulate a wide variety of cellular effects and physiologic conditions including breast cancer cell proliferation. Upon ligand binding, ERs dimerize to form an activated receptor and subsequently bind to specific DNA sequences (estrogen response elements), or interact with other protein/DNA complexes, to regulate target gene expression. Selective estrogen receptor modulators (SERMs) such as tamoxifen(Drug information on tamoxifen) disrupt ER activity in cells by blocking estrogen binding to the receptors. In the case of tamoxifen, administration for 5 years following breast cancer surgery almost halves the annual recurrence rate and reduces the breast cancer mortality rate by one-third in both pre- and postmenopausal women.
‘Jekyll and Hyde’ Phenomenon
Unfortunately, there are unpredictable aspects of tamoxifen therapy. In the clinical setting, oncologists sometimes observed “tumor flare” following administration of tamoxifen to women with metastatic breast cancer. Furthermore, in the adjuvant setting, a peak in the hazard rates for recurrence was consistently observed within the first 2 to 3 years in the adjuvant tamoxifen trials.[1] While the reasons for these clinical observations were not entirely clear, preclinical evidence suggested that tamoxifen could stimulate nongenomic growth-factor pathways and that in vivo administration of tamoxifen to mice bearing HER2-transfected ER-positive xenografts stimulated tumor growth.[2] However, the mystery surrounding this “Dr. Jekyll and Mr. Hyde” phenomenon with tamoxifen has continued to baffle most researchers, as the large randomized trials comparing tamoxifen and the aromatase inhibitiors (AIs) have yet to identify a tumor-specific biomarker (such as HER2) or gene profile predictive for response to tamoxifen.[3,4]
Aromatase Inhibitor Strategy
Aromatase inhibitors were developed based on the premise that reduction of the ligand responsible for stimulation of breast cancer growth would be a superior way to treat breast cancer. Randomized trials comparing tamoxifen with the third-generation AIs in the metastatic setting demonstrated that AIs had superior response rates in most[5-7] but not all studies.[8] This led to large “head to head” studies comparing tamoxifen and aromatase inhibitors in the adjuvant treatment of breast cancer, which demonstrated a small but statistically significant improvement in disease-free survival in favor of anastrozole(Drug information on anastrozole) (Arimidex) and letrozole(Drug information on letrozole) (Femara).[9,10] However, the small difference in disease-free survival comparing tamoxifen to AIs did not translate into a survival advantage. A recent meta-analysis representing data from over 20,000 women confirmed that the aromatase inhibitors offer no survival advantage.[11] This has left the oncology community in the strange position of trying to move from the pre-genomics era of “evidence based” science—in which thousands of patients are enrolled in clinical trials in order to proclaim a “winner” for the “average patient”—to 2009, when personalized medicine is a buzzword but not yet integrated into standard clinical practice in regard to the selection of hormonal therapy.
Genomics Revolution
While studies comparing tamoxifen and aromatase inhibitors were ongoing, a revolution was underway—the genomics revolution. Technologic advances allowed the rapid and accurate assessment of tumor gene expression and function, both at the level of individual genes and by global gene analysis. In the latter example, the expression patterns of thousands of tumor genes could be determined at one time through the use of tissue microarrays, which were critical in identifying specific biologic subsets of cancer more likely to relapse (prognostic) as well as the identification of genes or gene patterns associated with response (prediction).
In the well written paper by Ma et al, the authors summarize findings regarding multiple gene-expression profiles as well as prognostic factors associated with endocrine therapy outcome, mainly related to aromatase inhibitors. However, as of 2009, the study of the tumor genome has yet to identify a predictive marker that allows clinicians to discriminate which endocrine therapy, tamoxifen or AI, should be delivered to a specific patient.
In the specialty of oncology, it is becoming increasingly clear that clinicians must take into account all sources of genetic variation that influence drug effect. This includes both tumor (somatic) as well as host (germline) genetic variation. It should be noted that while the common final pathway of both tamoxifen and the AIs is to disrupt estrogen signaling, both classes of drugs undergo vastly different metabolic routes of elimination and/or activation, and each of these metabolic steps is under genetic control. Therefore, common genetic variations that alter the proteins involved in the metabolism, uptake, or distribution unique to each drug would be a leading candidate for a predictive factor. An obvious example is tamoxifen and the enzyme CYP2D6.
Tamoxifen (Mr. Hyde), Endoxifen (Dr. Jekyll)?
Tamoxifen is a weak antiestrogen (with the ability to stimulate breast cancer growth in some model systems), but metabolism leads to the activation of the drug. Although the tamoxifen metabolite endoxifen was first identified in 1989,[12] it was not until 2003 that its effects on breast cancer proliferation and the enzymes involved in its formation were characterized. Endoxifen is approximately 100-fold more potent than its parent drug (tamoxifen) as an antagonist of the estrogen receptor and is formed by the CYP2D6-mediated oxidation of N-desmethyl tamoxifen.[13,14] Common CYP2D6 genetic variation and/or drug-induced inhibition of CYP2D6 enzyme activity significantly reduces endoxifen plasma concentrations in humans.[15] Seven studies from different cohorts of tamoxifen-treated women in the adjuvant setting have established that patients with decreased CYP2D6 metabolism are at significantly increased risk of breast cancer recurrence compared to those with substantial 2D6 activity.[16-23]
Perhaps the most intriguing part of this story, however, is the recent “bedside-to-bench” discovery demonstrating that the mechanism of action of endoxifen differs substantially from that of either tamoxifen or 4-hydroxy tamoxifen. Endoxifen is a potent antiestrogen that mimics the actions of fulvestrant (Faslodex) with regard to its ability to target ER-alpha for degradation by the proteasome, to block ER-alpha–mediated transcriptional activation, and to inhibit estrogen-induced breast cancer cell proliferation.[24] Importantly, the mechanism by which endoxifen blocks ER signaling and its effect on proliferation is dose-dependent and is maintained even when cells are pretreated with the same concentrations of tamoxifen and 4-hydroxy tamoxifen seen in humans.[24] These findings for the first time identify the reason why metabolic activation by CYP2D6 is key to understanding tamoxifen drug effect, and are likely to unlock the “Dr. Jekyll and Mr. Hyde” mystery surrounding some of the unpredictable aspects of tamoxifen therapy.
In the case of AIs (and in contrast to tamoxifen), the parent drug is considered active, and metabolism inactivates the parent drug. Also in contrast to tamoxifen, CYP2D6 is not known to be involved in the metabolism of any of the three aromatase inhibitors. Large ongoing investigations, including genome-wide association studies, are testing the hypothesis that genetic variation in genes that encode proteins involved in the metabolism, uptake, and distribution of these drugs might be associated with clinical outcome and side effects. Like tamoxifen, this hypothesis is based on the knowledge that there are substantial differences in the frequency of these genetic changes across different populations.
Conclusions
Although the promise of pharmacogenetics and pharmacogenomics has yet to be fully achieved, health-care professionals increasingly have access to gene sequence information. When applied properly, this information has the potential to improve our ability to select the optimal drug at the optimal dose for each patient. We must be clear that the future of individualized medicine in oncology will require consideration of all sources of genetic variation—the tumor and the host. As a result, pharmacogenetics and pharmacogenomics promise to fundamentally alter rational drug therapy.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
