| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Division of Cancer Prevention National Cancer Institute Breast and Gynecologic Cancer Research Group Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: Karen A. Johnson, Division of Cancer Prevention, National Cancer Institute, Breast and Gynecologic Cancer Research Group, Bethesda, Maryland 20892. E-mail: kj80y{at}nih.gov.
Raloxifene is a selective estrogen receptor (ER) modulator (SERM) that received United States Food and Drug Administration (FDA) approval in 1997 for the prevention of osteoporosis and in 1999 for osteoporosis treatment. Currently, about 500,000 postmenopausal women in the United States are using raloxifene for these indications (1). Beyond the United States, raloxifene has been approved for use in more than 90 countries, and prescriptions have been issued to more than 10,000,000 women worldwide. The osteoporosis registration studies for raloxifene were noteworthy for providing evidence that its use is associated with the reduction of breast cancer risk. This benefit was recently confirmed in a population of women at high risk of developing breast cancer, as described in published results from the Study of Tamoxifen and Raloxifene (STAR) (2). In this study, tamoxifen served as the standard for comparison, having received U.S. approval for breast cancer risk reduction in 1998. The STAR trial was a double-blind, randomized clinical trial in postmenopausal women for whom raloxifene was found to be comparable to tamoxifen in reducing the rate of invasive breast cancer. In addition, thromboembolic events, cataracts, and uterine cancer occurred less often in the women assigned to raloxifene, whereas no substantive differences were seen in the occurrence of fractures, other invasive cancers, heart disease, or stroke. Results from STAR have prompted Eli Lilly, the pharmaceutical sponsor, to return to the United States FDA by the end of 2006 with a request to add breast cancer risk reduction to the raloxifene label. This anticipated milestone presages a surge of scientific effort to understand the manifold activities of raloxifene in greater detail and to further refine the use of SERMs in the clinic.
In aggregate, the pharmacological profile of raloxifene is favorable, leaving for debate whether a SERM of even greater worth could be developed for disease prevention. Discovered in 1979, the preclinical testing of raloxifene showed activity in multiple estrogen sensitive tissues (e.g. breast, bone, and endometrium) as well as an effect on lipids (3). In subsequent clinical testing, raloxifene demonstrated a limited effectiveness for breast cancer treatment but qualified for a major role in the management of osteoporosis (4). Now, after a head-to-head comparison of tamoxifen and raloxifene, the apparent advantage of raloxifene over tamoxifen in the postmenopausal STAR population is not based on a difference in breast cancer prevention or fractures, but rather in lower rates of several undesirable consequences, which are statistically significant for thromboembolism and cataracts and of borderline significance for uterine cancer.
Despite the favorable side-effect profile in STAR, concern about SERM side effects in premenopausal women has limited the opportunities for clinical evaluation of raloxifene in younger women. In this issue, Eng-Wong et al. (5) report results from a clinical assessment of raloxifene in a series of premenopausal women, who were selected for study because of their high risk of developing breast cancer. After 2 yr of raloxifene (60 mg daily with 1200 mg of calcium carbonate), the main finding was a decrease in bone mineral density as measured by dual energy x-ray absorptiometry. One year after drug cessation, the decline in bone mineral density appeared to reverse. This result is not surprising. Loss of bone mineral density in premenopausal women has also been observed with tamoxifen (6), thus suggesting that both raloxifene and tamoxifen block the usual effect of estrogen on bone homeostasis in premenopausal women. The paradoxical reduction of bone mineral density in premenopausal women by both raloxifene and tamoxifen awaits mechanistic clarification. Had raloxifene facilitated the preservation of premenopausal bone mineral density as hypothesized, that information would have added weight to a justification for pursuing the larger-scale studies needed to assess raloxifene for reducing breast cancer risk in younger women. At this point, tamoxifen remains the only drug approved for breast cancer risk reduction in premenopausal women. Although aromatase inhibitors are being studied for breast cancer prevention, they have not been developed for premenopausal women because of concern about osteoporosis and ovarian hyperstimulation. This situation leaves the door open for a new SERM that is sparing of bone mineral density in premenopausal women.
The study by Eng-Wong et al. (5) provided a rare opportunity to obtain clinical results from premenopausal women taking raloxifene. In addition to the loss of bone mineral density, other significant findings published here include a correlation between bone mineral density and decreasing serum osteocalcin, an increase in high-density lipoprotein cholesterol, and a decrease in apolipoprotein B and fibrinogen. For the same study population, a significant increase in serum estradiol, SHBG, IGF-binding protein-3, and leptin has been reported elsewhere (7, 8). These results partially overlap and are consistent with those from one other published study of short-term raloxifene use in premenopausal women (9). Review of these publications in detail reminds us that, in addition to the current focus on breast and bone, the other target organs of raloxifene that need to be considered for assessing net clinical benefit include, but are not limited to, endometrium, ovary, liver, hypothalamus, and pituitary.
An early hope that SERMs might be used to reduce mortality from coronary heart disease (CHD) remains unrealized. Like estrogen, raloxifene shows improvement in serum markers of cardiovascular risk, namely cholesterol, fibrinogen, and homocysteine. However, the Women’s Health Initiative clinical trial confirmed an unexpected increase in death from myocardial infarction and other coronary events in postmenopausal women using estrogen or hormone replacement therapy with estrogen plus progestin (10). Uncertainty about the role that raloxifene might play in postmenopausal cardiovascular events has been addressed by the recent publication of results from the Raloxifene Use for The Heart (RUTH) trial (11). In RUTH, even though breast cancer risk was reduced, raloxifene compared with placebo did not significantly reduce selected myocardial events. Given these results and others, it appears that the margin of opportunity for existing SERMs to reduce CHD is small (12), but the possibility of an interaction between exogenous hormone exposure and menopausal status in regard to CHD has not been fully explored. Observational studies, though fraught with the possibility of bias and confounding, suggest that oral contraceptive use may offer protection against CHD in later life (13). Clarification is obviously needed.
So what direction might SERM development take in the future? A number of issues have been identified. The ideal SERM for premenopausal women is likely to be different from the ideal SERM for postmenopausal women. Also, it is unclear whether there is a limit to major benefits that can be expected from a single drug, particularly in comparison with a combination of more selective agents. Is it unrealistic to hope for CHD reduction in addition to osteoporosis prevention and breast cancer risk reduction from a single, ideal SERM? Also there are intriguing suggestions in the literature that modulation of ERs may be of value in suppressing lung, colorectal, and ovarian carcinogenesis (14, 15, 16). These considerations bring to mind the vision of a super SERM that maintains bone density while preventing heart disease and multiple cancers. Time will tell if such a conjecture comes to fruition or otherwise represents the impossible dream. Recent clinical findings like the ones reviewed here can be of great importance to the translational research cycle, framing the mechanistic research questions that will lead back to the laboratory for the design of improved drugs.
Extensive work in the laboratory has established the basic, albeit complicated, model of SERM activity in estrogen signaling pathways. It is accepted that dimerization of ER follows the binding of its natural ligands, initiating a cascade of activating or repressing events that ultimately determine the transcriptional repertoire in a given tissue (17). With competitive binding to ER
, manmade ligands like tamoxifen and raloxifene block downstream activation mediated through various estrogen response elements (ERE) in the nucleus. Alternatively, binding of tamoxifen or raloxifene to ERß can block the usual repressive effects of estrogen through ERß at ERE. The nonclassic AP-1 signaling pathway provides a different route for ER-liganded SERMs to influence non-ERE-mediated gene transcription. With these factors alone, a deterministic multiplex model of SERM function in a given tissue type would start with the following constituents: two estrogen receptors (ER or ERß predominating but possible coordinating), multiple ligands or candidate SERMs in competition with natural estrogens at various concentrations, and two types of transcriptional docking stations for ER-SERM complexes. Having neglected to mention the numerous coactivators and corepressors of the ER-ligand complex, it is apparent that many additional components and interactions are available for testing in a more complex model. Could this approach be taken to a new level?
Progress in systems biology (18) suggests that a multiscale modeling approach might be used to advantage to simulate the dynamic and complex interactions that occur within the networks and pathways in an estrogen-sensitive tissue and across those tissues (19). In support of this approach, new information technologies allow computational models of SERM interactions with receptor isoforms (ER vs. ERß) (20) and other binding interactions. Experimental feedback from physiological systems is available from a variety of sources including animal models, cell cultures, and samples from clinical studies to inform the modeling process. Although modeling has its own set of difficulties, systems biology may offer a hedge against the risk and extensive but limited resources involved in mounting clinical trials with thousands of people (19,747 in STAR and 10,101 in RUTH). For the individuals who might benefit from SERMs and whose very survival or quality of life could hang in the balance, the time to test a systems approach is now.
Footnotes
Abbreviations: CHD, Coronary heart disease; ER, estrogen receptor; ERE, estrogen response elements; SERM, selective ER modulator.
Received August 9, 2006.
Accepted August 14, 2006.
References
and ß with potent selective ligands from structure-based design. Mol Endocrinol 18:1599–1609
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
