Long-Term Toxicities of Selective Estrogen-Receptor Modulators and Antiaromatase Agents

With the advent of aromatase inhibitor use in the adjuvant setting,[1] and the inception of trials examining their use for breast cancer prevention, it seems prudent to evaluate what we know to date about the long-term effects of these agents. Unfortunately, unlike selective estrogen-receptor modulators (SERMs)-in particular tamoxifen(Drug information on tamoxifen),[2] which has been used for over 15 years in patients with early-stage breast cancer- long-term data on the use of aromatase inhibitors are minimal. As outlined by Mortimer and Urban, the largest data set concerning the toxicities of aromatase inhibitors comes from the Arimidex, Tamoxifen Alone or in Combination (ATAC) trial, the results of which were recently updated at 4 years.[3] By comparison, tamoxifen was approved in the late 1980s for the treatment of early-stage breast cancer in both preand postmenopausal women, resulting in almost 15 years of follow-up. In addition, as outlined by the authors, we have data from the experience of well women using tamoxifen as a chemopreventive.[4] In general, tamoxifen has a good toxicity profile and was described in 1996 by the World Health Organization as an essential treatment for breast cancer.[5] However, most likely because of its estrogen-like effects, tamoxifen increases the incidence of endometrial cancer and thromboembolic events.[2] Aromatase inhibitors do not produce these estrogen-like effects, and the long-term toxicities associated with these agents will likely relate to profound estrogen suppression. Endometrial Cancer Risk
Tamoxifen increases the incidence of endometrial cancer approximately fourfold in postmenopausal women.[4] To date, raloxifene(Drug information on raloxifene) (Evista) has not been associated with an increase in the incidence of endometrial cancer.[6] However, the association between tamoxifen and endometrial cancer was not noted until the drug had been in use for almost 15 years. Despite the fact that raloxifene does not exhibit estrogen-like effects on the endometrium in clinical studies, it promotes the growth of endometrial cancers in athymic mice.[7] Longer follow-up of osteoporosis trials and the Study of Tamoxifen and Raloxifene (STAR) will be necessary before we can definitively state that raloxifene is not associated with an increase in endometrial cancer. Results of the ATAC trial to date demonstrate a significant increase in endometrial cancers among tamoxifen- treated patients compared to anastrozole(Drug information on anastrozole)-treated patients.[8] Bones and Bisphosphonates
Prolonged estrogen suppression is likely to result in a loss in bone mineral density and a possible increase in fractures. At a relatively short follow-up of the ATAC trial, the incidence of fractures in the anastrozole-treated group was increased compared to the tamoxifen- treated group.[1] How important is this bone loss, and can it be prevented? The Austrian Breast Cancer Study Group treated 1,200 premenopausal women with early-stage breast cancer with goserelin(Drug information on goserelin) (Zoladex), and then randomized them to tamoxifen or anastrozole, with or without zoledronate (Zabel, Zometa) given every 6 months. At 18 months, both tamoxifen and anastrozole reduced bone density, based on T-scores in the lumbar spine (Figure 1).[9] Zoledronate completely prevented the bone loss associated with both agents, resulting in no change in T-score at 18 months.[9] This question will be addressed directly by the Zometa/Femara Adjuvant Synergy Trial (Z-FAST), in which postmenopausal women with early-stage breast cancer are treated with letrozole (Femara) and randomized to receive zoledronate at the start of letrozole(Drug information on letrozole) therapy or when their T-score drops to -2 standard deviations. Preclinical data on exemestane(Drug information on exemestane) (Aromasin) suggest that it can prevent ovariectomy-induced bone loss in rats, possibly via its androgenic effects.[10] Limited clinical data to date confirm this finding in humans.[11] Lipids and the Heart
Despite the fact that tamoxifen has a beneficial effect on the lipid profile by reducing low-density lipoprotein (LDL) cholesterol levels, it has never been demonstrated to reduce the risk of coronary artery disease.[2,4] This may be explained by the lack of benefit of hormone-replacement therapy on coronary heart disease.[12] Raloxifene has effects on the lipid profile similar to those of tamoxifen,[6] and its effect on the risk of coronary artery disease is being addressed by the Raloxifene Use for The Heart (RUTH) trial, as pointed out by the authors. Data on the effects of the aromatase inhibitors on the lipid profile are sparse, although this issue will be addressed in a subprotocol of the ATAC trial. Data from patients with metastatic breast cancer demonstrate no adverse effects of anastrozole on the lipid profile.[13] Letrozole may increase LDL cholesterol levels, whereas exemestane produces no such effect in preclinical models. Chemoprevention
One of the primary reasons that tamoxifen was chosen as a possible chemopreventive was the finding that it reduces contralateral breast cancer by 50% in patients whose first primary is estrogen-receptor positive.[2] Based on the finding that raloxifene reduced the incidence of breast cancer threefold compared to placebotreated patients in the osteoporosis trials,[7] the STAR trial, which now requires 19,000 patients and is 50% accrued, is comparing raloxifene to tamoxifen as a chemopreventive. At the initial presentation of the ATAC data at 33 months' follow-up, the incidence of contralateral breast cancer was significantly lower in anastrozole-treated patients than in tamoxifen-treated patients (Table 1).[1] At the more recent follow-up of 47 months, this difference no longer achieved statistical significance, although there continues to be significantly less invasive contralateral breast cancer among anastrozoletreated patients.[3] Longer followup will determine whether there is a difference in contralateral events between tamoxifen and anastrozole. Meanwhile, a number of preventive trials comparing the aromatase inhibitors to placebo are planned. Estrogen and the Brain
One of the biggest concerns associated with the use of aromatase inhibitors as adjuvant and preventive agents relates to the effects of profound, prolonged estrogen suppression on cognitive function. As outlined by Mortimer and Urban, raloxifene may have a beneficial effect on memory and attention in women over age 70, and no effect in other women.[14] No data are available on the effects of aromatase inhibitors on cognitive function. This is an important concern and one of the factors that led the American Society of Clinical Oncology expert committee to recommend that tamoxifen remain the gold standard adjuvant treatment for postmenopausal women. Conclusions
In summary, we have many years of data concerning the "good and bad" effect of tamoxifen. Based on the similarities between the two SERMs, raloxifene, perhaps apart from the issue of endometrial cancer, is likely to have a similar risk/benefit profile to tamoxifen. In contrast, we know nothing about the long-term effects of aromatase inhibitors, and patients should be monitored closely for long-term toxicities.