The aromatase inhibitors (AIs) anastrozole (Arimidex), letrozole (Femara), and exemestane (Aromasin) are significantly more effective than the selective estrogen-receptor modulator (SERM) tamoxifen in preventing recurrence in estrogen receptor-positive early breast cancer. Aromatase inhibitors are likely to replace SERMs as first-line adjuvant therapy for many patients. However, AIs are associated with significantly more osteoporotic fractures and greater bone mineral loss. As antiresorptive agents, oral and intravenous bisphosphonates such as alendronate (Fosamax), risedronate (Actonel), ibandronate (Boniva), pamidronate (Aredia), and zoledronic acid (Zometa) have efficacy in preventing postmenopausal osteoporosis, cancer treatment-related bone loss, or skeletal complications of metastatic disease. Clinical practice guidelines recommend baseline and annual follow-up bone density monitoring for all patients initiating AI therapy. Bisphosphonate therapy should be prescribed for patients with osteoporosis (T score < -2.5) and considered on an individual basis for those with osteopenia (T score < -1). Modifiable lifestyle behaviors including adequate calcium and vitamin D intake, weight-bearing exercise, and smoking cessation should be addressed. Adverse events associated with bisphosphonates include gastrointestinal toxicity, renal toxicity, and osteonecrosis of the jaw. These safety concerns should be balanced with the potential of bisphosphonates to minimize or prevent the debilitating effects of AI-associated bone loss in patients with early, hormone receptor-positive breast cancer.
As a result of earlier diagnosis, more efficacious treatments, and longer survival, more patients than ever are receiving long-term treatment for breast cancer. Improved survival is due in part to the successful use of adjuvant hormonal therapies in estrogen and/or progesterone receptor-positive disease. The increasing breast cancer survivor population—estimated at slightly over 2 million women in the United States in 2005—may be at risk for long-term side effects of these beneficial treatments.
Recent randomized clinical trials have revealed the efficacy of aromatase inhibitors (AIs) as adjuvant therapy for early breast cancer in postmenopausal women.[3,4] However, the substantial benefits documented for this class of drug in the short term are associated with a significant, but manageable, increase in adverse events related to bone health. The cumulative incidence of AI-associated bone loss in patients with breast cancer in the long term is not known. The AIs anastrozole (Arimidex), letrozole (Femara), and exemestane (Aromasin) have been shown to be significantly more effective than the selective estrogen-receptor modulator (SERM) tamoxifen in preventing disease recurrence in estrogen receptor-positive early breast cancer. Aromatase inhibitors are likely to replace SERMs as first-line adjuvant therapy for many patients.[3-5] The growing use of AIs mandates vigilance about bone health in the oncology community. This article reviews the biology of AI-associated bone loss, summarizes recent safety data from clinical trials, and discusses options for preserving bone health in patients with breast cancer.
Pathophysiology of AI-Associated Bone Loss
The ongoing process of bone remodeling in the adult skeleton reflects a balance between the resorptive activity of osteoclasts and bone-forming action of osteoblasts, processes that are normally coupled and in equilibrium. Normal bone turnover is maintained through a complex regulatory system of both systemic and local factors. Molecular factors involved in bone resorption and remodeling include osteoprotegerin (OPG), receptor activator of nuclear factor kappa B (RANK) and RANK ligand (RANKL). RANKL activates and binds to the RANK receptor on osteoclasts and their precursor cells, promoting osteoclast formation and prolonged osteoclast lifespan. Osteoprotegerin acts as a decoy receptor of RANKL and inhibits RANKL binding to RANK. Altering the relative biological availabilities of OPG and RANKL has direct consequences for the regulation of normal bone remodeling and in the pathogenesis of cancer-induced bone loss.[7,8]
Estrogen and Bone Physiology
Although the importance of the steroid sex hormones in bone metabolism has been recognized for many decades, their cellular and molecular mechanisms of action are still being elucidated.[3,9] Broadly speaking, estrogen has suppressive, antiresorptive effects on osteoclasts during bone remodeling. These pleiotropic effects are mediated through intracellular and cell surface estrogen receptors expressed by osteoclasts and osteoblasts and are largely indirect. For example, estrogen modulates osteoblast-derived cytokines, resulting in decreased differentiation and maturation of osteoclasts from precursors and increased osteoclast apoptosis (programmed cell death).[6,9] Estrogen may also induce direct effects on osteoblasts, further favoring bone formation over resorption. In concert, these cellular responses have the net effect of dampening bone turnover under estrogen-replete conditions (Figure 1).
Estrogen deprivation that occurs with natural or treatment-associated menopause increases bone turnover and osteoclast activity, causing bone resorption and formation to become unbalanced. This imbalance results in net bone loss as well as decreased bone quality, a term that encompasses both bone microarchitecture and degree of mineralization. Bone quality is increasingly recognized as an important determinant of overall bone health, and studies have shown that substantial deterioration of bone microarchitecture can occur before bone density is affected.
Estrogen Blockade With AIs
In the postmenopausal state, the bulk of estrogen production switches from the ovaries to other sites such as adipose tissue, adrenal glands, smooth muscle, and bone.[3,9] Adrenally derived androgens circulate to peripheral tissues where they are converted to estrogen by the action of the enzyme aromatase (also called estrogen synthetase). This residual postmenopausal estrogen, although substantially reduced, remains critical for bone health.
The mechanism of action of the AIs is to block the peripheral conversion (aromatization) of estrogen from androgen precursors, effectively lowering tissue and circulating estrogen levels.[1,3] The AIs currently approved for adjuvant use in receptor-positive breast cancer in the United States belong to two distinct biochemical classes (Figure 2). The steroidal AI exemestane is a derivative of androstenedione that irreversibly occupies the aromatase substrate-binding site and may be considered an aromatase inactivator. Exemestane is approved as second-line treatment, after 2 to 3 years of tamoxifen, for early hormone receptor-positive breast cancer.
Nonsteroidal inhibitors such as anastrozole and letrozole bind reversibly to the cytochrome P450 domain of aromatase. Anastrazole is approved as first-line adjuvant treatment of hormone receptor-positive early breast cancer. Letrozole is currently approved as second-line treatment, within 3 months of completion of 5 years of tamoxifen, for early hormone receptor-positive breast cancer. Because they do not inhibit estrogen synthesis in functioning ovaries, AIs are not indicated for use in premenopausal or perimenopausal patients with breast cancer. All of these third-generation AIs inhibit aromatase activity by more than 98% and are more potent than earlier drug entities.[1,3]
Unique Aspects of AI-Associated Bone Loss
Treatment-associated bone loss after cancer therapies such as oophorectomy, cytotoxic chemotherapies (eg, doxorubicin and cyclophosphamide), and endocrine therapies may be distinct from normal postmenopausal bone loss. Estrogen deprivation that occurs after ovarian ablation or during AI therapy is generally abrupt compared with that occurring postmenopausally. In addition, treatment-related bone loss may occur at an accelerated rate. The rate of bone loss is estimated at 2% annually during the first years after the onset of menopause, leveling out to ~1% during the next decade and thereafter. In contrast, compared with untreated postmenopausal women, patients with breast cancer who receive AIs experience a rate of bone loss estimated at 2.6% per year. Premenopausal women who receive AIs in combination with a gonadotropin-releasing hormone (GnRH) agonist to effect complete ovarian suppression, or with ovarian failure secondary to chemotherapy, have even higher rates of bone loss—approximately 7% annually.[14,15]
Low bone mineral density (BMD) is a critical factor associated with pathologic fracture. The risk to patients inherent in bone loss is underscored by the exponential increase in vertebral and hip fracture with decreasing BMD, expressed as the T score (Figure 3 and Table 1). Decreases in bone density of 10% to 15% have been shown to approximately double the risk of fracture, and even modest BMD increases can offer substantial preventative benefit. Cumulative increases in BMD of at least 5% have been achieved with bisphosphonates in postmenopausal osteoporosis. Increased BMD with bisphosphonates may result from the remineralization of established bone formation units, rather than from an increase in new bone.
Bone Turnover Markers and AIs
Bone turnover markers are biochemical products of osteoblasts and osteoclasts that reflect bone turnover in the whole skeleton (Table 2).[18,19] Clinically useful osteoblast-derived formation markers include the bone-specific isoform of the enzyme alkaline phosphatase (BALP), the C- and N-terminal propeptides of type 1 collagen (PICP, PINP), and osteocalcin (OC). Markers of osteoclast-mediated bone resorption include the collagen breakdown products, C- and N-telopeptide cross-links (CTX, NTX), pyridinoline (PYD), deoxypyridinoline (DPD), and tartrate-resistant acid phosphatase (TRACP). Assays for the measurement of bone formation markers (BALP and OC) and bone resorption markers (NTX, PYD, and DPD) are commercially available and approved by the US Food and Drug Administration.
Increases in bone turnover markers approaching 100% may result from decreased estrogen levels at menopause. Similarly, postmenopausal patients with breast cancer who are treated with AIs have significantly higher levels of the urinary bone resorption markers PYD and DPD. Data from the placebo arms of two large randomized trials on the treatment of bone metastases secondary to prostate, non-small-cell lung cancer and other solid tumors with zoledronic acid found baseline levels of both NTX and BALP were highly predictive of subsequent skeletal complications.
Suppression of bone resorption and formation markers has been correlated with clinical outcomes and with clinical response to antiresorptive therapy. A recent study in 328 breast cancer patients with bone metastases found zoledronic acid therapy for 3 months resulted in normalization of bone marker levels in 76% of patients with elevated levels at baseline. Patients with normalized markers had significantly improved survival compared to patients whose bone marker levels were not normalized. Although these recent data suggest that bone marker data may someday be useful in predicting clinical outcome for patients at risk for skeletal events, bone marker monitoring is not yet considered reliable for screening or diagnostic purposes.[24,25]