This article is more than 5 years old. Information may no longer be current.
Update on Endocrine Therapy
The standard option for adjuvant endocrine therapy in premenopausal women is tamoxifen continued for 5 years. However, questions concerning the optimal duration of tamoxifen in premenopausal women exist. In postmenopausal women, the standard option is aromatase inhibitor (AI) therapy, either the up front, sequential, or extended use. Results from recent studies address some of the questions concerning the use of endocrine therapy in breast cancer.
Recent Results From Studies of Tamoxifen
Early results from the Adjuvant Tamoxifen Longer Against Shorter (ATLAS) trial, a major international study of adjuvant tamoxifen, suggest that the therapeutic benefit of tamoxifen therapy may not be limited to 5 years, as had previously been indicated by the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14 study.72 NSABP B-14 had previously demonstrated that the therapeutic benefit derived from 5 years of tamoxifen therapy was not improved by the extension of tamoxifen to 10 years of treatment.73 In the ATLAS trial, patients with a diagnosis of early stage breast cancer were randomized to further tamoxifen or no treatment after having completed 5 years of tamoxifen. The study enrolled 11,500 patients in 38 countries. A possible limitation of the study is that ER-testing was performed only in the 59% of patients that tested ER-positive, while 41% remained untested, resulting in an estimation of 90% of patients being ER-positive. Adherence to the regimen was believed to be approximately 80%. The mean follow-up was 4.2 years.
Results indicate that continuing tamoxifen therapy for an additional 5 years confers about a 12% relative reduction in risk of recurrence compared with tamoxifen discontinued after 5 years (HR: 0.88, P=.05) for years 5 to 9. Hazard ratios for recurrence for continued tamoxifen versus discontinued treatment were 1.0 for the first year after randomization, 0.90 for years 2 to 4, and 0.72 for year 5 and beyond. There were no significant differences in breast cancer-specific survival and overall survival between the two groups. Overall, there have been 739 recurrences for the group of patients randomized to continue treatment versus 835 recurrences for the group that stopped treatment. Of these recurrences, 221 were observed between years 10 and 14 of the study. Among patients who continued to receive tamoxifen for a second 5-year period, there has been a carry-over effect with an additional reduction in risk of 22%. Of those, 221 occurred among women in years 10 through 14 of the study, 96 in the tamoxifen continuation group, and 115 in the group that received tamoxifen for 5 years.73 These findings differ from those of the smaller NSABP B-14 trial, which had found that use of tamoxifen beyond 5 years might be worse than stopping after 5 years. Comparison of results for risk of recurrence from ATLAS with EBCTG data demonstrate the extent to which ATLAS may change perceptions of the usefulness of adjuvant tamoxifen beyond 5 years (Table 4). The size of the ATLAS study, which is the largest ever in adjuvant therapy in breast cancer, lends weight to these findings. Additionally, the unclear ER status of the population in ATLAS may mean that efficacy is being underestimated. However, no data on toxicities were presented, impeding a thorough risk to benefit assessment. These findings are significant because adjuvant aromatase inhibitor therapy has become accepted as the best means of prevention of recurrence. However, tamoxifen therapy is significantly less expensive and, based on results from ATLAS, tamoxifen may have a longer therapeutic benefit than previously thought. In addition, these data may be useful in determining the length of therapy with tamoxifen in patients who cannot tolerate aromatase inhibitor therapy.9 The Adjuvant Tamoxifen Treatment–Offer More? (aTTom) trial, another large-scale study of adjuvant tamoxifen, has recently finished patient accrual and may shed further light on the question of optimal duration of adjuvant tamoxifen therapy.
Results from the ATLAS study must also be seen in the context of data from the MA-17 trial, in which the aromatase inhibitor letrozole was given to patients who had completed 5 years of tamoxifen. Letrozole was well-tolerated and reduced recurrence risk by 42% and risk of distant metastases by 40% versus placebo. Because of the significant advantage in disease-free survival provided by letrozole, the study was unblinded early. Patients who elected to receive letrozole after unblinding experienced significant improvements in disease-free survival. The magnitude of benefit from letrozole in prevention of relapse after 5 years of tamoxifen makes a strong case for use of aromatase inhibitor therapy in this setting.74
The association of tamoxifen with increased incidence of uterine abnormalities (e.g., uterine cancer, sarcoma, endometrial hyperplasia, fibroids) is widely recognized. These effects are thought to be linked to the partial agonistic activity that tamoxifen exerts in the uterus. The Intergroup Exemestane Study (IES) is evaluating the effect of tamoxifen and the steroidal aromatase inhibitor exemestane on endometrial status in postmenopausal patients with early breast cancer who have been free of disease after 2 to 3 years of tamoxifen treatment. Patients (N=4724) were randomized to continued tamoxifen or exemestane treatment to complete 5 years of adjuvant therapy. During treatment, endometrial thickness remained stable in the tamoxifen group. Switching to exemestane resulted in rapid and significant improvements in endometrial thickness and uterine volume. At 6 months, the change from baseline was -1.9 mm (P=.003). At 2 years, there was a significant difference in the percentage of patients in the tamoxifen and exemestane groups with abnormal endometrial thickness (35.5% versus 61.8%, P=.004). Following treatment, endometrial thickness and uterine volume decreased in the tamoxifen group, suggesting that improvement in endometrial status resulted from withdrawal of tamoxifen rather than a protective effect of exemestane (Table 5).
Recent reports have indicated that polymorphism in genes associated with metabolism, including the CYP2D6 gene, may impact tamoxifen efficacy and tolerability. The CYD2D6 enzyme metabolizes tamoxifen to endoxifen, which is most likely responsible for much of the effect of tamoxifen. Approximately 8% of Caucasians are poor metabolizers of tamoxifen and will have poor outcomes if treated with the drug. The same mechanism may affect responses and side effects with aromatase inhibitors. Cuzick and colleagues evaluated the occurrence of hot flushes in a retrospective, exploratory analysis of results from the Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial.75 The analysis identified all patients who reported the side effect associated with either tamoxifen or anastrozole during the first 3-month follow-up visit. Hot flushes were reported in 35% of the women at 3 months (36% anastrozole versus 41% tamoxifen). Patients who experienced hot flushes had a 3.6% lower risk of recurrence than those who did not (A: HR 0.66 (P<.001), T: 0.77 (P=.006). This significant difference was seen in both treatment arms. These data suggest that, for both treatments, effectiveness is associated with occurrence of hot flushes. This symptom may be an indirect measure of the degree of suppression of estrogen or ER blockade achieved by these treatments.
Recent Studies of Anastrozole
In agreement with the MA-17 trial findings, results from the ATAC study also demonstrated a therapeutic advantage with aromatase inhibitor therapy over tamoxifen, while dispelling some of the concerns with adverse effects associated with this class. Forbes and colleagues presented results from the ATAC study with median follow-up of 100 months.76 The follow-up demonstrated the continued superior efficacy of the aromatase inhibitor anastrozole and no excess risk for fracture compared with tamoxifen upon completion of therapy. Earlier follow-up had shown that anastrozole was significantly more effective than tamoxifen in prevention of recurrences with better tolerability but with a higher risk for fractures.77 This large, randomized, double-blind phase 3 study of 5-year adjuvant hormonal treatment for early breast cancer randomized 3125 postmenopausal women to anastrozole and 3116 to tamoxifen. Anastrozole resulted in a significantly higher rate of DSF (primary endpoint) versus tamoxifen (HR: 0.85; 95% CI 0.76, 0.94; P=.003). There was also a significant advantage in favor of anastrozole with respect to time to recurrence (TTR), time to distant recurrence, and new contralateral breast cancer (Table 6). Over time, there was an increase in the absolute difference in TTR between the two arms: 2.7% at 5 years and 4% at 9 years. Rates of overall survival were comparable. Following completion of therapy, anastrozole and tamoxifen were associated with comparable rates of fractures.76
|
Another analysis of data from ATAC focused on risk factors for joint symptoms. Such symptoms are a recognized side effect of aromatase inhibitor therapy and are associated with postmenopausal status and low estrogen levels. However, tamoxifen therapy appears to have minimal impact on the development of joint symptoms. The analysis examined risk factors that had an influence on the development of joint symptoms and assessed whether there was an association between these symptoms and endocrine treatment.78 With a median follow-up of 68 months, 2,011 joint symptoms were observed. Rates of joint symptoms were higher in the anastrozole group (35.6%) than in the tamoxifen group (29.4%) (OR=1.32 [1.19-1.47]). In patients who received hormone replacement therapy (HRT) before the study, the rate of joint symptoms during the study was higher than for those who received no HRT. Joint symptoms were significantly more common in patients who received chemotherapy as part of treatment versus those without chemotherapy (OR=1.31 [1.15-1.48]). Patients with tumors that were hormone receptor negative had significantly fewer joint symptoms than those who were hormone receptor positive (OR=0.67 [0.55-0.83]). Other significant positive factors for development of joint symptoms included geographical origin in North America (United States and Canada), increased body weight, and cigarette smoking.78
Updated results from ATAC also included some previously unreported toxicities, including retinal hemorrhages associated with anastrozole therapy.79 The authors speculate that this finding may be explained by the recognized association of aromatase inhibitors with compromised systemic vascular integrity. Additionally, the retina itself may be directly impacted by depletion of estrogen due to the location of ER in the retina. Risk for retinal hemorrhages was evaluated by obtaining retinal fundus photographs for 35 patients who received anastrozole, 38 amenorrheic patients who received tamoxifen, and 38 amenorrheic control subjects who were not receiving HRT. The rate of retinal hemorrhages was significantly greater among anastrozole patients compared with tamoxifen and control patients. The prevalence of retinal hemorrhages for anastrozole was estimated at >4% (95% CI). These findings suggest that intraocular imaging techniques may be useful in determining whether aromatase inhibitor therapy may be leading to increased vitreoretinal traction in breast cancer patients receiving this type of treatment.79
Addressing Bone Loss with Endocrine Therapy
Gnant and colleagues reported results from an ABCSG-12 substudy that evaluated the use of the bisphosphonate zoledronic acid for prevention of bone loss induced by adjuvant endocrine therapy.80 The results reflected follow-up at 2 years after completion of adjuvant therapy in the main ABCSG-12 study. The ABCSG-12 study randomized 1801 patients to one of four treatment arms: tamoxifen and goserelin with or without zoledronic acid and anastrozole and goserelin with or without zoledronic acid. Patients received adjuvant therapy for 3 years. The bone substudy included 404 patients, 201 of whom received adjuvant zoledronic acid together with endocrine therapy and 203 of whom did not. After 3 years of treatment, patients who did not receive zoledronic acid experienced a significant loss of bone mineral density (11.3%) compared with baseline levels (P<.0001). This decrease in bone mineral density was greater in patients who received anastrozole versus tamoxifen (-13.6% versus -9%). Two years after completion of adjuvant therapy, patients who had not received zoledronic acid continued to have significantly decreased bone mineral density compared with baseline levels (-6.8%, P=.0005). For the group of patients who had received zoledronic acid, bone mineral density was unchanged at 36 months and significantly increased (+3.9%, P=.02) at 60 months compared with baseline levels.80
Aromatase inhibitor therapy results in almost complete ablation of estrogen production. This can lead to accelerated bone loss and increased risk for fracture. The Z-FAST study examined the effect of zoledronic acid on bone loss associated with aromatase inhibitor therapy. Brufsky and colleagues reported 36-month follow-up data from the study, which was conducted in 602 postmenopausal patients with ER+ and/or PR+ breast cancer who received letrozole.81 Patients were randomized to upfront zoledronic acid (n=189) or zoledronic acid delayed until T score (measure of osteopenia) reached less than -2 or fracture occurred (n=188). The primary endpoint for the study was percent change in lumbar spine bone mineral density at 12 months. Upfront zoledronic acid was associated with a mean increase in lumbar spine bone mineral density of 3.72% compared to a decrease of 2.95% in the delayed group, for a significant absolute difference of 6.7% (P<.001). There was an increase in mean total hip bone mineral density for the upfront group of 1.66% versus a mean decrease of 3.51% for the delayed group, for a significant absolute difference of 5.2% (P<.001). Two percent of patients in the upfront group experienced a decrease in T score compared with 18% in the delayed group. The overall difference in the percentage change in bone mineral density between the upfront and delayed zoledronic acid treatment groups progressively increased from baseline through 36 months. These results suggest that zoledronic acid is effective in the prevention of bone loss associated with aromatase inhibitor therapy.81
Modeling Relapse Risk With Endocrine Therapy
|
A model of relapse risk in patients who received endocrine therapy was developed based on results from a long-term study of neoadjuvant letrozole or tamoxifen in postmenopausal women with estrogen receptor positive (ER+) stage II and III breast cancer too large for breast conservation. Ellis and colleagues reported results from this development effort.82 A number of potential factors, including posttreatment ER status, Ki67 proliferation index, histological grade, pathological tumor size, node stage, and treatment response, were evaluated to identify independent predictors for relapse-free survival and overall survival. Since relapse-free survival and overall survival were similar between the neoadjuvant letrozole and tamoxifen treatment arms, the groups were pooled for analysis. The median follow-up for the study was 62.5 months. Multivariate analysis of posttreatment tumor characteristics found six independent predictors of relapse: pathological tumor stage, node status, Ki67 level, tumor grade, ER status, and clinical response. The multivariate predictive model based on these independent predictors was successful in distinguishing between four separate groups with relapse-free survival of 100%, 82%, 52%, and 0%. It provided additional information to discriminate between patients with the extremes of breast cancer outcomes in this setting and demonstrated that 4 months of neoadjuvant endocrine therapy, followed by a response assessment, is a promising approach for predicting the course of ER-positive breast cancer in postmenopausal women.82