Side Effects of Adjuvant Endocrine Therapy
Side effects of tamoxifen are well documented. In the NSABP P-1 trial, which compared tamoxifen to placebo in over 13,000 women with a high risk for developing breast cancer, women who received tamoxifen had increased hot flashes, vaginal discharge, and difficulties in some areas of sexual functioning.[46,47] There were no differences between the tamoxifen and placebo arms with regard to depression, overall physical or mental quality of life, and weight gain. The more serious side effects included pulmonary embolism, deep venous thrombosis, and endometrial cancer (Table 9). Tamoxifen reduced the risk of bone fractures (see Table 9). In the most recent EBCTCG overview analysis, there was a nonsignificant increase in stroke deaths (3 extra per 1000 during the first 15 years) balanced by a nonsignificant reduction in cardiac deaths (3 fewer per 1000 during the first 15 years), with a resulting minimal net effect of tamoxifen on overall cardiovascular mortality.
Compared to tamoxifen, AIs have a distinct toxicity profile (Table 10). This was demonstrated in a meta-analysis of seven randomized phase III trials that included 30,023 patients. There was a decreased incidence of venous thromboembolism and endometrial cancer for patients taking AIs compared with tamoxifen; however, the risk of cardiovascular disease was increased in patients taking AIs. This comparison may be confounded by the fact that tamoxifen may reduce cardiovascular events. One factor that may affect the risk of cardiovascular events is treatment-related hypercholesterolemia. Hypercholesterolemia was only assessed in four trials, it was not graded consistently, and fasting levels were not obtained in all trials.[24,28,33,49] Despite these caveats, there was a higher incidence of hypercholesterolemia in patients taking AIs. This was most apparent in the trials with AIs as initial therapy, and there was a suggestion that a shorter duration of AI therapy might reduce the odds of hypercholesterolemia.
MA.17 was the only large trial in which an AI was compared to placebo. When comparing the patients who remained on placebo after unblinding to those who chose to switch to letrozole(Drug information on letrozole), there was little difference in the rate of cardiovascular events (3.1% vs 4.2%, respectively; P = .17). The rate of bone fractures was increased in the patients who received letrozole (3.1% vs 5.2%; P = .02), as was the rate of newly diagnosed osteoporosis (1.6% vs 5.3%; P < .0001).
The most common side effects seen with AIs are myalgias and arthralgias. In the ATAC trial, 27.8% of women receiving anastrozole(Drug information on anastrozole) reported musculoskeletal disorders compared with 21.3% of those receiving tamoxifen. In the BIG 1-98 trial, 20.3% of patients receiving letrozole vs 12.3% of those receiving tamoxifen reported arthralgias of any grade. The incidence of musculoskeletal problems may actually be higher than the rates reported in the large trials. In a survey of 200 postmenopausal women receiving AIs for early-stage breast cancer, 47% reported joint pain and 44% reported joint stiffness. A placebo-controlled trial would be required to substantiate these findings.
AIs accelerate bone loss in postmenopausal women. There were more fractures while on treatment in the ATAC trials in patients receiving anastrozole (odds ratio [OR], 1.33; 95% CI, 1.15–1.55; P < .0001). After the completion of treatment, the incidence of fractures was similar (HR, 0.98; 95% CI, 0.74–1.30; P = .9). In the BIG 1-98 trial, a comparison of the monotherapy arms also showed an increase in bone fractures in women receiving letrozole, with a 60.3-month median follow-up (RR, 1.38; 95% CI, 1.13–1.69).
Nonadherence to Therapy
While studies have found that adjuvant hormonal therapy for hormone-sensitive breast cancer dramatically reduces recurrence and mortality, adherence to medications is suboptimal. In a study of 8769 patients with stage I-III breast cancer diagnosed from 1996 to 2007, only 49% of the patients took adjuvant hormonal therapy for the full duration on an optimal schedule. A recent study analyzing medical and pharmaceutical claims data from three national longitudinal databases found that adherence to adjuvant anastrozole therapy decreased from between 69% and 78% at year 1 to between 50% and 68% at year 3. The success of AI treatment depends on adherence to the regimen. Thus, it is important for the oncologist to ask a patient about adherence. One nonjudgmental way of doing this is to ask how many pills the patient has missed in the last month, rather than asking if she has been taking her medication. Switching to another AI may be an alternative for women who are nonadherent because they are unable to tolerate the side effects of a particular AI.
Ongoing Clinical Trials
Ongoing clinical trials are listed in Table 11. These trials are designed to answer a number of questions, including how one AI performs compared to another (MA.27 and FACE [Femara Anastrozole Clinical Evaluation]), the role of extended adjuvant therapy (ANZ [Australian New Zealand] 0501, NSABP B-42), and the optimal duration of extended adjuvant therapy (MA.17R, SALSA [Secondary Adjuvant Long-term Study with Anastrozole], SOLE [Study of Letrozole Extension]). Data from the trials in premenopausal women of ovarian suppression with an AI compared to tamoxifen or ovarian suppression with tamoxifen (SOFT, TEXT [Tamoxifen and Exemestane(Drug information on exemestane) Trial]) are eagerly awaited.
For many decades, tamoxifen has been the standard adjuvant endocrine treatment for HR-positive, early-stage breast cancer. For premenopausal women, it remains so. In premenopausal women, the role of ovarian suppression or ovarian ablation is not clear, and the results of studies examining these strategies in combination with tamoxifen or AIs—the SOFT and TEXT trials—are eagerly awaited.
In postmenopausal women, studies over the last decade have shown that AIs are more beneficial than tamoxifen in preventing disease recurrence. Trials have looked at tamoxifen directly compared to AIs upfront, at sequential therapy with tamoxifen and AIs, and at the extended use of AIs. It seems that the evidence from the sequenced therapy trials is strongest. However, in many of these trials, patients were randomized after having been disease-free for the first 2 to 3 years of tamoxifen therapy. The monotherapy trials have shown that the increased effectiveness of anastrozole and letrozole, compared to tamoxifen, seems to be greatest in the first 2 years. This benefit is not taken into account if the randomization occurs after that time. Among the questions still being studied are the optimal duration of extended adjuvant therapy with AIs, how one AI performs compared to another, and whether there is a benefit to intermittent extended adjuvant treatment.
Premenopausal women: Use tamoxifen for 5 years. If a patient becomes postmenopausal during treatment, consider extended adjuvant therapy with AI for 5 years.
• If considering using monotherapy, AIs are superior to tamoxifen.
• If considering a switch strategy (from AI to tamoxifen or from tamoxifen to AI, with a total duration of therapy of 5 years), randomized trials in which the randomization occurred at diagnosis showed no difference between switching and AI monotherapy.
• Extended AI therapy after 5 years of tamoxifen is superior to 5 years of tamoxifen alone.
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.