Breast cancer oncologists make choices between agents within the same therapeutic class every day-eg, paclitaxel vs docetaxel (Taxotere), doxorubicin vs epirubicin (Ellence), tamoxifen vs an aromatase inhibitor. In the case of chemotherapeutic agents, we do not yet have results from adequately powered direct comparisons, and so, decisions are based on indirect comparisons between trials, safety considerations, side-effect profiles, cost considerations, and clinical experience. In the case of adjuvant aromatase inhibitor therapy vs tamoxifen, the results of a huge trial are available to consider and, indeed, reconsider. In the years to come, the Arimidex, Tamoxifen Alone or in Combination (ATAC) trial experience will be augmented with results from multiple other trials that address almost all the worthwhile clinical questions (except 5 vs 10 years of an aromatase inhibitor).
Evaluating Adjuvant Endocrine Therapy
Before we get into the data, we would like to briefly discuss the standard by which adjuvant endocrine therapy trials should be judged. It is commonly accepted that for chemotherapy trials, the overall survival statistic is the key arbiter that drives practice standards. This makes sense, given that the burden of toxicity is high and the risk/benefit ratio is often narrow. Absolute survival advantages in the 3% to 5% range are commonly used as an indication for treatment or to add extra treatment. In the case of adjuvant tamoxifen therapy, however, the risk/benefit analysis is more complicated because the treatment has potential health benefits beyond a reduction in the risk of systemic breast cancer relapse and death, including a strong preventive effect on the contralateral breast and benefits in bone for postmenopausal women.
Moreover, these advantages must be set against the adverse effects of tamoxifen treatment. This has led to an analysis of the net health benefits of tamoxifen in the preventive setting that we think could be usefully applied to the ATAC data to help us make a balanced decision. We therefore support a broad view of the ATAC trial with composite end points such as "disease-free survival," which collectively examines locoregional relapse, distant relapse, contralateral breast cancer, and death as a first event. Similarly, we discourage the practice of censoring "non-breast cancer deaths," because these deaths could be treatment-related (pulmonary embolus, osteoporotic fractures, and endometrial cancer) and to ignore them moves away from the concept of net health benefit.
So what do the current data suggest? First, the small absolute advantage for anastrozole (Arimidex) in terms of disease-free survival has grown from 1.7% to 2.3% and is closer to 3% for the receptor-positive group in the latest analysis. We think this advantage is now at the point that one should discuss it with the patient to be logically consistent. After all, we often use small advantages to justify an intervention. (Adding paclitaxel to adjuvant chemotherapy or giving chemotherapy to patients with node-negative breast cancer are a couple of obvious examples.)
Second, the safety advantage of anastrozole over tamoxifen persists with a lower incidence of thromboembolism, endometrial cancer, and stroke, but the effect of estrogen deprivation on bone and perhaps other organs remains a concern. We do not think these safety considerations will change much, as the different sideeffect profiles are entirely predictable from the pharmacology of these agents. We suspect that the bone loss induced by estrogen deprivation will respond to oral or intravenous bisphosphonate therapy, and so aside from the added expense, osteoporosis is not an insolvable problem.
Perhaps what we really need to do is to encourage Dr. Visvanathan to develop a risk/benefit model " la Gail," with which we can integrate all the risks and benefits of anastrozole vs tamoxifen to make a good decision for each patient. Certainly, the cases outlined by Visvanathan and Davidson-ie, history of thromboembolism or use of tamoxifen or raloxifene (Evista) in the preventive setting-are "slam dunks" for the use of anastrozole. For most patients, however, the choice is less obvious. (Does the patient have an intact uterus? A family history of thromboembolism? A history of a transient ischemic attack?) In addition, what about patients who have difficulty affording the extra cost of anastrozole- when do we tell them the extra cost is not worth it and to save their cash?
A final consideration in this risk/ benefit analysis is the interesting possibility that HER1- and/or HER2-positive tumors are better treated with estrogen deprivation than with tamoxifen because HER1/2 signaling promotes the agonist effects of tamoxifen.[ 2] This hypothesis is now under intense investigation, and tissue blocks are being collected to examine HER1/2 status as a predictive biomarker in the ATAC trial. Although HER1 and HER2 are certainly of interest to us, the bigger issue concerns the general nature of resistance to endocrine therapy.
One does not have to be a breast oncologist long to realize that unraveling this problem is one of the key questions in breast cancer research. A new generation of trials is addressing this issue with HER1/2 kinase inhibitors, cyclooxygenase-2 (COX2) inhibitors, farnesyl transferase inhibitors, and mammalian target of rapamycin (mTOR) inhibitors. These signal transduction modulators all show promise as agents that might augment or prolong the response to endocrine intervention.
1. Gail MH, Costantino JP, Bryant J, et al: Weighing the risks and benefits of tamoxifen treatment for preventing breast cancer. J Natl Cancer Inst 91:1829-1846, 1999.
2. Ellis MJ, Coop A, Singh B, et al: Letrozole is more effective neoadjuvant endocrine therapy than tamoxifen for ErbB-1- and/or ErbB-2-positive, estrogen receptor-positive primary breast cancer: Evidence from a phase III randomized trial. J Clin Oncol 19:3808-3816, 2001.