[Lani]

EDITORIAL

Can Biology Trump Anatomy? Do All Node-Positive Patients With Breast Cancer Need Chemotherapy?

N. Lynn Henry, Daniel F. Hayes
Department of Internal Medicine; Breast Oncology Program, University of Michigan Comprehensive Cancer Center, Ann Arbor, MI

Mortality resulting from breast cancer is decreasing in the Western world,1 in part because of adjuvant chemotherapy,2 and most treatment guidelines for women with lymph node-positive disease dictate the use of adjuvant chemotherapy.3 Therefore in developed countries almost all such patients, regardless of the biology of their cancer, receive toxic and expensive chemotherapy. In this era of molecular medicine, one wonders whether we couldn't use biologic factors to be more selective for our therapies.

Perhaps the most clinically important biologic factor in all of oncology is the estrogen receptor (ER). ER expression is a powerful positive predictive factor for antiestrogen therapy.4 However, ER expression may also be a negative predictive factor for chemotherapy. This effect was first proposed more than 30 years ago,5 and a recent review of Cancer and Leukemia Group B trials suggested that the benefits of chemotherapy are substantially less in node-positive women with ER-positive cancers than in those with ER-negative cancers.6 These results raise a derivative to the question posed in the title of this editorial: Can ER expression be used to individualize chemotherapy as well as hormone therapy?

Historically, subset analyses have suggested that the benefits of adjuvant chemotherapy are greater in women under 50, and might be negligible in older patients.7 This age-specific observation is curious. Some have speculated that it can be completely explained by ER expression.8 Incidence of ER-positive breast cancer differs by age. Approximately 60% of women under the age of 50 have ER-positive disease, compared with 80% of those over the age of 50.9 If chemotherapy is more effective against ER-negative than ER positive tumors, then a greater proportion of younger women overall would benefit from its direct cytotoxic effects.6 In addition, those premenopausal women with ER-positive disease appear to gain secondary benefit from chemotherapy-induced ovarian failure, which does not occur in postmenopausal women.

These considerations suggest that ER might be useful not only to select hormone therapy but also to avoid chemotherapy, and multiple clinical trials have evaluated hormone therapy versus chemotherapy in ER-positive premenopausal women.10-14 In this issue of the Journal of Clinical Oncology, Schmid et al15 describe one such study, designated the Takeda Adjuvant Breast Cancer Study with Leuprorelin Acetate (TABLE) trial. Their results suggest that 2 years of adjuvant depot luteinizing hormone–releasing hormone agonist therapy is at least as or even more effective than 6 months of adjuvant chemotherapy (cyclophosphamide, methotrexate, and fluorouracil [CMF]). Disease-free survival was statistically similar in the two arms, whereas overall survival was, if anything, superior for the group that received endocrine treatment.

Strengths of this study include a homogenous patient population and exclusion of patients with known ER-negative breast tumors, and it is consistent with other, similarly designed trials of endocrine therapy versus chemotherapy in ER-positive women.10-13 However, the sample size is small and the follow-up is limited. Furthermore, the trialists used intravenous CMF chemotherapy, which may be inferior to classical oral CMF.16,17 Additionally, this trial did not account for effects from another important biologic factor, HER2. For example, in several studies, CMF chemotherapy has been shown, in HER2-positive patients, to be inferior to anthracycline-containing regimens.18,19 Thus, one might expect the HER2-positive subset, which may represent up to 20% of the ER-positive population, to do substantially better with more modern chemotherapy than CMF, regardless of how it is delivered. More recently, taxanes have been shown to add to anthracycline-based therapy, and these regimens may also be even more efficacious in patients with HER2-overexpressing tumors.20,21 These observations suggest that the results of the TABLE trial might not have been equivalent if the investigators had used more effective chemotherapy.

Taken together, these results are consistent with the hypothesis that ER expression might be used to avoid chemotherapy, but are they sufficient to change our guidelines and clinical practice for premenopausal women with node-positive breast cancer? We believe that the trial designs of the TABLE and other studies asked the wrong question. We propose that the issue is not chemotherapy alone versus hormone therapy alone. Rather, it seems that in ER-positive breast cancer, the more appropriate question is chemotherapy plus hormone therapy versus hormone therapy alone. We have no doubt that hormone therapy is effective in ER-positive patients and that all ER-positive patients should receive some form of endocrine therapy. The issue is whether all ER-positive patients with a poor prognosis based on an anatomic factor, such as positive lymph nodes, should be treated with chemotherapy in addition to endocrine therapy.

Few data are available to address this question. In the Southwest Oncology Group (SWOG) protocol 8814 (Intergroup 0100 trial), node-positive postmenopausal patients were randomly assigned to tamoxifen alone, tamoxifen plus concomitant CAF (cyclophosphamide, doxorubicin, fluorouracil), or CAF followed by tamoxifen.22 Patients treated with chemotherapy plus tamoxifen had a small but statistically significantly superior disease-free and overall survival compared with those treated with tamoxifen alone. In the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-20 trial, 2,306 node-negative patients were randomly assigned to tamoxifen alone or to CMF or MF plus concomitant and subsequent tamoxifen.23 Those patients randomly assigned to chemotherapy, especially those younger than 50 years of age, had superior disease-free and overall survival compared with those treated with tamoxifen alone. Conversely, results from a highly underpowered trial conducted by the International Breast Cancer Study Group (IBCSG; trial 11-93) failed to demonstrate a survival benefit for 174 premenopausal node-positive women randomly assigned to chemotherapy (doxorubicin and cyclophosphamide) plus goserelin and tamoxifen versus goserelin and tamoxifen alone.24

Overall, these results suggest that chemotherapy adds some additional benefit beyond endocrine therapy alone for the ER-positive, node-positive population, regardless of age. However, for the entire population, these benefits are small, and we agree with the TABLE investigators that most ER-positive patients might be treated with hormone therapy alone to avoid the toxicity of chemotherapy. Nonetheless, there appears to be a small subset of patients that do benefit from chemotherapy. Unfortunately, we are unable to identify these patients at present. Recent studies have suggested that additional biologic factors such as HER2 and/or measures of proliferation might identify patients with ER-positive cancers that actually behave more like ER-negative cancer in regard to hormone therapy resistance and/or chemotherapy sensitivity. Several recent studies have suggested that multiplex gene expression assays can distinguish patients with low- and high-risk cancers.25 For example, in node-negative, ER-positive patients, the OncotypeDX assay (Genomic Health Inc, Redwood City, CA) has been shown in retrospective studies to identify such patients who not only have a very low risk of recurrence on tamoxifen, but whose cancers also appear to be relatively resistant to chemotherapy.26,27 In at least one study, patients with a high recurrence score not only appear to have a high risk of recurrence, but also seem to have cancers that are very sensitive to the beneficial effects of chemotherapy.27 The North American Breast Cancer Intergroup is currently conducting a prospective randomized controlled trial, designated the TailorRx trial, to validate these exciting observations in node-negative women.

TailorRx is a bold and important clinical trial designed to use this new technology to individualize treatment in anatomically low-risk patients. We propose that it is now time to use our understanding of biology to similarly readdress the need for chemotherapy in women who are felt to have "high-risk" cancer on the basis of anatomic features (TNM staging). The era of individualized therapy for breast cancer, which really started with the understanding of ER biology in the 1970s and took a step forward with the entry of HER2 in the 1990s, is now ready to take another great leap with a new set of molecularly driven prospective randomized controlled trials.