The New Millennium for Adjuvant Therapy in Breast Cancer

This issue of ONCOLOGY is devoted to the memory of Dr. Martin Abeloff of Johns Hopkins University. Marty Abeloff was many things: a superb cancer researcher with a broad range of scientific interests, the director of one of the nation’s great cancer centers who oversaw its enormous growth, and a charming and decent human being who was loved and respected by all who knew him. He was also a spectacular mentor of junior faculty, both inside and outside of his institution. I came to know Marty when I joined the Breast Cancer Committee of the Eastern Cooperative Oncology Group, which he ably led in the late 1980s and early 1990s. His care for junior faculty deserves special comment. He considered them colleagues, to be respected and nurtured. He was a great listener who always seemed to have time for the often-naive thoughts of unseasoned juniors. I learned a great deal from him, and it is a matter of real sadness that I will not be able to share any more time with him.

Adjuvant therapy for breast cancer has changed immensely since the early 1990s, and these changes are the subject of this review. The research interests of those times are strikingly different from those of today. The dominant questions then being asked focused on questions of chemotherapy technique—in essence, how much drug could be administered, and did alterations in dose, dose intensity, or dose density affect outcome.

The questions of the new millennium reflect a strikingly different focus: How can our burgeoning understanding of the biology of breast cancer influence therapy? A related question suggests how far we have come from the 1990s: “Who doesn’t require treatment?” This review will examine the new biology of breast cancer and its therapeutic implications for adjuvant therapy.

The New Biology of Breast Cancer

Perhaps the most profound changes in our approach to adjuvant therapy in the past decade have been driven by changes in the understanding of breast cancer biology. In the not-too-distant past, breast cancer was viewed from a therapeutic standpoint as being essentially a single disease. While differentiations were made between estrogen receptor (ER)-positive and -negative breast cancer for the purpose of administering hormonal therapy, in essence all patients were considered potential candidates for cytotoxic chemotherapy. This therapeutic consensus, a legacy of population-based adjuvant trials of the 1980s, was formalized at the National Institute of Health’s 2000 Consensus Development Conference on adjuvant therapy for breast cancer. The conference panel recommended consideration of adjuvant chemotherapy for essentially all patients with tumors greater than or equal to 1 cm in size.[1]

At the same time that the Consensus Conference was advising physicians and patients, the Human Genome Project was using novel technology to revolutionize our understanding of the human genome. Spinoff technologies such as cDNA microarrays (“gene chips”) allowed investigators to examine large portions of the human genome in human breast cancers. Analysis of the results both reaffirmed and altered our understanding of breast cancer—reaffirmed, because genomic analyses suggested that some of our therapeutic interventions (such as hormonal and HER2-targeting therapy) had a solid basis in the cancer cell’s genetic makeup; altered, because it became clear that breast cancer should be considered a collection of diseases rather than a solitary disease. Breast cancer is now best thought of as a group of criminals sharing the same boarding house, requiring different forms of apprehension and punishment.

In broad terms, genomic analyses suggest the existence of luminal tumors (with at least two subtypes, A and B), which encompass the estrogen-sensitive breast cancers; basal tumors, which are in large part estrogen-, progesterone(Drug information on progesterone)-, and HER2-negative; and c-erbB2–positive, ER-negative tumors.[2-4] Breast cancer therapies largely parallel these subtypes: Hormonal manipulations are most likely to benefit luminal A tumors, many luminal B cancers benefit from chemotherapy, triple-negative (basal) tumors are currently best treated with chemotherapy, and c-erbB2–positive (HER2-positive) tumors benefit from HER2-targeting therapy. Systemic adjuvant therapies increasingly appear to have focused benefits related to genomic categories.

Treatment of Luminal Breast Cancers

As mentioned above, a significant proportion of human breast cancers fall into the broad general category of luminal cancers. These cancers are typically ER–positive. By the mid-1990s it had become apparent to clinical investigators that the use of adjuvant hormonal therapies (tamoxifen in pre- and postmenopausal women, and ovarian ablation in premenopausal women) provided significant, if incomplete, benefits with regard to disease-free and overall survival in both lymph node–negative and –positive disease. Similarly, trials randomizing patients with ER-positive tumors to hormonal therapy alone or hormonal therapy plus chemotherapy suggested that the addition of adjuvant chemotherapy improved disease-free and overall survival, though generally with modest overall benefits and not insignificant risks.[5]

New genomic analyses provide an intellectual basis for these empirical findings. Women with luminal A breast cancers (as variously defined by investigators using a variety of clinical assays)[6] have both a good overall prognosis and a high likelihood of benefit with hormonal manipulations (and, conversely, a low likelihood of benefit with cytotoxic agents). This is particularly true for patients with lymph node–negative tumors, where commercially available clinical assays (eg, OncotypeDx and Mammaprint) now play an established role in determining chemotherapy benefit.[7-10]

Recent genomic data from the Southwest Oncology Group (SWOG) 8814 trial suggest that lymph node–positive luminal A tumors may derive little or no benefit from the use of cytotoxic therapy when compared to the use of hormonal therapy alone.[11] In contrast, women with non–luminal A ER-positive tumors appear to derive significant benefit from the use of adjuvant chemotherapy, and may receive less benefit from the use of adjuvant hormonal therapy.[8] These tumors (currently commonly called luminal B, though other subtypes may exist) are characterized by a proliferative gene signature, and in many cases are both ER-positive and HER2-positive.

While the use of hormonal therapies for luminal cancers is well established, which hormonal therapies should be used (and how they should be used) remains a matter of controversy, or at least uncertainty. Until the current decade, the selective estrogen receptor modulator tamoxifen(Drug information on tamoxifen) represented the standard of care for ER-positive postmenopausal tumors. In recent years numerous prospective randomized controlled trials comparing tamoxifen to any of three commercially available aromatase inhibitors (anastrozole [Arimidex], letrozole(Drug information on letrozole) [Femara], and exemestane(Drug information on exemestane) [Aromasin]) have demonstrated statistically significant improvements in disease-free survival for aromatase inhibitor therapy. These trial results led to new guidelines by the American Society of Clinical Oncology, suggesting that aromatase inhibitor therapy should represent part of the standard of care in all postmenopausal women with ER-positive breast cancer.[12]

Overall survival benefits have proved lacking in many aromatase inhibitor trials, including the one with the longest overall follow-up.[13] These benefits are accompanied by an altered toxicity profile, with aromatase inhibition being associated with decreased rates of gynecologic and thromboembolic toxicities, and increased rates of musculoskeletal complaints and decreased bone mineral density. These altered toxicity profiles are often important for individual patients in the clinic, and to some extent may affect treatment decisions.