After peaking in 1990, the absolute number of deaths per year attributed to breast cancer has fallen steadily.[1] This decline occurred despite trends that would seem to increase breast cancer mortality (population growth, aging, increased obesity) and was mirrored even in countries lacking routine support for mammography. Systemic therapy is at least partly responsible for this mortality decline, and in support of this conclusion the predicted benefits (based on trials and metaanalyses) have been seen in population- based studies.[2] In this issue of ONCOLOGY, Mina and Sledge provide a timely and inspiring review of 2 decades of progress in systemic therapy for breast cancer. This leads to several questions, including: How did we get here and what is next? Divide and Conquer
Progress in systemic therapy has been based on several linear developments. First, our understanding of theProgress in systemic therapy has been based on several linear developments. First, our understanding of thebasic biology of the disease improved, and specifically, the recognition that microscopic, undetected, metastatic disease exists before surgery, drove the development of adjuvant treatment.[ 3] Mathematical models and laboratory experiments suggested the optimal time and content of adjuvant chemotherapy regimens at about the same time that the hormone dependence of the disease was increasingly exploited. These two developments allowed us to begin to divide and conquer breast cancer. We divided it into subsets based on anatomy and biology, and we "conquered" it (or at least made some progress) by beginning to develop and refine treatments for each type or setting. Going forward, it now seems inevitable that we will continue this process of division and conquest by more precisely identifying subgroups of patients and focusing specific therapies for narrow cohorts of patients. We first divided breast cancer anatomically to direct local control interventions. Distant disease was not amenable to cure via local control procedures, but these interventions were necessary (although often insufficient) to cure early-stage breast cancer. More recently, we divided the disease into hormone-responsive and -unresponsive subsets and learned to deliver better and safer hormonal therapies in the former group while pursuing other modalities for the latter. Women with hormone-unresponsive diseaseappear to benefit earlier and to a greater extent from chemotherapy, and we now struggle to determine-perhaps by using modern genomic techniques- if there is a subgroup of hormone-responsive disease with absolute chemotherapy resistance for whom chemotherapy provides little benefit. This will likely be the next target of our divide-and-conquer approach, as it now appears that we should be able to identify subsets of hormone-responsive breast cancer patients who may or may not receive an additional (marginal) benefit from chemotherapy. Rationale for Adjuvant Therapy
The motivation for broad use of adjuvant systemic therapy is in part the recognition that the presence of clinically detected metastatic growth identifies patients with incurable disease. Further, in the metastatic setting, the vast heterogeneity of breast cancer complicates counseling of individual patients even though on average, and with very few exceptions, metastatic breast cancer is fatal within years and is, at best, a chronic illness. Adjuvant therapy is intended to prevent fatal recurrences. For metastatic disease, hormone therapy offers a limited impact on quality of life and sometimes long periods without disease progression.[ 4] Sequential chemotherapy provides manageable toxicities and equivalent overall benefits comparedto more toxic combinations, and (again) molecular biology can provide further opportunities to divide and conquer.[5] The important recognition that there is a subtype of breast cancer dependent on HER2 signaling for growth led to the development of an effective targeted approach using a humanized monoclonal antibody with very limited toxicities.[6] The success of this agent (trastuzumab [Herceptin]) led to its successful testing in the adjuvant setting. This transition from palliative to curative use has been repeated several times over the decades. Past Is Prologue
One hopes that this story would be repeated again and again but, at the same time, the challenge to clinicians will be to keep up with an expanding body of knowledge and an evolving differential treatment approach as one disease becomes many. Using relatively simple dichotomous measures (eg, immunohistochemistry testing for the estrogen and/or progesterone(Drug information on progesterone) receptor, as well as for HER2), we divide breastcancer into four related diseases with somewhat different initial approaches (hormone therapy vs chemotherapy with or without trastuzumab(Drug information on trastuzumab)). However, the explosion in molecular profiling technologies may mean that we can subdivide much further, or that we find that breast cancer is perhaps a multitude of related malignancies.[7] In either case, our therapeutic options remain somewhat limited, and so we will have to continue to push optimal evolution in the use of the existing treatment options. In addition to targeted agents, this includes such things as dose-dense scheduling and rational sequences of hormonal agents. The present review of progress over the past 20 years in breast cancer, one hopes, is therefore a promising introduction to a rapidly evolving field and a future in which breast cancer becomes an ever decreasing cause of mortality. In particular, by trusting that past is prologue, we should be able to continue, and to accelerate, the remarkable advances of the past 2 decades.

—Clifford A. Hudis, MD