Anthracyclines and Early Breast Cancer: The End of an Era?

For the last 25 years, anthracyclines have been unchallenged as the backbone of all the most popular chemotherapy regimens for women with operable breast cancer. Over such a long period, there have been many improvements in adjuvant chemotherapy, as a result of incorporation of anthracyclines into sequential non–crossresistant regimens with cyclophosphamide, methotrexate, fluorouracil (CMF) or with the taxanes, new combinations with docetaxel, or novel scheduling approaches such as dose-dense administration. In this issue of Journal of Clinical Oncology, Jones et al update the results of the US Oncology 9735 (USO-9735) study that compared four cycles of docetaxel and cyclophosphamide (TC) with four cycles of doxorubicin and cyclophosphamide (AC). At 7 years, the report confirms superior disease-free survival with TC (81% TC v 75% AC, P .033; hazard ratio [HR] 0.74; 95% CI, 0.56 to 0.98). It also shows for the first time a significant survival benefit (87% TC v 82% AC, P .032; HR 0.69; 95% CI, 0.50 to 0.97). In many ways, the findings of USO-9735 question some tenets related to the popularity of the AC regimen in the United States and other regions of the world. In particular, the results are an integral part of the recent debate on the possibility that oncologists might undergo a radical departure from established practice by ending the era of anthracyclines in early breast cancer. Such a debate is one more reason to place the findings from USO-9735 into context. The simple design of USO-9735 allows for a straightforward interpretation of the main results and undoubtedly establishes TC as more effective than AC. In addition, the study shows that tolerability was qualitatively different (more nausea and vomiting and one cardiac death from congestive heart failure with AC, and more febrile neutropenia leading to one toxic death with TC), but TC was as feasible as AC, overall and in all subgroups of patients. These features prompted the investigators to propose that TC might be a preferred regimen for many patients, including older women and node-negative women. Overall, the data of USO-9735 support the role of TC as a new treatment option for early breast cancer, fully justifying its incorporation as the only nonanthracycline regimen in the long list of regimens supported by level I evidence in the guidelines of the National Comprehensive Cancer Network [Clinical Practice Guidelines in Oncology, v1.2008]. When proposing the use of the TC regimen, the USO-9735 investigators make much of unplanned and exploratory subset analyses. Exploratory analyses are now common and perceived as relevant in giving new insights and in generating new hypotheses. This method of eliciting new ideas might justify a departure from statistical purity, but it warrants great caution. In USO-9735, none of the unplanned subset analyses according to age, estrogen/progesterone receptor status, and human epidermal growth factor receptor 2 (HER-2) status shows anything but trends favoring TC, as illustrated by the fact that confidence limits cross the unity line of the forest plot. Therefore, the conclusion that TC performs equally well or better than AC in these subsets is not proven. This consideration is pertinent to implications for practice based on the results of USO-9735. Similarly pertinent is whether the superiority of TC over AC is enough to conclude that anthracyclines are not required for superior antitumor efficacy in early breast cancer. AC was originally tested in the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-15 study, in which it performed as well as classical oral CMF and was perceived as a better option mostly because of its shorter treatment duration (2 v 6 months), better acute tolerability, and lower logistical burden on patients and hospitals. Therefore, AC was selected on the basis of convenience, not superior efficacy. Indeed, however popular it is, AC is far from being one of the most effective anthracycline-based regimens. Several studies have demonstrated the superiority of fluorouracil, epirubicin, and cyclophosphamide (FEC) or cyclophosphamide, doxorubicin, and fluorouracil (CAF) regimens over CMF, leading to the reasonable inference that such anthracycline regimens are better than AC. A large body of evidence also supports the concept that anthracyclines perform best in adjuvant sequential regimens, and that using the taxanes in sequence with anthracyclines, rather than in their substitution, can be very effective. Stemming from the seminal work of Bonadonna et al, a recent trial showed the superiority of sequential epirubicin and CMF over CMF. The BIG98-02 trial and the European Cooperative Trial in Operative Breast Cancer (ECTO) study showed that incorporation of taxanes improved the benefit of sequential doxorubicin followed by CMF. Sequential addition of taxanes also resulted in superiority over AC in Cancer and Leukemia Group B (CALGB) 9344, over FEC in Grupo Espanol de Investigacion del Cancer de Mama (GEICAM) 9906, and over FEC in PACS-01. In three of the referenced trials, the addition of a taxane into the sequential treatment occurred in regimens of identical duration. In brief, there is a formidable array of available anthracyclineand anthacycline/taxane-containing regimens that are superior to AC. The better efficacy of TC over AC does not justify discarding any of them. JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 27 NUMBER 8 MARCH 1