Anthracycline vs Nonanthracycline Adjuvant Therapy for Breast Cancer
Anthracycline vs Nonanthracycline Adjuvant Therapy for Breast Cancer
The use of adjuvant systemic therapy represents a major advance in the treatment and cure of early-stage breast cancer. Depending on tumor characteristics, chemotherapy and/or hormonal therapy following local treatment for breast cancer is the standard of care worldwide. Data from numerous randomized adjuvant trials have shown that the risk of disease recurrence and death is significantly reduced when chemotherapy and/or hormonal therapy is added to treatment, which translates into many lives saved each year. Although substantial progress has been made, controversy over the best adjuvant regimen remains. One of the central questions has been whether anthracyclines should be part of adjuvant therapy, or whether it is appropriate to recommend the CMF regimen (cyclophosphamide [Cytoxan, Neosar], methotrexate, fluorouracil [5-FU]) or another chemotherapy- based regimen instead. As the optimal treatment continues to be debated, ongoing trials are under way to help answer the treatment questions that remain.
Benefit of Adjuvant Chemotherapy
One of the principal investigations that reported a benefit for adjuvant combination chemotherapy was from the Early Breast Cancer Trialists' Collaborative Group (EBCTCG). The EBCTCG meets at 5-year intervals, performing meta-analyses on all mature randomized clinical trials focusing on women with early-stage breast cancer. Trials must have at least 5 years of follow-up data to be included in the analysis, with many now reporting 10 to 15 years of follow-up. The 1995 overview analysis summarized the results of trials beginning prior to 1990 and included data from 47 randomized trials with over 18,000 patients. These trials all compared polychemotherapy with no therapy in the adjuvant setting. Compared to no therapy, adjuvant chemotherapy was associated with reductions in recurrence and mortality in both younger and older women. In women under age 50 receiving chemotherapy, improvement in the absolute 10-year survival rate increased from 71% to 78% among those with node-negative disease, and from 42% to 53% among those with node-positive disease, compared to women receiving no treatment. In women aged 50 to 69 years, the absolute 10-year overall survival rate improved from 67% to 69% for node-negative disease and from 46% to 49% for node-positive disease.
Data from the 2000 EBCTCG meeting have not yet been published, but are consistent with the above benefits published in 1998. Most of these trials addressed older types of chemotherapy and did not include the taxanes. However, there are excellent long-term data that can be used to reach some conclusions regarding appropriate adjuvant drug selection that will be outlined in this article. On the basis of current data,women with lymph node metastases or with primary breast cancers larger than 1 cm, regardless of nodal status, should have a discussion about adjuvant polychemotherapy. The choice of chemotherapy should be individualized for patients who have node-negative cancers less than 1 cm in diameter, as there are no prospective data demonstrating survival advantages for chemotherapy in this setting. Tumor grade is increasingly recognized as an additional factor to be taken into consideration in determination of prognosis. In women with small, node-negative cancers who have favorable histologic subtypes such as mucinous or tubular tumors, adjuvant chemotherapy may be avoided.
Long-term follow-up is now available for patients who were treated with adjuvant oral CMF chemotherapy. Data published in 1995 by Bonadonna et al reported significantly better rates of relapse-free and overall survival at 20-year follow-up in patients who received adjuvant CMF chemotherapy compared to no therapy. Overall, the benefit translated into a 34% reduction in the relative risk of relapse and a 26% reduction in the relative risk of death. With the exception of postmenopausal women, a benefit from adjuvant chemotherapy was evident in all patient subgroups. Overall survival at 20 years in the CMF group was 47% for premenopausal women and 22% for postmenopausal patients. Event-free survival at 20 years in the CMF group was 23% (18% for patients who did not receive therapy). In addition, a recent Cancer and Leukemia Group B (CALGB) study reported a natural history analysis of more than 20 years for node-positive breast cancer patients who had been treated with CMF-based adjuvant therapy. This study included 814 women with node-positive disease enrolled over 6 years, with a median followup of 22.6 years. Of the 599 patients who were known to have died, 80% died from metastatic breast cancer. Only 8.5% died of other causes and 1.3% of treatment-related causes. In addition, the disease-free survival rate at 20 years was only 23% and the 20-year overall survival was 28%. As expected, the greater the number of positive lymph nodes, the worse the outcome. In patients with more than 10 positive lymph nodes, disease-free survival at 15 years was only 9% (compared with 0% in the above Bonadonna study).
In summary, these data suggest that CMF therapy is less than optimal for patients such as those in the above trials (ie, node-positive) due to the poor disease-free and overall survival rates. This analysis stresses the need for improved adjuvant therapies in this population. It should also be mentioned that there is evidence in the metastatic setting that classical oral CMF is more effective than intravenous (IV) CMF, possibly because of the greater dose intensity of classical CMF. This was demonstrated in a randomized phase III trial conducted by the European Organization for Research and Treatment of Cancer (EORTC) Breast Cancer Cooperative Group (10808) from 1991 comparing "classical" oral CMF vs a 3-weekly IV CMF schedule in postmenopausal patients with advanced breast cancer. The response rate with classical CMF was 48% compared with 29% for IV CMF (P = .003). Response duration was similar at 11 months, but survival was longer for the classical schedule (17 vs 12 months, P = .016). Thus, if physicians select the IV adjuvant regimen with the intention of reducing toxicity or facilitating scheduling, they may be prescribing the less optimal CMF regimen in terms of antitumor activity.
CMF vs Anthracylines
Until the early 1990s, CMF chemotherapy had generally been considered the standard adjuvant therapy for many breast cancer patients. Based on the above CALGB study and other studies included in the EBCTCG meta-analysis, it has become clear that more effective regimens are needed. The EBCTCG overview analysis compared approximately 6,000 women from 11 randomized trials. Anthracycline- containing regimens were associated with a modest but significant reduction in disease recurrence (40.5% vs 43.2%) and death (28.8% vs 30.5%) compared to CMF adjuvant therapy. Further follow-up indicates that the benefit favoring anthracyclines is still evident at 10 years.
In addition, the 2000 National Institutes of Health (NIH) consensus concluded, "the inclusion of anthracyclines in adjuvant chemotherapy regimens produces a small but statistically significant improvement in survival over non-anthracycline-containing regimens." Several randomized studies have compared CMF to an anthracycline-based regimen as adjuvant therapy. Table 1 summarizes the key randomized trials, including number of patients, disease-free survival, and overall survival.[1, 7-17] The results of these trials demonstrate that overall survival either favors an anthracycline- based regimen over CMF, or that the two regimens are equivalent.
Doxorubicin-Based Therapy vs CMF
• INT 0102—One of the largest, single randomized trials comparing anthracycline- based chemotherapy with CMF was the Southwest Oncology Group (SWOG)/Intergroup trial (INT 0102) conducted by Hutchins et al and presented at the American Society of Clinical Oncology (ASCO) annual meeting in 1998. In this study, 4,400 women with node-negative breast cancer were stratified into highor low-risk groups. Women whose tumors were ≥ 2 cm or hormone-receptor- negative were considered high risk. Women with estrogen-receptor-positive tumors ≤ 2 cm could be classified as high risk based on S-phase fraction. The object of the study was to determine whether the CAF regimen (cyclophosphamide, doxorubicin [Adriamycin], 5-FU) is superior to CMF for high-risk, node-negative breast cancer patients. Not only was this a large study, but the two arms were very similar in regard to the number of cycles administered and doses given. All high-risk women were randomized to receive CAF or classical oral CMF chemotherapy for six cycles, with or without tamoxifen for 5 years; low-risk patients did not receive adjuvant treatment. Recurrence rates were 15% in the CAF group and 18% in the CMF arm (13% and 15%, respectively, with the addition of tamoxifen). Estimated 5-year overall survival in CAF-treated patients was 92%, compared with 90% in the CMF group. CAF chemotherapy, however, was associated with slightly more toxicity. Grade 4 neutropenia as well as grade 2 nausea and vomiting were increased in the CAF group, and alopecia was more common in the CAF group. Two fatal toxicities occurred in the CAF group, compared with one in the CMF group. The investigators concluded that CAF is slightly superior to CMF, with the risk of increased but manageable acute toxicity.
• NSABP Trials—Other trials have compared the efficacy of four cycles of AC (doxorubicin, cyclophosphamide) to six cycles of CMF chemotherapy. The National Surgical Adjuvant Breast and Bowel Project (NSABP) conducted two large randomized trials (B-15 and B-23), comparing AC (60 mg/m2, 600 mg/m2) every 3 weeks for four cycles with the classical oral CMF regimen for six cycles.[8,18] These two trials showed a similar outcome and efficacy for each regimen, without a statistically significant difference in disease-free or overall survival. (B-15 included lymph node-positive patien ts, and B-23 included only lymph node-negative patients.) The NSABP B-23 trial showed a relapse-free survival of 87% at 5 years in both groups and a similar overall 5-year survival of 89% vs 90% with CMF and AC, respectively. The question remains, however, whether the outcome would have been different (ie, favoring the use of AC chemotherapy) if the study had used six cycles of AC instead of four.
Epirubicin-Based Therapy vs CMF
Epirubicin (Ellence) is another anthracycline that has been used in combination chemotherapy for adjuvant breast cancer treatment. Several randomized trials from Canada and Europe have compared epirubicinbased chemotherapy with CMF.
• NCIC MA.5—The National Cancer Institute of Canada (NCIC) MA.5 trial was initially published in 1998 by Levine et al, and recently updated. The NCIC MA.5 trial randomized 710 premenopausal women with node-positive breast cancer to receive oral CMF or CEF (cyclophosphamide, 75 mg/m2 orally on days 1 to 14; epirubicin, 60 mg/m2 IV on days 1 and 8; and 5-FU, 500 mg/m2 IV on days 1 and 8) for six cycles or oral CMF (cyclophosphamide, 100 mg/m2 orally on days 1 to 14; methotrexate, 40 mg/m2 IV on days 1 and 8; and 5-FU, 600 mg/m2 IV on days 1 and 8). Tamoxifen was not given after chemotherapy in this study. In evaluations of hematologic toxicity, the rate of grade 3/4 neutropenia was 94% vs 60% for CEF and CMF, respectively (growth factors were not administered). Febrile neutropenia occurred in 9% of patients in the CEF group vs 1% of patients in the CMF group. Grade 3/4 thrombocytopenia was also slightly increased in the CEF arm. Grade ≥ 2 nausea and vomiting was reported in 51% and 42% of the patients receiving CEF vs 25% and 18% of patients in the CMF group (no 5-HT3 receptor antagonists were given). Stomatitis and alopecia also occurred more frequently in the CEF group. Despite the increased toxicity in the CEF group, 97% of patients in both the CEF and CMF groups completed all six cycles of therapy. The NCIC MA.5 trial reported a significantly better 5-year disease-free survival of 63% in the CEF group vs 53% in the CMF group (relative risk reduction = 29%). Overall survival at 5 years in the CEF group was 77%, compared to 70% with CMF (relative risk reduction = 19%). An update of this trial was presented at the San Antonio Breast Cancer Symposium in 2002 (median follow-up = 106 months). The update showed that patients who received CEF adjuvant chemotherapy continued to have an improved survival at 10-year followup, as well as no statistically significant difference in late toxicity between CEF and CMF recipients. Specifically, the 10-year disease-free survival rate was 52% for patients who received CEF compared to 45% for CMF patients, and the 10-year overall survival was 62% and 58% for CEF and CMF patients, respectively. The rates of acute myeloid leukemia (AML) were unchanged since the original report (1% AML incidence in the CEF group vs 0.3% in the CMF group), while the rates of congestive heart failure were 1.1% in the CEF group vs 0.3% in the CMF group. In conclusion, the MA.5 trial demonstrated a benefit for CEF over CMF adjuvant chemotherapy that is maintained with longer follow-up.
• European Trials—Anthracycline-based chemotherapy can be used concurrently or sequentially as adjuvant therapy for early-stage breast cancer. Epirubicin has recently been evaluated as sequential therapy with CMF for adjuvant treatment. The National Epirubicin Adjuvant Trial (NEAT) and BR9601 trial were a combined prospective meta-analysis of two multicenter phase III randomized trials from the United Kingdom and Scotland, the results of which were presented by C. Poole at the 2003 ASCO meeting. Patients were randomized to receive ECMF (epirubicin, 100 mg/m2, for four cycles, followed by classical CMF for four cycles) vs CMF for six or eight cycles (NEAT: classical day-1-and-8 CMF every 4 weeks for six cycles; BR9601: IV 3- weekly CMF for eight cycles). The two studies analyzed a total of 2,391 women with node-positive and node-negative cancers who were eligible for adjuvant therapy. Toxicities were higher in the ECMF group, including nausea, vomiting, and alopecia. Febrile neutropenia was slightly higher in the ECMF group, at 13% vs 10%. There was no difference between the two groups in the number of second primaries. It is important to note that a significantly higher number of deaths on treatment were noted in the CMF group compared to the ECMF group (13 vs 4 patients). The relapse-free survival rate was significantly better in the ECMF group-83% vs 77%-with a 31% reduction in recurrence. Overall survival was also significantly better in the ECMF group compared to the CMF group (88% vs 82.7%). In summary, sequential ECMF significantly prolongs relapse-free survival and overall survival compared to CMF adjuvant chemotherapy, and is justified as an option for standard anthracycline- based adjuvant therapy. Thus, this is another study that prospectively established the superiority of the anthracycline-containing regimen (compared to CMF alone).
• Cardiac Dysfunction—The long-term toxicities of the anthracyclines have been evaluated in several studies.[20,21] Doxorubicin is known to cause cardiac dysfunction; the incidence of cardiomyopathy and congestive heart failure secondary to this agent is dose-dependent, increasing substantially at cumulative doses greater than 500 mg/m2. One study found that congestive heart failure occurred in 4% of patients who received a cumulative dose of 500 to 550 mg/m2, compared to 36% of patients who received a dose greater than 601 mg/m2. Among women who receive the standard cumulative adjuvant doses of doxorubicin (≤ 300 mg/m2), congestive heart failure is uncommon (up to 1%), although radiation to the chest may increase the risk of cardiac toxicity from anthracyclines. Perez and colleagues recently reported data on the effect of four cycles of AC chemotherapy (doxorubicin at 60 mg/m2, cyclophosphamide at 600 mg/m2) in 1,458 patients who participated in the NCCTG N9831 Intergroup adjuvant trial. This study demonstrated that "standard" AC is associated with frequent asymptomatic decreases in left-ventricular ejection fraction (LVEF). This includes reductions ≥ 15% in 2.5% of patients, and 2.9% with decreases of ≤ 15% but which fell below the lower limit of normal, with a 6.6% rate of grade 2 LVEF toxicity by NCI-CTC criteria. Of note is the finding that the acute decrease in LVEF is reversible in the majority of patients receiving anthracyclines.
• Secondary Hematologic Disease—There is little evidence that the risk of second cancers, including treatment-related leukemia, is increased in women who receive CMF, and there is some information regarding treatment- related leukemia for anthracycline-based adjuvant regimens. In 2003, the NCIC reported the risk of acute leukemia following CMF and anthracycline-based adjuvant chemotherapy in 1,545 women. At a follow-up of 8 years, the conditional probability of secondary acute leukemia was 1.7% among the 539 patients treated with CEF, 0.4% among 678 women who received CMF, and 1.3% among 231 patients treated with AC. In addition, the NSABP recently published results from six trials evaluating the incidence of AML and myelodysplastic syndrome (MDS) after AC adjuvant chemotherapy. The development of AML/MDS was elevated in patients receiving the more intense regimens. In patients receiving two or four cycles of cyclophosphamide at 2,400 mg/m2 with granulocyte colony-stimulating factor (G-CSF [Neupogen]) support, the cumulative incidence of AML/MDS at 5 years was 1.01%, compared with 0.21% for patients who were given standard AC. Patients who received breast radiotherapy experienced more secondary AML/MDS than did those who did not receive radiotherapy (relative risk = 2.38, P = .006). Thus, breast cancer patients treated with standard doses of anthracyclines may have a similar or slightly higher rate of secondary acute leukemia compared to CMF, which is slightly above that of the general population for the development of leukemia. The addition of radiation therapy may increase this risk.
Overall, we can conclude that CMF adjuvant chemotherapy is better than no treatment for early-stage breast cancer. For low-risk patients, the above studies conclude that six cycles of oral CMF or four cycles of AC every 3 weeks are equivalent in efficacy. Clinical trials that have incorporated doxorubicin and epirubicin into polychemotherapy regimens (CAF and CEF for six cycles) show a modest but clear benefit for anthracycline-based therapy compared to six cycles of CMF in adjuvant breast cancer. Based on these results, anthracycline therapy with six cycles of CAF or CEF is recommended for higher-risk patients (ie, node-positive patients), and may be used as a benchmark for newer regimens incorporating taxanes and other novel agents. It is important to note that six cycles of CAF/CEF chemotherapy regimens have not been directly compared to four or six cycles of AC in the adjuvant setting. Also, there is no current evidence of excessive cardiac toxicity in women with normal heart function who receive anthracyclines at the cumulative doses utilized in standard adjuvant programs. The trade-off for a small survival benefit with anthracyclines is a different treatment-related toxicity profile including higher incidences of alopecia, vomiting, and cytopenias compared to CMF. With the availability of effective antiemetics (5-HT3 receptor antagonists and NK1 inhibitors) and the use of growth factor support, most of these acute toxicities can be fairly manageable. Data regarding the long-term toxicities of anthracyclines demonstrate a low risk of cardiomyopathy and treatmentrelated leukemia. Overall, we seldom recommend CMF as adjuvant therapy in our practice, and most often recommend anthracycline-based therapy, although we remain very interested in the development of nonanthracycline combinations other than CMF.