This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Antman, K. H. Search for Related Content PubMed PubMed Citation Articles by Antman, K. H.

Critique of the High-Dose Chemotherapy Studies in Breast Cancer: A Positive Look at the Data

From the Departments of Medicine and Biostatistics, Columbia University, Herbert Irving Comprehensive Cancer Center, New York, NY.

Address reprint requests to Karen Antman, MD, MHB 6N 435, 177 Ft Washington Ave, New York, NY, 10032; emailkha{at}columbia.edu

AT BOTH MILITARY academies and business schools, policy is divided into strategy (What is the right thing to do?) versus tactics or operations (What is the right way to do it?). The strategic question here is whether higher dose chemotherapy cures more patients. Whether to use stem cells or growth factor support is a tactic, along with single versus multiple cycles of therapy, sequential single agent chemotherapy versus combinations, and induction versus immediate high-dose chemotherapy.

Examples of each of four study designs are shown in Fig 1. The U.S. intergroup study recently presented by W.P. Peters et al¹ compared a single high-dose treatment cycle with a low-dose treatment cycle using the same drugs. The Scandinavian and Philadelphia trials studied cumulative dose. Dose rate was a major question in the National Surgical Adjuvant Breast and Bowel Project (NSABP) B22 and B25 studies.^2,3 Finally, high-dose sequential therapy was studied at the Institute Tumore (Milan, Italy) and at the Memorial Sloan-Kettering Cancer Center (MSKCC; New York, NY). Table 1 lists characteristics of all currently published randomized high-dose breast cancer studies. The three adjuvant studies are included in this supplement. The two smaller randomized Phase II studies have already been published.^4,5

View larger version (20K): [in this window] [in a new window]   Fig 1. Examples of each of four study designs (peak dose, cumulative dose, dose rate, and sequential dose).

Of the metastatic studies, only the Philadelphia trial is included in this supplement. The French study was presented as a poster at the 1999 American Society of Clinical Oncology (ASCO) meeting.⁶ The Duke study was presented at ASCO in 1996⁷ and the South African study has been published.⁸ Note that mortality in the randomized trials for both the high-dose and control groups is consistently in the 0 to 2.5% range, except for the U.S. intergroup study.¹

Adjuvant High-Dose Studies
Cancer and Leukemia Group B intergroup study.¹ On review of the adjuvant data, the most important point of the Intergroup Study headed by W.P. Peters is that the data are preliminary, with only 3.6 years of follow-up. The event-free survival (EFS) rate is approximately 70%, and, therefore, only one third of the predicted relapses have occurred.¹ This study was not scheduled to be analyzed for several more years. Comments have been made about the lack of "big differences." With the control group EFS rate at 70% and the high-dose chemotherapy (HDC) group toxic death rate at 7%, the maximum difference possible is 23%. Thus, big differences are impossible with the current survival rate of the control group, which was selected to have a tumor mortality rate of 85%. Survival in both arms will almost certainly fall with time, and differences may or may not emerge. Although low-dose cyclophosphamide, carmustine, and cisplatin (CBP) is not a standard regimen, this study, scientifically, is a pure comparison between high- and low-dose CBP.

This first generation carmustine; 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU)-containing regimen had a 7% mortality rate. W.P. Peters' has shown an analysis with mortality censored. Although we cannot necessarily assume that another regimen with a lower mortality would also be as effective, transplant registry data can detect no difference in efficacy among the most commonly used regimens.

The South African study. The South African study is a direct comparison of conventional cyclophosphamide, doxorubicin, and fluorouracil (CAF) versus two cycles of HDC.⁹ Family history as an eligibility criteria was unusual, but the number of such patients was small and equally distributed. Mortality was 1% on both the CAF and HDC arms. The median follow-up was 5.3 years. Disease-free survival (DFS) and overall survival are both significantly improved.

The Scandinavian trial. The Scandinavian trial probably included some patients with metastatic disease because positive marrow examinations and bone scans were allowed. The Scandinavian study did not compare HDC versus conventional-dose chemotherapy,¹⁰ but instead, compared one HDC cycle versus six moderately high–dose cycles. Follow-up was at just 2 years and OS rate was 60%. Again, like the U.S. adjuvant intergroup study, the Scandinavian trial is a very preliminary look at an interesting study.

The design of the Scandinavian study requires some comment. Step 4 of the tailored-dose arm gave the following doses per cycle: fluorouracil 600 mg/m², epirubicin 120 mg/m², and cyclophosphamide 1800 mg/m². The planned doses for Arm A and B are shown here (Table 2). The planned dose for tailored therapy significantly exceeds that for the bone marrow transplant (BMT) arm. The labels are not reversed. Thus, a superior DFS and OS for tailored therapy would support the importance of cumulative dose over early peak dose.

Three percent of patients on the tailored-dose arm developed leukemia or myelodysplasia. With a median follow-up of only 2 years, additional cases are likely. No patients have developed leukemias yet on the BMT arm. Leukemia risk was also increased in the NSABP B25 trial, which also escalated cyclophosphamide dose without stem-cell support.³ Perhaps stem cells collected after three cycles of chemotherapy are less damaged than those exposed to nine relatively high–dose cycles. If survival on the two arms proves equivalent, with better quality of life on the BMT arm, patients might prefer short intensive therapy over nine cycles of moderately high–dose chemotherapy.

Review of the Adjuvant High-Dose Studies The median follow-up times for the Scandinavian, American, and South African trials are projected on a survival curve from the 20-year follow-up of Bonnadona's cyclophosphamide, methotrexate, and fluorouracil adjuvant trial (Fig 2).¹¹

View larger version (15K): [in this window] [in a new window]   Fig 2. The median follow-up times of the Scandinavian, American, and South African trials projected on a survival curve from the 20-year follow-up of Bonadona's cyclophosphamide, methotrexate, and fluorouracil adjuvant trial.

Survival of breast cancer patients is inferior to age-matched controls for up to two decades after diagnosis. The survival in both arms fell significantly after 2 years of follow-up, and differences subsequently increased. In the Parma study, which randomized patients with relapsed lymphoma to conventional- or high-dose therapy, the curves were overlapping at the first interim analysis, P = .18 at the second analysis, but P = .04 and significant at the third analysis.^12,13 Even in lymphomas, which progress more rapidly than breast cancers, we had to wait for mature data.

The approximate size of a study required to show statistically significant differences with a power of 90%, as a function of control group mortality are shown in Fig 3. For the South African trial, with a control group mortality of 80%, a 30% improved survival in the HDC group is significant in a 150-patient study. In the U.S. intergroup study, the current control group mortality is 30%. Even with a 50% survival of the control group in this approximately 900-patient trial, only fairly large differences would be significant. If we had a very large trial, a 1% difference might be statistically significant, although not necessarily medically meaningful.

View larger version (49K): [in this window] [in a new window]   Fig 3. The approximate size of a study required to show statistically significant differences with a power of 90%, as a function of control group mortality (P = .05).

We need more data to get reliable answers. Four additional large randomized adjuvant studies of HDC are closed to accrual but are as yet unreported, and four more continue to accrue patients. Only one U.S. study is open (for patients with four to nine involved lymph nodes).

Based on the data so far, mortality in all but one study is 0% to 2.5 %. The only direct comparison with a standard CAF adjuvant regimen (South African trial) is unequivocally positive. The U.S. Intergroup trial, which is the purest scientific comparison of high-dose versus low-dose CBP, is early but interesting. And the Scandinavian trial studies a totally different aspect: one high-dose cycle versus six intermediate-dose cycles.

Metastatic Disease
The Philadelphia study. In the Philadelphia study, the number of patients randomized is 199, 36% of the patients enrolled onto the trial. An additional 18% of these randomized patients were ineligible or did not receive their assigned treatment.¹⁴ This seems to result in a randomized group with poorer than expected characteristics. Prior adjuvant chemotherapy, estrogen receptor—negative tumor and visceral disease are three characteristics shown in the U.S. transplant registry data¹⁵ to be significantly associated with poor transplant outcome (Table 3).

Three other metastatic studies from South Africa,⁸ the U.S.,⁷ and France⁶ show either significant differences or a trend for an advantage for high-dose therapy in comparison with conventional-dose therapy. Five randomized HDC studies are ongoing. No open randomized U.S. studies could be found and none of the studies listed seems near completion.

Based on the data presented, and in the interim before more data are available, what do we tell patients? We can tell patients that the mortality rate is 0% to 1% for most commonly used regimens and that the toxicity on cumulative-dose studies, such as the Scandinavian and Philadelphia trials, was not very different. However, we obviously need to encourage patients to participate in trials, although just recommending participation in trials is not fair because many patients do not have access to them. In fact, a problem for Americans is that only one randomized study is currently open, although a Southwest Oncology Group intergroup study of more than nine involved lymph nodes is planned, and pilot studies are open at various institutions. At this point, physicians need to provide a careful explanation of what we know and what we don't for patients considering HDC off-study. Additionally, a number of questions need to be asked in carefully designed studies, such as the magnitude of any benefit, the optimal regimen and schedule, what drugs, how many cycles, and combinations versus sequential single agents.

The data published to date support the conclusion that dose remains an important and promising strategy to explore in breast cancer. Although data need to be made publicly available as soon as possible, we probably need to remain cautious about interpreting abstracts without seeing all of the actual data. The current status of the field is probably best summed up by a quote by Winston Churchill (in a speech to the Canadian Senate and House of Commons, December 30, 1941): "Now is not the end. It is not even the beginning of the end, but it is, perhaps, the end of the beginning."

REFERENCES

1. Peters WP, Rosner G, Vredenburgh J, et al: A prospective, randomized comparison of two doses of combination alkylating agents as consolidation after CAF in high-risk primary breast cancer involving ten or more axillary lymph nodes: Preliminary results of CALGB 9082/SWOG 9114/NCIC MA-13. Proc Am Soc Clin Oncol 18:1a, 1999 (abstr 2)

2. Fisher B, Anderson S, Wickerham DL, et al: Increased intensification and total dose of cyclophosphamide in a doxorubicin-cyclophosphamide regimen for the treatment of primary breast cancer: Findings from National Surgical Adjuvant Breast and Bowel Project B-22. J Clin Oncol15:1858-1869, 1997[Abstract/Free Full Text]

3. Fisher B, Anderson S, DeCillis A, et al: Further evaluation of intensified and increased total dose of cyclophosphamide for the treatment of primary breast cancer: Findings from national surgical adjuvant breast and bowel project B-25. (in press)

4. Rodenhuis S, Richel KJ, Wall EVD, et al: Randomized trial of high-dose chemotherapy and hematopoietic progenitor cell support in operable breast cancer with extensive axillary lymph node involvement. Lancet352:515-521, 1998[Medline]

5. Hortobagyi GN, Buzdar AU, Champlin R, et al: Lack of efficacy of adjuvant high-dose tandem combination chemotherapy for high-risk primary breast cancer: A randomized trial. Proc Am Soc Clin Oncol17:123, 1998 (abstr 471)

6. Lotz JP, Cure H, Janvier M, et al: High-dose chemotherapy with hematopoietic stem cells transplantation for metastatic breast cancer: Results of the French protocol Pegase 04. Proc Am Soc Clin Oncol 18:43a, 1999 (abstr 161)

7. Peters W, Jones R, Vredenburgh J, et al: A large prospective randomized trial of high-dose combination alkylating agents (CPB) with autologous cellular support as consolidation for patients with metastatic breast cancer achieving complete remission after intensive doxorubicin-based induction therapy (AFM). Proc Am Soc Clin Oncol15:121, 1996 (abstr)

8. Bezwoda W, Seymour L, Dansey R: High-dose chemotherapy with hematopoietic rescue as primary treatment for metastatic breast cancer: A randomized trial. J Clin Oncol13:2483-2489, 1995[Abstract]

9. Bezwoda WR: Randomised, controlled trial of high dose chemotherapy versus standard dose chemotherapy for high risk, surgically treated, primary breast cancer. Proc Am Soc Clin Oncol 18:2a (abstr 4) 1999

10. Bergh J: Results from a randomized adjuvant breast cancer study with high-dose chemotherapy with CTCb supported by autologous bone marrow stem cells versus dose-escalated and tailored FEC therapy. Proc Am Soc Clin Oncol 18:2a, 1999 (abstr 3)

11. Bonadonna G, Valagussa P, Moliterni A, et al: Adjuvant cyclophosphamide, methotrexate, and fluorouracil in node-positive breast cancer: the results of 20 years of follow-up. N Engl J Med332:901-906, 1995[Abstract/Free Full Text]

12. Philip T, Guglielmi C, Hagenbeek A, et al: Autologous bone marrow transplantation as compared with salvage chemotherapy in relapses of chemotherapy-sensitive non-Hodgkin's lymphoma. N Engl J Med333:1540-1545, 1995[Abstract/Free Full Text]

13. Philip T, Guglielmi C, Hagenbeek A, et al: The Parma international randomized prospective study in relapsed nonHodgkin's Lymphoma; second interim analysis of 172 patients. Proc Eur Group Bone Marrow Transplant18:203, 1992 (abstr 402)

14. Stadtmauer EA, O'Neill A, Goldstein LJ, et al: Phase III randomized trial of high-dose chemotherapy and stem cell support shows no difference in overall survival or severe toxicity compared to maintenance chemotherapy with cyclophosphamide, methotrexate and 5-fluorouracil for women with metastatic breast cancer who are responding to conventional induction chemotherapy: The Philadelphia intergroup study. Proc Am Soc Clin Oncol 18:1a, 1999 (abstr 1)

15. Rowlings PA, Williams SF, Antman KH, et al: Factors correlated with progression-free survival after high-dose therapy and hematopoietic stem cell transplantation for metastatic breast cancer. JAMA (in press)

This article has been cited by other articles:

				M. E. Lippman High-Dose Chemotherapy plus Autologous Bone Marrow Transplantation for Metastatic Breast Cancer N. Engl. J. Med., April 13, 2000; 342(15): 1119 - 1120. [Full Text]

				S. Siena, R. Schiavo, P. Pedrazzoli, and C. Carlo-Stella Therapeutic Relevance of CD34 Cell Dose in Blood Cell Transplantation for Cancer Therapy J. Clin. Oncol., March 13, 2000; 18(6): 1360 - 1377. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Antman, K. H. Search for Related Content PubMed PubMed Citation Articles by Antman, K. H.