Prospective, Randomized Comparison of High-Dose Chemotherapy With Stem-Cell Support Versus Intermediate-Dose Chemotherapy After Surgery and
http://www.100md.com
《临床肿瘤学》
Adherex Technologies Inc
Duke Comprehensive Cancer Center
CALGB Statistical Center
Department of Medicine-JJV, Department of Radiation Oncology-LBM, Duke University Medical Center, Durham, NC
Department of Bone and Marrow Transplantation, M.D. Anderson Cancer Center, Houston, TX
Department of Medical Oncology and Hematology, Princess Margaret Hospital, Toronto, Canada
Dana-Farber Cancer Institute, Boston, MA
North Shore–Long Island Jewish Medical Center, Manhasset, NY
Division of Solid Tumor Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY
University of California San Francisco Cancer Center, San Francisco, CA
CALGB Central Office of the Chairman, Chicago, IL
Section of Hematology Oncology, Wake Forest University School of Medicine, Winston-Salem, NC.
ABSTRACT
PURPOSE: The prognosis for women with primary breast cancer involving multiple axillary nodes remains poor. High-dose chemotherapy with stem-cell support produced promising results in initial clinical trials conducted at single institutions.
PATIENTS AND METHODS: Seven hundred eighty-five women aged 22 to 66 years with stage IIA, IIB, or IIIA breast cancer involving 10 or more axillary lymph nodes were randomized after surgery and standard adjuvant chemotherapy to either high-dose cyclophosphamide, cisplatin, and carmustine (HD-CPB) with stem-cell support or intermediate-dose cyclophosphamide, cisplatin, and carmustine (ID-CPB) with G-CSF support but without stem cells. Planned treatment for all patients included locoregional radiation therapy. Hormone-receptor–positive patients were to receive 5 years of tamoxifen. Event-free survival (EFS) was the primary end point.
RESULTS: Median follow-up was 7.3 years. Event-free survival was not significantly different between the two treatment groups (P = .24). The probability of being free of an event at 5 years with HD-CPB was 61% (95% CI, 56% to 65%), and was 58% (95% CI, 53% to 63%) for ID-CPB.
Thirty-three patients died of causes attributed to HD-CPB, compared with no therapy-related deaths among women treated with ID-CPB. Overall survival for the two arms was identical at 71% at 5 years (P = .75).
CONCLUSION: HD-CPB with stem-cell support was not superior to ID-CPB for event-free or overall survival among all randomized women with high-risk primary breast cancer.
INTRODUCTION
Primary breast cancer that involves multiple axillary lymph nodes has a poor prognosis.1 Despite the introduction of new chemotherapy regimens and monoclonal antibody therapy, long-term results for these patients have not improved significantly over the past three decades.2
Single institution studies using high-dose chemotherapy and stem-cell support in high-risk patients suggested that it may be possible to achieve long-term disease-free survival but also demonstrated the significant morbidity and mortality associated with this approach.3-5 Conducting adequately powered randomized controlled trials of high-dose therapy has been difficult for multiple reasons including inadequate insurance coverage, physician and patient preferences, and difficulty in achieving consensus among investigators about the optimal treatment program.
In this article, we report results from a prospective, randomized trial evaluating high-dose chemotherapy in women with primary breast cancer involving multiple axillary lymph nodes. This trial evolved from preclinical observations demonstrating a steep dose-response relationship, non-overlapping toxicity, and potential synergy among selected alkylating agents, and from single institution, nonrandomized clinical observations demonstrating encouraging outcomes among women with high-risk primary breast cancer.4,6-8 At the time this trial was designed, a prevailing area of interest for Cancer and Leukemia Group B (CALGB) was the question of whether increasing the dose-intensity of treatment resulted in better outcomes for cancer patients. Therefore, rather than compare high-dose chemotherapy with stem-cell support to standard adjuvant chemotherapy, the design of this trial focused solely on the question of dose. The two treatment arms included the same drugs and the same number of cycles of treatment and differed in only one respect, the doses of the drugs employed. Many investigators were also concerned that accruing patients to a study that randomly assigned patients to receive or not receive transplantation would not be possible. This study was designed, therefore, to examine the question of dose in a rigorous fashion, and with treatment regimens felt to be acceptable to patients and physicians at the time.
PATIENTS AND METHODS
Patient Population and Study Teams
Women were eligible for the study if they were at least 18 years old, physiologically younger than 55 years old, and within 8 weeks of definitive surgery for breast cancer that involved 10 or more axillary lymph nodes. Patients were required to have stage IIA, IIB, or IIIA carcinoma of the breast, and were excluded if there was evidence of inflammatory breast cancer or metastatic disease when evaluated by computed tomography (CT) of the head, chest, abdomen, and pelvis, radionuclide scans of the bone, bilateral bone marrow aspirations and biopsies, and liver function tests. Adequate organ function was required and was defined as a leukocyte count of at least 4,000/mm3, platelet count of at least 150,000/mm3, hemoglobin level of at least 9.0 gm per dl, AST and ALT levels no higher than 100 IU per milliliter, serum creatinine level no higher than 1.2 mg per dl, creatinine clearance of at least 60 mL per minute, and forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and diffusion of carbon monoxide (DLCO) at 60% or greater than the predicted value.
Patients were required to provide written informed consent to participate in the study, and patients were required to provide evidence of adequate financial resources (insurance coverage or other arrangements) to cover the costs of protocol-specified treatments. The protocol was approved by the institutional review board of each participating center.
Because of the complexity of the protocol and the known morbidity of the high-dose chemotherapy regimen, teams had to demonstrate their proficiency by treating at least three patients. Doing so would also help ensure consistency in the administration of the protocol-specified therapy across multiple sites. One hundred fifty-one patients, designated "preliminary," were treated with the same high-dose treatment protocol by the transplant teams that were then evaluated on their ability to conduct the study in accordance with the protocol. Forty teams in the United States and Canada ultimately participated in the study, representing three cooperative groups.
The CALGB Data Audit Committee reviewed primary patient records for 23% of all patients entered on the protocol with on-site audits of the medical records to verify eligibility, treatment, toxicity, and outcome. Furthermore, an independent Data and Safety and Monitoring Committee (DSMC) reviewed primary toxicity and efficacy data from the study at 6-month intervals to determine if early stopping of the protocol was necessary.
Treatment
The treatment schema is shown in Figure 1. Patients initially received cyclophosphamide (600 mg/m2 intravenously on day 1 of cycle), doxorubicin (60 mg/m2 intravenously on day 1), and fluorouracil (600 mg/m2 on day 1 and day 8 of cycle; CAF). CAF was repeated at 28-day intervals for a total of four cycles. Between the third and fourth cycles of CAF, the patients were restaged with CT scans of the chest, abdomen, and pelvis, radionuclide bone scan, bone marrow examination, and serum chemistries to rule out recurrence of disease. After re-evaluation, patients free of disease were randomly assigned as described in the Chemotherapy Treatment section, and then treated with the fourth cycle of CAF. Randomization was delayed as long as possible to avoid potential bias caused by patients dropping out before the initiation of assigned therapy.9
Cellular Support
All patients treated with high-dose cyclophosphamide, cisplatin, and carmustine (HD-CPB) received both bone marrow and peripheral blood progenitor cell support.1,10 Bone marrow was collected if the leukocyte count was greater than 3,000/mm3 after either cycle 3 or cycle 4 of CAF, depending on whether verification of insurance coverage had been obtained at the time of scheduled marrow collection. Bone marrow was collected from the posterior iliac crests under general or regional anesthesia. A buffy coat of bone marrow was prepared in accordance with institutional standards. The bone marrow was stored in a liquid phase of nitrogen until used. At the appropriate time, the bone marrow was thawed rapidly at 37°C in a water bath and infused through a central catheter for 10 minutes without further processing.
In addition to bone marrow, patients received granulocyte colony-stimulating factor (G-CSF) –primed peripheral-blood progenitor cells. After full hematopoietic recovery from the last cycle of CAF, patients received G-CSF (filgrastim, Amgen Inc, Thousand Oaks, CA) at 5 μg/kg/day subcutaneously for 8 days. Patients underwent leukapheresis the morning after the fifth, seventh, and eighth doses of G-CSF. Each leukapheresis was performed via a central venous catheter following standard institutional procedures. The peripheral-blood progenitor cells were cryopreserved on a daily basis in accordance with institutional practice and then stored in the liquid phase of nitrogen until used. CD34 data were not routinely obtained following leukapheresis.
Chemotherapy Treatment
High-dose CPB. All patients had a triple lumen central venous catheter placed for venous access. Before initiation of the high-dose chemotherapy, bladder irrigation was begun with 1 L/hr urologic saline containing 2 ml of neomycin and polymyxin B. This was continued until 24 hours after the last dose of cyclophosphamide to minimize hemorrhagic cystitis. Aggressive hydration was used and urine output was replaced on a mililiter-for-mililiter basis throughout chemotherapy. If urine output fell below 200 mL/hr, additional hydration was provided until urine output returned to the level of greater than 200 mL/hr.
The high-dose chemotherapy, administered over 4 days, consisted of cyclophosphamide (1,875 mg/m2/d) administered as a 1-hour intravenous infusion on 3 successive days, cisplatin (55 mg/m2/d) administered as a 72-hour continuous intravenous infusion for a total dose of 165 mg/m2, and carmustine (600 mg/m2) administered on the last day of chemotherapy as an intravenous infusion at a rate of 5 mg/m2/min, unless hypotension, not responsive to intravenous fluids, occurred. If patients were more than 20% over ideal body weight, the administered doses of high-dose cyclophosphamide, cisplatin, and carmustine (CPA/cDDP/BCNU) in the high-dose arm of this program were calculated using the average of the body-surface area calculated on the basis of actual and ideal body weight.11
Patients received transfusions of packed RBCs until a hematocrit of more than 42% was reached and patients were maintained at this level by transfusion if necessary until hematopoietic recovery. We have previously reported that HD-CPB induces a functional platelet defect, which is partially corrected by transfusion of allogeneic platelets.12 All patients were therefore transfused twice with single-donor platelets 24 hours after completion of chemotherapy, regardless of the measured platelet count. Throughout the remainder of the treatment course, single-donor platelets were administered to maintain a platelet count greater than 25,000/μL.
Intermediate-dose CPB. Patients randomly assigned to intermediate-dose CPB (ID-CPB) received cyclophosphamide (900 mg/m2/d) as a 1-hour intravenous infusion on 3 successive days, cisplatin (90 mg/m2) as a 72-hour continuous intravenous infusion, and, on the last day, carmustine (90 mg/m2) as an intravenous infusion at a rate of 5 mg/m2/min. Patients were prescribed G-CSF at 5 μg/kg/d subcutaneously from the time the WBC fell to less than 1,000 cells/mm3 until the polymorphonuclear leukocyte (PMN) count was more than 1,000 cells/mm3, for 2 consecutive days.
Radiation Therapy
The protocol prescribed locoregional radiation therapy following recovery from either dose-intensive regimen. Patients typically received 45 to 50 Gy to the chest wall/breast and regional lymph nodes (supraclavicular ± internal mammary, ± axillary) with a 10 to 15 Gy scar/tumor bed boost at 1.8 to 2.0 Gy per fraction, for a period of approximately 6 to 7 weeks, using standard radiation techniques. The details of administration and the effects of radiation therapy are described elsewhere.1,13
Tamoxifen
Patients with hormone-receptor–positive tumors (measured estrogen- or progesterone-receptor level > 7 fmol/mg protein) or patients for whom receptor status was unknown were prescribed tamoxifen 10 mg by mouth twice daily for 5 years following the completion of chemotherapy.
Protocol Compliance
Patients were declared ineligible if they did not have all protocol-specified restaging tests before randomization. Overall, 122 patients were declared ineligible because of missing laboratory tests—predominantly, either a missing resting or exercise multigated blood pool imaging of the heart (MUGA). Ten of these 122 patients were not randomly assigned for other reasons (listed in Table 1), and are excluded from the analysis. The remaining 112 patients (54 randomly assigned to the HD-CPB arm, and 58 randomly assigned to the ID-CPB arm) were considered minor violations, were randomly assigned, and are included in the intention-to-treat analysis of 785 randomly assigned patients.
Data on the use of tamoxifen were not rigorously collected during the study. However, the available data indicate that 489 (89%) of 548 (HD-CPB, 86%; ID-CPB, 92%) of the patients who had hormone-receptor–positive or unknown tumors began tamoxifen, and 155 (32%) of the 489 patients (HD-CPB, 28%; ID-CPB, 35%) received tamoxifen for 5 years.
Subsequent Therapy
After first relapse, all patients could undergo HD-CPB plus transplant (ie, the HD-CPB regimen). Postrelapse transplantation was not part of the protocol, however, and data were not systematically collected regarding postrelapse therapy.
Evaluations Performed After Transplant
Six weeks after receiving high-dose or intermediate-dose chemotherapy, patients were fully evaluated using the same tests as pretreatment, except for bone marrow aspiration, biopsy, and bone scan. Patients were then monitored at 6-week intervals with physical examination and at 12-week intervals with CT of the chest, abdomen, and pelvis for the first 2 years. Evaluations after 2 years were performed at 6-month intervals, or more often as clinically indicated.
Quality-of-Life Evaluation
Patients on this study were eligible to participate in a companion quality of life protocol (CALGB 9066). Patients received mailed questionnaires, and data collection was completed with follow-up telephone interviews. The questionnaire consisted of the Functional Living Index Cancer (FLIC) and the Symptom Distress Scale (SDS). The results of the quality-of-life study are described elsewhere.14
Statistical Considerations
The primary end point of the study was event-free survival (EFS), defined as the time from randomization until the first event from the following: breast cancer recurrence, death due to any cause, or diagnoses of secondary acute myeloid leukemia/myelodysplastic syndrome (AML/MDS) or breast cancer. Patients alive and event free at the time of this analysis are censored at the last date they were known to be event free.
The primary comparison of the two treatments is with the log-rank test,15 including the version for adjusting for the stratification variables used in randomization (ie, hormone-receptor status, menopausal status, and stage). The Cox proportional hazards regression model15 provides estimates of the effect of therapy on outcome accounting for patient characteristics. Estimates of the probability of being free from outcomes of interest over time are based on the method of Kaplan and Meier.15 All P values are two-sided. Treatment comparisons are by intention to treat.
Initially, the study sought 80% power to detect an improvement in median EFS from 2 years to 3 years, leading to a goal of randomizing 250 patients over a period of 5 years, with an additional 3 years of follow-up. In 1994, based on results of a phase II study of high-dose chemotherapy1 and the results of CALGB 8541,16 an adjuvant chemotherapy trial, we felt that the EFS would likely be better than originally hypothesized. With median EFS expected to be beyond 5 years, we amended the protocol to increase accrual.
The new sample-size calculations were based on computer simulations17 of the clinical trial, assuming various hypothesized underlying survival curves. We allowed for 10 interim analyses18 by the DSMC beginning after 20% of the anticipated events occurred. The interim analyses used the log-rank test. Two thousand simulations allowed estimation of statistical power to be accurate within 1% to 2% of the true power (with 95% probability).
The treatment difference we wanted to detect included crossing treatment-specific hazard functions. Women receiving HD-CPB would initially have a higher risk of treatment-related mortality. The group-specific EFS curves would come together by 1 year with 80% EFS, and would then diverge. Five-year EFS would be 70% for HD-CPB and 55% for ID-CPB. The simulations indicated that randomly assigning and treating 760 eligible patients (with at least 3 years' follow-up after randomization) would provide 90% power to detect the EFS difference of interest. Assuming that 5% of the patients entering the study would refuse further therapy or might be found ineligible, we sought to register 800 patients over an expected 6.67 years at 120 patients per year. All analyses are based on the study database that was locked on June 30, 2003.
RESULTS
Between January 1991 and May 1998, 885 patients enrolled onto the study, 10 patients were canceled, and 785 were randomized per protocol. Forty transplant centers participated; two centers enrolled 40% of the patients.
The reasons 100 patients were not randomly assigned are listed in Table 1. The primary reasons were denial of insurance coverage and relapse during the induction chemotherapy with CAF. Of the 24 women randomly assigned to HD-CPB who did not receive assigned treatment, nine refused the assigned treatment, two women experienced CAF-related mortality, two women had medical reasons (decreasing left ventricular ejection fraction and increasing liver function test), one woman's blood counts did not return to normal following CAF, one woman had disease progresssion, two women received nonprotocol therapy, and no reason was given for seven women. Of the 18 women randomly assigned to ID-CPB who did not receive assigned treatment, six withdrew, two had medical reasons (bipolar disorder and diverticulitis), one patient's blood counts did not return to normal following CAF, two received nonprotocol therapy, and seven women did not give a reason. Of the 394 patients randomly assigned to the high-dose chemotherapy arm (HD-CPB), 370 patients (94%), received the high-dose cyclophosphamide, cisplatin, and carmustine with stem-cell support; there were two patients randomly assigned to HD-CPB who were treated with ID-CPB (a major protocol violation), but were analyzed as HD-CPB recipients as part of the intent-to-treat analysis. Of the 391 patients randomly assigned to the intermediate-dose arm, 373 patients (95%) received ID-CPB.
Table 2 lists the patient characteristics for the 785 randomly assigned patients. The study was well balanced for age, ethnicity, and disease characteristics. The median age of the patients was 44 years (range, 22 to 66 years); 25% of the patients were at least 50 years of age. The median tumor size was 3.0 cm (range, 0.1 to 18.0 cm), and the median number of involved lymph nodes was 14 (range, 10 to 55), but 19% of patients had 20 or more lymph nodes involved at the time of surgery. Twenty-one percent of the patients had tumors larger than 5 cm in size (stage IIIA), and 69% of tumors were hormone-receptor positive.
EFS
The median follow-up time for EFS was 7.3 years and included 372 events (99% of the 375 events expected under the assumptions underlying the sample-size calculations). There have been 179 events (relapses or deaths) among the 394 randomized patients on the high-dose chemotherapy arm, and 193 events among the 391 patients on the ID-CPB chemotherapy arm. One hundred forty-eight and 146 patients, respectively, have died on the HD-CPB and ID-CPB chemotherapy arms.
For the 785 randomly assigned patients, 5-year EFS was 61% for HD-CPB (95% CI, 56% to 65%) and 58% for ID-CPB (95% CI, 53% to 63%). EFS was not significantly different between the two treatment groups (P = .24); nor did it differ (P = .21) after adjusting for the stratification variables used in randomization (hormone-receptor status, menopausal status, and stage; Fig 2A). The hazard ratio for EFS was 0.89 (95% CI, 0.72 to 1.09).
Age by itself was not a significant predictor of EFS (P = .23 for continuous age and P = .99 comparing women 50 years old and older to women younger than 50 years). An unplanned analysis comparing the effect of HD-CPB among women 50 years old and older with the effect among women who were 49 years and younger showed a significant interaction (P = .02). Figures 3A and 3B show the Kaplan-Meier EFS curves for patients younger than 50 years of age and 50 years or older, respectively. When this analysis is stratified with respect to the three factors used in the randomization (stage, hormone status, and menopausal status), the interaction between the treatment effect and a woman's age being 50 years or older is just statistically significant (P = .05). The interaction between treatment and menopausal status (premenopausal versus peri- or postmenopausal) is not significant, however (P = .49).
Relapses
There were 320 relapses among the 785 randomly assigned patients. One hundred thirty-seven (35%) of the 394 patients on the HD-CPB arm relapsed (95% CI, 30% to 40%), compared with 183 (47%) of the 391 women assigned to the ID-CPB arm (95% CI, 42% to 52%). Among women younger than 50 years randomly assigned to the HD-CPB arm, a smaller percentage relapsed during the follow-up period, compared with women younger than 50 years randomly assigned to the ID-CPB arm. However, the same fraction of women 50 years and older relapsed regardless of treatment assignment. Table 3 lists the treatment allocation, frequency of relapses, and treatment-related mortality for patients younger than 50 years compared with those who were 50 years or older. The cumulative hazard of relapse by treatment is shown in Figure 4. The figure shows that the women who received ID-CPB relapsed at a greater rate during the first 2 years (ie, steeper slope). After 2 years of follow-up, however, there seems to be less of a difference in the rate of relapse between the two groups.
Overall Survival
There was no significant difference in survival between the two treatment groups (P = .75). The probability of surviving 5 years (from randomization) was 71% (95% CI, 66% to 75%) for patients on each treatment arm. Figure 2B shows the Kaplan-Meier curves for survival. The hazard ratio for overall survival was 1.04 (95% CI, 0.82 to 1.30).
Toxicity
Treatment-related mortality. Thirty-five patients died of treatment-related causes; two patients died as a result of toxicity from the CAF induction chemotherapy and the remaining 33 treatment-related deaths occurred following HD-CPB and stem-cell transplantation (9% of 370 transplantations). The causes of HD-CPB mortality are listed in Table 4. Infections,19 pulmonary toxicity,20 and hemolytic-uremic syndrome21 were the primary causes of treatment-related mortality.
Fourteen of the 370 patients who received HD-CPB died within 100 days of the start of the high-dose therapy. The 100-day transplantation-related mortality is 3.8% (95% CI, 2.1% to 6.3%, exact binomial method). Thirty of the 33 therapy-related deaths occurred within the first year following high-dose chemotherapy; one of the three late deaths was due to AML secondary to treatment, and the other two deaths were caused by treatment-related pulmonary fibrosis, which was diagnosed based on clinical findings and confirmed at autopsy.
The overall TRM varied considerably, ranging from 5.6% to 28.6% among the centers participating in the study. Centers performing more than 10 transplantations tended to have a lower TRM (7%; 95% CI, 4% to 12%) compared with those performing fewer than 10 transplantations (16%; 95% CI, 8% to 27%).
Of the 370 patients who underwent transplant, 268 were younger than 50 years and 102 patients were 50 years or older. The TRM was 6.7% for the former group and 14.7% for the latter group. The TRM for the 111 women younger than 40 years was 5.4%.
Treatment-related toxicity. Leukopenia and thrombocytopenia were common among patients receiving either HD-CPB or ID-CPB, but were more severe and persistent among HD-CPB patients. Hepatic, pulmonary, and nervous system toxicity of grade 3 or greater occurred in more than 10% of patients receiving HD-CPB chemotherapy, whereas these toxicities were infrequent among patients receiving ID-CPB.
Hemolytic uremic syndrome (HUS) or thrombotic thrombocytopenic purpura (TTP) were severe complications of HD-CPB. HUS or TTP occurred in 14 patients receiving HD-CPB and were primary or associated causes of death in 10 patients. These syndromes have been previously reported to complicate high-dose therapy and proved resistant to usual treatment approaches.21
Thirty-six randomly assigned patients, 16 on the HD-CPB arm and 20 on the ID-CPB arm, developed secondary cancers. The most common sites were AML/MDS (HD-CPB, seven patients; ID-CPB, four patients) and breast (HD-CPB, five patients; ID-CPB, eight patients). Median onset of AML/MDS was approximately 45 months in both groups. Nine of the 13 breast cancers were considered new primary tumors.
DISCUSSION
There are four conclusions from this study: (1) that for all randomized patients EFS was not significantly different between patients randomly assigned to ID-CPB versus patients assigned to HD-CPB; (2) that transplant-related toxicity is severe and particularly adverse for HD-CPB among women 50 years and older; (3) that overall survival for patients on both treatment regimens was encouraging; and (4) that among women younger than 50 years, HD-CPB appears to be associated with fewer relapses.
The difference in EFS at 7 years of follow-up was approximately 4% (54% for HD-CPB and 50% for ID-CPB) in absolute terms or an 8% relative reduction in event rate for all patients. Whereas differences of this magnitude are not uncommon in trials of standard-dose treatments for breast cancer, the difficulty of administering the HD-CPB procedure and the magnitude of the toxicity led us to design a trial seeking a larger therapeutic advantage if HD-CPB was to be recommended. The trial was designed with a sample size sufficient to detect a 15% difference in 5-year EFS.
The EFS was dominated by the events of relapses in the intermediate-dose arm and the combination of therapy-related deaths and relapses in the high-dose arm. The rapid advance of stem-cell and hematopoietic–growth factor technologies22,23 prompted the extension of the study to older patients. Furthermore, the limitation to age 50 years was thought to be arbitrary and the protocol was therefore written to permit those "physiologically" under the age of 55 years to be enrolled. The consequence of these shifts in emphasis resulted in 25% of the patients entered being at least 50 years old.
Women 50 years and older appeared to be at higher risk of treatment-related mortality than younger women if randomly assigned to HD-CPB. However, the results suggest a benefit of HD-CPB for younger women (younger than 50 years). These results are similar to those obtained in other randomized trials in this population. The Dutch multicenter trial24 demonstrated a statistically significant improvement in disease-free survival among similar breast cancer patients younger than 40 years who received high-dose chemotherapy and among patients with HER2-neu negative tumors. In the current study, however, there is no significant interaction between treatment and age when split at age 40 years (P = .46).
Results suggesting benefits for high-dose therapy in younger women have also been reported from randomized trials conducted in Italy,25 Germany, and France.26,27 The Anglo-Celtic study did not show an advantage for high-dose therapy.28 A recently reported Eastern Cooperative Oncology Group trial did not report on the younger patient subgroup.29 The results of these studies collectively provide some insight into the role of high-dose therapy. No study demonstrates an overall or disease-free survival advantage for high-dose therapy, but in several studies there is either a strong trend or statistically significant advantage for high-dose therapy among younger women. It is notable that the pilot trial on which the design of this trial was based included predominantly younger women.1 Biologic differences in the effectiveness of dose-intense therapy between age groups and other parameters (HER2 status24) have been described previously and may provide an important lead for subsequent studies.
The ID-CPB treatment used in this study produced better outcomes than expected, compared with other commonly used regimens, and particularly when compared with the CALGB 8541 study of CAF, results of which were used to estimate requisite sample sizes for this trial.15,30-31 Overall survival for all patients in this study (71% overall survival at 5 years from randomization) is encouraging and surprising. The reason for the favorable results is undoubtedly multifactorial and may include the use of additional chemotherapeutic agents not usually part of standard adjuvant therapy, stage migration in breast cancer, delay until completion of three cycles of CAF before randomly assigning patients, or extensive pretreatment staging resulting in better patient selection, as well as the use of radiation and hormonal therapies. It is noteworthy that the results with HD-CPB in this study are essentially identical to the results from the phase II studies, which prompted this prospective trial. However, caution is necessary when making comparisons to other adjuvant studies, since randomization in our study occurred after three cycles of adjuvant CAF therapy. We estimate that the probability of a relapse or death by median time to randomization is approximately 3%. Still, to our knowledge, no conventional treatment program with similar follow-up and size has produced superior survival results in this patient population.
Interpreting the results of this trial compared with standard treatment regimens is complicated by the use of ID-CPB, which had previously been evaluated in only a preliminary fashion. We chose to compare HD-CPB to ID-CPB to focus on dose and to avoid introducing confounding interpretations because of different drugs in the arms of the study. The interpretation is further confounded by the observation that the HD-CBP patients were less likely to complete the prescribed locoregional radiation therapy than were the ID-CBP patients.32 Since such locoregional therapy appears to have an impact on disease-free, and possibly overall survival,33,34 the HD-CBP patients may have been negatively affected by the reduced use of radiation therapy. The absence of a true control arm in this study does not permit us to draw conclusions regarding whether either HD-CPB or ID-CPB produce better outcomes in this patient population than might be achieved with CAF alone. As mentioned previously, however, the overall results in this trial appear superior to those previously reported with conventional adjuvant chemotherapy regimens.
The trend toward fewer relapses among younger women on HD-CPB does not appear to be related to the induction of a chemical menopause. Of the randomized patients who were premenopausal at study entry, 42% of those taking HD-CPB became amenorrheic compared with 43% of those taking ID-CPB. The difference in relapse frequency between ID-CPB and HD-CPB was similar for women with hormone-receptor–positive and hormone-receptor–negative tumors.
Women with breast cancer have more treatment options now than ever before with the introduction of aromatase inhibitors, trastuzumab, and new cytotoxic agents such as capecitabine and gemcitabine. Nevertheless, our study suggests that clinical trials should continue to examine the role of high-dose chemotherapy with stem-cell support in high-risk women younger than 50 years to determine unequivocally whether this approach provides clinical benefit.
Appendix
The following institutions participated in the study: Brook Army Medical Center/Wilford Hall Medical Center, San Antonio, TX, Mitchell A. Garrison, MD, supported by CA76447; CALGB Statistical Office, Durham, NC, Stephen George, PhD, supported by CA33601; Cancer Care Manitoba–Winnipeg, Manitoba, Tsiporah Shore, MD; Central IL CCOP, Springfield, IL, James L. Wade, MD, supported by CA45807; Cleveland Clinic, Cleveland, OH, George Thomas Budd, MD, supported by CA04919; Columbia River CCOP, Portland, OR, Keith S. Lanier, MD, supported by CA45377; Dana-Farber Cancer Institute, Boston, MA, George P. Canellos, MD, supported by CA32291; Dartmouth Medical School–Norris Cotton Cancer Center, Lebanon, NH, Marc Ernstoff, MD, supported by CA04326; Duke University Medical Center, Durham, NC, Jeffrey Crawford, MD, supported by CA47577; Grand Rapids CCOP, Grand Rapids, MI, Kathleen J. Yost, MD, supported by CA35178; Henry Ford Hospital, Detroit, MI, Robert A. Chapman, MD, supported by CA58416; Massachusetts General Hospital, Boston, MA, Michael Grossbard, MD, supported by CA12449; McGill Cancer Center, Montreal, Quebec, Canada, Brian Leyland-Jones, MD, supported by CA31809; Medical University of South Carolina, Charleston, SC, Mark Green, MD, supported by CA03927; Memorial Sloan-Kettering Cancer Center, New York, NY, George Bosl, MD, supported by CA77651; Mount Sinai School of Medicine, New York, NY, Lewis Silverman, MD, supported by CA04457; North Shore–Long Island Jewish Medical Ctr, Manhasset, NY, Daniel R. Budman, MD, supported by CA35279; Oregon Health Science University, Portland, OR, Craig R. Nichols, MD, supported by CA46113; Rhode Island Hospital, Providence, RI, William Sikov, MD, supported by CA08025; Roswell Park Cancer Institute, Buffalo, NY, Ellis Levine, MD, supported by CA02599; St Louis University, St Louis, MO, Paul J. Petruska, MD, supported by CA76462; SUNY Upstate Medical University, Syracuse, NY, Stephen L. Graziano, MD, supported by CA21060; Temple University, Temple, TX, supported by CA58415; The Toronto Hospital–General Division Toronto, Ontario, Canada, Michael Crump, MD; Toronto SunnyBrook Regional Cancer Centre–Toronto, Ontario, Canada, Kathleen Pritchard, MD; University of Arkansas, Little Rock, AR, Laura F. Hutchins, MD, supported by CA37981; University of California at San Diego, San Diego, CA, Stephen Seagren, MD, supported by CA11789; University of California San Francisco, San Francisco, CA, Alan Venook, MD, supported by CA60138; University of Chicago Medical Center, Chicago, IL, Gini Fleming, MD, supported by CA41287; University of Colorado, Denver, CO, Scott I. Bearman, MD, supported by CA42777; University of Hawaii CCOP, Honolulu, HI, Brian F. Issell, MD, supported by CA63844; University of Iowa, Iowa City, IA, Gerald Clamon, MD, supported by CA47642; University of Kentucky, Lexington, KY, Michael A. Doukas, MD, supported by CA46136; University of Maryland Cancer Center, Baltimore, MD, David Van Echo, MD, supported by CA31983; University of Minnesota, Minneapolis, MN, Bruce A. Peterson, MD, supported by CA16450; University of Missouri/Ellis Fischel Cancer Center, Columbia, MO, Michael C. Perry, MD, supported by CA12046; University of North Carolina at Chapel Hill, Chapel Hill, NC, Thomas C. Shea, MD, supported by CA47559; University of Oklahoma, Oklahoma City, OK, Howard Ozer, MD, PhD, supported by CA58686; Upstate Carolina, Spartanburg, SC, James D. Bearden, MD, supported by CA35119; Vancouver Cancer Centre–Vancouver, British Columbia, Canada, Joseph Ragaz, MD, and Michael Barnett, MD; Virginia Mason CCOP, Seattle, WA, Andrew D. Jacobs, MD, supported by CA35192; Wake Forest University School of Medicine, Winston-Salem, NC, David D. Hurd, MD, supported by CA03927; Walter Reed Army Medical Center, Washington, DC, Joseph J. Drabeck, MD, supported by CA26806; Washington University School of Medicine, St Louis, MO, Nancy Bartlett, MD, supported by CA77440; Wayne State University, Detroit, MI, Lawrence E. Flaherty, MD, supported by CA14028; Weill Medical College of Cornell University, New York, NY, Scott Wadler, MD, supported by CA07968.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We would like to express our appreciation to Dr Raymond Weiss and to the members of the CALGB Data Audit Committee for their efforts to help ensure the quality and validity of these data. Above all, we express our gratitude to the patients and their families who participated in this study.
NOTES
Supported in part by grants from the National Cancer Institute (CA31946) to the Cancer and Leukemia Group B (Richard L. Schilsky, Chairman), and by grants CA33601, CA14028, CA33601, CA47577, CA42777, CA77202, CA32291, CA35279, CA77651, CA60138, CA41287, and CA03927.
The manuscript contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Peters WP, Ross M, Vredenburgh JJ, et al: High-dose chemotherapy and autologous bone marrow support as consolidation after standard-dose adjuvant therapy for high-risk primary breast cancer. J Clin Oncol 11:1132-1143, 1993
Peters WP, Dansey RD, Klein JL, et al: High-dose chemotherapy and peripheral blood progenitor cell transplantation in the treatment of breast cancer. Oncologist 5:1-13, 2000
Antman K, Eder JP, Elias A, et al: High-dose combination alkylating agent preparative regimen with autologous bone marrow support: The Dana-Farber Cancer Institute/Beth Israel Hospital experience. Cancer Treat Rep 71:119-125, 1987
Baynes RD, Dansey RD, Klein JL, et al: High-dose chemotherapy and autologous stem cell transplantation for breast cancer. Cancer Invest 18:440-455, 2000
Rodenhuis S, Richel DJ, van der Wall E, et al: Randomized trial of high-dose chemotherapy and haemopoietic progenitor-cell support in operable breast cancer with extensive axillary lymph-node involvement. Lancet 352:515-521, 1998
Peters WP: High-dose chemotherapy with autologous bone marrow transplantation for the treatment of breast cancer: Yes. Important Adv Oncol 215-230, 1995
Antman K, Ayash L, Elias A, et al: High-dose cyclophosphamide, thiotepa, and carboplatin with autologous marrow support in women with measurable advanced breast cancer responding to standard-dose therapy: Analysis by age. J Natl Cancer Inst Monogr 16:91-94, 1994
Antman K, Eder JP, and Frei E 3rd: High-dose chemotherapy with bone marrow support for solid tumors. Important Adv Oncol 221-235, 1987
Durrleman S, Simon R: When to randomize J Clin Oncol 9:116-122, 1991
Shpall EJ: The utilization of cytokines in stem cell mobilization strategies. Bone Marrow Transplant 23:S13-S19, 1999 (suppl 2)
Petros WP, Broadwater G, Berry D, et al: Association of high-dose cyclophosphamide, cisplatin, and carmustine pharmacokinetics with survival, toxicity, and dosing weight in patients with primary breast cancer. Clin Cancer Res 8:698-705, 2002
Panella TJ, Peters W, White JG, et al: Platelets acquire a secretion defect after high-dose chemotherapy. Cancer 65:1711-1716, 1990
Marks LB, Rosner GL, Prosnitz LR, et al: The impact of conventional plus high dose chemotherapy with autologous bone marrow transplantation on hematologic toxicity during subsequent local-regional radiotherapy for breast cancer. Cancer 74:2964-2971, 1994
Winer EP, Lindley C, Hardee M, et al: Quality of life in patients surviving at least 12 months following high dose chemotherapy with autologous bone marrow support. Psychooncology 8:167-176, 1999
Klein JP, Moeschberger ML: Survival Analysis: Techniques for Censored and Truncated Data. New York, NY, Springer-Verlag Inc, 1997
Wood WC, Budman DR, Korzun AH, et al: Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 330:1253-1259, 1994
Halpern J, Brown BW Jr: A computer program for designing clinical trials with arbitrary survival curves and group sequential testing. Control Clin Trials 14:109-122, 1993
O'Brien PC, Fleming TR: A multiple testing procedure for clinical trials. Biometrics 35:549-556, 1979
Williams SF, Mick R, Desser R, et al: High-dose consolidation therapy with autologous stem cell rescue in stage IV breast cancer. J Clin Oncol 7:1824-1830, 1989
Patz EF Jr, Peters WP, Goodman PC: Pulmonary drug toxicity following high-dose chemotherapy with autologous bone marrow transplantation: CT findings in 20 cases. J Thorac Imaging 9:129-134, 1994
Fisher DC, Sherrill GB, Hussein A, et al: Thrombotic microangiopathy as a complication of high-dose chemotherapy for breast cancer. Bone Marrow Transplant 18:193-198, 1996
Rosti G, Ferrante P, Ledermann J, et al: High-dose chemotherapy for solid tumors: Results of the EBMT. Crit Rev Oncol Hematol 41:129-140, 2002
Simnett SJ, Stewart LA, Sweetenham J, et al: Autologous stem cell transplantation for malignancy: A systematic review of the literature. Clin Lab Haematol 22:61-72, 2000
Rodenhuis S, Bontenbal M, Beex LV, et al: High-dose chemotherapy with hematopoietic stem-cell rescue for high-risk breast cancer. N Engl J Med 349:7-16, 2003
Gianni AM: Standard vs High-Dose Chemotherapy in the Adjuvant Treatment of Breast Cancer with > 3 LN+: 5 year results. Proc Am Soc Clin Oncol 20:21a, 2001 (abstr 80)
Roche H, Viens P, Biron P, et al: PEGASE Group High-dose chemotherapy for breast cancer: The French PEGASE experience. Cancer Control 10:42-47, 2003
Roche H, Pouillart P, Meyer N, et al: Adjuvant High Dose Chemotherapy Improves Early Outcome for High Risk (N > 7) Breast Cancer Patients: The Pegase 01 Trial. Proc Am Soc Clin Oncol 20:26a, 2001 (abstr 102)
Crown JP, Lind M, Gould A, et al: Failure of adjuvant peripheral blood progenitor-supported high-dose chemotherapy to improve the prognosis of patients with high-risk breast cancer in a randomized trial with low treatment-related mortality: The Anglo-Celtic I Study. Ann Oncol 13:2, 2002 (suppl 5, abstr 20)
Tallman MS, Gray R, Robert NJ, et al: Conventional adjuvant chemotherapy with or without high-dose chemotherapy and autologous stem-cell transplantation in high-risk breast cancer. N Engl J Med 349:17-26, 2003
Budman DR, Berry DA, Cirrincione CT, et al: Dose and dose intensity as determinants of outcome in the adjuvant treatment of breast cancer: The Cancer and Leukemia Group B. J Natl Cancer Inst 90:1205-1211, 1998
Roach M 3rd, Cirrincione C, Budman D, et al: Race and survival from breast cancer: Based on Cancer and Leukemia Group B trial 8541. Cancer J Sci Am 3:107-112, 1997
Marks LB, Cirrincione C, Peterson B, et al: The impact of local/regional radiotherapy (RT), and its timing, on survival following lumpectomy/mastectomy and systemic chemotherapy in patients with 10 positive axillary nodes: Analysis of CALGB 9082. Int J Radiat Oncol Biol Phys 60:S136, 2004 (suppl)
Ragaz J, Jackson SM, Le N, et al: Adjuvant radiotherapy and chemotherapy in node positive premenopausal women with breast cancer. N Engl J Med 337:956-962, 1997
Overgaard M, Hansen PS, Overgaard J, et al: Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. N Engl J Med 337:949-955, 1997(William P. Peters, Gary L)
Duke Comprehensive Cancer Center
CALGB Statistical Center
Department of Medicine-JJV, Department of Radiation Oncology-LBM, Duke University Medical Center, Durham, NC
Department of Bone and Marrow Transplantation, M.D. Anderson Cancer Center, Houston, TX
Department of Medical Oncology and Hematology, Princess Margaret Hospital, Toronto, Canada
Dana-Farber Cancer Institute, Boston, MA
North Shore–Long Island Jewish Medical Center, Manhasset, NY
Division of Solid Tumor Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY
University of California San Francisco Cancer Center, San Francisco, CA
CALGB Central Office of the Chairman, Chicago, IL
Section of Hematology Oncology, Wake Forest University School of Medicine, Winston-Salem, NC.
ABSTRACT
PURPOSE: The prognosis for women with primary breast cancer involving multiple axillary nodes remains poor. High-dose chemotherapy with stem-cell support produced promising results in initial clinical trials conducted at single institutions.
PATIENTS AND METHODS: Seven hundred eighty-five women aged 22 to 66 years with stage IIA, IIB, or IIIA breast cancer involving 10 or more axillary lymph nodes were randomized after surgery and standard adjuvant chemotherapy to either high-dose cyclophosphamide, cisplatin, and carmustine (HD-CPB) with stem-cell support or intermediate-dose cyclophosphamide, cisplatin, and carmustine (ID-CPB) with G-CSF support but without stem cells. Planned treatment for all patients included locoregional radiation therapy. Hormone-receptor–positive patients were to receive 5 years of tamoxifen. Event-free survival (EFS) was the primary end point.
RESULTS: Median follow-up was 7.3 years. Event-free survival was not significantly different between the two treatment groups (P = .24). The probability of being free of an event at 5 years with HD-CPB was 61% (95% CI, 56% to 65%), and was 58% (95% CI, 53% to 63%) for ID-CPB.
Thirty-three patients died of causes attributed to HD-CPB, compared with no therapy-related deaths among women treated with ID-CPB. Overall survival for the two arms was identical at 71% at 5 years (P = .75).
CONCLUSION: HD-CPB with stem-cell support was not superior to ID-CPB for event-free or overall survival among all randomized women with high-risk primary breast cancer.
INTRODUCTION
Primary breast cancer that involves multiple axillary lymph nodes has a poor prognosis.1 Despite the introduction of new chemotherapy regimens and monoclonal antibody therapy, long-term results for these patients have not improved significantly over the past three decades.2
Single institution studies using high-dose chemotherapy and stem-cell support in high-risk patients suggested that it may be possible to achieve long-term disease-free survival but also demonstrated the significant morbidity and mortality associated with this approach.3-5 Conducting adequately powered randomized controlled trials of high-dose therapy has been difficult for multiple reasons including inadequate insurance coverage, physician and patient preferences, and difficulty in achieving consensus among investigators about the optimal treatment program.
In this article, we report results from a prospective, randomized trial evaluating high-dose chemotherapy in women with primary breast cancer involving multiple axillary lymph nodes. This trial evolved from preclinical observations demonstrating a steep dose-response relationship, non-overlapping toxicity, and potential synergy among selected alkylating agents, and from single institution, nonrandomized clinical observations demonstrating encouraging outcomes among women with high-risk primary breast cancer.4,6-8 At the time this trial was designed, a prevailing area of interest for Cancer and Leukemia Group B (CALGB) was the question of whether increasing the dose-intensity of treatment resulted in better outcomes for cancer patients. Therefore, rather than compare high-dose chemotherapy with stem-cell support to standard adjuvant chemotherapy, the design of this trial focused solely on the question of dose. The two treatment arms included the same drugs and the same number of cycles of treatment and differed in only one respect, the doses of the drugs employed. Many investigators were also concerned that accruing patients to a study that randomly assigned patients to receive or not receive transplantation would not be possible. This study was designed, therefore, to examine the question of dose in a rigorous fashion, and with treatment regimens felt to be acceptable to patients and physicians at the time.
PATIENTS AND METHODS
Patient Population and Study Teams
Women were eligible for the study if they were at least 18 years old, physiologically younger than 55 years old, and within 8 weeks of definitive surgery for breast cancer that involved 10 or more axillary lymph nodes. Patients were required to have stage IIA, IIB, or IIIA carcinoma of the breast, and were excluded if there was evidence of inflammatory breast cancer or metastatic disease when evaluated by computed tomography (CT) of the head, chest, abdomen, and pelvis, radionuclide scans of the bone, bilateral bone marrow aspirations and biopsies, and liver function tests. Adequate organ function was required and was defined as a leukocyte count of at least 4,000/mm3, platelet count of at least 150,000/mm3, hemoglobin level of at least 9.0 gm per dl, AST and ALT levels no higher than 100 IU per milliliter, serum creatinine level no higher than 1.2 mg per dl, creatinine clearance of at least 60 mL per minute, and forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and diffusion of carbon monoxide (DLCO) at 60% or greater than the predicted value.
Patients were required to provide written informed consent to participate in the study, and patients were required to provide evidence of adequate financial resources (insurance coverage or other arrangements) to cover the costs of protocol-specified treatments. The protocol was approved by the institutional review board of each participating center.
Because of the complexity of the protocol and the known morbidity of the high-dose chemotherapy regimen, teams had to demonstrate their proficiency by treating at least three patients. Doing so would also help ensure consistency in the administration of the protocol-specified therapy across multiple sites. One hundred fifty-one patients, designated "preliminary," were treated with the same high-dose treatment protocol by the transplant teams that were then evaluated on their ability to conduct the study in accordance with the protocol. Forty teams in the United States and Canada ultimately participated in the study, representing three cooperative groups.
The CALGB Data Audit Committee reviewed primary patient records for 23% of all patients entered on the protocol with on-site audits of the medical records to verify eligibility, treatment, toxicity, and outcome. Furthermore, an independent Data and Safety and Monitoring Committee (DSMC) reviewed primary toxicity and efficacy data from the study at 6-month intervals to determine if early stopping of the protocol was necessary.
Treatment
The treatment schema is shown in Figure 1. Patients initially received cyclophosphamide (600 mg/m2 intravenously on day 1 of cycle), doxorubicin (60 mg/m2 intravenously on day 1), and fluorouracil (600 mg/m2 on day 1 and day 8 of cycle; CAF). CAF was repeated at 28-day intervals for a total of four cycles. Between the third and fourth cycles of CAF, the patients were restaged with CT scans of the chest, abdomen, and pelvis, radionuclide bone scan, bone marrow examination, and serum chemistries to rule out recurrence of disease. After re-evaluation, patients free of disease were randomly assigned as described in the Chemotherapy Treatment section, and then treated with the fourth cycle of CAF. Randomization was delayed as long as possible to avoid potential bias caused by patients dropping out before the initiation of assigned therapy.9
Cellular Support
All patients treated with high-dose cyclophosphamide, cisplatin, and carmustine (HD-CPB) received both bone marrow and peripheral blood progenitor cell support.1,10 Bone marrow was collected if the leukocyte count was greater than 3,000/mm3 after either cycle 3 or cycle 4 of CAF, depending on whether verification of insurance coverage had been obtained at the time of scheduled marrow collection. Bone marrow was collected from the posterior iliac crests under general or regional anesthesia. A buffy coat of bone marrow was prepared in accordance with institutional standards. The bone marrow was stored in a liquid phase of nitrogen until used. At the appropriate time, the bone marrow was thawed rapidly at 37°C in a water bath and infused through a central catheter for 10 minutes without further processing.
In addition to bone marrow, patients received granulocyte colony-stimulating factor (G-CSF) –primed peripheral-blood progenitor cells. After full hematopoietic recovery from the last cycle of CAF, patients received G-CSF (filgrastim, Amgen Inc, Thousand Oaks, CA) at 5 μg/kg/day subcutaneously for 8 days. Patients underwent leukapheresis the morning after the fifth, seventh, and eighth doses of G-CSF. Each leukapheresis was performed via a central venous catheter following standard institutional procedures. The peripheral-blood progenitor cells were cryopreserved on a daily basis in accordance with institutional practice and then stored in the liquid phase of nitrogen until used. CD34 data were not routinely obtained following leukapheresis.
Chemotherapy Treatment
High-dose CPB. All patients had a triple lumen central venous catheter placed for venous access. Before initiation of the high-dose chemotherapy, bladder irrigation was begun with 1 L/hr urologic saline containing 2 ml of neomycin and polymyxin B. This was continued until 24 hours after the last dose of cyclophosphamide to minimize hemorrhagic cystitis. Aggressive hydration was used and urine output was replaced on a mililiter-for-mililiter basis throughout chemotherapy. If urine output fell below 200 mL/hr, additional hydration was provided until urine output returned to the level of greater than 200 mL/hr.
The high-dose chemotherapy, administered over 4 days, consisted of cyclophosphamide (1,875 mg/m2/d) administered as a 1-hour intravenous infusion on 3 successive days, cisplatin (55 mg/m2/d) administered as a 72-hour continuous intravenous infusion for a total dose of 165 mg/m2, and carmustine (600 mg/m2) administered on the last day of chemotherapy as an intravenous infusion at a rate of 5 mg/m2/min, unless hypotension, not responsive to intravenous fluids, occurred. If patients were more than 20% over ideal body weight, the administered doses of high-dose cyclophosphamide, cisplatin, and carmustine (CPA/cDDP/BCNU) in the high-dose arm of this program were calculated using the average of the body-surface area calculated on the basis of actual and ideal body weight.11
Patients received transfusions of packed RBCs until a hematocrit of more than 42% was reached and patients were maintained at this level by transfusion if necessary until hematopoietic recovery. We have previously reported that HD-CPB induces a functional platelet defect, which is partially corrected by transfusion of allogeneic platelets.12 All patients were therefore transfused twice with single-donor platelets 24 hours after completion of chemotherapy, regardless of the measured platelet count. Throughout the remainder of the treatment course, single-donor platelets were administered to maintain a platelet count greater than 25,000/μL.
Intermediate-dose CPB. Patients randomly assigned to intermediate-dose CPB (ID-CPB) received cyclophosphamide (900 mg/m2/d) as a 1-hour intravenous infusion on 3 successive days, cisplatin (90 mg/m2) as a 72-hour continuous intravenous infusion, and, on the last day, carmustine (90 mg/m2) as an intravenous infusion at a rate of 5 mg/m2/min. Patients were prescribed G-CSF at 5 μg/kg/d subcutaneously from the time the WBC fell to less than 1,000 cells/mm3 until the polymorphonuclear leukocyte (PMN) count was more than 1,000 cells/mm3, for 2 consecutive days.
Radiation Therapy
The protocol prescribed locoregional radiation therapy following recovery from either dose-intensive regimen. Patients typically received 45 to 50 Gy to the chest wall/breast and regional lymph nodes (supraclavicular ± internal mammary, ± axillary) with a 10 to 15 Gy scar/tumor bed boost at 1.8 to 2.0 Gy per fraction, for a period of approximately 6 to 7 weeks, using standard radiation techniques. The details of administration and the effects of radiation therapy are described elsewhere.1,13
Tamoxifen
Patients with hormone-receptor–positive tumors (measured estrogen- or progesterone-receptor level > 7 fmol/mg protein) or patients for whom receptor status was unknown were prescribed tamoxifen 10 mg by mouth twice daily for 5 years following the completion of chemotherapy.
Protocol Compliance
Patients were declared ineligible if they did not have all protocol-specified restaging tests before randomization. Overall, 122 patients were declared ineligible because of missing laboratory tests—predominantly, either a missing resting or exercise multigated blood pool imaging of the heart (MUGA). Ten of these 122 patients were not randomly assigned for other reasons (listed in Table 1), and are excluded from the analysis. The remaining 112 patients (54 randomly assigned to the HD-CPB arm, and 58 randomly assigned to the ID-CPB arm) were considered minor violations, were randomly assigned, and are included in the intention-to-treat analysis of 785 randomly assigned patients.
Data on the use of tamoxifen were not rigorously collected during the study. However, the available data indicate that 489 (89%) of 548 (HD-CPB, 86%; ID-CPB, 92%) of the patients who had hormone-receptor–positive or unknown tumors began tamoxifen, and 155 (32%) of the 489 patients (HD-CPB, 28%; ID-CPB, 35%) received tamoxifen for 5 years.
Subsequent Therapy
After first relapse, all patients could undergo HD-CPB plus transplant (ie, the HD-CPB regimen). Postrelapse transplantation was not part of the protocol, however, and data were not systematically collected regarding postrelapse therapy.
Evaluations Performed After Transplant
Six weeks after receiving high-dose or intermediate-dose chemotherapy, patients were fully evaluated using the same tests as pretreatment, except for bone marrow aspiration, biopsy, and bone scan. Patients were then monitored at 6-week intervals with physical examination and at 12-week intervals with CT of the chest, abdomen, and pelvis for the first 2 years. Evaluations after 2 years were performed at 6-month intervals, or more often as clinically indicated.
Quality-of-Life Evaluation
Patients on this study were eligible to participate in a companion quality of life protocol (CALGB 9066). Patients received mailed questionnaires, and data collection was completed with follow-up telephone interviews. The questionnaire consisted of the Functional Living Index Cancer (FLIC) and the Symptom Distress Scale (SDS). The results of the quality-of-life study are described elsewhere.14
Statistical Considerations
The primary end point of the study was event-free survival (EFS), defined as the time from randomization until the first event from the following: breast cancer recurrence, death due to any cause, or diagnoses of secondary acute myeloid leukemia/myelodysplastic syndrome (AML/MDS) or breast cancer. Patients alive and event free at the time of this analysis are censored at the last date they were known to be event free.
The primary comparison of the two treatments is with the log-rank test,15 including the version for adjusting for the stratification variables used in randomization (ie, hormone-receptor status, menopausal status, and stage). The Cox proportional hazards regression model15 provides estimates of the effect of therapy on outcome accounting for patient characteristics. Estimates of the probability of being free from outcomes of interest over time are based on the method of Kaplan and Meier.15 All P values are two-sided. Treatment comparisons are by intention to treat.
Initially, the study sought 80% power to detect an improvement in median EFS from 2 years to 3 years, leading to a goal of randomizing 250 patients over a period of 5 years, with an additional 3 years of follow-up. In 1994, based on results of a phase II study of high-dose chemotherapy1 and the results of CALGB 8541,16 an adjuvant chemotherapy trial, we felt that the EFS would likely be better than originally hypothesized. With median EFS expected to be beyond 5 years, we amended the protocol to increase accrual.
The new sample-size calculations were based on computer simulations17 of the clinical trial, assuming various hypothesized underlying survival curves. We allowed for 10 interim analyses18 by the DSMC beginning after 20% of the anticipated events occurred. The interim analyses used the log-rank test. Two thousand simulations allowed estimation of statistical power to be accurate within 1% to 2% of the true power (with 95% probability).
The treatment difference we wanted to detect included crossing treatment-specific hazard functions. Women receiving HD-CPB would initially have a higher risk of treatment-related mortality. The group-specific EFS curves would come together by 1 year with 80% EFS, and would then diverge. Five-year EFS would be 70% for HD-CPB and 55% for ID-CPB. The simulations indicated that randomly assigning and treating 760 eligible patients (with at least 3 years' follow-up after randomization) would provide 90% power to detect the EFS difference of interest. Assuming that 5% of the patients entering the study would refuse further therapy or might be found ineligible, we sought to register 800 patients over an expected 6.67 years at 120 patients per year. All analyses are based on the study database that was locked on June 30, 2003.
RESULTS
Between January 1991 and May 1998, 885 patients enrolled onto the study, 10 patients were canceled, and 785 were randomized per protocol. Forty transplant centers participated; two centers enrolled 40% of the patients.
The reasons 100 patients were not randomly assigned are listed in Table 1. The primary reasons were denial of insurance coverage and relapse during the induction chemotherapy with CAF. Of the 24 women randomly assigned to HD-CPB who did not receive assigned treatment, nine refused the assigned treatment, two women experienced CAF-related mortality, two women had medical reasons (decreasing left ventricular ejection fraction and increasing liver function test), one woman's blood counts did not return to normal following CAF, one woman had disease progresssion, two women received nonprotocol therapy, and no reason was given for seven women. Of the 18 women randomly assigned to ID-CPB who did not receive assigned treatment, six withdrew, two had medical reasons (bipolar disorder and diverticulitis), one patient's blood counts did not return to normal following CAF, two received nonprotocol therapy, and seven women did not give a reason. Of the 394 patients randomly assigned to the high-dose chemotherapy arm (HD-CPB), 370 patients (94%), received the high-dose cyclophosphamide, cisplatin, and carmustine with stem-cell support; there were two patients randomly assigned to HD-CPB who were treated with ID-CPB (a major protocol violation), but were analyzed as HD-CPB recipients as part of the intent-to-treat analysis. Of the 391 patients randomly assigned to the intermediate-dose arm, 373 patients (95%) received ID-CPB.
Table 2 lists the patient characteristics for the 785 randomly assigned patients. The study was well balanced for age, ethnicity, and disease characteristics. The median age of the patients was 44 years (range, 22 to 66 years); 25% of the patients were at least 50 years of age. The median tumor size was 3.0 cm (range, 0.1 to 18.0 cm), and the median number of involved lymph nodes was 14 (range, 10 to 55), but 19% of patients had 20 or more lymph nodes involved at the time of surgery. Twenty-one percent of the patients had tumors larger than 5 cm in size (stage IIIA), and 69% of tumors were hormone-receptor positive.
EFS
The median follow-up time for EFS was 7.3 years and included 372 events (99% of the 375 events expected under the assumptions underlying the sample-size calculations). There have been 179 events (relapses or deaths) among the 394 randomized patients on the high-dose chemotherapy arm, and 193 events among the 391 patients on the ID-CPB chemotherapy arm. One hundred forty-eight and 146 patients, respectively, have died on the HD-CPB and ID-CPB chemotherapy arms.
For the 785 randomly assigned patients, 5-year EFS was 61% for HD-CPB (95% CI, 56% to 65%) and 58% for ID-CPB (95% CI, 53% to 63%). EFS was not significantly different between the two treatment groups (P = .24); nor did it differ (P = .21) after adjusting for the stratification variables used in randomization (hormone-receptor status, menopausal status, and stage; Fig 2A). The hazard ratio for EFS was 0.89 (95% CI, 0.72 to 1.09).
Age by itself was not a significant predictor of EFS (P = .23 for continuous age and P = .99 comparing women 50 years old and older to women younger than 50 years). An unplanned analysis comparing the effect of HD-CPB among women 50 years old and older with the effect among women who were 49 years and younger showed a significant interaction (P = .02). Figures 3A and 3B show the Kaplan-Meier EFS curves for patients younger than 50 years of age and 50 years or older, respectively. When this analysis is stratified with respect to the three factors used in the randomization (stage, hormone status, and menopausal status), the interaction between the treatment effect and a woman's age being 50 years or older is just statistically significant (P = .05). The interaction between treatment and menopausal status (premenopausal versus peri- or postmenopausal) is not significant, however (P = .49).
Relapses
There were 320 relapses among the 785 randomly assigned patients. One hundred thirty-seven (35%) of the 394 patients on the HD-CPB arm relapsed (95% CI, 30% to 40%), compared with 183 (47%) of the 391 women assigned to the ID-CPB arm (95% CI, 42% to 52%). Among women younger than 50 years randomly assigned to the HD-CPB arm, a smaller percentage relapsed during the follow-up period, compared with women younger than 50 years randomly assigned to the ID-CPB arm. However, the same fraction of women 50 years and older relapsed regardless of treatment assignment. Table 3 lists the treatment allocation, frequency of relapses, and treatment-related mortality for patients younger than 50 years compared with those who were 50 years or older. The cumulative hazard of relapse by treatment is shown in Figure 4. The figure shows that the women who received ID-CPB relapsed at a greater rate during the first 2 years (ie, steeper slope). After 2 years of follow-up, however, there seems to be less of a difference in the rate of relapse between the two groups.
Overall Survival
There was no significant difference in survival between the two treatment groups (P = .75). The probability of surviving 5 years (from randomization) was 71% (95% CI, 66% to 75%) for patients on each treatment arm. Figure 2B shows the Kaplan-Meier curves for survival. The hazard ratio for overall survival was 1.04 (95% CI, 0.82 to 1.30).
Toxicity
Treatment-related mortality. Thirty-five patients died of treatment-related causes; two patients died as a result of toxicity from the CAF induction chemotherapy and the remaining 33 treatment-related deaths occurred following HD-CPB and stem-cell transplantation (9% of 370 transplantations). The causes of HD-CPB mortality are listed in Table 4. Infections,19 pulmonary toxicity,20 and hemolytic-uremic syndrome21 were the primary causes of treatment-related mortality.
Fourteen of the 370 patients who received HD-CPB died within 100 days of the start of the high-dose therapy. The 100-day transplantation-related mortality is 3.8% (95% CI, 2.1% to 6.3%, exact binomial method). Thirty of the 33 therapy-related deaths occurred within the first year following high-dose chemotherapy; one of the three late deaths was due to AML secondary to treatment, and the other two deaths were caused by treatment-related pulmonary fibrosis, which was diagnosed based on clinical findings and confirmed at autopsy.
The overall TRM varied considerably, ranging from 5.6% to 28.6% among the centers participating in the study. Centers performing more than 10 transplantations tended to have a lower TRM (7%; 95% CI, 4% to 12%) compared with those performing fewer than 10 transplantations (16%; 95% CI, 8% to 27%).
Of the 370 patients who underwent transplant, 268 were younger than 50 years and 102 patients were 50 years or older. The TRM was 6.7% for the former group and 14.7% for the latter group. The TRM for the 111 women younger than 40 years was 5.4%.
Treatment-related toxicity. Leukopenia and thrombocytopenia were common among patients receiving either HD-CPB or ID-CPB, but were more severe and persistent among HD-CPB patients. Hepatic, pulmonary, and nervous system toxicity of grade 3 or greater occurred in more than 10% of patients receiving HD-CPB chemotherapy, whereas these toxicities were infrequent among patients receiving ID-CPB.
Hemolytic uremic syndrome (HUS) or thrombotic thrombocytopenic purpura (TTP) were severe complications of HD-CPB. HUS or TTP occurred in 14 patients receiving HD-CPB and were primary or associated causes of death in 10 patients. These syndromes have been previously reported to complicate high-dose therapy and proved resistant to usual treatment approaches.21
Thirty-six randomly assigned patients, 16 on the HD-CPB arm and 20 on the ID-CPB arm, developed secondary cancers. The most common sites were AML/MDS (HD-CPB, seven patients; ID-CPB, four patients) and breast (HD-CPB, five patients; ID-CPB, eight patients). Median onset of AML/MDS was approximately 45 months in both groups. Nine of the 13 breast cancers were considered new primary tumors.
DISCUSSION
There are four conclusions from this study: (1) that for all randomized patients EFS was not significantly different between patients randomly assigned to ID-CPB versus patients assigned to HD-CPB; (2) that transplant-related toxicity is severe and particularly adverse for HD-CPB among women 50 years and older; (3) that overall survival for patients on both treatment regimens was encouraging; and (4) that among women younger than 50 years, HD-CPB appears to be associated with fewer relapses.
The difference in EFS at 7 years of follow-up was approximately 4% (54% for HD-CPB and 50% for ID-CPB) in absolute terms or an 8% relative reduction in event rate for all patients. Whereas differences of this magnitude are not uncommon in trials of standard-dose treatments for breast cancer, the difficulty of administering the HD-CPB procedure and the magnitude of the toxicity led us to design a trial seeking a larger therapeutic advantage if HD-CPB was to be recommended. The trial was designed with a sample size sufficient to detect a 15% difference in 5-year EFS.
The EFS was dominated by the events of relapses in the intermediate-dose arm and the combination of therapy-related deaths and relapses in the high-dose arm. The rapid advance of stem-cell and hematopoietic–growth factor technologies22,23 prompted the extension of the study to older patients. Furthermore, the limitation to age 50 years was thought to be arbitrary and the protocol was therefore written to permit those "physiologically" under the age of 55 years to be enrolled. The consequence of these shifts in emphasis resulted in 25% of the patients entered being at least 50 years old.
Women 50 years and older appeared to be at higher risk of treatment-related mortality than younger women if randomly assigned to HD-CPB. However, the results suggest a benefit of HD-CPB for younger women (younger than 50 years). These results are similar to those obtained in other randomized trials in this population. The Dutch multicenter trial24 demonstrated a statistically significant improvement in disease-free survival among similar breast cancer patients younger than 40 years who received high-dose chemotherapy and among patients with HER2-neu negative tumors. In the current study, however, there is no significant interaction between treatment and age when split at age 40 years (P = .46).
Results suggesting benefits for high-dose therapy in younger women have also been reported from randomized trials conducted in Italy,25 Germany, and France.26,27 The Anglo-Celtic study did not show an advantage for high-dose therapy.28 A recently reported Eastern Cooperative Oncology Group trial did not report on the younger patient subgroup.29 The results of these studies collectively provide some insight into the role of high-dose therapy. No study demonstrates an overall or disease-free survival advantage for high-dose therapy, but in several studies there is either a strong trend or statistically significant advantage for high-dose therapy among younger women. It is notable that the pilot trial on which the design of this trial was based included predominantly younger women.1 Biologic differences in the effectiveness of dose-intense therapy between age groups and other parameters (HER2 status24) have been described previously and may provide an important lead for subsequent studies.
The ID-CPB treatment used in this study produced better outcomes than expected, compared with other commonly used regimens, and particularly when compared with the CALGB 8541 study of CAF, results of which were used to estimate requisite sample sizes for this trial.15,30-31 Overall survival for all patients in this study (71% overall survival at 5 years from randomization) is encouraging and surprising. The reason for the favorable results is undoubtedly multifactorial and may include the use of additional chemotherapeutic agents not usually part of standard adjuvant therapy, stage migration in breast cancer, delay until completion of three cycles of CAF before randomly assigning patients, or extensive pretreatment staging resulting in better patient selection, as well as the use of radiation and hormonal therapies. It is noteworthy that the results with HD-CPB in this study are essentially identical to the results from the phase II studies, which prompted this prospective trial. However, caution is necessary when making comparisons to other adjuvant studies, since randomization in our study occurred after three cycles of adjuvant CAF therapy. We estimate that the probability of a relapse or death by median time to randomization is approximately 3%. Still, to our knowledge, no conventional treatment program with similar follow-up and size has produced superior survival results in this patient population.
Interpreting the results of this trial compared with standard treatment regimens is complicated by the use of ID-CPB, which had previously been evaluated in only a preliminary fashion. We chose to compare HD-CPB to ID-CPB to focus on dose and to avoid introducing confounding interpretations because of different drugs in the arms of the study. The interpretation is further confounded by the observation that the HD-CBP patients were less likely to complete the prescribed locoregional radiation therapy than were the ID-CBP patients.32 Since such locoregional therapy appears to have an impact on disease-free, and possibly overall survival,33,34 the HD-CBP patients may have been negatively affected by the reduced use of radiation therapy. The absence of a true control arm in this study does not permit us to draw conclusions regarding whether either HD-CPB or ID-CPB produce better outcomes in this patient population than might be achieved with CAF alone. As mentioned previously, however, the overall results in this trial appear superior to those previously reported with conventional adjuvant chemotherapy regimens.
The trend toward fewer relapses among younger women on HD-CPB does not appear to be related to the induction of a chemical menopause. Of the randomized patients who were premenopausal at study entry, 42% of those taking HD-CPB became amenorrheic compared with 43% of those taking ID-CPB. The difference in relapse frequency between ID-CPB and HD-CPB was similar for women with hormone-receptor–positive and hormone-receptor–negative tumors.
Women with breast cancer have more treatment options now than ever before with the introduction of aromatase inhibitors, trastuzumab, and new cytotoxic agents such as capecitabine and gemcitabine. Nevertheless, our study suggests that clinical trials should continue to examine the role of high-dose chemotherapy with stem-cell support in high-risk women younger than 50 years to determine unequivocally whether this approach provides clinical benefit.
Appendix
The following institutions participated in the study: Brook Army Medical Center/Wilford Hall Medical Center, San Antonio, TX, Mitchell A. Garrison, MD, supported by CA76447; CALGB Statistical Office, Durham, NC, Stephen George, PhD, supported by CA33601; Cancer Care Manitoba–Winnipeg, Manitoba, Tsiporah Shore, MD; Central IL CCOP, Springfield, IL, James L. Wade, MD, supported by CA45807; Cleveland Clinic, Cleveland, OH, George Thomas Budd, MD, supported by CA04919; Columbia River CCOP, Portland, OR, Keith S. Lanier, MD, supported by CA45377; Dana-Farber Cancer Institute, Boston, MA, George P. Canellos, MD, supported by CA32291; Dartmouth Medical School–Norris Cotton Cancer Center, Lebanon, NH, Marc Ernstoff, MD, supported by CA04326; Duke University Medical Center, Durham, NC, Jeffrey Crawford, MD, supported by CA47577; Grand Rapids CCOP, Grand Rapids, MI, Kathleen J. Yost, MD, supported by CA35178; Henry Ford Hospital, Detroit, MI, Robert A. Chapman, MD, supported by CA58416; Massachusetts General Hospital, Boston, MA, Michael Grossbard, MD, supported by CA12449; McGill Cancer Center, Montreal, Quebec, Canada, Brian Leyland-Jones, MD, supported by CA31809; Medical University of South Carolina, Charleston, SC, Mark Green, MD, supported by CA03927; Memorial Sloan-Kettering Cancer Center, New York, NY, George Bosl, MD, supported by CA77651; Mount Sinai School of Medicine, New York, NY, Lewis Silverman, MD, supported by CA04457; North Shore–Long Island Jewish Medical Ctr, Manhasset, NY, Daniel R. Budman, MD, supported by CA35279; Oregon Health Science University, Portland, OR, Craig R. Nichols, MD, supported by CA46113; Rhode Island Hospital, Providence, RI, William Sikov, MD, supported by CA08025; Roswell Park Cancer Institute, Buffalo, NY, Ellis Levine, MD, supported by CA02599; St Louis University, St Louis, MO, Paul J. Petruska, MD, supported by CA76462; SUNY Upstate Medical University, Syracuse, NY, Stephen L. Graziano, MD, supported by CA21060; Temple University, Temple, TX, supported by CA58415; The Toronto Hospital–General Division Toronto, Ontario, Canada, Michael Crump, MD; Toronto SunnyBrook Regional Cancer Centre–Toronto, Ontario, Canada, Kathleen Pritchard, MD; University of Arkansas, Little Rock, AR, Laura F. Hutchins, MD, supported by CA37981; University of California at San Diego, San Diego, CA, Stephen Seagren, MD, supported by CA11789; University of California San Francisco, San Francisco, CA, Alan Venook, MD, supported by CA60138; University of Chicago Medical Center, Chicago, IL, Gini Fleming, MD, supported by CA41287; University of Colorado, Denver, CO, Scott I. Bearman, MD, supported by CA42777; University of Hawaii CCOP, Honolulu, HI, Brian F. Issell, MD, supported by CA63844; University of Iowa, Iowa City, IA, Gerald Clamon, MD, supported by CA47642; University of Kentucky, Lexington, KY, Michael A. Doukas, MD, supported by CA46136; University of Maryland Cancer Center, Baltimore, MD, David Van Echo, MD, supported by CA31983; University of Minnesota, Minneapolis, MN, Bruce A. Peterson, MD, supported by CA16450; University of Missouri/Ellis Fischel Cancer Center, Columbia, MO, Michael C. Perry, MD, supported by CA12046; University of North Carolina at Chapel Hill, Chapel Hill, NC, Thomas C. Shea, MD, supported by CA47559; University of Oklahoma, Oklahoma City, OK, Howard Ozer, MD, PhD, supported by CA58686; Upstate Carolina, Spartanburg, SC, James D. Bearden, MD, supported by CA35119; Vancouver Cancer Centre–Vancouver, British Columbia, Canada, Joseph Ragaz, MD, and Michael Barnett, MD; Virginia Mason CCOP, Seattle, WA, Andrew D. Jacobs, MD, supported by CA35192; Wake Forest University School of Medicine, Winston-Salem, NC, David D. Hurd, MD, supported by CA03927; Walter Reed Army Medical Center, Washington, DC, Joseph J. Drabeck, MD, supported by CA26806; Washington University School of Medicine, St Louis, MO, Nancy Bartlett, MD, supported by CA77440; Wayne State University, Detroit, MI, Lawrence E. Flaherty, MD, supported by CA14028; Weill Medical College of Cornell University, New York, NY, Scott Wadler, MD, supported by CA07968.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We would like to express our appreciation to Dr Raymond Weiss and to the members of the CALGB Data Audit Committee for their efforts to help ensure the quality and validity of these data. Above all, we express our gratitude to the patients and their families who participated in this study.
NOTES
Supported in part by grants from the National Cancer Institute (CA31946) to the Cancer and Leukemia Group B (Richard L. Schilsky, Chairman), and by grants CA33601, CA14028, CA33601, CA47577, CA42777, CA77202, CA32291, CA35279, CA77651, CA60138, CA41287, and CA03927.
The manuscript contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Peters WP, Ross M, Vredenburgh JJ, et al: High-dose chemotherapy and autologous bone marrow support as consolidation after standard-dose adjuvant therapy for high-risk primary breast cancer. J Clin Oncol 11:1132-1143, 1993
Peters WP, Dansey RD, Klein JL, et al: High-dose chemotherapy and peripheral blood progenitor cell transplantation in the treatment of breast cancer. Oncologist 5:1-13, 2000
Antman K, Eder JP, Elias A, et al: High-dose combination alkylating agent preparative regimen with autologous bone marrow support: The Dana-Farber Cancer Institute/Beth Israel Hospital experience. Cancer Treat Rep 71:119-125, 1987
Baynes RD, Dansey RD, Klein JL, et al: High-dose chemotherapy and autologous stem cell transplantation for breast cancer. Cancer Invest 18:440-455, 2000
Rodenhuis S, Richel DJ, van der Wall E, et al: Randomized trial of high-dose chemotherapy and haemopoietic progenitor-cell support in operable breast cancer with extensive axillary lymph-node involvement. Lancet 352:515-521, 1998
Peters WP: High-dose chemotherapy with autologous bone marrow transplantation for the treatment of breast cancer: Yes. Important Adv Oncol 215-230, 1995
Antman K, Ayash L, Elias A, et al: High-dose cyclophosphamide, thiotepa, and carboplatin with autologous marrow support in women with measurable advanced breast cancer responding to standard-dose therapy: Analysis by age. J Natl Cancer Inst Monogr 16:91-94, 1994
Antman K, Eder JP, and Frei E 3rd: High-dose chemotherapy with bone marrow support for solid tumors. Important Adv Oncol 221-235, 1987
Durrleman S, Simon R: When to randomize J Clin Oncol 9:116-122, 1991
Shpall EJ: The utilization of cytokines in stem cell mobilization strategies. Bone Marrow Transplant 23:S13-S19, 1999 (suppl 2)
Petros WP, Broadwater G, Berry D, et al: Association of high-dose cyclophosphamide, cisplatin, and carmustine pharmacokinetics with survival, toxicity, and dosing weight in patients with primary breast cancer. Clin Cancer Res 8:698-705, 2002
Panella TJ, Peters W, White JG, et al: Platelets acquire a secretion defect after high-dose chemotherapy. Cancer 65:1711-1716, 1990
Marks LB, Rosner GL, Prosnitz LR, et al: The impact of conventional plus high dose chemotherapy with autologous bone marrow transplantation on hematologic toxicity during subsequent local-regional radiotherapy for breast cancer. Cancer 74:2964-2971, 1994
Winer EP, Lindley C, Hardee M, et al: Quality of life in patients surviving at least 12 months following high dose chemotherapy with autologous bone marrow support. Psychooncology 8:167-176, 1999
Klein JP, Moeschberger ML: Survival Analysis: Techniques for Censored and Truncated Data. New York, NY, Springer-Verlag Inc, 1997
Wood WC, Budman DR, Korzun AH, et al: Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 330:1253-1259, 1994
Halpern J, Brown BW Jr: A computer program for designing clinical trials with arbitrary survival curves and group sequential testing. Control Clin Trials 14:109-122, 1993
O'Brien PC, Fleming TR: A multiple testing procedure for clinical trials. Biometrics 35:549-556, 1979
Williams SF, Mick R, Desser R, et al: High-dose consolidation therapy with autologous stem cell rescue in stage IV breast cancer. J Clin Oncol 7:1824-1830, 1989
Patz EF Jr, Peters WP, Goodman PC: Pulmonary drug toxicity following high-dose chemotherapy with autologous bone marrow transplantation: CT findings in 20 cases. J Thorac Imaging 9:129-134, 1994
Fisher DC, Sherrill GB, Hussein A, et al: Thrombotic microangiopathy as a complication of high-dose chemotherapy for breast cancer. Bone Marrow Transplant 18:193-198, 1996
Rosti G, Ferrante P, Ledermann J, et al: High-dose chemotherapy for solid tumors: Results of the EBMT. Crit Rev Oncol Hematol 41:129-140, 2002
Simnett SJ, Stewart LA, Sweetenham J, et al: Autologous stem cell transplantation for malignancy: A systematic review of the literature. Clin Lab Haematol 22:61-72, 2000
Rodenhuis S, Bontenbal M, Beex LV, et al: High-dose chemotherapy with hematopoietic stem-cell rescue for high-risk breast cancer. N Engl J Med 349:7-16, 2003
Gianni AM: Standard vs High-Dose Chemotherapy in the Adjuvant Treatment of Breast Cancer with > 3 LN+: 5 year results. Proc Am Soc Clin Oncol 20:21a, 2001 (abstr 80)
Roche H, Viens P, Biron P, et al: PEGASE Group High-dose chemotherapy for breast cancer: The French PEGASE experience. Cancer Control 10:42-47, 2003
Roche H, Pouillart P, Meyer N, et al: Adjuvant High Dose Chemotherapy Improves Early Outcome for High Risk (N > 7) Breast Cancer Patients: The Pegase 01 Trial. Proc Am Soc Clin Oncol 20:26a, 2001 (abstr 102)
Crown JP, Lind M, Gould A, et al: Failure of adjuvant peripheral blood progenitor-supported high-dose chemotherapy to improve the prognosis of patients with high-risk breast cancer in a randomized trial with low treatment-related mortality: The Anglo-Celtic I Study. Ann Oncol 13:2, 2002 (suppl 5, abstr 20)
Tallman MS, Gray R, Robert NJ, et al: Conventional adjuvant chemotherapy with or without high-dose chemotherapy and autologous stem-cell transplantation in high-risk breast cancer. N Engl J Med 349:17-26, 2003
Budman DR, Berry DA, Cirrincione CT, et al: Dose and dose intensity as determinants of outcome in the adjuvant treatment of breast cancer: The Cancer and Leukemia Group B. J Natl Cancer Inst 90:1205-1211, 1998
Roach M 3rd, Cirrincione C, Budman D, et al: Race and survival from breast cancer: Based on Cancer and Leukemia Group B trial 8541. Cancer J Sci Am 3:107-112, 1997
Marks LB, Cirrincione C, Peterson B, et al: The impact of local/regional radiotherapy (RT), and its timing, on survival following lumpectomy/mastectomy and systemic chemotherapy in patients with 10 positive axillary nodes: Analysis of CALGB 9082. Int J Radiat Oncol Biol Phys 60:S136, 2004 (suppl)
Ragaz J, Jackson SM, Le N, et al: Adjuvant radiotherapy and chemotherapy in node positive premenopausal women with breast cancer. N Engl J Med 337:956-962, 1997
Overgaard M, Hansen PS, Overgaard J, et al: Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. N Engl J Med 337:949-955, 1997(William P. Peters, Gary L)