Bone Marrow Transplantation (2003) 31, 655–661 & 2003 Nature Publishing Group All rights reserved 0268-3369/03 $25.00 www.nature.com/bmt Breast Cancer Adjuvant treatment of high-risk stage II breast cancer with doxorubicin followed by high-dose chemotherapy and autologous stem-cell transplantation: a single-institution experience with 132 consecutive patients

SM Stemmer1, I Hardan1, H Raz1, AK Adamou2, M Inbar3, M Gottfried4, Y Merrick5, Y Cohen6, A Sulkes7, N Ben-Baruch8, RP Pfeffer1, HJ Brenner1 and S Rizel1

1Bone Marrow Transplant Service, Department of Oncology and Radiotherapy, Chaim , Tel Hashomer, ; 2Bank of Cyprus Oncology Center, Nicosia, Cyprus; 3Department of Oncology, Sourasky Medical Center, , Israel; 4Oncology Unit, Meir , Sapir Medical Center, , Israel; 5Oncology Unit, , , Israel; 6Department of Oncology, , Beer-Sheva, Israel; 7Institute of Oncology, , Petach Tikva, Israel;and 8Oncology Institute, , , and Sackler Faculty of , Tel-Aviv University, Israel

Summary: tality markedly influenced the outcome. For this high-risk patient population, further testing of intensive chemother- Several studies have shown conflicting results with the use apy regimens with a lower toxicity profile is warranted. of intensive consolidation chemotherapy for breast cancer. Bone Marrow Transplantation (2003) 31, 655–661. The aim of the present study was to investigate the doi:10.1038/sj.bmt.1703856 efficacy, feasibility and toxicity of high-dose chemother- Keywords: breast cancer; high-dose chemotherapy; apy with stem cell support in patients with high-risk stage toxicity; efficacy II breast cancer. From February 1994 to November 1998, 132 consecutive patients with multinode positive breast cancer were entered to the study. In total, 86 patients had The use of adjuvant chemotherapy for intermediate and X10 positive axillary lymph nodes, and 46 had 4–9 high-risk breast cancer has led to a modest but consistent positive axillary lymph nodes with at least two additional improvement in disease-free and overall survival. The predetermined risk factors at diagnosis. All patients were number of positive axillary lymphnodes is a major factor offered adjuvant chemotherapy (doxorubicin, 75 mg/ in predicting the prognosis of these patients. 2 m  4) followed by high-dose chemotherapy (cyclopho- Intensive consolidation chemotherapy is currently being 2 2 sphamide 6000 mg/m , carboplatin 800 mg/m and thio- investigated as a therapeutic option for aggressive, locally 2 tepa 500 mg/m ) and autologous stem cell support with advanced breast cancer. To date results are controversial. growth factor. In all, 131 patients also received local Two prospective, randomized studies presented at the radiation therapy and tamoxifen based on receptor status. ASCO 2001 meeting showed no survival benefit.1,2 The After a median follow-up of 51 months (range 27–87), the study by Peters et al1 was impaired by the high incidence of disease-free and overall survival rates were 72 and 81%, treatment-related mortality in the high-dose arm, empha- respectively. There was no difference in the outcome for sizing the need for less toxic high-dose chemotherapy high-risk patients with 4 or o than 10 positive axillary regimens in the adjuvant setting and for longer follow-up lymph nodes. On Cox regression analysis only progeste- and more mature data. Gianni et al2 used an unusual rone receptor status was predictive of disease-free, but not approach: very-short-term (o90 days) therapy consoli- overall survival. There were no treatment-related deaths; dated by high-dose melphalan. An earlier report however, grades III–IV toxicity was relatively low. This combined by Rodenhuis et al3 (ASCO 2000), noted promising results approach of doxorubicin followed by high-dose che- for high-dose chemotherapy in a similar setup, with a very motherapy and stem-cell support, followed by locoregional low treatment-related mortality. Unfortunately, the differ- radiotherapy, was safe and seems to be effective in ent regimens used in the various studies, the small sample patients with multinode positive stage II breast cancer. In sizes, and the different staging techniques make compar- previous trials of adjuvant high-dose therapy in this isons difficult. The aim of the present prospective study was patient population, treatment-related morbidity and mor- to investigate the efficacy, feasibility and toxicity of doxorubicin followed by intensive chemotherapy with stem-cell support in patients with high-risk stage II breast Correspondence: Dr SM Stemmer, Institute of Oncology, Rabin cancer withfour or more positive axillary lymphnodes. Medical Center, Beilinson Campus, PetachTikva 49100, Israel In view of the controversy surrounding this approach, Received 18 March2002; accepted 13 October 2002 this large prospective trial from a single center using a Stem cell transplant for hight risk stage II breast cancer SM Stemmer et al 656 regimen witha low toxic profile may significantly pectomy (60%) or mastectomy (40%) and axillary lymph contribute to the information regarding this high-risk node dissection. group of patients. At 3–5 weeks after the last cycle of adjuvant chemo- therapy, autologous peripheral stem cells were collected Patients and methods (Table 1) after induction withgranulocyte colony-stimulating factor (G-CSF) (Neupogen; Amgen Biologicals, Thousand Oaks, In all, 132 consecutive patients withhistologicallyproven CA, USA), 5 mg/kg/day subcutaneously. The cells were American Joint Committee on Cancer (AJCC)4 stage II collected on days +5 and +6, and also +7, if necessary. breast carcinoma with high-risk factors who were treated at The amount of cells required for consolidation with high- the Chaim Sheba Medical Center between January 1994 dose chemotherapy was 5  108 mononuclear cells/kg and and October 1998 were entered into a prospective study of 2.5  106 CD34-positive cells/kg. An aliquot of reserve stem an intensive treatment protocol. The protocol was com- cell was preserved and stored. Collection was adequate in prised of induction chemotherapy and high-dose che- all patients. All patients received high-dose chemotherapy motherapy with autologous stem-cell support, followed withtheStamp V protocol as previously described. 6 In by locoregional radiotherapy; tamoxifen was administered brief, cyclophosphamide (1500 mg/m2 continuous infusion on the basis of receptor status. The end points of the study per day for a total of 96 h), thiotepa (125 mg/ m2 were disease-free and overall survival in addition to short- continuous infusion per day for a total of 96 h) and and long-term transplant-related toxicity. The feasibility carboplatin (200 mg/m2 continuous infusion per day for a and toxicity of radiotherapy was previously described in total of 96 h) were given on days À7toÀ3. Stem cells were Cancer.5 High-risk stage II disease was defined as either reinfused on day 0 and G-CSF was administered the same X10 positive axillary lymphnodes ( n ¼ 86/132, 60%) or 4– day (n ¼ 119 patients) and continued until engraftment 9 positive axillary lymphnodes ( n ¼ 46/132, 40%) withtwo (absolute neutrophil count 41000/ml) was achieved. In 13 or more of the following poor prognostic features: age 35 p patients, G-CSF was started on days +2, +4, +5 and +6 years, histologic grade 3, less than 10 axillary lymph nodes after stem-cell transplantation, as part of an ongoing sampled, tumor involvement of the axillary fat, or large protocol. Prophylactic oral quinolone was given from day tumor-involved axillary lymphnodes ( 42 cm). Other À2. Antibacterial and antifungal treatment was added for eligibility criteria were age 65 years and normal organ p neutropenic fever as per routine protocols. Antibiotic function, namely, normal radionuclide cardiac scan (multi- coverage was stopped on engraftment or after resolution gated angiogram (at rest and exercise) 450%), carbon of a documented bacterial/fungal infection. Patients under- monoxide diffusing capacity of the lungs (DL) 460%, went transfusion withirradiated red blood cells and creatinine clearance 460 ml/min and normal liver function platelets to maintain hemoglobin levels of greater than tests. All patients had a negative metastatic work-up that 8 g/dl and platelet counts of greater than 5000/ml. Con- included brain, chest, abdominal and pelvic computed tinuous infusion of morphine was given for severe tomography scans; a normal radionuclide bone scan and mucositis. Oral cultures for herpes were performed only normal findings on bilateral bone marrow biopsies (hema- for patients withmucositis. None of thepatients received toxylin and eosin stain). Written informed consent was total parenteral nutrition. After discharge, no prophylactic obtained from all patients. antibiotic therapy was given. The protocol specified adjuvant chemotherapy with a Radiotherapy was planned after recovery from high-dose doxorubicin-based regimen. Single-agent doxorubicin chemotherapy and within 70 days of stem cell transplant. (75 mg/m2 every 3 weeks for a total of four cycles) was The radiotherapy protocol used in our center, delivered to the preferred regimen, used in 111 patients (84%). The 72% of the patients, has been previously described.5 In brief, remaining 21 patients (16%) received cyclophosphamide 5040 cGy (180 cGy fractions, 5 per week) were delivered (500–600 mg/m2), doxorubicin (50–60 mg/m2) and 5-fluoro- witha 6 MeV linear accelerator to thebreast/chest wall uracil (500–600 mg/m2) every 21 days (four to six cycles). through two opposed tangential fields. The supraclavicular Surgery, performed by different surgeons, included lum- and axillary areas were treated through an anterior field withthesame doses and fractions. An additional boost of Table 1 Patients characteristics 1400 cGy (9–12 MeV electrons, 200 cGy fractions) was given X10+ALN 4–9+ALN P-value by a direct electron field to the area of the primary disease to No. of patients 86 46 patients after lumpectomy (based on the surgical scar and Median age (years) 46 (25–63) 44 (33–59) NS information obtained from the patients). The remaining Median no. of positive nodes (range) 14 (10–47) 8 (4–9) 28% of the patients were irradiated at the referring centers. Only one patient did not receive radiotherapy because of Receptor status transplant-related lung injury. Not known 4 (5%) 0 (0%) NS ER-positive (%) 48 (56%) 20 (40%) NS Patients withhormonereceptor-positive tumors were PR-positive (%) 46 (53%) 20 (40%) NS treated withtamoxifen (20 mg/day) for 5 years after completing radiotherapy. Breast surgery Mastectomy 34 (40%) 18 (40%) NS Lumpectomy 51 (60%) 28 (60%) NS Statistical analysis

ALN = axillary lymphnodes; ER = estrogen receptor; PR = progesterone Means and standard deviations were computed. Student’s receptor. t-test was used for comparisons of continuous variables

Bone Marrow Transplantation Stem cell transplant for hight risk stage II breast cancer SM Stemmer et al 657 between the two groups (49 vs 4–9 positive axillary lymph heparin. The other six patients had infectious complica- nodes). tions, mostly bacterial pneumonia. Disease-free and overall survival curves for all patients Documented infections were noted in 15% of the were calculated using the Kaplan–Meier product limit transplant patients. Of these, 40% had Gram-positive method, applied from the day of transplant, as in bacteremia related to line infection and the other 60% had transplant protocols. When multivariate Cox regression Gram-negative bacterial infections. In 32% of the patients, analysis was performed, the regression coefficients were oral culture was positive for herpes. estimated by maximum-likelihood criteria, and their difference was determined by Wald’s test. All tests were Supportive care performed using SPSS statistical analysis software version 10. The supportive care provided after transplantation is shown in Table 3. All patients had grade IV bone marrow Results toxicity. The median duration of neutropenia was 9 days (range 5–15). Almost all patients (98%) required red blood Treatment toxicity cell transfusions (median 3 units, range 0–10), and 95% required platelet transfusions (median 12 units (two adult There were no treatment-related deaths. Toxicity was random donor pack), range 0–54). Single-donor platelets similar in patients with X10 positive axillary lymphnodes were given only for refractory patients. In addition, 95% of and patients with4–9 positive axillary lymphnodes (Table the patients received intravenous antibacterial antibiotics 2). The most common side effect of the STAMP V protocol for febrile neutropenia, for a median of 7 days (range 0–20). was mucositis grade III, which occurred in 59 patients Median hospitalization time was 20 days (range 16–42). (45%) and was treated witha morphinedrip for a median There were no significant differences between the two high- of 8 days (1–14 days). None of the patients received total risk groups in any of these parameters. parenteral nutrition. Acute (grade 4II) cardiac toxicity was noted in three patients and acute renal and liver Radiotherapy toxicities in one patient each; all resolved completely within Of the 132 patients in the present sample, 96 received 3 months of transplantation. Severe lung toxicity including radiotherapy at our institution. Overall, early post-trans- infection was seen in eight patients (6%) after transplanta- plant radiotherapy was very effective for local control, well tion. One patient had diffuse alveolar hemorrhage requiring tolerated and safe, as previously described.6 One patient respiratory intensive care support for 10 days. Her who did not receive radiotherapy had diffuse alveolar pulmonary function (pulmonary function tests including hemorrhage peritransplantation and required prolonged diffusion capacity) fully recovered 1 year after the ventilation. At 4 years after the transplant, she had a single transplant. Another patient suffered from pulmonary bony lesion that was treated with radiotherapy (reported as emboli during transplantation and was treated with relapse).

Table 2 Acute transplant-related toxicity, grade >II

4–9+ALN (N=46) 10+ALN (N=86) % of total (N=132) P-value Mucositis 22 (48%) 37 (43%) 59 (45%) NS Lung 1 PE, 1 DAH, 8 (6%) NS 1 Pneumonia 5 Pneumonia Cardiac 1 2 3 (2%) NS Liver 0 1 1 (1%) NS Genitourinary 1 0 1 (1%) NS Bacteremia/sepsis 9 (19%) 10 (12%) 19 (15%) NS Oral herpes 14 (30%) 28 (32%) 42 (32%) NS

ALN = axillary lymphnodes; PE = pulmonary embolism; DAH = diffuse alveolar hemorrhage.

Table 3 Supportive care of transplant patients

Parameter median (range) 4–9+ALN (N=46) 10+ALN (N=86) P-value Days of neutropenia 9 (6–15) 9 (5–12) NS No. of RBC transfusions (units) 3 (1–6) 2 (0–10) NS No. of platelet transfusions (units) 12 (0–54) 10 (0–48) NS Days of IV antibiotics 7 (3–16) 7 (0–20) NS Engraftment day 9 (7–15) 9 (8–14) NS Days of G-CSF 11 (7–18) 11 (6–18) NS Days of hospitalization 19 (17–28) 20 (16–42) NS Days withfever 6 (3–13) 5 (0–20) NS

ALN = axillary lymphnodes; RBC = red blood cells; G-CSF = granulocyte colony-stimulating factor.

Bone Marrow Transplantation Stem cell transplant for hight risk stage II breast cancer SM Stemmer et al 658 Less than 5% of the patients had late clinical pneumo- (n ¼ 2), peritoneum o1% (n ¼ 1), endometrium o1% nitis. All were treated witha 6 to 10-week course of steroids (n ¼ 1), and cerebrospinal fluid o1% (n ¼ 1). In three withcomplete functional recovery. In one patient, there patients, no data on site of first relapse were available. This was one relapse of pneumonitis, which responded com- pattern appears to be similar to that observed in patients pletely to a second course of steroids. This patient treated with conventional chemotherapy. continues to do well more than 5 years after transplanta- tion. Discussion Late complications After a median follow-up of 51 months (range 27–86 The results of the present study indicate that for patients months), no events of late cardiac morbidity, secondary withmultinode positive stage II breast carcinoma, adjuvant leukemia, myelodysplastic syndrome or other malignancy chemotherapy followed by high-dose chemotherapy and were noted in any of the 132 treated patients. autologous stem cell transplantation is feasible and safe. Further, the patient outcome is promising. After a median Outcome follow-up of 51 months from the end of therapy (range 27– 86), the actuarial disease-free survival is 72% and the All 132 patients were eligible for the final analysis. After a overall survival 81%. median follow-up of 51 months (range 27–87), the disease- As compared with conventional-dose chemotherapy, free and overall survival rates were 72 and 81%, high-dose chemotherapy is believed to increase tumor cell respectively. The disease-free and overall survival rates kill. In the adjuvant setting, it may improve the chance of were similar for the patients with 4–9 positive axillary eradicating micrometastases in patients at high risk of lymphnodes and thepatients with10 or more positive relapse, thus increasing the impact of adjuvant therapy on axillary lymphnodes ( P ¼ 0.19). Cox regression analysis survival. was performed to evaluate the predictive power of several Several phase II trials7,8 have reported encouraging prognostic features on the outcome. The number of results with intensive chemotherapy and autologous stem- positive nodes, status of estrogen receptor, type of surgery, cell support, showing 60–75% progression-free and overall young age, regimen of induction chemotherapy and survival rates. Other studies, however, presented in abstract number of chemotherapy cycles pre high-dose therapy form at the 2001 meeting of the American Society of were not predictive for disease-free and overall survival. Clinical Oncology, failed to confirm these findings.1,2 In a Only progesterone receptor status was predictive of disease- large CALGB trial, Peters et al1 compared the survival of free, but not overall survival (Figures 1 and 2). patients treated with adjuvant CAF followed by a high- The sites of first relapse after high-dose chemotherapy dose STAMP I regimen (cyclophosphamide, cisplatin and were liver 9% (n ¼ 12), bone 8% (n ¼ 10), lung 4% (n ¼ 5), carmustine) and PSC support, and patients given the same brain 3% (n ¼ 4), chest wall 2% (n ¼ 3), lymphnodes 1% induction chemotherapy followed by a low-dose similar

1.0

0.9

0.8 Group 1

0.7

0.6 Group 2

0.5 Cumulative Proportion 0.4

0.3

0.2 0 102030405060708090100 Months from Transplant

Figure 1 Disease-free survival of the two (4–9 positive axillary lymph nodes; N ¼ 46 Group 1, and X10 positive axillary lymphnodes; N ¼ 86 Group 2) transplanted groups. There was no statistical difference in DFS in the two groups of patients (P ¼ 0.189).

Bone Marrow Transplantation Stem cell transplant for hight risk stage II breast cancer SM Stemmer et al 659 1.0

Group 1 0.9

0.8

0.7 Group 2 0.6

0.5 Cumulative Proportion

0.4

0.3

0.2 0 102030405060708090100 Months from Transplant

Figure 2 Overall survival of the two (4–9 positive axillary lymph nodes; N ¼ 46 Group 1, and X 10 positive axillary lymphnodes; N ¼ 86 Group 2) transplanted groups. There was no statistical difference in overall survival in the two groups of patients (P ¼ 0.198).

regimen. At a median follow-up of 5.1 years, the intent-to- An initial analysis by a Dutchgroup 3 of 284 patients treat event-free survival was 61% for the patients given the randomized to receive either high-dose chemotherapy with high-dose regimen and 60% for the low-dose BCNU autologous stem cell support or conventional chemother- regimen. Although the transplant group had fewer relapses apy showed a statistical benefit in the high-dose arm with (28.9 vs 39.1%), there were 32 transplant-related deaths, as regard to disease-free survival, and borderline significant opposed to none in the control group. At 5 years, overall improvement in overall survival. Similar results were survival in the two groups was 70 and 72%, respectively. reported by a French10 group. Bothstudies reported very The authors concluded that for this patient population the low rates of treatment-related mortality. overall outcome exceeded that of studies with standard- The length of follow-up in our study was similar to that dose therapy alone, but there were no significant differences reported by Peters et al,1 Gianni2 and Rodenhuis et al,3 and in either event-free or overall survival between the high- like the latter two studies, we had no treatment-related dose regimen (transplant) and the low-dose regimen mortality. The duration of therapy as well as the transplant (control). Nevertheless, the high transplant-related mortal- protocol used here were similar to those used by Rodenhuis ity of 7.4% (6% in an experienced center and 11% in et al3 and we report of a similar disease-free survival of inexperienced centers) renders these results too premature 70% and overall survival of more than 80%. for interpretation, even after 5 years of follow-up. Studies with high-dose chemotherapy that carry a high Gianni and Bonadonna2 compared a non-crossresistant transplant-related mortality should be interpreted carefully chemotherapy regimen delivered over a very short period of when follow-up is relatively short. The longer the follow- time, withtheclassical doxorubicin regimen followed by up, the less the impact of treatment-related mortality on CMF. After a median follow-up of 52 months, the 5-year overall survival. Thus, initial reports from the Parma study progression-free survival rates were 65% in the transplant of high-dose chemotherapy for non-Hodgkin’s lymphoma group and 62% in the conventionally treated group. The showed similar survival for the transplant and conventional overall survival rates were 76 and 77%, respectively. arms,11 and the benefit of transplantation became clear However, when the 112 patients younger than 36 years only after a longer follow-up. and the 147 patients with 4–9 positive lymph nodes were The absence of mortality in our series reflects stringent considered separately, those who received the high-dose patient selection, patient care by a single experienced team regimen had a trend towards an improved progression-free in one hospital, selection of a noncarmustine-based survival. It is possible that the very short-term therapy, transplant protocol, and the use of peripheral blood stem- known to be less effective than longer therapy,9 delivered in cell support. This is in line with other studies showing a the high-dose arm (o90 days) was inadequate for the very- marked reduction in transplant-related mortality when high-risk population (410 positive axillary lymphnodes), using peripheral stem-cell support.12 but sufficient for the intermediate-high-risk population (4–9 With regard to the specific role of the adjuvant regimen positive axillary lymphnodes). in these trials Rodenhuis et al,3 Hortobagyi,13,14 Gianni and

Bone Marrow Transplantation Stem cell transplant for hight risk stage II breast cancer SM Stemmer et al 660 Bonadonna,2 and Peters et al,1 used a cyclophosphamide- conducted carefully because of differences in patient based protocol for induction, which theoretically could selection, supportive care, and other. have induced resistance to a high-dose cyclophosphamide- In summary, patients with intermediate and high-risk containing transplant protocol. We elected to use single- stage II breast cancer consolidated with high-dose chemo- agent doxorubicin as adjuvant chemotherapy, followed by therapy and autologous stem cell transplantation had an a high-dose alkylating protocol as consolidation. We failed actuarial disease-free survival of 72% and an overall to show that patients treated with doxorubicin alone survival of 81% after a median follow-up of 51 months did better than patients who received CAF. Nevertheless, from transplantation. On the basis of these findings, we this was not the aim of this study, and the two groups believe that intensive chemotherapy with stem cell support (84% of patients who received single-agent doxorubicin vs may play a role in the comprehensive treatment of this 16% who received CAF) were too small for statistical patient population. Further studies are needed to reach a analysis. definite conclusion. No difference was found in disease-free and overall survival between the two high-risk groups: patients with X10 positive axillary lymphnodes, and thosewith4–9 positive axillary lymphnodes and otherpoor prognostic Acknowledgements features. Our results for patients in the 4–9 axillary lymph We thank the staff of the Hemato-Oncology Unit, the node subgroup were similar to those of Bearman et al,15 Department of Oncology, Sheba Medical Center, the referring who achieved a 71% disease-free survival after a median physicians, and all the patients who participated in this study. follow-up of 31 months. Marks et al16 reported that patients with high-risk breast carcinoma treated with high-dose chemotherapy and autologous bone marrow transplantation had a significant References incidence of locoregional recurrence when locoregional irradiation was omitted. To obtain the maximum benefit 1 Peters WP, Roser G, VredenbrughJ et al. Updated results of a from locoregional radiotherapy, it is important to ensure prospective, randomized comparison of two doses of combina- tion alkylating agents as consolidation after CAF in high-risk that it is safe and has a minimal risk of long-term primary breast cancer involving ten or more axillary lymph morbidity, particularly when it follows chemotherapy, nodes. Proceedings of the Annual Meeting of the American which is expected to increase treatment-related morbidity. Society of Clinical Oncology, #81, San Francisco, CA, May The high-dose carmustine protocol used by Peters et al,1 is 2001 CALGB 9082/SWO 9114/NCIC Ma-13. known to cause severe lung injury in more than 40% of 2 Gianni A, Bonadonna G. Five-year results of the randomized patients. The lung injury precluded the early initiation of clinical trial comparing standard versus high-dose myeloabla- radiotherapy in the transplant arm. Thus, only 78% of the tive chemotherapy in the adjuvant treatment of breast cancer patients allocated to the transplant arm received radio- with 43 positive nodes. Proceedings of the Annual Meeting of therapy compared to 89% in the control arm.17 This may the American Society of Clinical Oncology, #80, San Francisco, explain in part the nonsignificant findings in disease free CA, May 2001. 1 3 Rodenhuis S, Bontenbal M, Beex LVAM et al. Randomized and survival in Peters study. In our study, the low rate of phase III study of high-dose chemotherapy with cyclophos- lung injury (o5%), probably because of a noncarmustine- phamide, thiotepa and carboplatin in operable breast cancer based transplant protocol, enabled initiation of radio- with4 or more axillary lymphnodes. Proceedings of the Annual therapy as scheduled. In all, 99% of patients were given Meeting of the American Society of Clinical Oncology, #286, radiation therapy with minimal radiation-related toxicity, New Orleans, LA, May 2000. because of a strict radiation procedure keeping lung volume 4 Fleming ID, Cooper JS, Henson DE et al. (Eds.) American to a minimum. Joint Committee on Cancer Staging Manual, 5thedn. Cox multivariate analysis demonstrates that only pro- Lippincott Raven: Philadelphia, 1997. gesterone receptor status was an independent predictor of 5 Stemmer SM, Pfeffer MR, Rizel S et al. Feasibility and low disease-free survival, but not overall survival. These data toxicity of early radiotherapy after high-dose chemotherapy 18 and autologous stem cell transplantation for patients with are in agreement withSomlo et al, who found PR to be a high-risk stage II–III and locally advanced breast carcinoma. predictor of overall survival as well. Cancer 2001; 91: 1983–1991. The issue of stem cell contamination by breast cancer 6 Antman K, AyashL, Elias A et al. A phase II study of high- cells and the effect of purging were not addressed in this dose cyclophosphamide, thiotepa, and carboplatin with auto- report. Nevertheless, one inclusion criterion for the study logous marrow support in women withmeasurable advanced was negative bilateral bone marrow biopsies withhemato- breast cancer responding to standard-dose therapy. J Clin xylin–eosin staining. One preliminary report has shown a Oncol 1992; 10: 102–110. possible reduction in relapse in patients withCD34 positive 7 Peters WP, Ross M, VredenburghJJ et al. Highdose stem-cell transplantation.19 chemotherapy and autologous bone marrow support as This study is one of the largest single-institution series consolidation after standard dose adjuvant therapy for high- risk primary breast cancer. J Clin Oncol 1993; 11: 1132–1143. withprolonged follow-up (median of 51 months)reported 8 Gianni AM, Siena S, Bregni M et al. Efficacy, toxicity, and in the literature. Our results compare favorably with results applicability of high-dose sequential chemotherapy as adjuvant using conventional-dose adjuvant chemotherapy with a treatment in operable breast cancer with10 or more involved median follow-up of 3–10 years (20–60% disease-free axillary nodes: Five-year results. J Clin Oncol 1997; 15: survival).20,21 Although such comparisons need to be 2312–2321.

Bone Marrow Transplantation Stem cell transplant for hight risk stage II breast cancer SM Stemmer et al 661 9 International Breast Cancer Study Group. Duration and cell support for primary breast cancer in patients with4–9 reintroduction of adjuvant chemotherapy for node-positive involved axillary lymphnodes. Bone Marrow Transplant 1997; premenopausal breast cancer patients. J Clin Oncol 1996; 14: 20: 931–937. 1885–1894. 16 Marks LB, Halperin EC, Prosnitz LR et al. Post-mastectomy 10 Roche HH, Pouillart P, Meyer N et al. Adjuvant high dose radiotherapy following adjuvant chemotherapy with autolo- chemotherapy (HDC) improves early outcome for high risk gous bone marrow transplant for breast cancer patients with (N47) breast cancer patients: the Pegase 01 trial. Proceedings 410 positive axillary lymphnodes. Int J Radiat Oncol Biol of the Annual Meeting of the American Society of Clinical Phys 1992; 23: 1021–1026. Oncology, #102, San Francisco, CA, May 2001. 17 Marks LB, Fitzgerald TJ, Laurie F et al. Preliminary analysis 11 Gisselbrecht C. Autologous stem cell transplantation in of radiotherapy data from CALGB 9082: variability of aggressive non-Hodgkin’s lymphoma. Recent Results Cancer treatment fields for local/regional breast cancer and the impact Res 1998; 144: 15–26. of high dose chemotherapy to deliver radiation therapy 12 Fisher DC, Vredenburgh JJ, Petros WP et al. Reduced [Abstract]. Int J Radiat Oncol Biol Phys 1999; 195a (Suppl): 45. mortality following bone marrow transplantation for breast 18 Somlo G, Doroshow JH, Forman SJ et al. High-dose cancer with the addition of peripheral blood progenitor cells is chemotherapy and stem cell rescue in the treatment of high- due to a marked reduction in veno-occlusive diseases of the risk breast cancer: prognostic indicators of progression-free liver. Bone Marrow Transplant 1998; 21: 117–122. and overall survival. J Clin Oncol 1997; 15: 2882–2893. 13 Hortobagyi GN, Buzdar AU, Theriault RL et al. Randomized 19 Cornetta K, YanovichS, Rosenfeld C et al. High-dose trial of high-dose chemotherapy and blood cell autografts for chemotherapy and stem cell transplant in breast cancer high-risk primary breast carcinoma. J Natl Cancer Inst 2000; patients: a randomized multicenter study of CD 34 selection 92: 225–233. [abstract]. Biol Blood Marrow Transplant 2000; 141a:6. 14 Bolwell BJ, Andresen SW, Pohlman BL et al. The prognostic 20 Abeloff MD, Beveridge RA, Donehower RC et al. Sixteen importance of the axillary lymph node ratio in autologous week dose intense chemotherapy in the adjuvant treatment of transplantation for high-risk stage II–III breast cancer. breast cancer. J Natl Cancer Inst 1990; 82: 570–574. Proceedings of the Annual Meeting of the American Society of 21 Buzdar AV, Kau SW, Hortobagyi GN et al. Clinical course of Clinical Oncology, #57, New Orleans, LA, May 2000. patients withbreast cancer withten or more positive nodes 15 Bearman SI, Overmoyer BA, Bolwel BJ et al. High-dose who were treated with doxorubicin containing adjuvant chemotherapy with autologous peripheral blood progenitor therapy. Cancer 1992; 69: 448–452.

Bone Marrow Transplantation