High-Dose Busulfan, Melphalan, Thiotepa and Peripheral Blood Stem Cell Infusion for the Treatment of Metastatic Breast Cancer

Home , Busulfan

High-dose busulfan, melphalan, thiotepa and peripheral blood stem cell infusion for the treatment of metastatic breast cancer

WI Bensinger1,2,3, KS Schiffman1, L Holmberg1, FR Appelbaum1,2,3, R Maziarz3, P Montgomery3, E Ellis3, S Rivkin3, P Weiden1,3, K Lilleby1, S Rowley1,2,3, S Petersdorf1,3, JP Klarnet3, W Nichols3, A Hertler3, R McCroskey3, CH Weaver4 and CD Buckner1,4

1Fred Hutchinson Cancer Research Center, Seattle; 2The University of Washington School of Medicine, Seattle; 3Puget Sound Oncology Consortium, Seattle, WA, USA

Summary: with chemotherapy-sensitive disease where 35–50% achieve complete responses (CR) and 15–20% survive The purpose of this study was to determine the outcome progression-free 3–5 years after treatment.1–8 Treatment- of patients with metastatic breast cancer treated with related mortality (TRM) was low with the primary reason high-dose busulfan (Bu), melphalan (Mel) and thiotepa for poor outcome of patients receiving HDC being relapse, (TT) followed by peripheral blood stem cell (PBSC) demonstrating the need for more effective regimens. infusion. Fifty-one patients with chemotherapy refrac- A regimen of busulfan (Bu), melphalan (Mel) and metastatic breast thiotepa (TT) was developed in a phase I trial. These drugs (19 ؍ or responsive (n (32 ؍ tory (n cancer received Bu (12 mg/kg), Mel (100 mg/m2) and TT have marrow ablation as their dose-limiting toxicity, (500 mg/m2) followed by PBSC collected after chemo- allowing significant dose escalation with the infusion of or growth factor autologous hematopoietic stem cells.9 Each of these drugs (43 ؍ therapy and growth factor (n The 100 day treatment-related mortality exhibits significant anti-tumor activity in patients with .(8 ؍ alone (n was 8% including one death from cytomegalovirus breast cancer either alone or in combination with other pneumonia, one from aspiration pneumonia and two alkylating agents.2,3,10–12 The maximum-tolerated dose from regimen-related toxicity (RRT). Seven of 28 (MTD) of TT that could be added to fixed doses of refractory (25%) and 5/7 (71%) responsive patients 12 mg/kg of Bu and 100 mg/m2 of Mel was 500 mg/m2.9 with evaluable disease achieved a complete response of In a phase I–II trial of Bu/Mel/TT treatment-related all measurable disease or all soft tissue disease with at mortality was 4% among 25 patients with metastatic breast least improvement in bone lesions (PR*). Fifteen of 51 cancer.13 The purpose of this study was to determine the patients (29%) are alive and progression-free a median outcome for patients with metastatic breast cancer treated of 423 days (range 353–934) after treatment, 5/32 (16%) with a fixed dose of Bu/Mel/TT and peripheral blood stem with refractory disease and 10/19 (53%) with responsive cell (PBSC) infusion. disease. The probabilities of progression-free survival ؍ PFS) at 1.5 years for the patients with refractory (n) ,disease were 0.24 and 0.53 (19 ؍ and responsive (n (32 respectively. These preliminary data suggest that high- Patients and methods dose Bu/Mel/TT has significant activity in patients with advanced breast cancer and may be superior to some Between May 1993 and 15 June 1995, 51 women with previously published regimens. refractory (n ϭ 32) or responsive (n ϭ 19) stage IV breast Keywords: busulfan; melphalan; thiotepa; PBSC; breast cancer were treated with high-dose Bu/Mel/TT followed by cancer PBSC infusion (Table 1). Patients were eligible for study if they had a bilirubin Ͻ2 mg/dl, a creatinine clearance Ͼ50 ml/min and a Karnofsky score of у70. Patients with active or untreated CNS disease were excluded. Twenty- There have been many reports of clinical trials evaluating eight patients were treated at the Fred Hutchinson Cancer high-dose chemotherapy (HDC) and autologous stem cell Research Center (FHCRC) or a separate but affiliated aca- support in women with advanced stage IV breast cancer. demic center (University of Washington, Oregon Health Although these trials have reported high response rates, 50– Sciences University) and 23 were treated at one of seven 70% of patients who achieve complete responses will participating community cancer centers listed in the Appen- 1,2 relapse. The best results have been achieved in patients dix. All patients were treated on a FHCRC or Puget Sound Oncology Consortium protocol approved by the insti- tutional review board of the hospital where the therapy was Correspondence: WI Bensinger, Fred Hutchinson Cancer Research Center, administered and a protocol specific consent was signed. 1124 Columbia, Mailstop M185, Seattle, WA 98104, USA 4Current address: Response Oncology, 600 Broadway, Suite 230, Seattle, Twenty-six patients have been previously reported as part 9,13 WA 98122, USA of phase I or phase II clinical trials. Date of last follow- Received 7 November 1996; accepted 2 March 1997 up for patients in the present report was 15 June 1996. High-dose therapy for breast cancer WI Bensinger et al 1184 Table 1 Pre-treatment patient characteristics (n ϭ 51) taxol (n ϭ 1) or cyclophosphamide and etoposide (n ϭ 1) for further cytoreduction before HDC. No. of patients (%) High-dose chemotherapy Disease status The protocol required a minimum 4 week interval between Chemotherapy refractory 32 (63) Chemotherapy responsive 19 (37) administration of chemotherapy for PBSC collection or De novo stage IV disease 11 (22) cytoreduction and beginning oral Bu. Patients were treated Doxorubicin in adjuvant regimen 22 (43) with phenytoin beginning 24 h prior to the first dose, and Hormone receptor status continued until 24 h after the final dose of Bu. All patients ERϩ/PRϩ 27 (53) ERϪ/PRϩ 5 (10) received Bu 1 mg/kg/dose p.o.q. 6 h on days Ϫ8, Ϫ7, Ϫ6 2 ERϪ/PRϪ 12 (23) for a total dose of 12 mg/kg. Mel (50 mg/m /day/i.v.) was ERϩ/PRϪ 4 (8) given on days Ϫ5 and Ϫ4 for a total of 100 mg/m2.TT Unknown 3 (6) (250 mg/m2/day/i.v.) was administered on days Ϫ3 and Ϫ2 Prior hormone therapy for a total dose of 500 mg/m2. Peripheral blood stem cells As adjuvant therapy 14 (27) For treatment of metastatic disease 14 (27) were infused 48 h after the last dose of TT. Disease status NED 12 (24) Measurable disease 28 (55) Supportive care Bone-only 11 (22) Disease sites Following the administration of HDC, all patients received Lung 12 (24) prophylactic antibiotics and fluconazole when the absolute Liver 11 (22) neutrophil count (ANC) was Ͻ0.5 ϫ 109/l. Patients who Bone 24 (47) were serologically positive for herpes simplex virus Lymph node/chest wall 18 (35) received prophylactic acyclovir. Eleven of 51 patients CNS (remission) 1 (2) No. of sites received G-CSF, 5 ␮g/kg/day subcutaneously, after PBSC 136infusion at the discretion of the attending physician, while 211the remaining 40 patients did not receive growth factors. 34

Median age, years (range) 45 (25–66) Treatment after high-dose chemotherapy and PBSC Median time from diagnosis to metastasis 34 (4–125) infusion (n ϭ 40), months (range) Ͻ2 years (n ϭ 14) 15 (4–23) Patients who were estrogen and/or progesterone receptor Ͼ2 years (n ϭ 26) 26 (25–125) positive were given tamoxifen, 20 mg/day, until disease Prior conventional dose regimens for stage 1 (1–3) progression or for at least 1 year. Five patients who achi- IV disease (n ϭ 43), median (range) eved complete response after HDC with PBSC infusion received chest wall radiation and ﬁve received local radiation therapy to areas of bone metastasis.

Collection, cryopreservation and infusion of PBSC Regimen-related toxicity grading Peripheral blood stem cells were collected after the admin- Regimen-related toxicity (RRT) was graded to day 100 istration of recombinant granulocyte colony-stimulating after PBSC infusion using a previously described system factor (G-CSF) alone (n ϭ 8),14 after intermediate-dose to assess toxicities to the oral, renal, hepatic, cardiac, pul- chemotherapy and G-CSF or recombinant granulocyte– monary, gastrointestinal (GI), bladder, skin and central ner- macrophage colony-stimulating factor (GM-CSF) (n ϭ 40) vous systems.13,20 Infection was excluded from this grading or after conventional doxorubicin-based chemotherapy and system. Veno-occlusive disease (VOD) of the liver was G-CSF (n ϭ 3). Intermediate-dose chemotherapy for 40 defined as liver toxicity secondary to HDC. VOD was patients consisted of cyclophosphamide 4 g/m2 (n ϭ 8),15 defined clinically by hepatomegaly and/or liver tenderness, cyclophosphamide 4 g/m2 and taxol 170–250 mg/m2 (n ϭ weight gain Ͼ2% of baseline, and an elevated serum biliru- 14),16,17 cyclophosphamide 4 g/m2 and etoposide bin Ͼ2 mg/dl and/or histologically by damaged endothelial 600 mg/m2 (n ϭ 2),18 cyclophosphamide 4 g/m2, etoposide cells of the terminal hepatic venules, dilatation of the sinus- 600 mg/m2 and cisplatin 105 mg/m2 (CEP) (n ϭ 15)18 or oids and necrosis of hepatocytes.21 Idiopathic pneumonia taxol 250 mg/m2 (n ϭ 1). The techniques of PBSC cryopre- syndrome (IPS) was defined as pulmonary infiltrates not servation, thawing and reinfusion have been reported.14,19 associated with pulmonary edema or an identifiable infec- The day of PBSC infusion was designated day 0. tious cause.22

Second courses and intermediate-dose chemotherapy Evaluation of response Thirteen patients received a second course of intermediate- Patients were considered to have refractory disease if they dose chemotherapy with cyclophosphamide, etoposide, and did not achieve a partial remission (PR) or progressed fol- cisplatin (n ϭ 8), cyclophosphamide and taxol (n ϭ 3), lowing initial chemotherapy for stage IV disease. Patients High-dose therapy for breast cancer WI Bensinger et al 1185 who achieved a 50% reduction in the spread of all measur- 1 able metastatic disease (PR) or complete response (CR) following initial chemotherapy for metastatic disease or who 0.8 were rendered disease-free following surgery or radiation therapy without receiving chemotherapy were classiﬁed as 0.6 having responsive disease. Bone-only disease was categor- 0.4 PFS

ized as responsive if sclerosis of prior lesions was demon- Probability strated with no new lesions and normalization of a tumor 0.2 OS marker following initial chemotherapy occurred. Patients who entered HDC without measurable disease were classi- 0 fied as having no evidence of disease (NED). De novo 0 0.5 1 1.5 2 2.5 3 patients were defined as those who presented with stage IV Years after HDC disease at the time of initial diagnosis. Local/regional dis- Figure 1 Refractory disease: PFS progression-free survival; OS ϭ over- ease included ipsilateral or contralateral nodal, chest wall all survival; HDC high-dose chemotherapy. or breast recurrences. Visceral disease was defined as liver, lung or abdominal metastases. 1 PFS Baseline staging was carried out within 10 weeks of HDC using imaging studies (CT or MRI scans of brain, 0.8 OS chest, abdomen and pelvis and radionuclide bone scans or 0.6 skeletal MRIs), tumor markers (CA 15-3, CA 27-29, CEA or CA 125) and bilateral bone marrow biopsies. Patients 0.4 had tumor markers repeated within 6 weeks of HDC and Probability those with measurable visceral or soft tissue disease were 0.2 restaged by CT or MRI scans at previous sites of disease. All patients were restaged at 60–80 days, at 6 months and 0 yearly after HDC. 00.5 1 1.5 2 2.5 3 Responses after HDC and PBSC infusion for patients Years after HDC with measurable disease were defined as CR if there was Figure 2 Responsive disease: PFS ϭ progression-free survival; OS ϭ disappearance of all tumor for greater than 30 days, and overall survival; HDC ϭ high-dose chemotherapy. PR if there was a reduction of 50% or more in the size of measurable disease for at least 30 days. Responses of less 1) or chemotherapy (n ϭ 7) prior to entry on to the study. than PR or progressive disease were considered no response All 11 patients with bone-only disease had failed hor- (NR). Patients with measurable disease and bone metastasis monal therapy. or bone-only disease were defined as PR* if there was a CR of measurable disease in non-bone areas and sclerosis Intermediate-dose chemotherapy of prior lytic bone lesions but continuing activity by bone scan in areas of prior uptake without the development of Thirty-three patients received one course and 13 patients new bone lesions.6 Actuarial probabilities were calculated received two courses of intermediate-dose therapy before from day 0 using the method of Kaplan and Meier.23 End- HDC and PBSC infusion. There were no grade 3–4 RRT points for survival were duration from PBSC infusion until following intermediate-dose chemotherapy. death censored by the date of last contact. The endpoints for progression-free survival (PFS) were death and relapse High-dose chemotherapy with Bu/Mel/TT and PBSC or progression censored by the date of last contact. infusion All patients received the prescribed doses of Bu, Mel, TT. Results Peripheral blood stem cell infusion and hematologic Patient characteristics recovery Table 1 shows the characteristics of the 51 patients treated. The median number of CD34+ cells infused was 8.20 ϫ Eleven patients (seven with refractory and four with 106 CD34ϩ cells/kg (range 2.56–59.09). One patient died responsive disease) presented with de novo metastatic dis- on day 6 before recovery of neutrophils or platelets. The ease. Forty-three patients received a median of one conven- remaining 50 patients achieved an ANC Ͼ0.5 ϫ 109/l at a tional chemotherapy regimen (range 1–3) for stage IV dis- median of 10 days (range 8–14) and a platelet level Ͼ20 ease. The eight patients who had not received conventional ϫ 109/l at a median of 11 days (range 1–26) after infusion chemotherapy for stage IV disease were given one cycle of of PBSC. intermediate-dose chemotherapy for mobilization (n ϭ 6) or two cycles for mobilization and debulking (n ϭ 2). All Treatment-related morbidity and mortality (Table 2) eight were classified as chemotherapy sensitive. Twelve patients, 10 with responsive and two with refractory disease There were six episodes of grade III/IV RRT occurring in (24%) were rendered NED by surgery (n ϭ 4), RT (n ϭ six (12%) patients. Four patients experienced grade III tox- High-dose therapy for breast cancer WI Bensinger et al 1186 Table 2 Regimen-related toxicity (n ϭ 51)

Category Toxicity grade - No. of patients (%)

1234a

Pulmonary 1 (2) 4 (8) — 1 (2) Cardiac 2 (4) 1 (2) — 1 (2) Skin 16 (31) 8 (16) — — Veno-occlusive disease 17 (33) 3 (6) 1 (2) — Stomatitis 7 (14) 44 (86) — — Renal 1 (2) — — — Diarrhea 23 (45) 9 (18) 2 (4) — Hemmorrhagic cystitis 1 (2) 1 (2) 1 (2) — Total No. of patients 51 (100) 44 (86) 4 (8) 2 (4)

aFatal.

icities (two gastrointestinal, one VOD and one hemorrhagic chemotherapy were 0.33 and 0.38, respectively. The prob- cystitis). Two patients (4%) died of RRT within 100 days abilities of survival and PFS at 1.5 years for the 13 patients of treatment. One died from IPS at 100 days post-HDC and (10 with refractory and three with responsive disease) who one patient with a pre-treatment history of acute pulmonary received two courses of intermediate-dose therapy were edema and a decreased left ventricular ejection fraction of 0.21 and 0.27, respectively. 40% developed an acute arrythmia and died from cardiac arrest on day 16. One patient died from CMV pneumonia Outcome for refractory patients on day 100 and one patient with vomiting died from aspiration pneumonia on day 6. Thus, a total of four patients Among 32 patients with refractory disease, two were NED (8%) died from complications of HDC with a 100 day non- pre-HDC and two died before day 100 from treatment- relapse mortality of 0.06. related complications. Thus, 28 were evaluable for There were four episodes of non-fatal bacteremia second- response. Seven of 28 (25%) achieved a CR/PR* and four ary to Staphylococcus epidermidis, one non-fatal candidal (14%) are alive progression-free. One of two patients who pneumonia, one CMV gastritis and one fatal CMV pneu- were NED remains progression-free at 353 days. Sites of monia at 100 days. disease involvement for the ﬁve survivors (16%) with One of 13 patients who received two courses of inter- refractory disease included: bone-only (n ϭ 2); lung (n ϭ mediate-dose chemotherapy, 2/33 who received one and 1/6 2) and local (n ϭ 1). The probabilities of survival and PFS who did not receive any intermediate-dose chemotherapy for the 32 patients with refractory disease at 1.5 years were prior to HDC died of TRM. 0.25 and 0.24, respectively (Figure 1).

Response and survival (Table 3) Outcome for responsive patients Of 51 patients, 12 were NED pre-HDC and four died before Nineteen patients were defined as having responsive dis- day 100 from treatment-related complications, leaving 35 ease. Nine of these patients had measurable disease, of patients evaluable for response. Twelve of 35 (34%) achi- whom five achieved a PR*. Overall, 10 of the 19 (53%) eved a CR/PR* (five with responsive and seven with refrac- (four in PR* and six NED) are alive and progression-free tory disease) of whom eight are alive and progression-free a median of 612 (range 364–934) days after HDC. The a median of 589 days (range 339–771) after HDC. Seven probabilities of survival and PFS at 1.5 years for patients of the 12 patients (24%) who were NED (10 with respon- with responsive disease were 0.53 and 0.53, respectively sive and two with refractory disease) remain progression- (Figure 2). One patient with responsive disease in PR* died free a median of 367 days (range 353–934) after HDC. from suicide 440 days after PBSC infusion. Thus, 15 patients (29%) are alive without progression with a median follow-up of 423 days (range 339–934). Survival and PFS for all 51 patients at 1.5 years was 0.26 and 0.24, Discussion respectively. PFS at 1.5 years for the 24 patients who were NED or achieved CR/PR* post-HDC was 0.64. A regimen of high-dose Bu/Mel/TT followed by PBSC Four of 22 patients (one CR and three NED) who had infusion in patients with metastatic breast cancer was asso- failed doxorubicin as adjuvant therapy, and two with more ciated with an 8% 100 day treatment-related mortality. than one site of metastatic involvement are alive and pro- Death occurred in 2/32 patients with refractory disease and gression-free a median of 581 days (range 364–687) after 2/19 with responsive disease. The major non-hematologic HDC. morbidity was gastrointestinal with all patients developing The probabilities of survival and PFS at 1.5 years for significant stomatitis with grade III gastroenteritis occurring the 33 patients (21 with refractory and 12 with responsive in two patients. disease) who received one course of intermediate-dose Response rates were high with CR/PR* and PR rates of High-dose therapy for breast cancer WI Bensinger et al 1187 Table 3 Response and outcome (n ϭ 51)

Disease Response PR Progression-free status survivors NED ED CR/PR*

Bone Refractory 10 1 2 1 2 2 Responsive 3 1 1 1 — 2 Visceral Refractory 7 — — 3 1 2 Responsive 1 1 — — — 0 Local/Regional Refractory 5 — — 2 1 1 Responsive 10 8 1 — 1 6 Bone ϩ Visceral Refractory 7 — — 1 2 0 Responsive 4 — — 3 1 2 Bone ϩ Local Refractory — — — — — 0 Responsive 1 — — 1 — 0

Local/Visceral 3 1 — — — 0 Refractory — — — — — 0 Responsive

Total 51 12 4 12/35 (34%) 8/35 (23%) 15/51 (29%) Refractory 32 2 2 7/28 (25%) 6/28 (21%) 5/32 (17%) Responsive 19 10 2 5/7 (71%) 2/7 (29%) 10/19 (53%)

NED ϭ no evaluable disease; ED ϭ early death; CR ϭ complete response; PR* ϭ complete response in the setting of bone disease; PR ϭ partial response.

34 and 23%, respectively, in patients with evaluable responsive disease who were not in CR became long-term disease. Of the 12 patients evaluable for response who survivors following HDC.2 Despite the small numbers in achieved a CR/PR*, eight (four with refractory and four the present study (n ϭ 19), a greater than 50% PFS for with responsive disease) remain in remission a median of responsive patients is encouraging and suggests that 581 (range 339–771) days after HDC. The probabilities of Bu/Mel/TT should be compared to other established treat- PFS at 1.5 years were 0.53 and 0.23 for responsive and ment regimens. refractory patients, respectively. Certain prognostic factors have been associated with A variety of high-dose chemotherapy regimens with long-term DFS following HDC for patients with responsive autologous marrow or PBSC support for treatment of metastatic breast cancer. Patients who achieve CR with patients with metastatic breast cancer who failed chemo- initial therapy for stage IV disease and have a single meta- therapy have been evaluated, including: cyclophosphamide, static site of disease had improved PFS following HDC carmustine and cisplatin;5 cyclophosphamide, thiotepa and with cyclophosphamide, thiotepa and carboplatin.6 An carboplatin; cyclophosphamide and thiotepa;12 busulfan and analysis of 114 patients treated at diagnosis for stage IV cyclophosphamide;11 or cyclophosphamide, etoposide, and disease with AFM induction, intermediate-dose chemo- cisplatin.24 For patients who have failed chemotherapy, CR therapy mobilization of PBSC, and HDC with cyclophos- rates as high as 48% have been reported but the median phamide, TT and carboplatin revealed that the presence of time to progression has been 6–10 months with few or no liver or lung metastasis or a history of receiving doxorub- survivors.1–3,5,11,24 In the current study, five patients (16%) icin as adjuvant therapy were adverse risk factors with no with refractory disease are surviving progression-free with patient surviving progression-free with any one of these a minimum follow-up of 1 year. These results justify factors.8 further attempts to develop new HDC treatment regimens In the current study, there were four patients with lung or strategies for patients with refractory breast cancer. involvement, two with more than one site of disease and When HDC was administered to patients with stage IV four who failed adjuvant doxorubicin who are surviving breast cancer who were responding to initial induction ther- progression-free. Although numbers are small, these results apy, other studies have reported CR rates of 30–60% with suggest that further evaluation of the Bu/Mel/TT regimen 10–25% of patients remaining in CR for more than 2 should be carried out in patients with adverse risk factors. years.12,25 In the present study, 10/19 patients (53%) with Intermediate-dose chemotherapy given for mobilization responsive disease are surviving without progression a of PBSC and/or for tumor reduction prior to HDC may median of 612 days. In a recent review of the literature, 20 have improved the outcome for some patients. Based on of 85 (27%) patients with metastatic breast cancer became the observations that patients who are in CR prior to HDC long-term survivors after HDC when treatment was carried have a superior outcome,2 13 patients received a second out in CR.2 However, only 15/174 (9%) patients with cycle of intermediate-dose chemotherapy. There were no High-dose therapy for breast cancer WI Bensinger et al 1188 apparent differences in the occurrence of life-threatening or therapy with hematopoietic rescue as primary treatment for fatal toxicities in this small group of 13 patients and no metastatic breast cancer: a randomized trial. J Clin Oncol difference in PFS when compared to patients who received 1995; 13: 2483–2489. a single cycle of intermediate-dose chemotherapy prior to 8 Weaver CH, West WH, Schwartzberg LS et al. Induction, HDC. This suggests that more intensive therapy could be mobilization of peripheral blood stem cells (PBSC), high-dose chemotherapy and PBSC infusion in patients with untreated given safely before Bu/Mel/TT. In order to determine the stage IV breast cancer: outcomes by intent to treat analyses. effect of intermediate-dose chemotherapy prior to HDC on Bone Marrow Transplant 1997; 19: 661–670. response and PFS, a randomized trial would be required in 9 Weaver CH, Bensinger WI, Appelbaum K et al. Phase I study a homogeneous population of patients. The doses of drugs of high-dose busulfan, melphalan, thiotepa with autologous used for mobilization or debulking, while higher than con- stem cell support in patients with refractory malignancies. ventional doses, are lower than the doses used in the high- Bone Marrow Transplant 1994; 14: 813–819. dose CEP regimen. Dunphy et al26 reported that two cycles 10 Lazarus HM, Reed MD, Spitzer TR et al. High-dose IV thi- of high-dose CEP produced a 15% PFS at 3 years in otepa and cryopreserved autologous bone marrow transplan- chemotherapy-responsive stage IV patients. This indicates tation for therapy of refractory cancer. Cancer Treat Rep that it is unlikely that the intermediate-dose chemotherapy 1987; 71: 689–695. 11 Demirer T, Buckner CD, Appelbaum FR et al. High-dose bus- administered prior to HDC accounts for the improved out- ulfan and cyclophosphamide followed by autologous trans- come with Bu/Mel/TT. plantation in patients with advanced breast cancer. Bone Mar- Although the number of patients treated with Bu/Mel/TT row Transplant 1996; 17: 769–774. was relatively small, outcome for patients with refractory 12 Williams S, Gilewski T, Mick R et al. High dose consolidation or responsive disease are at least comparable and may be therapy with autologous stem cell rescue in stage IV breast superior to other treatment regimens. This can only be veri- cancer: follow-up report. J Clin Oncol 1992; 10: 1743–1747. fied by an expanded phase II or a phase III trial comparing 13 Schiffman KS, Bensinger WI, Appelbaum FR et al. Phase II outcomes with commonly used HDC regimens. study of high-dose busulfan, melphalan and thiotepa with autologous peripheral blood stem cell support in patients with malignant disease. Bone Marrow Transplant 1996; 17: 943– 950. Acknowledgements 14 Bensinger WI, Singer J, Appelbaum F et al. Autologous transplantation with peripheral blood mononuclear cells collected This work was supported by grants CA-18029, CA-47748, CA- after administration of recombinant granulocyte stimulating 18221, CA-15704, CA-58576, CA-09319 and CA-09515 from the factor. Blood 1993; 81: 3158–3163. National Cancer Institute, National Institutes of Health, Bethesda, 15 Demirer T, Buckner CD, Lilleby K et al. Failure of a single MD, The Jose Carreras Foundation Against Leukemia, Barcelona, cycle of high dose cyclophosphamide followed by intensive Spain and The Joseph Steiner Krebstifftung, Bern, Switzerland. myeloablative therapy and autologous stem cell transplantation to improve outcome in relapsed disease. Cancer 1994; 74: 715–721. References 16 Demirer T, Rowley S, Buckner CD et al. Peripheral blood stem-cell collections after paclitaxel, cyclophosphamide and 1 Antman KH. Dose-intensive therapy in breast cancer. In: Arm- recombinant human granulocyte colony-stimulating factor in itage JO, Antman KH (eds). High-dose Cancer Therapy patients with breast and ovarian cancer. J Clin Oncol 1995; (Pharmacology, Hematopoietins, Stem Cells). Williams & 13: 1714–1719. Wilkins: Baltimore, 1992, pp 701–717. 17 Demirer T, Buckner CD, Storer B et al. The effect of different 2 Weaver C, Birch R, Schwartzberg L et al. High-dose chemo- chemotherapy regimens on peripheral blood stem cell collec- therapy and autologous stem cell transplantation for breast tions in patients with breast cancer receiving granulocyte-col- cancer. In: Buckner CD, Clift R (eds). Technical and Biologi- ony stimulating factor. J Clin Oncol 1997; 15: 684–690. cal Components of Marrow Transplantation. Kluwer Aca- 18 Bensinger W, Appelbaum F, Rowley S et al. Factors that demic Publishers: Boston, 1995, pp 59–85. influence collection and engraftment of autologous peripheral- 3 Antman K, Ayash L, Elias A et al. A phase II study of high- blood stem cells. J Clin Oncol 1995; 13: 2547–2555. dose cyclophosphamide, thiotepa, and carboplatin with auto- 19 Rowley SD, Bensinger WI, Gooley TA et al. The effect of logous marrow support in women with measurable advanced cell concentration on bone marrow and peripheral blood stem breast cancer responding to standard-dose therapy. J Clin cell cryopreservation Blood 1994; 83: 2731–2736. Oncol 1992; 10: 102–110. 20 Bearman SI, Appelbaum FR, Buckner CD. Regimen-related 4 Kennedy MJ, Beveridge RA, Rowley SD et al. High-dose toxicity in patients undergoing bone marrow transplantation. chemotherapy with reinfusion of purged autologous bone mar- J Clin Oncol 1988; 6: 1562–1568. row following dose-intense induction as initial therapy for 21 Shuhart MC, McDonald GB. Gastrointestinal and hepatic metastatic breast cancer. J Natl Cancer Inst 1991; 83: 920– complications. In: Forman SJ, Blume KG, Thomas ED (eds). 926. Bone Marrow Transplantation. Blackwell Scientific: Boston, 5 Peters WP, Shpall EJ, Jones RB et al. High-dose combination 1994, pp 454–481. alkylating agents with bone marrow support as initial treat- 22 Crawford SW. Critical care and respiratory failure. In: Thomas ment for metastatic breast cancer. J Clin Oncol 1988; 6: ED, Forman SJ, Blume K (eds). Bone Marrow Transplan- 1368–1376. tation, Chapter 37. Blackwell Scientific: Boston, 1994, 6 Ayash LJ, Wheeler C, Fairclough D et al. Prognostic factors pp 513–526. for prolonged progression-free survival with high-dose 23 Kaplan EL, Meier P. Nonparametric estimation from incom- chemotherapy with autologous stem-cell support for advanced plete observations. J Am Stat Assoc 1958; 53: 457–481. breast cancer. J Clin Oncol 1995; 13: 2043–2049. 24 Dunphy FR, Spitzer G, Buzdar AU et al. Treatment of estro- 7 Bezwoda WR, Seymour L, Dansey RD. High-dose chemo- gen receptor-negative or hormonally refractory breast cancer High-dose therapy for breast cancer WI Bensinger et al 1189 with double high-dose chemotherapy intensification and bone 26 Dunphy FR, Spitzer G, Fornoff JE et al. Factors predicting marrow support. J Clin Oncol 1990; 8: 1207–1216. long-term survival for metastatic breast cancer patients treated 25 Ayash LJ, Elias A, Wheeler C et al. Double dose-intensive with high-dose chemotherapy and bone marrow support chemotherapy with autologous marrow and peripheral-blood (published erratum appears in Cancer 1994; 74: 773). Cancer progenitor-cell support for metastatic breast cancer: a feasi- 1994; 73: 2157–2167. bility study. J Clin Oncol 1994; 12: 37–44.

Appendix: Collaborating community institutions

St Lukes Regional Medical Center, Boise, ID (1) ACoS Designation: Swedish Hospital Medical Center, Seattle, WA (2) (1) Comprehensive Cancer Center; Virginia Mason, Seattle, WA (1) (2) Teaching Center Hospital Cancer Program; St Joseph Hospital, Tacoma, WA (1) (3) Comprehensive Community Cancer Center; Community Medical Center, Missoula, MT (3) (4) Community Cancer Center. Good Samaritan Hospital, Puyallup, WA (4) Sacred Heart Hospital, Spokane, WA (1)