Cyclophosphamide and Paclitaxel As Initial Or Salvage Regimen for the Mobilization of Peripheral Blood Progenitor Cells

Home , Cyclophosphamide, Paclitaxel

Bone Marrow Transplantation, (1999) 24, 959–963  1999 Stockton Press All rights reserved 0268–3369/99 $15.00 http://www.stockton-press.co.uk/bmt Cyclophosphamide and paclitaxel as initial or salvage regimen for the mobilization of peripheral blood progenitor cells

JL Klein, PM Rey, R Dansey, C Karanes, E Abella, L Cassells, C Hamm, M Flowers, C Couwlier, WP Peters and RD Baynes

Bone Marrow Transplantation, Division of Hematology and Oncology, Barbara Ann Karmanos Cancer Institute, and Wayne State University, Detroit, MI, USA

Summary: accelerate the tempo of engraftment compared to bone marrow cells, thereby reducing morbidity, mortality, support Peripheral blood progenitor cells are now commonly measures required and economic costs.1–4 In addition, used for hematologic reconstitution after myelosup- PBPCs can be collected in an outpatient setting with mini- pressive chemotherapy for hematologic and solid malig- mal morbidity avoiding the operative procedure of BM har- nancies. The purpose of this study was to evaluate the vest. Hematopoietic growth factors alone can provide an activity of paclitaxel 170 mg/m2 and cyclophosphamide adequate yield of stem cells to ensure hematological recon- 2 g/m2 (CP) with filgrastim (human G-CSF) for mobiliz- stitution after HDC. Collection of PBPCs after CT followed ation of PBPCs as the first or second maneuver after by cytokines takes advantage of the additional mobilizing failure with filgrastim alone. Sixty-four patients with effect of the myelosuppressive drugs. The use of chemo- stage II–IV breast cancer received (CP) followed by fil- therapy has been recognized to increase PBPC yields com- grastim (10 ␮g/kg/day). In 35 (55%) this was the first pared to cytokines alone, therefore requiring fewer apher- maneuver while it was for salvage in 29 (45%) patients. esis procedures.5,6 In spite of the extensive experience The median number of aphereses was two (range, 1–7). acquired in this area, it is not known which is the best In 83% of the patients apheresis was initiated on days regimen to mobilize hematopoietic progenitor cells. Mobil- 10–11 following chemotherapy. The median numbers of ization with disease-oriented chemotherapy together with CD34+ cells/kg, CD34+ cells/apheresis/kg and total cytokines may be also combine the advantage of antitumor nucleated cells/kg collected were 8.7 × 106 (2.11–73.5), activity with PBPC mobilization.7,8 Paclitaxel and cyclo- 3.97 × 106 (0.3–36.75) and 164.15 × 108 (9–660), respect- phosphamide are both active in breast cancer and share a ively. All the patients yielded at least 2 × 106 CD34+ similar pattern of hematological recovery. The value of a cells/kg. CP mobilization salvaged the 29 patients who combination of paclitaxel and cyclophosphamide as a mob- failed mobilization with filgrastim alone. When used as ilization regimen in patients with ovarian and breast cancer first-line mobilization the yield of CD34+ cells × 106/kg has already been described.9,10 was higher than in the salvage group (16.93 vs 3.94, One of the problems associated with the use of PBPCs P Ͻ 0.001). Patients receiving CP as salvage reached the is the variability in the number of cells collected between target of 5 × 106 CD34+ cells/kg in only 45% (13/29) of patients, partly because of the amount of previous treat- cases vs 94.3% as first maneuver. CP followed by fil- ment.11–14 Moreover, cytokines by themselves will fail to grastim is a safe and effective regimen for the mobiliz- mobilize stem cells in 10% of patients.5,11,15 Some of these ation of PBPCs in patients with breast cancer and shows patients still will be salvaged adding chemotherapy, higher significant activity in patients who failed to mobilize doses of growth factors or both, but the ideal salvage with filgrastim, suggesting a higher mobilization regimen is controversial.16 potential. Our report studies the efficacy of paclitaxel 170 mg/m2 Keywords: PBPC; mobilization; CD34 selection; Isolex and cyclophosphamide 2 g/m2 (CP) with filgrastim, and 300i particularly analyzes its value as a salvage maneuver after failure to mobilization by filgrastim alone.

Within the last decade mobilized peripheral blood progeni- Patients and methods tor or stem cells (PBPC) have emerged as the most common source of hematopoietic cells for support following high- dose chemotherapy (HDC). Various combinations of cyto- Patients kines or myelosuppressive chemotherapy (CT) and cyto- Between November 1996 and January 1998 64 women with kines have been used to mobilize PBPCs. PBPC infusions stage II–IV breast cancer underwent PBPC collection after paclitaxel 170 mg/m2 and CP 2 g/m2 followed by subcut- Correspondence: Dr JL Klein, Karmanos Cancer Institute, 3990 John R, aneous filgrastim. Thirty-five (55%) patients were mobil- 4 Brush South, Detroit, Michigan 48201, USA ized in this fashion as first maneuver, and 29 (45%) after Received 11 March 1999; accepted 23 June 1999 a prior, unsuccessful attempt with filgrastim 10 ␮g/kg alone PBPC mobilization with cyclophosphamide/paclitaxel JL Klein et al 960 (45%) (Table 1). Thirty-two patients who had either bone 20 mg i.v., diphenhydramine 50 mg i.v. and cimetidine marrow involvement or multiple skeletal metastasis were 300 mg i.v. 20 min before paclitaxel administration. Cyclo- involved in a protocol for CD34 selection as a means of phosphamide 2.0 g/m2 was given intravenously 1 h after the tumor purging and had CD34 positive enrichment of the completion of the paclitaxel. Cyclophosphamide was PBPC production after collection using the Isolex 300i infused over 1 h. All patients received MESNA for the pre- (Baxter Healthcare Corporation, Irvine, CA, USA). Patients vention of hemorrhagic cystitis. The MESNA dose was could have received prior chemotherapy for metastatic dis- 2.0 g/m2 as equally divided administrations prior to ease. As part of this protocol, patients underwent initial infusion of CP and 4 and 8 h after. Following chemotherapy PBPC mobilization with CP. Three additional patients with patients received filgrastim 10 ␮g/kg/day subcutaneously. high risk primary breast cancer either stage II (one) or III Patients also received ciprofloxacin, 500 mg three times a (two) received CP as the initial mobilization regimen. day as bacterial prophylaxis. These patients had mobilization with CP because of delays in obtaining insurance coverage and received chemotherapy Collection and cryopreservation of PBPC because of the length of time since their last chemotherapy. Patients who were initially mobilized with filgrastim but Following the administration of the chemotherapy and fil- did not produce 2.5 × 106 CD34+/kg underwent salvage grastim, patients had complete blood counts measured mobilization. Patients given filagrastim had received every other day and began apheresis when the WBC count 10 ␮g/kg for 4 days. On the fifth day PB CD34+ cells were was at least 1.5 × 109/ml. Daily apheresis procedures were enumerated by FACS analysis. Patients with Ͼ10/␮l were performed using a Fenwall CS 3000 plus cell separator started on leukapheresis and continued for 3 days. If the (Fenwall, Baxter Healthcare Corporation, Round Lake, IL, PB CD34+ cell count was Ͻ10 ␮l, patients were continued USA). Sixteen liters were processed per procedure, using on filgrastim and PB CD34+ cells were again checked. If continuous flow centrifugation at a flow rate of 70 ml/min. the PB CD34+ cells remained below 10/␮l for 3 days they The target was to collect у5.0 × 106 CD34+ cells/kg, but were considered mobilization failures and proceeded to CP attempts were made to collect a minimum of 2.5 × 106 mobilization. Patients who had Ͼ10 CD34+ cells/␮l and cells/kg. The final product containing 10% DMSO and started PBPC collection but did not yield Ͼ2.5 × 106 CD34+ autologous plasma was frozen in a liquid nitrogen con- cell/kg within 3 days and had a falling PB CD34+ cell count trolled-rate freezer and was maintained in the liquid phase were also considered mobilization failures and proceeded of liquid nitrogen. to CP mobilization. No patients received a second mobilization with filgrastim at either the same or an increased dose. High-dose chemotherapy regimens Five of the women who received CP, as the second mobilization regimen had stage II disease and 24 (83%) had stage All but three patients received cisplatin, CP and BCNU IV disease. The median number of chemotherapy cycles (STAMP I) followed by PBPC infusion. either adjuvantly and/or for metastatic disease was 10 (range, 3–26) and eight (range, 3–24), for patients under- Cyclophosphamide, cisplatin and BCNU: Sixty-one going mobilization as a first or salvage maneuver, patients received cyclophosphamide (1875 mg/m2/day) respectively. days −6to−4, cisplatin (165 mg/m2) continuous i.v. infusion days −6to−4, and BCNU (600 mg/m2) on day − Mobilization 3, as follows. Actual body weight was used to calculate the therapeutic doses of all agents unless the actual body On day 1 paclitaxel 170 mg/m2 was diluted in 500 cc D5W weight was 20% greater than the ideal body weight. In the in a glass container with a polyethylene-line tubing and a event that the actual body weight was 20% greater the aver- 0.22 ␮m in-line filter and infused over 24 h intravenously. age of the ideal and actual body weight was used to calcu- Patients were premedicated with dexamethasone 20 mg late the therapeutic doses. PBPCs were reinfused 3 days orally 12 and 6 h before paclitaxel and dexamethasone after the completion of the regimen. Patients received fil-

Table 1 Patient characteristics

Overall Initial maneuver Salvage regimen

n (%) 64 35 (55%) 29 (45%) Age (median and range) 47 (22–71) 45 (22–59) 48 (26–71) Stage II 6 (9.5%) 1 (3%) 5 (17%) III 2 (3%) 2 (6%) 0 IV 56 (87.5%) 32 (91%) 24 (83%) Prior radiotherapy 36 (56%) 16 (46%) 20 (69%) No. of prior cycles of chemotherapy 9.5 (3–26) 10 (3–26) 8 (3–24) No. of prior chemotherapy regimens 2 (1–3) 2 (1–3) 2 (1–3) Bone marrow involvement 14 (22%) 12 (34%) 2 (7%) Bone metastasis (Ͼ3 sites) 35 (55%) 25 (71%) 10 (34%) PBPC mobilization with cyclophosphamide/paclitaxel JL Klein et al 961 grastim 5 ␮g/kg/day s.c. starting 3 h after the completion 100 of the PBPC infusion. Patients were discharged from hospi- tal after the PBPBC infusion and followed daily in the out- 75 patient transplantation clinic. 50 Ifosphamide, dacarbazine and melphalan: Three patients % Patients 25 received ifosphamide (6000 mg/m2), melphalan 2 2 (120 mg/m ) and dacarbazine (1000 mg/m ). 0 123 45 678 + Measurement of CD34 cells Days of apheresis + Peripheral blood was analyzed prior to apheresis for CD34 Overall First Salvage cell counts using flow cytometry. An aliquot of each apher- × 6 + esis product was collected and the CD34+ cell content Figure 1 Probability of reaching a target of 2.5 10 CD34 cell/kg for patients undergoing first or salvage mobilization. measured. CD34+ cell enumeration was performed as previously described.17 100 Statistics 75 Statistical comparisons were made with SPSS software. The Mann–Whitney, Wilcoxon and Kruskal–Wallis tests 50 were used for non-parametric analyses. Proportions were compared using chi-square or Fisher’s exact method, as % Patients 25 appropriate. Multivariate associations were assessed using 0 a multiple regression model. Factors were included in a 123 45 6 stepwise manner. Time to collect at least 2.5 and 5 × 106 CD34+ cells and time to initiate apheresis were estimated Days of apheresis using the product-limit method according to Kaplan and Meier and comparisons made by Breslow test. Overall First Salvage Figure 2 Probability of reaching a target of 5.0 × 106 CD34+ cell/kg for patients undergoing first or salvage mobilization. Results

PBPC collections the first mobilization procedure and patients who received CP as a salvage procedure after failing PBPC collection PBPC collection was initiated when the WBC was with filgrastim alone. The median collection was 3.94 × 106 Ͼ1.5 × 109/l. The median day for starting apheresis was CD34+ cells/kg in patients receiving CP as salvage, com- day 10 after receiving the mobilizing chemotherapy (range, pared with 16.93 × 106 CD34+ cells/kg when CP was used 7–17). The date of initiating leukapheresis following CP as first mobilization (P Ͻ 0.001). Four patients in the sal- was very consistent, with more than 80% of patients start- vage group (13.7%) did not achieve the target number of ing the collections on days 10 or 11. There was no statisti- 2.5 × 106 and 16 did not collect 5 × 106, vs 0 and 2, respect- cally significant difference in the day to start the apheresis ively, in the initial mobilization group (P Ͻ 0.001). between the two groups. On the first day of collection the median WBC and CD34+ cell counts were 4.65 × 109/ml Analysis of prognostic factors of mobilization (1.4–59.8) and 23/␮l (range, 1–340), respectively. Patients underwent a median of two (range, 1–8) apheresis pro- Patients’ characteristics were analyzed for factors that cedures which provided a median yield per patient of influenced CD34+ cell collection including age, stage of 164.15 × 108 total nucleated cells (120.5–1489.8) and disease at the time of HDC, bone marrow involvement, 8.7 × 106 (2.11–73.5) CD34+ cells/kg. There was a strong number of prior regimens and chemotherapy cycles, prior correlation between the number of preleukapheresis circul- radiotherapy, and order of mobilization. Only the last was ating CD34+ cells/␮l and the CD34+ cells per kilogram in associated with worse PBPC yields and was a predictive the apheresis collection (r = 0.82, P Ͻ 0.001). All patients factor for failure to collect at least 5 × 106 CD34+ cells reached a minimum collection of 2 × 106 CD34+ cells/kg. (Table 3). Sixty-two per cent of patients collected у2.5 × 106 and 52% у5 × 106 CD34+ cells/kg with the first apheresis Engraftment kinetics after PBPCT (Figures 1 and 2). The number of CD34+ cells collected on the first day of apheresis was superior to the second or All but one patient who progressed during induction pro- subsequent days. Comparisons between the primary mobil- ceeded to high-dose chemotherapy and PBPCT. In 25 ization and salvage mobilization groups are shown in patients who received the selected CD34+ PBPC, the Table 2. There was a significant difference in the number median day to reach an absolute neutrophil count of of CD34+ cells collected between the group that had CP as 0.5 × 109/l and unsustained platelet count of 20 × 109/l were PBPC mobilization with cyclophosphamide/paclitaxel JL Klein et al 962 Table 2 Collection results by order of mobilization

Overall Initial mobilization Salvage mobilization

No. aphereses 2 (1–8) 2 (1–5) 2 (1–8) CD34 × 106/kg 8.72 (2.11–73.5) 16.93 (2.69–73.5) 3.94 (2.11–19.8) CD34 × 106/kg/apheresis 3.97 (0.3–36.75) 7.37 (0.89–36.75) 1.65 (0.3–19.8) First apheresis CD34+ cells × 106/kg 3.16 (0.03–44.4) 7.35 (0.32–44.4) 1.05 (0.03–19.8) Collection of у2.5 × 106 CD34+ cells/kg 60/64 (93%) 35/35 (100%) 24/29 (83%) Collection of у5 × 106 CD34+ cells/kg 46/64 (72%) 33/35 (94.3%) 13/29 (45%)

Table 3 Regression analysis of prognostic factors influencing the total fulfilled another important requirement of a mobilization number of CD34+ cells/kg collected protocol, namely, predictable timing eliminating one of the criticisms of the use of chemotherapy as mobilization. This Factor r P is influenced by the amount of previous chemotherapy and radiation treatment. CP accomplishes both safe and predict- Ͻ Timing (first or salvage mobilization) 0.56 0.001 able mobilization in our poor prognostic group for mobiliz- Age −0.155 0.15 Bone marrow invasion −0.17 0.2 ation. In addition, leukapheresis could be started on day 10 No. of prior CT cycles 0.16 0.2 or 11 in 90% of patients as first attempt. There is increasing evidence for faster recovery with higher doses of PBPCs infused.15,26,27 This emphasizes the practical importance of effective mobilization in breast 9 (range, 8–11) and 17 (range, 13–29), respectively, after cancer patients in order to improve the CD34+ cell the PBPC infusion. For patients receiving the unselected yields.5,11–14 However, even with no or limited prior anti- PBPC infusion the median days to reach an absolute neutro- neoplastic treatment, the number of PBPCs collected is phil count of 0.5 × 109/l and platelet count of 20 × 109/l variable among patients and 10% of them will fail to mobil- were 9 (range, 8–11) and 16 (range, 11–60), respectively, ize successfully. There is no agreement on the best regimen after the PBPC infusion. after initial failure to mobilize PBPCs by cytokines. Higher doses of filgrastim, cocktails of growth factors or chemotherapy and growth factors provided mobilization rates Toxicity of the regimen below 50%.16 In our series, CP successfully salvaged the Nine patients (14%) required admission because of febrile 29 patients who failed a first mobilization with filgrastim neutropenia (two catheter infections caused by S. epider- alone. These results are more important if we consider the midis and by S. aureus, one patient presented with sinusitis adverse prognostic factors for mobilization exhibited in our and bronchitis and six patients had no documented bacterio- series, such as a large amount of prior chemotherapy exten- logical or clinical infection). No differences were observed sive bone disease, marrow involvement and prior radio- between first and salvage mobilizations. No procedure- therapy.5,11–14,23 Interestingly, none of these factors were related deaths occurred. significantly worse in salvage mobilizations than in patients mobilized as first attempt. With initial CP in the current study, all but two patients reached the target of 5 × 106 Discussion CD34+ cells/kg with a median of two (1–7) collections, as compared with 13/29 patients who received salvage CP. Paclitaxel is now a well-established agent in the treatment The yield of CD34+ cells and average CD34+ of breast cancer, with significant activity even in doxorub- cells/kg/apheresis collected were significantly lower icin-resistant tumors.18 Paclitaxel causes less thrombocyto- (P Ͻ 0.001) in the salvage patients. Moreover, the only penia, does not have cumulative hematological toxicity and prognostic factor associated with the number of CD34+ seems to spare the pool of primitive stem cells.19 These cells collected per apheresis after CP was the failure to factors make it an optimal tool to mobilize PBPCs in breast mobilize adequate PBPCs with filgrastim alone. This infor- cancer patients. Paclitaxel has been used alone20 or com- mation suggests that various factors affect the mobilization bined with either CP9,10,21–23 or epirubicin24,25 for PBPC of PBPC whether the patients received growth factor alone mobilization. CP has been tested with different schedules or CP plus growth factor. Because of the success of this and doses, ranging for paclitaxel between 150–300 mg/m2 regimen for patients on our CD34+ cell selection protocol and 2–4 g/m2 for CP. Paclitaxel in combination with CP we elected to use the CP mobilization rather than a second improves the mobilization of PBPCs vs CP alone, without course of higher dose filgrastim. We were successful in col- a major increase in toxicity, particularly febrile neutropenia. lecting an adequate PBPC product in all patients salvaged We have used a lower dose of CP (2 g/m2) in order to dim- with this regimen. The PBPC product was able to provide a inish the associated toxicity without compromising the rapid hematological recovery after transplant in all patients. efficacy. The resulting combination provided successful We have used a lower dose of CP than was originally mobilizations in all of the patients to reach a dose of 2 × 106 reported for mobilization with paclitaxel and CP.5 There CD34+ cells/kg, as our results and others confirm.10 CP also were no episodes of hematuria associated with our regimen PBPC mobilization with cyclophosphamide/paclitaxel JL Klein et al 963 and the incidence of febrile neutropenia (14%) was lower 12 Haas R, Mo¨hle R, Fruëhauf S. Patient characteristics associa- than reported with CP (4 g/m2) and paclitaxel ted with successful mobilizing and autografting of peripheral (170 mg/m2),5 although this may have been related to the blood progenitor cells in malignant lymphoma. Blood 1994; use of MESNA and prophylactic antibiotics during the 83: 3787–3794. neutropenic period. 13 Kotasek D, Shepherd KM, Sage RE et al. Factors affecting blood stem cell collections following high-dose cyclophos- In summary, CP is a highly effective method for mobiliz- phamide mobilization in lymphoma, myeloma and solid ing PBPCs, both as initial approach and as a salvage tumors. Bone Marrow Transplant 1992; 9: 11–17. maneuver after failure to yield adequate PBPC product with 14 Schwartzberg LS, Birch R, Hazelten B et al. Peripheral blood filgrastim alone. It provided a sufficient number of hemato- stem cell mobilization by chemotherapy with and without r- poietic progenitor cells to support HDC in all patients, sug- h-filgrastim. J Hematother 1992; 1: 317–327. gesting a higher efficacy of CP plus filgrastim over filgras- 15 Weaver CH, Hazelton B, Birch R et al. An analysis of tim alone. engraftment kinetics as a function of the CD34 content of peripheral blood progenitor cell collections in 692 patients after the administration of myeloablative chemotherapy. Blood References 1995; 86: 3961–3969. 16 Weaver CH, Tauer K, Zhen B et al. Second attempts at mobilization of peripheral blood stem cells in patients with initial 1 Henon R, Liang H, Beck-Wirth G et al. Comparison of hema- low CD34+ cell yields. J Hematother 1998; 7: 241–249. topoietic and immune recovery after autologous bone marrow 17 Sutherland DR, Anderson L, Keeney M et al. The ISHAGE or blood stem cell transplants. Bone Marrow Transplant 1992; guidelines for CD34+ cell determination by flow cytometry. 9: 285–291. J Hematother 1996; 5: 213–226. 2 To LB, Roberts MM, Haylock DN et al. Comparison of haematological recovery times and supportive care require- 18 Seidman AD, Crown JPA, Reichman B et al. Lack of clinical ments of autologous recovery phase peripheral blood stem cell cross resistance of paclitaxel with anthracycline in the treat- transplants, autologous bone marrow transplants and allo- ment of metastatic breast cancer. Proc Am Soc Clin Oncol geneic bone marrow transplants. Bone Marrow Transplant 1993; 12: 63 (Abstr. 53). 1992; 9: 277–284. 19 Shea TC. Mobilization of peripheral blood progenitor cells 3Lo´pez M, Mortel O, Pouillart P et al. Acceleration of hemato- with paclitaxel-based chemotherapy. Semin Oncol 1997; 107 poietic recovery after autologous bone marrow transplantation (Suppl. 2): 105–107. by low doses of peripheral blood stem cells. Bone Marrow 20 Holmberg LA, Demirer T, Rowley S et al. High-dose busul- Transplant 1991; 7: 173–181. fan, melphalan and thiotepa followed by autologous peripheral 4 Peters WP, Rosner G, Ross M et al. Comparative effects of blood stem cell (PBSC) rescue in patients with advanced stage GM-CSF and G-CSF on priming peripheral blood progenitor III/IV ovarian cancer. Bone Marrow Transplant 1998; 22: cells for use with autologous bone marrow after high dose 651–659. chemotherapy. Blood 1993; 81: 1709–1719. 21 Raptis G, Tiersten A, Fennelly D et al. The addition of paclit- + 5 Bensinger W, Appelbaum F, Rowley S et al. Factors that axel to high dose cyclophosphamide G-CSF markedly + influence collection and engraftment of autologous peripheral- enhances the mobilization of CD34 peripheral blood progeni- blood stem cells. J Clin Oncol 1995; 13: 2547–2555. tor cells in patients with responding metastatic breast cancer 6 Alegre A, Tomas JF, Martinez-Chamorro C et al. Comparison undergoing high dose chemotherapy. Proc Am Soc Clin Oncol of peripheral blood progenitor cell mobilization in patients 1995; 14: 313 (Abstr.). with multiple myeloma: high-dose cyclophophamide plus 22 Pedrazzoli P, Perotti C, Da Prada GA et al. Collection of cir- GM-CSF vs G-CSF alone. Bone Marrow Transplant 1997; 20: culating progenitor cells after epirubicin, paclitaxel and G- 211–217. CSF in patients with metastatic breast cancer. Br J Cancer 7 Venturini M, DelMartro L, Melioli G. Release of peripheral 1997; 75: 1368–1372. blood progenitor cells during standard dose cyclophospham- 23 Demirer T, Buckner D, Storer B et al. Effect of different ide, epidoxorubicin, 5-fluorouracil regimen plus granulocyte chemotherapy regimens on peripheral-blood stem-cell collec- colony stimulating factor for breast cancer therapy. Cancer tions in patients with breast cancer receiving granulocyte col- 1994; 74: 2300–2306. ony-stimulating factor. J Clin Oncol 1997; 15: 684–690. 8 Olivieri A, Offidani M, Ciniero L et al. DHAP regimen plus 24 Conte PF, Gennari A, Salvadori B et al. Paclitaxel plus epirub- G-CSF as salvage therapy and priming for blood progenitor icin in advanced breast cancer. Oncology 1998; 12 (Suppl. 1): cell collection in patients with poor prognosis lymphoma. 40–44. Bone Marrow Transplant 1995; 16: 85–93. 25 Bengala C, Pazzagli I, Tibaldi C et al. Mobilization, collec- 9 Demirer T, Rowley S, Buckner CD et al. Peripheral-blood tion, and characterization of peripheral blood hemopoietic pro- stem-cell collections after paclitaxel, cyclophosphamide, and genitors after chemotherapy with epirubicin, paclitaxel, and recombinant human granulocyte colony-stimulating factor in granulocyte-colony stimulating factors administered to patients with breast and ovarian cancer. J Clin Oncol 1995; patients with metastatic breast carcinoma. Cancer 1998; 82: 13: 1714–1719. 867–873. 10 Weaver CH, Schwartzberg LS, Birch R et al. Collection of 26 Watts MJ, Sullivan AM, Jamieson E et al. Progenitor-cell peripheral blood stem cells following administration of paclit- mobilization after low-dose cyclophosphamide and granulo- axel, cyclophosphamide and G-CSF in patients with breast and cyte colony-stimulating factor: an analysis of progenitor-cell ovarian cancer. Biol Blood Marrow Transplant 1997; 3: 83– quantity and quality and factors predicting for these para- 90. meters in 101 pretreated patients with malignant lymphoma. 11 Bensinger WI, Longin K, Appelbaum F et al. Peripheral blood J Clin Oncol 1997; 15: 535–546. stem cells collected after recombinant granulocyte colony sti- 27 Pecora AL, Preti RA, Gleim GW et al. CD34+CD33− cells mulating factor: an analysis of factors correlating with the influence days to engraftment and transfusion requirements in tempo of engraftment after transplantation. Br J Haematol autologous blood stem-cell recipients. J Clin Oncol 1998; 16: 1994; 87: 825–831. 2093–2104.