Chemotherapy-induced mobilization of karyotypically normal PBSC for autografting in CML

Transcription

1 Bone Marrow Transplantation, (1998) 21, Stockton Press All rights reserved /98 $ Chemotherapy-induced mobilization of karyotypically normal PBSC for autografting in CML T Fischer 1, A Neubauer 2, J Mohm 2, D Huhn 3, C Busemann 3, H Link 4, L Arseniev 4,BBüssing 4, J Novotny 4, A Ganser 4, J Duyster 5, D Bunjes 5, S Kreiter 1, W Aulitzky 1, R Hehlmann 6 and C Huber 1 1 III Medizinische Klinik, Johannes-Gutenberg-Universität, Mainz; 2 Medizinische Klinik I, Technische Universität, Dresden; 3 Virchow- Klinikum, Abteilung für Innere Medizin und Poliklinik, Humboldt-Universität, Berlin; 4 Medizinische Hochschule Hannover, Abteilung Hämatologie und Onkologie; 5 Universität Ulm, Abteilung Innere Medizin III; and 6 III Medizinische Klinik, Klinikum Mannheim, Universität Heidelberg, Germany Summary: High-dose chemotherapy with autologous transplantation of in vivo purged PBSC is a new and interesting therapeutic option for CML patients not eligible for allogeneic transplantation. We investigated the feasibility and toxicity of this approach in 57 patients with Ph-positive CML. For mobilization of Ph-negative PBSC, patients were treated either with 5 + 2/ type chemotherapy or with mini-ice/ice chemotherapy followed by administration of G-CSF. Fourteen patients were in early chronic phase, 30 patients in late chronic phase and 13 patients in accelerated phase (AP) or blast crisis (BC). Cytogenetic responses in the PBSC harvests were dependent on both disease stage and type of chemotherapy: in late chronic phase and AP/BC, a complete or major cytogenetic response could be obtained in nine out of 13 patients treated with mini- ICE/ICE but only in three out of 23 patients treated with 5 + 2/7 + 3 chemotherapy. However, in early chronic phase a Ph-negative autograft could be obtained in three out of eight patients upon mobilization with chemotherapy. Thirty-one patients underwent PBSC transplantation and all of them successfully engrafted. Post-transplant cytogenetic analysis was available on 21 cases, of whom seven achieved a complete or major cytogenetic response, with two minor cytogenetic remissions. One patient (1/57) in blast crisis died during mobilization therapy (1.8%). Transplantation related mortality was 0%. This study demonstrates that mobilization of Ph-negative PBSC after myelosuppressive chemotherapy is feasible in CML patients and is associated with acceptable toxicity. Autologous transplantation of in vivo purged PBSC is a safe procedure with rapid and complete hematopietic recovery. Keywords: CML; in vivo purging; autologous transplantation Correspondence: Dr T Fischer, University of Mainz, Medical Clinic III, Langenbeckstr 1, Mainz, Germany Received 16 September 1997; accepted 18 January 1998 Chronic myeloid leukemia (CML) is a clonal hematopoietic disease resulting from malignant transformation and expansion of a single pluripotent progenitor cell. It is characterized pathogenetically by the Philadelphia chromosome (Ph) translocation. Clinically, CML has a grave prognosis since the initial indolent chronic phase evolves into an accelerated phase and eventually into acute blastic transformation after a median of 5 years. Treatment of CML represents a therapeutic dilemma; conventional therapy with hydroxyurea and interferon- (IFN- ) prolongs survival but does not cure. In addition, long-term treatment with interferon may be cumbersome due to side-effects and the need for injections. In contrast allogeneic bone marrow transplantation (BMT) offers the prospect of cure in CML, but only 30% of all patients with CML are eligible for BMT and this modality is associated with significant morbidity and early mortality. Overall, it appears that current therapeutic approaches are not satisfactory for many patients. Therefore there is a strong need for novel therapeutic strategies to improve survival in CML. Among new treatment modalities, autologous PBSC transplantation is an interesting investigational approach. Present studies suggest a potential survival benefit for patients treated with this modality. 1 4 However, a major problem is the massive contamination of blood and marrow with malignant cells in CML. Only a minor fraction of the hematopoietic progenitor cells do not have the Ph-chromosome. Since transfusion of Ph-positive cells present in the autograft is at least in part responsible for recurrence of CML after autotransplant, methods to select for Ph-negative stem cells are required. Recently, in vivo purging of PBSC with chemotherapy has been proposed as a suitable experimental approach. 5 8 We have investigated this method in a cohort of patients with chronic phase CML and in patients with accelerated phase or blast crisis using either a 5 + 2/ type chemotherapy or mini-ice/ice chemotherapy followed by application of G-CSF. The and chemotherapy regimens represent combinations of a relatively low dose of cytarabine with an anthracycline whereas mini- ICE and ICE chemotherapy consist of an intermediate dose of cytarabine plus an anthracycline plus etoposide.

2 1030 Patients and methods Patients Between April 1995 and July 1997, 57 patients with Phpositive CML were entered into this prospective, multicenter pilot study (Table 1). The majority of patients had received prior IFN- therapy with a mean duration of 15 months (range 1 85 months). These patients were cytogenetically refractory to IFN- therapy (Table 1) or had to stop IFN- treatment because of intolerable side-effects. Other approaches of premobilization chemotherapy, cytogenetic responses prior to PBSC mobilization, Kantarjian s index at mobilization, and where available, Sokal s index at diagnosis are also depicted in Table 1. The study was performed according to the Declaration of Helsinki. The protocol was approved by the local Ethics Committees. Patients were recruited from five centers in Germany. The median age was 49 years (range years). Forty-four patients were in first chronic phase (CP1) and 13 patients were in accelerated phase or blast crisis. For patients in first chronic phase the duration of CP1 varied widely, with a mean duration of 29.5 months (range months). Mobilization therapy and harvesting Four chemotherapy regimens were used as mobilizing chemotherapy (Table 2): : cytarabine 100 mg/m 2 /day by continuous i.v. infusion over 24 h, days 1 5; idarubicin 12 mg/m 2 /day, i.v. days 1 2; : cytarabine 200 mg/m 2 /day, by continuous i.v. infusion over 24 h, days 1 7; daunorubicin 50 mg/m 2 /day, i.v., days 1 3; mini-ice : cytarabine 800 mg/m 2 /day, i.v., days 1 3; idarubicin 8 mg/m 2 /day, i.v., days 1 3; etoposide 150 mg/m 2 /day, i.v. days 1 3; 2 ICE : cytarabine 800 mg/m 2 /day, i.v., days 1 5; idarubicin 8 mg/m 2 /day, i.v., days 1 5; etoposide 150 mg/m 2 /day, i.v., days Each center had selected one of the treatment arms as active protocol. This represented the sole selection process of patients to one of the chemotherapy protocols. Patients received granulocyte colony stimulating factor (G-CSF) 5 10 g/kg/day from day +8 following chemotherapy to the end of leukaphereses. Eight patients received a second cycle of mobilizing chemotherapy followed by application of G-CSF (5 10 g/kg/day) plus IL-3 (5 g/kg). 10 In two patients this was done to improve the cytogenetic response in leukapheresis collections. Six patients received the combination of G-CSF plus IL-3 because of insufficient mobilization of CD34- positive cells after the first cycle of chemotherapy. In two out of six patients successful mobilization of CD34-positive cells could be documented upon administration of G-CSF plus IL-3. PBSC harvesting was performed using either a COBE Spectra (Cobe Laboratories, Lakewood, CO, USA) or Baxter CS3000+ (Baxter Deutschland, Unterschleißheim, Germany) continuous flow centrifuge. Leukaphereses were started when the absolute leukocyte count reached 800 to 1000/ l and when a distinct population of CD34-positive cells ( 2/ l) could be identified by immunofluorescence analysis. Additional harvests were performed on subsequent days and leukapheresis products were cryopreserved in DMSO and stored in liquid nitrogen. Samples were taken from each individual leukapheresis prior to cryopreservation for the analysis of CFU-GM colonies and cytogenetics. In the majority of patients, back-up PBSC harvests were taken before mobilization chemotherapy and cryopreserved. PBSC transplantation A total of 31 patients subsequently underwent autologous PBSC transplantation between April 1995 and July Patients not transplanted in this period, received autografting thereafter. Twenty-two patients were transplanted in CP1 and nine patients were transplanted in accelerated phase or blast crisis. Nine patients were conditioned using a standard Bu/Cy regimen (busulphan: total dose 16 mg/kg plus cyclophosphamide: total dose 120 mg/kg), nine received Bu/Mel (busulphan: total dose 16 mg/kg plus melphalan: total dose 140 mg/m 2 ), seven received busulphan alone (total dose 16 mg/kg) and six cyclophosphamide (total dose 120 mg/kg) plus TBI G-CSF was given at a dose of 5 g/kg from day +8 following reinfusion of stem cells. Upon hematopoietic recovery the majority of patients received IFN- therapy. Response criteria The definition of CML phases and of hematologic responses followed the Houston criteria. 16,17 A cytogenetic response was considered as complete (0% Ph-positive metaphases = CCR), major (1 34% Ph-positive metaphases = MCR) or minor (35 95% Ph-positive metaphases = mcr). More than 95% Ph-positive metaphases were considered as no response (NR). 16,17 Analysis of PBSC harvests Leukapheresis collections were considered successful when the following criteria were met: CD34-positive cells /kg and CFU-GM /kg. Cytogenetic analysis (G-banding) was carried out on metaphases prepared from fresh PBSC samples. Where possible, at least 20 metaphases were analyzed from each harvest sample. Overall cytogenetic responses of PBSC autografts were assessed according to the cumulative number of Ph-negative metaphases in the leukapheresis samples needed to meet the cumulative number of CD34-positive progenitors /kg and of CFU-GM /kg. For example, the patient with UPN 19 (Table 3) received three leukapheresis collections containing a total of CD34-positive cells. Cytogenetic analysis revealed that PBSC harvests LP1 and LP2 (Table 3) were Ph-negative (0/22 and 0/14 Ph-positive metaphases, respectively) whereas in LP3 3/24 metaphases were Ph-positive. Since LP1 and LP2 contained sufficient CD34-positive cells ( ) and CFU-GM to meet the above criteria this patient could receive a Ph-negative autograft for transplantation. In several patients cytogenetic analysis was not performed in all leukapheresis samples. In these patients, cytogenetic analysis was considered either as not done or as NR if analysis of the first leukapheresis sample showed NR.

3 Table 1 Patient characteristics 1031 UPN Age/ Diagnosis to Stage Previous Cytogenetic Kantarjian 1 / Cytogenetic Mobilization sex mobilization of therapy response on Sokal 2 response in chemotherapy (months) disease (months on IFN (% Ph+) index autograft IFN) 1 28/F 120 CP HU, IFN (9) ND 4 1 NR /M 84 AP HU, IFN (6) MR mini-ice 3 54/F 65 CP HUF, IFN (38) mr mini-ice 4 56/M 64 CP HU, IFN (60) MR mini-ice 5 34/F 37 CP HU, IFN (1) no mobil /M 34 CP HU, IFN (22) mr /F 30 CP HU, IFN (18) mr /F 29 CP HU, IFN (37) no mobil /F 37 CP HU, IFN (14) ND 1 1 MR mini-ice 10 24/M 20 CP HU 100 a 1 1 mr /M 35 CP HU, IFN (5) MR mini-ice 12 56/M 15 CP HU, IFN (1) 73 a 1 1 MR /M 22 AP HU, IFN (19) mr /M 12 CP HU, IFN (9) / CR /M 5 CP HU, IFN (13) mr /F 14 BC HU, IFN (12) CR mini-ice 17 37/M 10 CP HU ND 4 1 MR /M 8 CP HU 100 a 1 1 / CR /M 10 CP HU, IFN (3) ND 1 1 / CR /F 7 CP HU 100 a 1 1 / no mobil /M 5 CP HU, IFN (4) mr /M 6 CP HU, IFN (4) / mr /F 2 CP HU, CA 100 a 1 1 / no mobil. mini-ice 24 54/M 54 CP HU, IFN (49) no mobil. mini-ice 25 54/M 52 CP HU, IFN (40) CR mini-ice 26 53/F 9 CP HU, IFN (8) / no mobil. mini-ice 27 57/M 3 CP HU 100 a 1 1 mr mini-ice 28 35/M 89 CP Bu, IFN (85) ND /M 48 CP HU, IFN (16) ND /F 40 CP HU, CA, IFN (19) ND /M 24 CP Bu, HU, IFN (10) NR /F 180 CP Bu, HU, IFN 100 NA NR /F 61 CP HU, IFN (54) 57 NA no mobil /F 32 CP HU ND 1 1 mr /F 3 CP HU, IFN (36) ND 1 1 MR /M 56 CP HU, IFN (55) ND /M 48 BC HU, CA, IFN 100 NA NR /M 34 CP HU 100 a 1 1 MR mini-ice 39 20/F 29 CP HU, IFN (23) ND /F 175 BC Bu, HU, IFN (2) 100 a 3 1 death mini-ice 41 49/F 98 AP Bu, HU 100 a 4 1 NR mini-ice 42 32/F 4 CP HU 100 a 1 1 / CR mini-ice 43 45/F 2 CP HU 100 a 1 1 / mr mini-ice 44 58/M 13 CP HU 100 a 1 1 / mr mini-ice 45 40/F 7 AP Bu, HU, IFN (8) MR /M 42 AP HU 100 a 4 1 mr /M 50 AP HU, CA, IFN (1) 100 a 1 1 NR /F 74 BC Bu, HU, IFN NR /F 42 AP HU, CA, IFN (52) mr /M 7 AP HU 100 a 4 1 NR /M 44 CP HU, CA, IFN (13) NR /F 24 CP HU, IFN (23) mr /F 25 BC HU, IFN (16) NR /M 24 CP HU, IFN NA NA NA ICE 55 51/M 46 CP NA NA NA CR ICE 56 55/M 6 CP HU NA NA MR ICE 57 35/M 54 CP NA NA NA CR ICE UPN = unique patient number; CP = chronic phase; AP = accelerated phase; BC = blast crisis; NR = no response; mr = minor cytogenetic response; MR = major cytogenetic response; CR = complete cytogenetic response; no mobil = no mobilization of CD34-positive cells; LP = leukapheresis; NA = not available; ND = not done; HU = hydroxyurea; CA = cytosine-arabinoside; Bu = busulfan. a Cytogenetic response (prior to mobilization) of patients who have not received a trial of IFN-.

4 1032 Table 2 Hematopoietic recovery and toxicity following PBSC mobilization Chemotherapy Patients WBC PLT Fever (d) Antibiotics Ampho-B TC (U) EC (U) Mortality n = (T 38.5 C) (d) median (d) median median median (n) (d) median (d) median median (range) (range) (range) (range) (range) (range) (range) mini ICE/ICE (11 22) (13 44) (0 35) (0 47) (0 31) (0 26) (0 19) 7+3/ (0 27) (0 43) (0 13) (0 27) (0 17) (0 22) (0 30) PLT = platelets; ampho-b = amphotericin-b; TC = thrombocyte concentrate; EC = erythrocyte concentrate; U = units; d = day. Statistical analysis For comparison between groups of patients, Fisher s test was applied (Tables 4 and 5). Probability of survival of transplanted patients was estimated according to the method of Kaplan Meier. 18 Results Mobilization of PBSC For mobilization of Ph-negative PBSC we used four chemotherapy regimens which can be separated into two groups according to the type and dose of chemotherapy applied: and chemotherapy regimens represent combinations of a relatively low dose of cytarabine with an anthracycline, whereas mini-ice and ICE chemotherapy consist of an intermediate dose of cytarabine plus an antracycline plus etoposide. Hematopoietic recovery following mobilizing chemotherapy was related to dose of chemotherapy applied (Table 2). Recovery of the WBC to /l occurred at a median of 17 days (range 11 22) and 15 days (range 0 23) following the end of mini- ICE/ICE and 7 + 3/5 + 2, respectively (Table 2). Table 2 gives an overview on hematopoietic recovery and toxicity. One patient in blast crisis died following mobilization chemotherapy due to sepsis giving an overall mortality of 1.8% (1/57 patients). Upon recovery of the WBC 800/ l monitoring of CD34-positive cells in the peripheral blood was commenced. Leukaphereses were started when 2/ l CD34- positive cells could be identified. The median number of mononuclear cells (MNC) collected was (range ) following 5 + 2/ type chemotherapy and (range ) after mini-ice/ice. The median number of harvests performed was four (range 1 11) and three (range 2 8), respectively. Following 5 + 2/ type chemotherapy, CD34-positive cells (median ) and CFU-GM (median ) could be harvested. Upon mobilization with mini-ice/ice, CD34-positive cells (median ) and CFU-GM (median ) could be collected. Table 3 shows the cytogenetic responses in PBSC collections. In 50/57 patients PBSC harvests were evaluated genetically. A complete cytogenetic response was documented in 8/50 patients, with a further 10/50 showing a major cytogenetic response. The overall complete and major cytogenetic response rate was 36% (18/50 patients). Our results show that the probability to mobilize cytogenetically improved grafts depends on both disease stage and type of chemotherapy: Using 5 + 2/7 + 3 chemotherapy, a complete cytogenetic response was observed in 38% of patients treated in early chronic phase ( 12 months post-diagnosis) compared with 0% in patients treated later in chronic phase (P = 0.049) (Table 4). Late chronic phase and AP/BC were refractory to 5 + 2/7 + 3 but were sensitive to mini- ICE/ICE chemotherapy. 19 Table 5 illustrates that a CCR and MCR was observed in 13% of patients treated with 5 + 2/7 + 3 vs 69% in patients treated with mini-ice/ice (P = ). One patient who underwent PBSC mobilization in CP1 and one in BC died at 7 and 8 months post-mobilization because of uncontrolled blast crisis. PBSC transplantation Thirty-one patients underwent PBSC transplantation and all of them successfully engrafted. Twenty-two patients were transplanted in CP1 whereas nine patients were in accelerated phase or blast crisis. According to our protocol, patients could proceed to autografting 8 weeks post-pbsc mobilization. There was no additional selection process for patients transplanted in chronic phase. Pretransplant cytogenetic analysis has not been performed routinely due to the short interval between PBSC mobilization and transplantation. Median time to achieve a white blood cell (WBC) count of /l was 12 days (range 6 20) and to regenerate platelets to /l was 20 days (range ). Table 6 demonstrates hematopoietic recovery and toxicity following PBSC transplantation. No patient died of transplant-related complications. The outcome of transplantation has been updated at a median follow-up for surviving patients of 12 months (range 4 32) for patients transplanted in CP1 and of 11 months (range 7 19) for patients transplanted in AP/BC. Nineteen patients transplanted in CP1 are alive and well, either in hematologic remission or in chronic phase (Figure 1). Three patients transplanted in CP1 died at 10, 14 and 21 months post-transplant due to transformation to blast crisis. Four patients transplanted in accelerated phase and blast crisis are in complete hematologic remission or chronic phase, one patient relapsed into blast crisis, one patient received MUD transplantation because of relapse and on

5 Table 3 Cytogenetic studies on harvested cells (per cent Ph-positive cells in each harvest) 1033 UPN LP1 LP2 LP3 LP4 LP5 LP6 LP7 LP8 LP9 LP10 LP11 Overall cytogenetic response 1 100% NR 2 NE 29% NE NE MR 3 NE 80% NE mr 4 NE 10% MR 5 NE no mobil. 6 NE 93% NE mr 7 NE 60% ND ND ND ND mr 8 NE no mobil. 9 5% NA 0% MR 10 33% 44% 0% 0% 80% 86% mr 11 20% 16% MR 12 ND 35% NE NE 30% NE NE NE MR 13 ND ND ND 73% ND 100% mr % 0% 0% CR 15 13% 46% 40% 68% 70% 66% 100% 85% 95% NE NE mr 16 0% 0% CR 17 3% 5% 6% 4% 0% 3% MR 18 21% 0% CR 19 0% 0% 12.5% CR 20 NE no mobil. 21 NE 92% 75% 69% 83% mr 22 NE 90% 93% 91% mr 23 NE no mobil. 24 NE no mobil. 25 0% NE 0% CR 26 NE no mobil. 27 NE 31% 51% mr 28 40% ND ND ND ND 29 30% ND 31% ND ND 30 ND 87.5% ND ND ND % ND ND NR % ND ND NR 33 NE no mobil. 34 ND ND ND 76% ND mr 35 33% 17% ND MR 36 ND ND ND ND ND ND ND % NR 38 NE 32% MR 39 ND ND ND ND ND ND 40 NE death 41 ND 100% NR 42 0% CR 43 0% 78% 5% ND 13% 43% 69% ND mr 44 ND 43% ND ND mr 45 0% 40% 33% MR 46 14% 19% 90% mr % 100% 72% NR % 100% 100% NR 49 NE 10% 30% 40% mr % 95% NR % 100% 100% 100% 100% 90% NR 52 NE 100% 40% mr % 100% 100% 100% 100% NR 54 NA NA NA NA NA 55 0% 5% 29% CR 56 11% 0% 5% 15% MR 57 0% 0% 0% 0% 5% CR UPN = unique patient number; NR = no response; mr = minor cytogenetic response; MR = major cytogenetic response; CR = complete cytogenetic response; no mobil. = no mobilization of CD34-positive cells; LP = leukapheresis; NE = not evaluable; ND = not done. one patient data on current hematologic response are not available. Two patients died 7 and 10 months post-transplant due to uncontrolled blast crisis. Currently there is only limited data on the cytogenetic status of the patients posttransplant. Results are currently available on bone marrow samples of 21 patients analyzed at a median of 6 months (range 2 20) post-transplant. In three patients there was a complete cytogenetic response and in four patients a major

6 1034 Table 4 Effect of duration of first chronic phase on outcome of PBSC mobilization with 5 + 2/ type chemotherapy Cytogenetic response CP1 12 CP1 12 months months Complete (%) 3 (38) 0 (0) Major/minor/nil (%) 5 (62) 14 (100) Total 8 14 (P = 0.049) Table 5 Effect of type of chemotherapy on outcome of PBSC mobilization in late CP and AP/BC of CML Cytogenetic 5+2/7+3 ICE/mini-ICE response Complete/major (%) 3 (13) 9 (69) Minor/nil (%) 20 (87) 4 (31) Total (P = ) % Surviving Months from transplantation Figure 1 Survival of chronic phase CML patients autografted with in vivo purged PBSC. Probability of survival of transplanted patients was estimated according to the method of Kaplan Meier. cytogenetic response, accounting for a rate of 33% significant cytogenetic responses. Another two patients had a minor cytogenetic response and 12 patients did not show any Ph-negative metaphases post-transplant. Discussion Our results confirm previous reports demonstrating that mobilization of Ph-negative PBSC with myelosuppressive chemotherapy and autologous transplantation is feasible in CML. In addition we addressed the question of the impact of dose intensity of chemotherapy on graft quality. For mobilization of PBSC we used a low dose or an intermediate dose chemotherapy regimen. Both types of chemotherapy protocols showed acceptable and comparable toxicity. Recovery of WBC was slightly shorter following 5 + 2/7 + 3 chemotherapy but this was not associated with an advantage in other clinical endpoints as median number of days with T 38.5 C, median number of days on i.v. antibiotics or median number of blood products transfused. The majority of patients included in this study were cytogenetically refractory to IFN- and were in late chronic phase (median 30 months) or AP/BC. This needs to be taken into consideration when interpreting our results. Thus, the overall complete and major cytogenetic response rate of 36% was observed in a selected group of high risk patients. Our data show that the outcome of PBSC mobilization is dependent on both disease stage and type of chemotherapy. Using 5 + 2/7 + 3 chemotherapy we demonstrated that duration of first chronic phase is an important predictor for successful mobilization of Ph-negative stem cells. This is in accordance with data of two studies on mobilization of Ph-negative stem cells in CML. 3,4 In early chronic phase a significant cytogenetic response (CCR and MCR) could be obtained in 50% of cases. It is important to note that the rate of complete and major cytogenetic responses in early chronic phase has been described to be significantly better upon using mini-ice/ice chemotherapy or higher doses of cytarabine and idarubicin. 5,6,9 Late chronic phase and AP/BC were largely refractory to 5 + 2/ type chemotherapy with a complete response rate of 0% and a major response rate of only 13% but were sensitive (69% of patients with a CCR and MCR) to mini- ICE/ICE chemotherapy. This difference was highly significant in the statistical analysis. Thus, for mobilization of PBSC in CML mini-ice/ice -type chemotherapy appears to be superior to 5 + 2/7 + 3 chemotherapy. Genetic marking experiments showed that Ph-positive hematopoietic stem cells contribute to relapse post-autologous transplantation. 20 Therefore, on theoretical grounds it can be assumed that use of Ph-negative hematopoietic stem cells is advantageous. However cytogenetic remissions post-transplantation have been observed using 100% Phpositive bone marrow harvests. 21,22 Talpaz et al 23 showed that the chance of generating cytogenetic remissions posttransplant depends on the percentage of Ph-negative cells in the stem cell harvests. However, previous studies could not show a significant correlation of overall survival with the cytogenetic response post-transplant. 4 Table 6 Hematopoietic recovery and toxicity following PBSC transplantation WBC 1000 PLT Fever (d) Antibiotics Ampho-B TC (U) EC (U) Mortality (d) median (T 38.5 C) (d) median (d) median median (range) median (n) (range) (d) median median (range) (range) (range) (range) (range) (6 20) (10 123) (2 21) (0 28) (0 36) (0 28) (0 22) PLT = platelets; ampho-b = amphotericin-b; TC = thrombocyte concentrate; EC = erythrocyte concentrate; U = units; d = day.

7 Partial cytogenetic remissions in leukapheresis products may facilitate additional in vitro purging. Recently, a number of new procedures for in vitro purging have been investigated with interesting results. 24 Future studies will analyze a possible advantage of in vitro purging using long-term in vitro culture or immunofluorescence-based selection of Phnegative hematopoietic stem cells. Other approaches may employ specific inhibitors of the bcr/abl tyrosine kinase activity or bcr/abl targeting with antisense constructs. 24 We used various conditioning protocols before reinfusion of PBSC. According to the preliminary results of this and other studies there was no obvious difference regarding their efficacy in autologous transplantation. 12,13 Busulphan monotherapy presented low toxicity and was well tolerated. Cyclophosphamide does not appear to be particularly useful in the autologous setting since its application in allogeneic transplantation is primarily as an immunosuppressive agent. Relapses post-allogeneic transplantation have been observed in all published studies. 1,16 This shows that current conditioning regimens are not able completely to eliminate the malignant stem cell pool. There is considerable evidence that immune-mediated anti-leukemia effects are important in achieving cure of the disease. Therefore it appears that anti-leukemic maintenance therapy with IFN- or IL-2 is necessary post-autologous transplantation. 25 Hematopoietic recovery post-autologous transplantation was complete and in most patients as rapid as in autologous transplantation for other hematopoietic malignancies. Toxicity was acceptable and there was no transplant-related mortality. Longer follow-up will be required to draw conclusions about the significance of cytogenetic responses in leukapheresis collections on the post-transplant hematologic and cytogenetic response or on a potential prolongation of overall survival. However, it has been shown recently that the outcome of mobilization therapy correlates with the cytogenetic response post-transplant. 23 It can be hypothesized that this may result in prolongation of chronic phase by inhibition of clonal evolution. Current uncontrolled studies suggest a survival benefit for autologous transplantation in comparison to conventional therapy. However, it is certainly not possible to draw firm conclusions from this highly selected cohort of patients. Therefore, this question needs to be addressed in a randomized prospective phase III study which will be included in the forthcoming German CML IIIA trial. References 1 Butturini A, Keating A, Goldman J, Gale RP. Autotransplants in chronic myelogenous leukaemia: strategies and results. Lancet 1990; 335: Daley GQ, Goldman JM. Autologous transplant for CML revisited. Exp Hematol 1993; 21: Hoyle C, Gray R, Goldman J et al. Autografting for patients with CML in chronic phase: an update. Br J Haematol 1994; 86: McGlave PB, De Fabritiis P, Deisseroth A et al. Autologous transplants for chronic myelogenous leukaemia: results from eight transplant groups. Lancet 1994; 343: Carella A, Podesta M, Frassoni F et al. Collection of normal blood repopulating cells during early hemopoietic recovery after intensive conventional chemotherapy in chronic myelogenous leukemia. Bone Marrow Transplant 1993; 12: Chalmers EA, Franklin IM, Kelsey SM et al. Treatment of CML in first chronic phase with idarubicin and cytarabine: mobilization of Philadelphia-negative peripheral blood stem cells. Br J Haematol 1997; 96: Kantarjian HM, Talpaz M, Hester J et al. Collection of peripheral blood diploid cells from chronic myelogenous leukemia patients early in the recovery phase from myelosuppression induced by intensive-dose chemotherapy. J Clin Oncol 1995; 13: Mehta J, Mijcovic A, Powles R et al. Myelosuppressive chemotherapy to mobilize normal stem cells in chronic myeloid leukemia. Bone Marrow Transplant 1996; 17: Carella AM, Chimirri F, Podestà M et al. High-dose chemoradiotherapy followed by autologous Philadelphia chromosome-negative blood progenitor cell transplantation in patients with chronic myelogenous leukemia. Bone Marrow Transplant 1996; 17: Kolbe K, Peschel C, Rupilius B et al. Peripheral blood stem cell (PBSC) mobilization with chemotherapy followed by sequential IL-3 and G-CSF administration in extensively pretreated patients. Bone Marrow Transplant 1997; 20: Cony-Makhoul P, Marit G, Boiron JM et al. Busulphan and melphalan prior to autologous transplantation for myeloid malignancies. Bone Marrow Transplant 1995; 16: O Brien SG, Szydlo R, Rule SA et al. Busulphan only as cytoreduction prior to autografting for CML. Bone Marrow Transplant 1996; 17 (Suppl. 1): 315 (S68). 13 Reiffers J, Goldman J, Meloni G, Cahn JY. Autologous stem cell transplantation (ASCT) for chronic myeloid leukemia (CML) in chronic phase: a report of the European Blood and Marrow Transplant (EBMT) Group. Bone Marrow Transplant 1996; 17 (Suppl. 1): 56 (S3). 14 Rule SA, Savage DG, O Brien SG, Goldman JM. Outpatient autografting for advanced phase CML. Bone Marrow Transplant 1996; 17 (Suppl. 1): 312 (S67). 15 Srivastava A, Bradstock KF, Szer J et al. Busulphan and melphalan prior to autologous bone marrow transplantation. Bone Marrow Transplant 1993; 12: Katarjian HM, O Brien S, Anderlini P, Talpaz M. Treatment of chronic myelogenous leukemia: current status and investigational options. Blood 1996; 87: Talpaz M, Kantarjian H, Kurzrock R et al. Interferon-alpha produces sustained cytogenetic responses in chronic myelogenous leukemia. Ann Intern Med 1991; 114: Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1976; 53: Fischer T, Neubauer A, Huhn D et al. Outcome of peripheral blood stem cell mobilization in advanced phases of CML is dependent on dose of chemotherapy applied (submitted for publication). 20 Deisseroth AB, Zu Z, Claxton D et al. Genetic marking shows that Ph+ cells present in autologous transplants of chronic myelogenous leukemia (CML) contribute to relapse after autologous bone marrow in CML. Blood 1994; 83: Reiffers J, Goldman J, Meloni G et al. Autologous stem cell transplantation in chronic myelogenous leukemia: a retrospective analysis of the European Group for Bone Marrow Transplantation. Bone Marrow Transplant 1994; 14: Reiffers J, Trouette R, Marit G et al. Autologous blood stem cell transplantation for chronic granulocytic leukaemia in transformation: a report of 47 cases. Br J Haematol 1991; 77: Talpaz M, Kantarjian H, Liang J et al. Percentage of Philadel- 1035

8 1036 phia chromosome (Ph)-negative and Ph-positive cells found after autologous transplantation for chronic myelogenous leukemia depends on percentage of diploid cells induced by conventional dose chemotherapy before collection of autologous cells. Blood 1995; 85: O Brien SG, Goldman JM. Current approaches to hematopoietic stem-cell purging in chronic myeloid leukemia. J Clin Oncol 1995; 13: Carella AM, Cunningham I, Lerma E et al. Mobilization and transplantation of Philadelphia-negative peripheral-blood progenitor cells early in chronic myelogenous leukemia. J Clin Oncol 1997; 15: