& 2005 Nature Publishing Group All rights reserved /05 $

Transcription

1 (25), & 25 Nature Publishing Group All rights reserved 28-39/5 $ Graft clonogenicity and intensity of pre-treatment: factors affecting outcome of autologous peripheral hematopoietic cell transplantation in patients with acute myeloid leukemia in first remission D Strodtbeck 1, M Bornhäuser 1,MHa nel 2, L Lerche 1, M Schaich 1, T Illmer 1, C Thiede 1, G Geissler 2, R Herbst 2, G Ehninger 1 and U Platzbecker 1 1 Medizinische Klinik und Poliklinik I, Universitätsklinikum Carl Gustav Carus, Dresden, Germany; and 2 Klinik für Innere Medizin III, Klinikum Chemnitz, Germany Summary: A total of 22 patients with acute myeloid leukemia (AML) in first complete remission receiving autologous blood stem cell transplantation (ABSCT) were investigated in order to determine factors affecting outcome. All but two patients had a normal karyotype and received the same high-dose chemotherapy followed by G-CSF-mobilized peripheral blood stem cells after the second (n ¼ 5) or third (n ¼ 17) course of induction and post-remission chemotherapy, respectively. With a median follow-up of months, the median disease-free survival is 24.1 months. Univariate analysis showed that three chemotherapy cycles before ABSCT were associated with a significant better disease-free survival (P ¼ 18) and overall survival (P ¼ 33), whereas the presence of an FLT3- mutation (n ¼ ) showed no impact. The number of megakaryocytic progenitors (CFU-MK) infused tended to correlate with primary platelet engraftment (P ¼ 7) and were predictive for neutrophil (P ¼ 11) and platelet counts (P ¼ 9) 18 days after transplantation. Patients receiving a higher amount of CFU-MK had a better event-free survival (P ¼ 2). Our data suggest that the content of CFU-MK within the graft predicts the quality of hematological recovery and long-term disease control. Additionally, a minimum of three chemotherapy cycles before ABSCT seems to be associated with an improved outcome. (25), doi:138/sj.bmt.17517; published online 1 October 25 Keywords: megakaryocytic progenitors; AML; autologous transplantation; reconstitution; FLT3 A recent meta-analysis has shown that there is only limited scientific evidence for the use of autologous bone marrow transplantation as consolidation treatment in patients with acute myeloid leukemia in first complete remission. 1 Nevertheless, within the past years autologous blood stem cell transplantation (ABSCT) has become a new and widely used treatment strategy in patients with acute myeloid leukemia (AML). Although it has been shown to be associated with a moderate toxicity and lower treatmentrelated mortality compared to bone marrow, its value in subgroups of AML patients is still a question of debate. In fact, patients with favourable or intermediate cytogenetics seem to benefit from this procedure, 2 although randomized trials are missing. Recently, activating FLT3 receptor mutations have been shown to distinguish a poor-risk patient population out of those with normal karyotype. Preliminary data suggest that FLT-mutation status is associated with a higher risk for relapse even after autologous and allogeneic transplantation. 3 It is not known whether the increased relapse risk can also be observed after ABSCT. There are also conflicting data whether the number of chemotherapy courses affects outcome after ABSCT. 4,5 Additionally, several studies have defined distinct parameters influencing the hematopoietic reconstitution and long-term disease-free survival in these patients. 9 This includes the effects of progenitor cell count within the graft on engraftment speed. There is only one study suggesting that the amount megakaryocytic progenitors (CFU-MK) infused affects disease-free survival after ABSCT, whereas other lineage-associated colony assays had no impact. 1 This is important since several patients with AML do not show complete hematological recovery after autologous transplantation 11 and CFU-MK are decreased in diseases with marrow failure like MDS. Therefore, it might be of interest to correlate megakaryocytic colony formation with hematopoietic reconstitution after ABSCT. Therefore, we prospectively investigated parameters affecting outcome in 22 patients with AML receiving an ABSCT after two or three courses of induction chemotherapy. Correspondence: Dr U Platzbecker, Medizinische Klinik I, Universita tsklinikum Carl Gustav Carus, Fetscherstrasse 74, Dresden 17, Germany; Uwe.Platzbecker@uniklinikum-dresden.de Received 31 May 25; accepted 2 August 25; published online 1 October 25 Patients and methods Patient characteristics are shown in Table 1. All patients were treated within the AML 9 protocol of the DSIL

2 184 study group. The study was approved by the local institutional review board. Patients below the age of 5 years with de novo AML, as defined by the French American British (FAB) classification, were eligible for the study. All included patients gave written informed consent. Patients were considered for ABSCT if they had intermediate-risk cytogenetics (normal karyotype, inv1) or high-risk cytogenetic features in case of the absence of an HLA-matched related or unrelated donor. The FLT3- mutation status was performed as described, 13 and had no impact on risk stratification in this trial. Chemotherapy Every patient received two induction courses (IC) consisting of MAV (mitoxantrone 1 mg/m 2 once daily intravenously (i.v.) days 4 8, total dose 5 mg/m 2 ; cytarabine 1 mg/m 2 by continuous i.v. infusion days 1 8, total dose 8 mg/m 2 ; etoposide 1 mg/m 2 once daily i.v. days 4 8, total dose 5 mg/m 2 ) followed by MAMAC (m-amsa 1 mg/m 2 once daily i.v. days 1 5, total dose 5 mg/m 2 ; cytarabine twice daily 1 g/m 2 i.v. days 1 5, total dose 1 g/m 2 ). Within a median of 24.7 (1 44.9) weeks after diagnosis, five patients (group A) directly went onto Table 1 Patient characteristics n 22 Sex (male/female) 14/8 Median age in years (range) 5 (range 21 1) Karyotype Normal 2 Aberrations (+8; 7) 2 FAB type M1 5 M2 9 M4 3 M5 2 M5a 1 M 1 RAEB-t 1 FLT3 mutated FLT3 wild type 1 Induction therapy I MAV 22 Induction therapy II MAMAC 22 Post-remission therapy I-MAC 8 H-MAC 9 None 5 No. of chemotherapy cycles before APBSCT 2 (group A) 5 3 (group B) 17 Conditioning regimen Bu/Cy/Eto 22 ABSCT¼ autologous blood stem cell transplantation; Bu/Cy/Eto ¼ busulfan/cyclophosphamide/etoposide. transplant. In the remaining 17 patients (group B), IC was followed by one course of post-remission consolidation chemotherapy (CT) with I-MAC (mitoxantrone 1 mg/m 2 once daily i.v. days 4, total dose mg/m 2 ; cytarabine twice daily 1 g/m 2 i.v. days 1 ; total dose g/m 2, n ¼ 8) or H-MAC (mitoxantrone 1 mg/m 2 once daily i.v. days 2 4, total dose mg/m 2 ; cytarabine twice daily 3 g/m 2 i.v. days 1 4; total dose 24 g/m 2, n ¼ 9). During chemotherapyinduced cytopenia, G-CSF-induced peripheral blood stem cell (PBSC) mobilization was performed after the second IC in all 22 patients and in patients within group B a second time after CT. The latter group received only the graft obtained after the consolidation course. Leukapheresis procedure During chemotherapy-induced cytopenia, patients received recombinant human G-CSF at a dosage of 1 mg/kg/day. In case of an increase of the percentage of CD34 þ cells over a certain threshold in the peripheral blood, leukapheresis was started. CD34 þ cells were quantified by flow cytometric analysis. 14 Venous access was obtained through venipuncture of both arms. A large volume apheresis comprising four times the calculated patient s blood volume was performed using a continuous-flow cell separator (Cobe Spectra, Lakewood, USA). After collection, the cells were placed in 1% (v/v) dimethylsulfoxide (DMSO) and stored in liquid nitrogen at 191C until use. Analysis of colony-forming unit-granulocyte macrophage (CFU-GM) and CFU-MK For quantification of progenitor content, mononuclear cells diluted in PBS Dulbecco s w/o Ca and Mg (Bio Whittaker, Walkersville, Maryland, USA) were plated in a complete methylcellulose medium containing IMDM with % fetal bovine serum, 3 U/ml erythropoietin, 5 ng/ml SCF, 2 ng/ml GM-CSF, 2 ng/ml IL-3, 2 ng/ml IL- and 2 ng/ml G-CSF (Methocult GF H4435, Stem Cell Technologies, Vancouver, Canada). Cultures were incubated at 371C and 5% CO 2. The cultures were assessed at day 14 for the presence of burst-forming uniterythroid (BFU-E), CFU-GM and mixed colony-forming unit (CFU-GEMM). The content of CFU-MK was determined using the Stem a.ii-mega Kit (Tebu-Bio GmbH, Offenbach, Germany) according to the manufacturer s instructions. In brief, MNC were suspended in 1 ml of a complete methylcellulose medium containing IMDM with % fetal bovine serum, 2 ng/ml IL-3, 2 ng/ml IL-, 5 ng/ml IL-11, 2 ng/ml SCF, 2 ng/ml bfgf and ng/ml rhtpo. Colonies were scored after days of culture at the conditions mentioned above. The counted numbers were calculated and expressed as the number of colonies per patient body weight at the time of ABSCT. Conditioning regimen All patients were conditioned with busulfan 4 mg/kg p.o. in divided doses daily for 4 days (total dose 1 mg/kg) on days

3 7 to 4, etoposide once mg/kg i.v. on day 3 and cyclophosphamide mg/kg once daily i.v. on days 3 and 2 (total dose mg/kg). On day, the cryopreserved stem cells were thawed and reinfused. Neutrophil engraftment was defined as the first of three consecutive days with an absolute neutrophil count (ANC) of greater than 1 9 /l independently of the administration of G-CSF. Platelet recovery was defined as the first of three consecutive days of a sustained absolute platelet count of greater than /l without any transfusion support. Statistical analysis Quantitative parameters are given as median with maximum and minimum. Comparisons of the median between both groups were performed with the Wilcoxon test. An univariate analysis was used in order to estimate the effect of covariates like number of transplanted CD34 þ cells, total cell number collected, patient age, therapy before transplantation, cytogenetics, CFUs, FLT3 receptor status and timing of stem cell harvest (LPH) on overall survival, disease-free survival and probability of relapse. Kaplan Meier estimates of survival and log-rank tests for comparisons are provided whenever appropriate. Correlation coefficients between transplanted cells, CFU- GM and MK-colonies and platelet and neutrophil counts,, 1, 18 and days after ABSCT were calculated using the Pearson s statistics. The analyses were performed using the SPSS software programme (Chicago, IL, USA). Results Transplantation and engraftment A median of /kg CD34 þ cells (range ) were infused on day (Table 2). There was a significant (P ¼ 48) difference between the number of CD34 þ cells in the graft between group A (3.2 1 /kg (range )) and B (2.2 1 /kg (range )). The number of CFU- GM in the graft between group A (7 1 5 /kg (range )) and B ( 1 5 /kg (range 1.13)); P ¼ 2 did not differ just like the number of CFU-MK between group A (4 1 5 /kg (range 3 )) and B ( 1 5 /kg (range 1 8)); P ¼. The speed of hematopoietic recovery is shown in Table 2. The median time to reach an ANC of /l and platelets 1 9 /l was 11 days (range 9 19) and 13.5 days (range 1 159), respectively. There was no statistically significant difference between both groups. Factors influencing hematopoietic reconstitution The median ANC on days, 18 and after transplant was /l (range 4 4.2), /l (range ) and /l (range ), while a median platelet count of /l (range ), /l (range 4 177) and /l (range 23 25) was documented, respectively. In fact, there were three patients failing to achieve normal blood counts (ANC /l, platelets /l) 18 and days after ABSCT. The factors significantly affecting neutrophil and platelet recovery in the univariate analysis were the CFU-MK, CFU-GM dose and the number of transplanted CD34 þ cells infused (Table 3). Age and total cell number infused had no impact on the kinetics of engraftment. Table 2 Variable Graft characteristics P-value CD34+ ( 1 /kg) All 2.3 (range ) Group A 3.2 (range ) Group B 2.2 (range ) 48 CFU-GM ( 1 5 /kg) All 3 (range 1.53) Group A 7 (range ) Group B (range 1.13) 2 CFU-MK ( 1 5 /kg) All (range 1 8) Group A 4 (range 3 ) Group B (range 1 8) Engraftment Neutrophils in days 11 (range 9 19) Platelets in days 13.5 (range 1 159) All values are provided as median (range), group A (transplantation after second cycle of induction chemotherapy), group B (transplantation after third cycle of induction/post-remission chemotherapy). 185 Table 3 Factors correlating with platelet and neutrophil recovery ANC ANC 18 ANC PLT PLT 18 PLT CD34+ cells transplanted ( 1 /kg) P-value Correlation coefficient CFU-GM ( 1 5 /kg) P-value Correlation coefficient CFU-MK ( 1 5 /kg) P-value Correlation coefficient All values are P-values and the correlation coefficient; PLT ¼ platelets; ANC ¼ absolute neutrophil count; the number given is the day after transplantation.

4 18 a a b b c Figure 2 (a and b) Impact of the number of therapies (two ¼ dashed line vs three ¼ solid line) on disease-free (a, P ¼ 18), overall survival (b, P ¼ 33) and probability of relapse (c, P ¼ 13) Figure 1 (a c) Kaplan Meier plot of the probability of disease-free (a), overall survival (b) and relapse (c). After a median follow-up period of months, the median disease-free and overall survival of all patients (n ¼ 22) is 22. and 24.1 months, respectively. Factors influencing outcome With a median follow-up of months, the median overall survival had not been reached and the disease-free survival was 24.1 months (Figure 1a and b). Only one patient died while being in remission due acute cardiac failure 11 days after ABSCT. Nine patients relapsed a median of 8. months after ABSCT (Figure 1c), three of them had received only two cycles of induction chemotherapy, none had high-risk cytogenetics and three out of nine were FLT3 positive. The number of therapies before ABSCT (2 vs 3) affected overall and disease-free survival (Figure 2a and b) as well as the probability of relapse (P ¼ 33, 18 and 13). In line with this, observation time of stem cell harvesting (stem cells obtained after the second or the third IC) had an impact on overall and disease-free survival (P ¼ 82 and 29). Besides, CFU-MK had also a significant influence on the duration of disease-free survival and the probability of relapse (P ¼ 274 and 45; Figure 3).

5 18 24 Figure 3 Impact of the numbers of CFU-MK in the graft (more than median ¼ dashed line; less than median ¼ solid line) on disease-free (P ¼ 274) survival. Interestingly, FLT3-mutation status had no impact on all parameters (data not shown), with a total of 1 and 5 FLT3- positive patients treated in groups A and B, respectively. Discussion With the availability of PBSC as stem cell source, there is growing interest in applying ABSCT as treatment strategy in patients with newly diagnosed AML. Our data show that the procedure is associated with a low treatment-related mortality and can result in disease control in patients with intermediate-risk AML defined by cytogenetics. It is also promising that both patients with high-risk cytogenetics did not relapse, although data from other studies suggest that ABSCT is no option for this patient population. Nevertheless, we confirm that ABSCT after two cycles of IC is insufficient to achieve long-term disease control. 15 This suggests an insufficient in vivo purging as a result of only two courses of induction chemotherapy. There is a long tradition of graft-purging using autologous bone marrow grafts in patients with AML. Although prospective trials could never be performed, retrospective analyses suggested purging of clonogenic leukemia progenitors from the graft could have a positive impact on disease-free survival. 1 This experience also argues for the importance to obtain an autologous graft with as less as possible leukemic contamination. On the other hand, the favourable outcome of the subgroup of patients mobilizing enough CD34 þ cells after the third chemotherapy cycle could be a selection bias. We observed a comparable outcome of FLT3-positive patients. Since these mutations have been shown to negatively affect outcome, ABSCT might be the treatment of choice for these patients. 17 In contrast to other reports, we could not show that the CD34 þ concentration had an influence on the hematopoietic recovery. This might be due to the low number of patients studied in this analysis. On the other hand, the intensity of induction chemotherapy containing intermediate-dose cytosine arabinoside in our protocol might have led to successful mobilization only in a limited percentage of patients. We confirm the data showing the influence of CFU-GM on hematopoietic recovery. Nevertheless, little data are available on the role of CFU-MK. We show here that there is indeed a correlation, especially with long-term platelet and neutrophil reconstitution, which is not infrequently incomplete in patients after transplant. 18 It is of note that we were able to confirm the data by Reichle et al 19 on the impact of CFU-MK on long-term disease control. Therefore, this assay might be a simple tool in order to estimate the prognosis of an individual patient. Of course, the colonies quantified with the commercial assay used in this series are not merely restricted to CFU-MK, but perhaps provide a more accurate assessment of the clonogenicity of the autologous graft compared to the CFU-GM assay used. Given the relatively short follow-up in our analyses, the prognostic role of graft clonogenicity and cycles of pretreatment on disease-free survival with respect to other risk factors has to be regarded with caution. In patients receiving three cycles prior to ABSCT, the disease-free survival reached 49%. Although the patient number is limited in our study, these data at least provide some rationale to believe that ABSCT may be a reasonable strategy in this patient population. 2 Nevertheless, randomized studies comparing chemotherapy with ABSCT will have to confirm our observation since the majority of published trials used bone marrow as stem cell source. References 1 Nathan PC, Sung L, Crump M, Beyene J. Consolidation therapy with autologous bone marrow transplantation in adults with acute myeloid leukemia: a meta-analysis. J Natl Cancer Inst 24; 9: Vey N, Bouabdallah R, Stoppa A et al. Autologous stem cell transplantation for acute myelogenous leukemia in first complete remission: a -year follow-up study of 11 patients from a single institution. Bone Marrow Transplant 24; 33: Kottaridis PD, Gale RE, Linch DC. Prognostic implications of the presence of FLT3 mutations in patients with acute myeloid leukemia. Leukemia Lymphoma 23; 44: Feller N, Schuurhuis GJ, van der Pol MA et al. High percentage of CD34-positive cells in autologous AML peripheral blood stem cell products reflects inadequate in vivo purging and low chemotherapeutic toxicity in a subgroup of patients with poor clinical outcome. Leukemia 23; 17: de la Rubia J, Sanz GF, Martin G et al. Autologous blood stem cell transplantation for acute myeloblastic leukemia in first complete remission. Intensification therapy before transplantation does not prolong disease-free survival. Haematologica 1999; 84: Carral A, de la Rubia J, Martin G et al. Factors influencing hematopoietic recovery after autologous blood stem cell transplantation in patients with acute myeloblastic leukemia and with non-myeloid malignancies. Bone Marrow Transplant 22; 29: Reiffers J, Stoppa AM, Attal M et al. Allogeneic vs autologous stem cell transplantation vs chemotherapy in patients with 187

6 188 acute myeloid leukemia in first remission: the BGMT 87 study. Leukemia 199; 1: Gorin NC, Labopin M, Laporte JP et al. Importance of marrow dose on posttransplant outcome in acute leukemia: models derived from patients autografted with mafosfamidepurged marrow at a single institution. Exp Hematol 1999; 27: Gorin NC, Labopin M, Pichard P et al. Feasibility and recent improvement of autologous stem cell transplantation for acute myelocytic leukaemia in patients over years of age: importance of the source of stem cells. Br J Haematol 2; 11: Reichle A, Rothe G, Krause S et al. Transplant characteristics: minimal residual disease and impaired megakaryocytic colony growth as sensitive parameters for predicting relapse in acute myeloid leukemia. Leukemia 1999; 13: Lowenthal RM, Faberes C, Marit G et al. Factors influencing haemopoietic recovery following chemotherapy-mobilised autologous peripheral blood progenitor cell transplantation for haematological malignancies: a retrospective analysis of a 1-year single institution experience. Bone Marrow Transplant 1998; 22: Schaich M, Illmer T, Seitz G et al. The prognostic value of Bcl- XL gene expression for remission induction is influenced by cytogenetics in adult acute myeloid leukemia. Haematologica 21; 8: Thiede C, Steudel C, Mohr B et al. Analysis of FLT3- activating mutations in 979 patients with acute myelogenous leukemia: association with FAB subtypes and identification of subgroups with poor prognosis. Blood 22; 99: Platzbecker U, Prange-Krex G, Bornhauser M et al. Spleen enlargement in healthy donors during G-CSF mobilization of PBPCs. Transfusion 21; 41: Feller N, Schuurhuis GJ, van der Pol MA et al. High percentage of CD34-positive cells in autologous AML peripheral blood stem cell products reflects inadequate in vivo purging and low chemotherapeutic toxicity in a subgroup of patients with poor clinical outcome. Leukemia 23; 17: Gorin NC. Autologous stem cell transplantation for adult acute leukemia. Curr Opin Oncol 22; 14: Levis M, Small D. FLT3: ITDoes matter in leukemia. Leukemia 23; 17: Lowenthal RM, Faberes C, Marit G et al. Factors influencing haemopoietic recovery following chemotherapy-mobilised autologous peripheral blood progenitor cell transplantation for haematological malignancies: a retrospective analysis of a 1-year single institution experience. Bone Marrow Transplant 1998; 22: Reichle A, Rothe G, Krause S et al. Transplant characteristics: minimal residual disease and impaired megakaryocytic colony growth as sensitive parameters for predicting relapse in acute myeloid leukemia. Leukemia 1999; 13: Jourdan E, Rigal-Huguet F, Marit G et al. One vs two highdose cytarabine-based consolidation before autologous stem cell transplantation for young acute myeloblastic leukaemia patients in first complete remission. Br J Haematol 25; 9: