GM-CSF in Addition to Chemotherapy of ALL for Kinetics Based Protection of Stem Cells and Stimulation of Haemopoiesis. A randomized Study.
Jerzy Holowiecki    
In: Zander et al. (eds) Gene Technology, Stem Cell and Leukemia Research,Nato Asi Series H:
Cell Biology, Vol 94,Springer-Verlag, Berlin Heidelberg New York London

for the Polish Acute Leukemia Group*
Department of Haematology, Siiesian Medical Academy Reymonta Str.8,40029 Katowice Poland

*Polish Acute Leukemia Study Group: Holowiecki J ,Cedrych 1.,Krzemien S.,Rudzka E.,Holowiecka B.,Jagoda K., Krawczyk M., Hellmann A.,Maj S.,Konopka L.,Dmoszynska A.,Krykowski E.,Robak T.,Kuratowska Z., Skotnicki A.

Intensive induction chemotherapy regimens has led to considerable improvement in the prognosis of adult acute lymphoblastic leukemia ( ALL ), but the results are still unsatisfactory and only about 35% of patients can be cured with current available therapy (7,10,11,14, ) .In conclusions from the studies of large series of adult ALL the complete remission (CR) rate ranges from 70 to 85% ( 2,6,7, 9,10, 11, 14,16,23,27,30 ). A further intensification of chemotherapy will increase CR rates but it is limited by hematopoietic toxicity and therefore the use of recombinant human growth factors in addition to induction chemotherapy has been attempted ( 13,18,19,22,23,25,29 ). The studies on bone marrow cells kinetics during GM-CSF treatment revealed, that 48 to 96 hours after GM-CSF discontinuation the proliferative activity of normal stem cells and progenitors decreases to levels lower than that observed prior to GM-CSF administration ( 1,5, 12,28 ), so this represents a period of partial refractoriness of these cells to the cell cycle dependent cytostatics. Based on the above data as well as on pharmacokinetics of anthracyclines and after successful completing of the pilot study ( 12 ), an original cell kinetics based protocol using rhGM-CSF during induction and consolidation treatment of adult ALL has been designed and evaluated in a multicenter, randomized, prospective study. The aims of this study were as follows: 1) to evaluate the efficacy and tolerability of recombinant human GM-CSF in adult ALL patients, 2) to prove the concept that sequential administration of cell cycle specyfic cytostatics and GM-CSF during induction can enhance the myeloprotection by modulation of normal stem cell kinetics, 3) to evaluate the efficacy of GM-CSF for prevention of neutropenia after intensive consolidation therapy. Material and methods. The multicenter, prospective, randomized trial, using GM-CSF in adult ALL patients has been carried out in 8 departments of haematology. Recombinant human GM-CSF (E.coli derived, Leucomax ) had been supplied by Sandoz/Schering Plough. Fifty two consecutive, previously untreated ALL adult patients were enrolled to the study ( F=22,M=30,median age 29 (range16-59), immunological phenotypes: common=35, early pre 8=7, pre TIT type=1 0, median WBC count=9,4 G\L, median Hb=9,35 g/l, median platelets count=59 G\L, hepatomegaly presented 63% of patients, splenomegaly-65%, Iymphomegaly- 71% of patients) Patients below 16 years and over 60 years old, FAB L3,ALL with "myeloid" suface antigens, historyof anaphylaxis to human proteins, with severe heart, liver,lung 0r kidney impairment, were the exclusion criteria. After informed consent had been obtained, patients were stratified according to risk groups the high risk one (age>35, WBC>30 G\L, "common" negative, phenotype Ph+) and the standard risk one (none of the above criteria). Within each group patients were randomized to receive either chemotherapy plus GM-CSF (GM arm) or chemotherapy alone (control arm). The patient's characteristics in each group is presented in the table 1.

tab. 1 Material

The patients treated according to the schedule presented in fig. 1.

Fig. 1. Scheme of the treatment protocol

Recombinat human GM-CSF ( Leukomax ) , was subcutaneously, in daily dose of 5 ug/kg body weight , during induction, in five day cycles, each one was startet 36 hours after Farmorubicin and Vincristin injection and discontinued 48 hours before the next injection of cytostatics. After completion of the induction, as well as during consolidation with Cyclophosphamide and HO ARA-C, GM-CSF was continued until absolute neutrophil count (ANC) in peripheral blood reached 1 G\l during three consecutive days. The peripheral blood morphology with differential count, reticulocytes and platelets were monitored three times a week. In addition, the monitoring of peripheral blood CO34,CO38 positive cells was performed in 13 patients; 8 receiving GM-CSF and 5 treated in the control arm (each time on the last day of GM-CSF administration -days 3,20,27, and next after 48 hours, just before cytosta-tics dministration -days 1,8,15,22. lmunophenothyping of leukemic blasts at the start of induction and next peripheral blood active stem cells measurements during induction performed using flow cytometry immunoenzymatic APAAP staining. The statistal analysis of survival was performed using original database programme leukos 7 and Biomedical Package Software BMDP.


The overall CR rate equalled 73%, with 90% CR in the standard risk group and 62,5% in the high risk one. A higher CR rate of 80% has been obtained in patiente treated with GM-CSF (GM arm) as compared to the control group -66%. All patients in the "GM arm" achieved CR after 1 induction cycle, whereas 33% of patients in the control arm received more than 1 cycle of induction. The time to reach CR in the "GM arm" was significantly shorter (median 29 days) than in the control one (median 41 days) p<0.05. The particular results obtained in each group of patients are schown in table 2.

tab. 2 Results of induction treatment

Analysis of bone marrow regeneration times revealed, that in the GM arm both the cytoreduction and the subsequent regeneration of particular haematopoetic lines occured significantly earlier than in the control arm (median of days to nadir granulocyte counts equaled 9 versus 15 days, p<0.05 and to nadir platelets 6.5 versus 12,5 days respectively, p<0.05 ). Times to granulocyte recovery and surprisingly also to platelets recovery were also significantly shorter in the "GM arm" than in the control one; median days to ANC >0.5 G\L 11 versus 28, to > 1 G\L 21,5 versus 33,5, to > 1,5 G\L 24 versus 34 days, p<0.05 , median of days to reach platelet count >50 G\L 8.5 versus 12 days, > 100 G/I 26,5 versus 33,5 days (p<0.05. The time to clearance of blasts from peripheral blood was also shorter in the GM arm (median 7 days) as compared to the control group (10 days).

fig.2 Cytoreduction and regeneration times

The use of GM-CSF shortened significantly the neutropenic period during induction, the mean of days with ANC < 0.5 G\L in GM arm equaled 10,6 as compared to 21,4 in the control arm (p<0.05) and only 59% of GM-CSF treated patients displayed agranulocytosis, whereas 85% in the control group. The insidence and severity of infections during induction were reduced in the GM-CSF treated group (only 9% of patients in GM arm developed infection of WHO grade 4, compared to 23% in the control arm). As a consequence also the antibiotic treatment was significantly reduced in the GM arm (p<0.05) fig 3.

fig.3 Number of days of agranulocytosis, antibiotics treatment and fever compared in both groups

The need of RBC and platelets support during induction appeared to be lower in the GM arm than in the control one; median RBC liters equaled 1 ,5 versus 22,3, median platelets units 3 versus 3,8 respectively) . During consolidation, the GM-CSF treatment as an adjunct to high dose chemotherapy enabled a close adherence to the intensive therapy regimen (the average dalay of the subsequent cycle equaled 0.5 day in the MM arm versus 6.6 days in the control group, p<0.5). In the "GM arm" the frequency of infection, the fever rate, and the usage of antibiotics were also lower than in the control group. The current follow up, reveales comparative results in both groups with not significant differences between Kaplan Meyer DFS curves (Wilcoxon Breslow-p=0.9,Mantel-Cox-p=0.6). The median CCR has not been reached at the mean observation time of 601 days in GM arm and 525 days in the control arm, although the tendency suggesting the better outcome in GM-CSF arm appears ( 75 th quantile has not been reached yet versus 298 days in control) -fig 4

fig. 4 Kaplan Meyer curves of disease free survival in both groups of patients

Repeated dual color staining for CD34CD38 flow cytometry analyses of a significant increase in the absolute count and percent mononeulear and granulomonocytic fraction, during GM-CSF administration, followed by the rapid decrease 48 hours after the cessation of the cytokine. This phenomenon has been never observed in patients obtaining chemotherapy only, who displayed a distinct decrease of CD34CD38 positive cell count during the induction treatment. (fig.5)

fig. 5 The kinetics of peripheral blood CD34+CD38+cells during GM-CSF and chemotherapy treatment compared to chemotherapy alone



The current results of the study presented here support the idea of the beneficial use of growth factors in combination with chemotherapy in acute lymphoblastic leukemia. The results of multiple studies reflect the impact of the use intensive chemotherapy regimens in induction of ALL (2,6,7,9,10,11,13,14,16,18,22,23,25,27,30) to achieve rapid cytoreduction, but it is important balance the antileukemic effect with the minimal myelotoxicity (14). In the German Multicentre Adult ALL Trial the granulocytopenia has appeared the major dose limiting toxicity, leading to delay or discontinuation of therapy and most investigators have demonstrated that an induction failure is equally due to the toxicity and to the refractoriness t0 chemotherapy (14). The use of recombinant growth factos such G-CSF and GM-CSF in the management of lymphoid malignancies have demonstrated their effectiveness in supportion of myelopoiesis, a good tolerability (4,8,11,13,18,24,25) with no evidence of malignant lymphoid clones stimulation (1,3,15,17,20,26). In addition to supportive function of growth factors in cytopenia, new perspectives have been offered based on prewious data suggesting that 48-96 hour period after discontinuation of GM-CSF administration represents a period of relative refractoriness of normal progenitors to cell cycle specyfic chemotherapy (1,3,5,12,26,28). Based on these findings we developed a novel remission induction protocol, and the data presented here confirm its efficacy. Our flow cytometric findings of kinetic charnges of peripheral blood CD34 CD38 posivite cells during and after GM-CSF treatment appear to confirm that a period of relative refractoriness of normal stem and progenitor cells may be obtained using properly sequenced administration of rhGM-CSF and cytostatic drugs. Rapid granulopoietic reconstitution, as well as a rather unexpected shorter platelets recovery time, achieved in GM-CSF group should be considered as an another evidence of cytoprotection on multilineage stem cell level. Complete remission has been achiewed significantly more rapidly in patients receiving GM-CSF in addition to chemotherapy and this finding is of great importance. At first it confirs the value of the above discussed policy and secondly it has a prognostic value predicting for a longer remission duration. Data from large series of ALL patients demontrated that a short induction treatment duration , is an one of the most important parameters predicting long term disease free survival ( 9, 10, 11 ). Further investigations are needed to explain the accelerated reduotion of leukemic cells during GMCSF administration. A priming effect of secondary secreted cytokines as well as activation of immunological mechanisms of tumor control should be taken into consideration. In summary, the protocol employing GM-CSF has following advantages in comparison to chemotherapy alone: shorter time to CR, (p<0.05), CR in all cases obtained after 1 chemotherapy cycle, accelerated blasts cytoreduction (p=0.05), accelerated hematopoietic reconstitution ( p<0.05 ), better outcome of treatment with lower incidence and severity of infections. The use GM-CSF during high dose consolidation treatment permits close adherence to planned chemotherapy without delay. Our clinical data supported by flow cytometric findings confirm the idea of a cell kinetics based protection of stem cells in ALL, using properly sequenced chemotherapy and GM-CSF .

The study protocol received the Local Ethics Commitee agreement.
This trial was subvenced in part by Polish Research Comitee, Grant No.406669101

Participating centers Dept of Hematology Medical Academy Katowice, Deptof Hematology MA Gdansk,lnstitute of Hematology Warsaw, Deptof Hematology M.A Lublin, Deptof Hematology M.A Lodz, Deptof Hematolgy MA Warsaw, Dept of Hematology MA Krakow


Aglietta M. Piacibello al.Kinetics of Human Hematopoietic Cells After in Vivo Administration of Granulocyte Macrophage Colony Stimulating Factor J.Clin.lnvest.1989.83.

Anglin P.Keating A.Adult Acute Lymphoblastic Leukemia Treated with a Pediatric High Risk Protocol. Single Centre Report of 50 Consecutive Cases.Blood.1993.83(10)SuppI1

Bernstein I.D.,Andrews R.G.Recombinant Human Stem Cell Factor Enhances the Formation of Colonies by CD 341in- Cells and the Generation of the Colony Forming Cell Progeny from CD 34+lin-Cells Cultured with Interleukin 3,Granulocyte Macrophage Colony Stimulating Factor or Granulocyte Colony Stimulating Factor .Blood.1991.77.

Blaise D,Vernant JP,Fiere D,Gluckman E, Reiffers J,et al.A Randomized Controlled Multicentre Trial of Filgrastim in Patients Treated by Bone Marrow Transplantation with Total Body Irradiation for Acute Lymphoblastic Leukemia or Lymphoblastic Lymphoma.Congress Abstract Review,34 th Annual Meeting of American Society of Hematology"Growth Factors in Hematology "December, 1992

Broxmeyer HE,Benninger L,PateIS,Benjamin RS, Vadhan-Rai S,Differences in Kinetics Responses of Human Marrow Myeloid Progenitor Cells to in Vivo Treatment of G-CSF and GMCSF.Congress Abstract Review,34 th Annual Meeting of American Society of Hematology"Growth Factors in Hematology "December,1992

Casssileth P.A.Andersen J.W.Bennett al.Adult Acute Lymphocytic Leukemia:The Eastern Cooperative Oncology Group experience. Leukemia. 1992. 6 Suppl 2

Gale R.P.Butturini A.Controversies in Therapy of Acute Lymphoblastic Leukemia. Leukemia. 1992. 6, Suppl 2

Gerhartz H.H.,Englehard M,Meusers P,et al.Randomized Double Blind,Placebo Controlled Phase III Study of Recombinant Human Granulocyte Macrophage Colony Stimulating Factor as Adjunct to Induction Treatment of High Grade Malignant Non Hodkin Lymphomas.Blood, Vol 82,No 8,1993,pp.2329

Hoelzer D. Therapy of Newly Diagnosed Adult Acute Lymphoblastic Leukemia.Hematol-Oncol-Clin North-Am.1993 Feb:7(1) 139-60

Hoelzer D., Therapy of Acute Lymphoblastic Leukemia in Adults.Leukemia.1992.6,Suppl 2.

Hoelzer D.Thiel E.Ludwig WD.Loffler H.,Buchner TH.,Freud M.,et al.The German Multicentre Trials for Treatment of Acute Lymphoblastic Leukemia.Leukemia.1992 6.Suppl2

Holowiecki J.,Cedrych al.Efficacy of an Original Cell Kinetics Based Chemotherapy Regimen Using rHuGM-CSF for Induction and Consolidation Treatment of ALL. Preliminary Analysis.Acute Leukemias IV.Springer-Verlag Berlin Heidelberg 1993

Link H,Boogaerts MA,Carella AM,et al.A Controlled Trial of Recombinant Human GranulocyteMacrophage Colony Stimulating Factor after Total Body Irradiation, High Dose Chemotherapy and Autologous Bone Marrow Transplantation for Acute Lymphoblastic Leukemia or Malignant Lymphoma Blood 1992, Nov 1,80(9) 2188-95

Linker C.A.Levitt L. al. Treatment of Adult Acute Lymphoblastic Leukemia with Intensive Cyclical Chemotherapy:A Follow uo Report. Blood.1991. 78.

Lord B.I.,Molineux al.Myeloid Cell Kinetics in Mice Treated with Recombinant Interleukin 3,Granulocyte Macrophage Colony Stimulating Factor or Granulocyte Stimulating Factor.Blood.1991.77.

Mandelli F.Annino L. Vegna M.L.GIMEMA ALL 0288 : a Multicentric Study on Adult Acute Lymphoblastic Leukemia. Preliminary Results. Leukemia. 1992 6.SuppI2.

Ode D.L.Zhou al.The Effect of Recombinant GM-CSF and G-CSF on the Bone Marrow Cells of Children with Acute Lymphoblastic Leukemia. Leukemia.1992.6(11)

Ottman OG,Hoelzer D,.Concomitant rmetHuG-CSF(Filgrastim) and Intensive Chemoradiotherapy as Induction Treatment in Adult ALL.A Randomized Multicenter Phase III trial.Blood.1993.83(10)SuppI1

Reynoso E.E.Duenas S.Granulocyte Macrophage Colony Stimulating Factor ( GM-CSF ) to Atenuate the Hematological Toxicity of Phase II Induction of Adult Acute Lymphoblastic Leukemia. Blood.1993. 83 ( 10) Suppl 1

Ridgway D.Borzy M.S.Differential Effect of Continous versus Cyclic Maintenance Chemotherapy on Cytokine Induced Myeloid Colony Growth.Hematol. Pathol.1990.4.87.

Ridgway D.Borzy M.S.Defective Production of Granulocyte Macrophage Colony Stimulating Factor and Interleukin 1 by Mononuclear Cells from Children Treated for Acute Lymphoblastic Leukemia.Leukemia 1992.6.

Scherrer R.Geissler al.Granulocyte Colony Stimulating Factor as an Adjunct to Induction Chemotherapy of Adult Acute Lymphoblastic Leukemia.Ann.Hematol.1993.66.(6)  

Schiffer C.A. Larson R.A. Bloomfield C.D.Cancer and Leukemia Study Group B ( CALGB) Studies in Adult Acute Lymphocytic Leukemia. Leukemia. 1992. 6,Suppl2

Schulz G.Frish al.New Therapeutic Modalities for the Clinical Use of th GM-CSF in Patients with Malignancies.Am.J. Clin.Oncol.1991.14.

Smedmyr B.Glycosylated rHuG-CSF as Supportive Treatment During Remission InductionTherapy for Adult Patients with Newly Diagnosed ALL\AUL.Blood.1993

Strauss L.C.Welsh al.lnitiation of DNA Synthesis by Colony Stimulating Factors in Subsets of Human CD34+Marrow Cells. Exp. Hematol.1991.19.

Stryckmans P.De Witte al. Therpay of Adult ALL:overview of 2 succesive EORTC studies :( ALL -2 & ALL -3 ) .Leukemia. 1992 6 Suppl 2

Vadhan Rai S.Hal. E. et al.Abrogating Chemotherapy Induced Myelosupression by Recombinant Granulocyte Macrophage Colony Stimulating Factor in Patients with Sarcoma. Protection at the Progenitor Cell Level.J .Clin. Oncol.1992.1 ).

Welte K.Mernpel K.Reiter A,et al.A Randomized Phase III Study of Recombinant Human Granulocyte Colony Stimulating Factor ( rmet Hu G-CSF) in Childchood High risk acute lymphoblastic leukemia.Blood 1993.83 (10) Suppl1

Weiss M.Berman E.Rapid Remission Induction Therpay in Adult Acute Lymphoblastic Leukemia Without VCR or Prednisone.Blood 1993.83 (10) Suppl 1