Leukemia Research Vol. 16, No. 10, pp. 973-977, 1992. Printed in Great Britain.

0145-2126/92 $5.00 + .DO © 1992 Pergamon Press Ltd

I N V I T R O P U R G I N G OF CLONOGENIC LEUKEMIC CELLS FROM H U M A N BONE M A R R O W BY HEAT: SIMULATION EXPERIMENTS FOR A U T O L O G O U S BONE M A R R O W T R A N S P L A N T A T I O N YOSHIAKI MORIYAMA, SHIGEO HASHIMOTO, TAKAO GOTO, TATSUO FURUKAWA, KENJI KISHI, MASUHIRO TAKAHASHI and AKIRA SHIBATA First Department of Internal Medicine, Niigata University School of Medicine, 1-754 Asahimachi-Dori, Niigata 951, Japan (Received 27 March 1992. Revision accepted 13 June 1992)

Abstract--In order to apply a simple purging method by heat to autologous bone marrow transplantation (ABMT), we have revaluated the ability to purge clonogenic leukemic cells from the simulated marrow mixture of normal marrow cells and leukemic cell lines (HL-60, Molt-3 and HEL) in vitro by heat, using two different clonogenic assays for normal granulocyte-macrophage progenitors (CFU-GM) and leukemic cell lines. It appeared that in vitro hyperthermia (42°C for 120 min) is able to selectively remove clonogenic leukemic ceils from simulated tumor cell-normal marrow mixtures even when leukemic cell concentrations are increased up to 3 × 10 6 cells/ml in vitro, and results in a 4--6 log destruction of clonogenic leukemic cells/ml according to a limiting dilution assay, while leaving half of normal CFU-GM surviving. The hyperthermic purging of clonogenic leukemic cells was not affected in the presence of normal marrow cells in vitro. This high level of clonogenic leukemic cell depletion by heat correlated with that of immunologic and pharmacologic studies. These results suggest that in vitro hyperthermia could be applied effectively and safely for the elimination of residual clonogenic leukemic cells in autologous marrow grafts before ABMT. Key words: Thermal sensitivity, clonogenic leukemic cells, limiting dilution assay, cryopreservation, simulated marrow mixture.

INTRODUCTION

general, purging of leukemic cells by pharmacological methods have been more extensively investigated than purging by immunologic methods [3-6]. Our previous studies [7-9] showed that human L - C F U appear to be much more sensitive than normal granulocytemacrophage progenitors ( C F U - G M ) to heat (4144°C). From our results [7-9] and others [10-12], in vitro hyperthermia could be used to purge residual leukemic cells, especially L-CFU from remission marrow in human acute leukemia prior to A B M T . However, little is known about the hyperthermic purging of clonogenic tumor cells in the presence of normal bone marrow cells in vitro. Specifically, there is a lack of an appropriate normal a n d / o r leukemic counterpart that would provide a relative ratio of the effect on normal and leukemic cells. To apply this simple purging m e t h o d to A B M T , in this study, we have revaluated the in vitro purging effect of heat in a tumor cell-normal marrow mixture model which mimicked remission and relapsed marrows, using two different assay systems for C F U - G M and clonogenic tumor cells in vitro.

bone marrow transplantation ( A B M T ) has the potential advantage of being available to a larger population of patients with acute leukemia and lymphoma while avoiding many of the associated immunobiological effects of allogenic graft as well as the age restriction. The major drawback of A B M T is contamination with microscopically undetectable clonogenic tumor cells which are able to self-renew and proliferate [1]. Thus, the residual clonogenic tumor cells must be purged from harvested marrow if the A B M T is to be curative [2-4]. Because of the lack of the specific antigens for human leukemic progenitor cells (L-CFU), in AUTOLOGOUS

Abbreviations: A B M T , autologous bone marrow transplantation; L-CFU, leukemic progenitor cells; CFU-GM, .granulocyte-macrophage progenitor cells; FBS, fetal bovme serum; MNC, mononuclear cells; GM-CSF, granulocyte-macrophage colony stimulating factor. Correspondence to: Yoshiaki Moriyama, M.D., 1st Department of Internal Medicine, 1-754 Asahimachi-Dori, Niigata 951, Japan.

973

Y. MORIYAMA et al.

974

MATERIALS

AND METHODS

Leukemic cell lines Two human myeloid leukemic cell lines ( H E L and HL60) and one T lymphoblastic cell line (Molt-3) were used in this study. Cells were grown in RPMI-1640 medium containing 10% heat-inactivated fetal bovine serum (FBS) and 1% penicillin and streptomycin. Bone marrow cells Normal marrow cells used as controls were collected from donors of BMT after obtaining appropriate informed consent. Bone marrow mononuclear cells (MNC) were then separated by density gradient (density 1.077, Lymphoprep, Nyegaared, Oslo). Cells at the interface were collected, washed three times and resuspended in RPMI1640 medium.

0 vl

(J

¢m

cs -~

-3 cs

c

o

-4

L

cm _5 >

Tumor cell-normal marrow mixture The simulated marrow mixture was formed by mixing normal MNC (1 x 10 7) with 5 x 105, 106, 3 x 106 and 5 × 10 6 leukemic cell line cells in ratios of 20:1, 10:1, 10:3 and 10:5, respectively. Control cells contained only normal MNC (107/ml) or leukemic cells (106/ml). For the clonogenic tumor cell assay, normal MNC were irradiated to 4000 cGy [13] before mixing, and for the C F U - G M assay leukemic cell line cells were also irradiated to 4000 cGy, prior to adding to normal MNC that were not irradiated. Treatment of cells with heat Tumor cell-normal marrow mixtures were suspended in RPMI-1640 medium without FBS [14]. The cell suspension in Falcon 3033 culture tubes was then immersed in a constant water bath (+ 0.1°C) and exposed to a temperature of 42°C for 120 min. Control cells were kept at room temperature for the same time period. After hyperthermic treatment, cell mixtures were immediately cooled in water (4°C) and prepared for limiting dilution and C F U - G M assays. Limiting dilution assay After heating, each sample was serially diluted to concentrations varying from 5 x 10 6 to 0.5 tumor cells per 100 gl in 10% FBS-RPMI medium; 48-96 aliquots of each dilution were plated in flat bottom microculture wells (Becton Dickinson Labware, Oxnard, California). The cells were fed every 4 days and incubated at 37°C with 5% CO2 for 14-18 days. Growth at serial dilution was assessed in an 'all-or-nothing' fashion under an inverted phase microscope [13]. CFU-GM assay The C F U - G M assay was performed according to a modification of the method of Burgess et al, [15]. After heating, normal MNC (105/ml) used as controls and irradiated tumor cell-normal marrow mixtures were cultured in a final concentration of 0.3% agar, 15% FBS and recombinant GM-CSF (100ng/ml) (Behringwerke A G , Germany) in McCoy's 5A modified medium [8]. All cultures were done in triplicate. After 7-10 days of incubation, colonies containing 40 or more cells were counted with an inverted microscope. For determining C F U - G M survival in vitro, the ratio of the colony count in each heated sample to the mean control colony count was calculated.

i

I

30

60

I

I

120

Time(min) a[ 42 ° C

FIG. 1. A comparison between the thermal sensitivity of normal C F U - G M (O) and clonogenic H E L ( A ) , Molt-3 (O) and HL-60 (11) cells in vitro at 42°C. The survival of C F U - G M was estimated as the ratio of colony counts at each point to the count at 0 time. The growth of clonogenic leukemic cell lines was assayed according to a limiting dilution assay to measure log depletion of tumor cells in vitro. Bars (n -- 6); SD, number of colonies at room temperature (controls); CFU°GM, 164 -+ 30/1 x 105 MNC. Significant difference between C F U - G M and all leukemic cell lines at all temperatures studied, and between H E L cells and HL-60 cells, respectively. *p