Synergistic activity of the new ABL-specific tyrosine kinase ... - Nature

Leukemia (2001) 15, 342–347  2001 Nature Publishing Group All rights reserved 0887-6924/01 $15.00 www.nature.com/leu

Synergistic activity of the new ABL-specific tyrosine kinase inhibitor STI571 and chemotherapeutic drugs on BCR-ABL-positive chronic myelogenous leukemia cells J Topaly1, WJ Zeller1 and S Fruehauf2 1

German Cancer Research Center (DKFZ), D0200, Heidelberg; and 2Department of Internal Medicine V, University of Heidelberg, Heidelberg, Germany

The ABL-specific tyrosine kinase inhibitor STI571 (formerly CGP57148B) induced cytogenetic remissions in 33% of chronic myelogenous leukemia (CML) patients in a phase I trial (Druker et al 1999). Combination therapy may increase this proportion. We tested whether combinations of STI571 and cytarabine or other chemotherapeutic agents such as hydroxyurea, mafosfamide or etoposide would display synergistic activity in BCRABL-positive chronic myelogenous leukemia (CML) cell lines derived from patients in blast crisis. In addition, the toxicity of these combinations on BCR-ABL-negative cells was investigated. A tetrazolium-based MTT assay was used to quantify growth inhibition after 48 h of exposure to cytotoxic agents alone and in simultaneous combination with STI571. The drug interactions were analyzed using the median-effect method of Chou and Talalay. The combination index (CI) was calculated according to the classic isobologram equation. At growth inhibition levels of over 50%, STI571 + cytarabine as well as STI571 + etoposide were significantly synergistic (CI ⬍ 1, P ⬍ 0.05) in the BCR-ABL-positive cell lines evaluated. At 60% inhibition or higher, a similar synergistic pattern became apparent for STI571 + mafosfamide (P ⬍ 0.05), while STI571 + hydroxyurea showed ambiguous, cell line-dependent synergism (BV173), additivity (EM-3) or antagonism (K562) in CML cell lines. Furthermore, the BCR-ABL-negative HL-60, KG1a and normal CD34+ progenitor cells were not affected by 0.8 ␮M STI571, a concentration which produced more than 50% growth inhibition in all BCR-ABL-positive cells tested, and no potentiation of growth inhibition was observed in these BCR-ABL-negative cells when STI571 was combined with chemotherapeutic agents. Our in vitro data with CML blast crisis cell lines strongly suggest that combinations of STI571 with cytarabine or etoposide be rapidly considered for clinical testing. Leukemia (2001) 15, 342–347. Keywords: chronic myelogenous leukemia; STI571; combination drug therapy; median effect method of Chou and Talalay; synergism

BV173, EM-3, K562, KG1a, and HL-60 human leukemic cell lines were obtained from the German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany). BV173, EM-3 and K562 represent BCR-ABL-positive cells derived from patients in the blast crisis of CML. KG1a and HL60 are BCR-ABL-negative acute myeloid leukemia cell lines. The BCR-ABL status of all cell lines in our tests was confirmed using reverse transcriptase-polymerase chain reaction (data not shown). Cells were grown in RPMI-1640 medium supplemented with 10% (K562, BV173, HL-60) or with 20% (KG1a) heat inactivated fetal calf serum (FCS), 2 mm l-glutamine, and penicillin/streptomycin (Life Technologies, Eggenstein, Germany) at 37°C in a fully humidified atmosphere of 95% air and 5% CO2.

Introduction

Patient and donor cells

Chronic myelogenous leukemia (CML) is a clonal disorder of the pluripotent stem cell with involvement of all hemopoietic lineages and is characterized by preferential expansion of myeloid cells. The clinical features and therapy of CML were recently reviewed by Sawyers.1 Despite the considerable progress that has been made in the treatment of CML over the past decades, the majority of CML patients achieve only a partial or no cytogenetic response while on conventional therapy and eventually succumb to blast crisis. One of the new and promising therapeutic strategies for this group of patients is the targeting of the BCR-ABL tyrosine kinase. The recently reported phase I trial with the ABL-specific tyrosine kinase inhibitor STI571 in interferon alfa refractory CML patients showed encouraging results.2 Complete hemato-

Leukapheresis samples (from two healthy donors and one non-leukemic patient assigned for autologous stem cell transplantation) and peripheral blood (from one patient with a confirmed newly diagnosed and untreated CML with 150 × 109/l leukocytes in the peripheral blood), further designated as nCD34 and cmlCD34 respectively, were obtained with written informed consent according to procedures approved by the Ethics Committee of the Medical Faculty of the University of Heidelberg. Mononuclear cells were isolated by Ficoll density gradient centrifugation. CD34+ cells were then selected using a MidiMACS CD34 Isolation Kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s instructions. The purity of CD34 cells ranged between 90 and 96% in all samples as determined by flow cytometry. In the CML sample, less than 5 × 106 selected CD34 cells were obtained so that only testing of STI571 as single agent was possible. Combination therapies could not be tested in the primary CML sample. Cells were cultured at an initial concentration of 2.5 ×

Correspondence: S Fruehauf, Department of Internal Medicine V, University of Heidelberg, Hospitalstrasse 3, D-69115, Heidelberg, Germany; Fax: +49–6221–565721 Received 15 September 2000; accepted 13 November 2000

logical remission was achieved in 96% of patients treated with more than 300 mg/day of STI571. Within 2 months of therapy, 33% of patients achieved different levels of cytogenetic response including some cases of complete cytogenetic remission. Based on our own experience with STI571 in vitro,3 we wondered whether the therapeutic index of chemotherapeutic agents can be increased by STI571 which in turn could translate to still higher remission rates for patients. Materials and methods

Cell lines

Combination drug therapy of CML employing ST1571 J Topaly et al

105/ml in Iscove’s modified Dulbecco’s medium (Life Technologies) containing 10% heat inactivated FCS, penicillin/streptomycin and a cytokine mix consisting of 50 ng/ml of Flt-3 ligand (R&D Systems, Wiesbaden, Germany), 50 ng/ml of stem cell factor (R&D Systems), and 20 ng/ml of thrombopoietin (R&D Systems) as previously reported.4

Drugs For this study we chose the chemotherapeutic drugs hydroxyurea and cytarabine, which constitute part of the first line conventional drug treatment of CML,5 etoposide which is included in some regimens for treatment of myeloid blast crisis of CML and mafosfamide (active form of cyclophosphamide) which has been used for ex vivo marrow purging in CML blast crisis.6 STI571 and mafosfamide were kindly provided by Novartis Pharma (Basel, Switzerland) and ASTA Medica (Frankfurt am Main, Germany), respectively. Hydroxyurea (Sigma, Steinheim, Germany), cytarabine (ARA-cell; Cell Pharm, Hannover, Germany), and etoposide (Vepesid J; Bristol Arzneimittel, Munich, Germany) were purchased from the manufacturers. Stock solutions of all drugs at a concentration of 10 mmol/l were prepared by dissolving the compounds in sterile Dulbecco’s phosphate-buffered saline (Dulbecco’s PBS; Life Technologies). Aliquots were stored at −20°C until use. The stock solutions were thawed and serial dilutions of each drug were prepared freshly prior to the start of incubation. Simultaneous exposure to STI571 in combination with the chemotherapeutic drug was chosen to ensure a continuous inhibition of the BCR-ABL tyrosine kinase. Cells were treated with five increasing concentrations (doubling with each increment) of the chemotherapeutic drugs alone, then in combination with five different concentrations of STI571 (0.05, 0.1, 0.2, 0.4, and 0.8 ␮m), and with STI571 alone.

Table 1 Inhibitory activity of single agents on the proliferation of CML cells and estimation of equitoxic dose ratios for the combination treatment testing

Drug/Cell type

Concentration range testeda (␮M)

STI571 BV173 0.05–0.8 EM-3 0.05–0.8 K562 0.05–0.8 Cytarabine BV173 0.00125–0.02 EM-3 0.25–4.0 K562 10–160 Hydroxyurea BV173 10–160 EM-3 50–800 K562 200–3200 Mafosfamide BV173 0.2–3.2 EM-3 1–16 K562 8–128 Etoposide BV173 0.025–0.4 EM-3 0.5–8 K562 4–64

Equitoxic ratiob (ST1571: drug 2)

0.32 ± 0.08 0.17 ± 0.06 0.26 ± 0.05 0.0082 ± 0.0005 1.1 ± 0.2 n/ec 72 ± 11 160 ± 20 1040 ± 80

40:1 1:5 1:200 1:1000 1:4000

1.1–0.02 3.0 ± 0.1 31 ± 8

1:4 1:20 1:160

0.13 ± 0.01 1.8 ± 0.9 16 ± 7

2:1 1:10 1:80

Mean values ± 1 s.d. are given, n = 4 each. a Concentrations were doubled with each increment. b Equitoxic ratios for testing the drug combinations were chosen on the basis of single drug concentrations that lead to a 50% growth inhibition in the target cell line (IC50). c Not evaluable, insufficient dose–response correlation in the median effect plot (r ⬍ 0.95), this cell line was therefore excluded from further testing of ara-C + STI571. Table 2 Inhibitory activity of single agents on the proliferation of BCR-ABL-negative cells

Drug

Cell type

MTT assay The MTT assay is based on the cleavage of the yellow tetrazolium salt 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT; Sigma) to purple formazan crystals by viable cells.7 A high correlation between the viable cell number and formazan production has been reported.8,9 Cells were washed and resuspended in culture medium at a concentration of 3 × 105 cells/ml. To achieve logarithmic cell growth, cells were preincubated for 24 h in tissue culture flasks. Then, 90 ␮l aliquots of the cell suspension were dispensed into 96-well flat-bottomed microtiter plates (Becton Dickinson, Heidelberg, Germany) containing 10 ␮l/well of serial drug dilutions in Dulbecco’s PBS or Dulbecco’s PBS alone as control. The range of final drug concentrations is given in Tables 1 and 2. The drugs remained in the plates until the end of experiments. The plates were incubated for 48 h at 37°C in a humidified atmosphere containing 5% CO2. Then, 15 ␮l of a 5 mg/ml solution of MTT in RPMI was added to each well and the plates were incubated for another 4 h at the same conditions. Next, 100 ␮l/well of 10% sodium dodecyl sulfate solution in 0.01 m hydrochloric acid was added to solubilize the formazan crystals. After incubation at 37°C overnight the absorbance (A) was measured at 540 nm (reference wavelength 690 nm) with a Multiskan Bichromatic plate reader (Labsystems, Helsinki, Finland). The absorbance

IC50 (␮M)

343

STI571b Cytarabine Hydroxyurea Mafosfamide Etoposide

HL-60 KG1a nCD34 HL-60 KG1a nCD34 HL-60 KG1a nCD34 HL-60 KG1a nCD34 HL-60 KG1a nCD34

Concentration range testeda (␮M) 0.2–51.2 0.2–51.2 0.2–51.2 2–32 1–16 0.05–8.0 50–800 100–1600 2.5–640 1–16 1–16 0.05–8.0 0.05–0.8 1–16 0.0625–3.2

IC50 (␮M)

17.1 26.7 3.94 6.32 2.41 0.114 160 654 147 3.64 5.85 1.49 0.157 7.10 0.340

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.7 3.3 1.14 0.83 0.97 0.085 26 228 27 0.46 1.63 0.14 0.020 3.03 0.173

a

Concentration was doubled with each increment. Concentration range used to determine the IC50 values is given; in combination treatment experiments the same concentration range of STI571 was used as in BCR-ABL-positive cells (cf. Table 1).

b

of blank controls containing the respective drug dilutions in culture medium but no cells was subtracted. The mean of four (cell lines) or three (patient and donor cells) replicate wells for each drug dilution was used to calculate growth inhibition defined as affected fraction Leukemia


344

(Fa = 1 − [mean Atreated sample/mean Auntreated sample]). Fa values of 0.25, 0.5 or 0.75, therefore, correspond to 25%, 50% or 75% growth inhibition.

for any effect level (Fa) and, vice versa, the effect (Fa) for any dose level (D) may be derived from equation C. This method was applied to calculate the dose–effect relationships (eg IC25, IC50 or IC75) for each drug and cell type.

Analysis of dose–effect curves The dose–effect curves were analyzed by the median-effect method of Chou and Talalay10 using the Calcusyn Software (Biosoft, Cambridge, UK). The sigmoid dose-effect curves were linearized using the median effect equation: Fa/Fu = (D/Dm)m or log(Fa/Fu) = m log(D) − m log(Dm)

(A)

In this equation, D is the dose of drug, Dm is the median– effect dose (IC50) signifying the potency, Fa is the fraction affected by dose D, Fu is the unaffected fraction (Fu = 1 − Fa), and m is a coefficient related to the shape of the dose– effect curve. The median-effect plot x = log(D) vs y = log(Fa/Fu), which is based on the logarithmic form of equation A, yielded a straight line when the dose–effect relationship followed the principle of mass action (Figure 1). The linear correlation coefficient (r) was generated for each curve to determine the applicability of the data to this method of analysis. Of notice, Dm is represented by the x-intercept of the median effect plot. The equation A may be rearranged as follows: D = Dm[Fa/1 − Fa)]1/m

(B)

Fa = 1/[1 + (Dm/D)m]

(C)

Equation B may thus be solved providing the drug dose (D)

Figure 1 An example of the dose–response curve of ST1571 in BCR-ABL-positive BV173 cells (a) and the data transformation to allow calculation of the level of growth inhibition at a given dose according to the median-effect equation of Chou and Talalay (b)19 are given. The dose–response relationship is displayed as a linear regression line (r = 0.997). The x-intercept of the median effect-plot (b) represents the median effect dose at which a 50% growth inhibition occurs (Dm, Fa = 0.5 or IC50). Fa affected fraction; Fu unaffected fraction; D dose). Leukemia

Statistics Results are shown as the mean value ± standard deviation (s.d.) of three or four separate experiments. Statistical significance of the data were calculated by Student’s t-tests. In BCRABL-positive cells it was of interest whether CI values were significantly lower than 1 (CI = 1 is mere additive effect) and a one-sided t-test was used. In BCR-ABL-negative cells the IC50 values in paired experiments with and without STI571 were compared in a two-sided paired t-test to recognize a possible potentiation or inhibition of the chemotherapy effect. Significance levels of P ⬍ 0.05 and P ⬍ 0.01 were chosen.

Results

Single agent therapy IC50 values of STI571 alone were assessed following 48 h of incubation for all cell lines or primary cell samples (Tables 1 and 2). BCR-ABL-positive cells were highly susceptible to the inhibitory effects of STI571. IC50 values ranged from 0.17 to 0.32 ␮m in the BCR-ABL-positive cell lines BV173, EM-3 and K562. In CD34+ hemopoietic cells obtained from a CML patient the IC50 was 0.68 ␮m (not shown in Table). Interestingly, the BCR-ABL-negative CD34+ hemopoietic cells proved to be more susceptible to STI571-induced growth inhibition as compared to the BCR-ABL-negative cell lines HL-60 and KG1a. The corresponding IC50 values (mean ± s.d.) were 3.94 ± 1.14 vs 17.1 ± 0.7 and 26.7 ± 3.3 ␮m, respectively (Table 2). As the CD34+ cells were cultured in SCF (c-kit ligand) and STI571 is known to inhibit c-kit, this could account for the higher susceptibility of the normal (BCR-ABL-negative) CD34+ cells to STI571 as compared to the cell lines that were cultured without growth factors. The therapeutic index of STI571 defined as the ratio of IC50 in BCR-ABL-negative cells (doselimiting side-effect) to IC50 in BCR-ABL-positive cells (desired therapeutic effect), therefore, ranged from 5.8 in primary cells to 157 in cell lines. In a row of serial 2-fold dilutions, 0.8 ␮m of STI571 proved to be the highest concentration that failed to measurably affect any of the BCR-ABL-negative cells including CD34+ progenitors (data not shown). On the other hand, this concentration caused more than 50% of growth inhibition in all BCR-ABL-positive cells tested and was, for these two reasons, set as the highest STI571 concentration for our combination treatment experiments in BCR-ABL-negative cells. The sensitivity of cells to chemotherapeutic drugs varied widely depending on the cell type tested, eg the IC50 of cytarabine in the EM-3 cell line was 134-fold and the IC50 of mafosfamide in the K562 cell line was 28-fold higher than in the BV173 cell line. In all experiments, apart from exposure of K562 to cytarabine, the linear correlation coefficient of the median-effect plot (r) was ⬎0.95. The combination STI571 + cytarabine in K562 cells was therefore excluded from further analysis.


345

Combination drug treatment, BCR-ABL-positive cells The approximately equitoxic concentrations of the drug combinations were chosen for assessment of drug interactions. Because each cell line displayed different sensitivities to the drugs tested, the fixed molar ratios were adjusted individually for each BCR-ABL-positive cell type (Table 2). An example of the experimental design is given in Table 3. We considered a maximum dose of 0.8 ␮m STI571 sufficient for the combination therapy effect testing because growth inhibition levels of up to 99% were achieved in BCR-ABL-positive cells (Table 3). The combination index (CI) was used to express synergism (CI ⬍ 1), additivity (CI = 1) or antagonism (CI ⬎ 1). The CI was calculated according to the classic isobologram equation: CI = d1/D1 + d2/D2

(D)

In this equation, D1 and D2 represent the doses of drug 1 and drug 2 alone, required to produce x% effect, and d1 and d2 are the doses of drugs 1 and 2 in combination required to produce the same effect (Figure 2). In Figure 2 the synergism of STI571 and cytarabine becomes apparent because the concentrations of both drugs in combination resulting in an inhibition level of, eg 75% is lower than expected from the testing of single agents. Equitoxic ratios for combination testing of STI571 and chemotherapeutic drugs were chosen on the basis of single drug concentrations that led to a 50% growth inhibition (Fa = 0.5) in the target cell line (Table 1). Different CI values were obtained for solving the equation D for different effect levels (eg 25%, 50% or 75% of growth inhibition). The results are therefore shown in a CI vs Fa plot (Figure 3). When STI571 was combined with cytarabine a significant (P ⬍ 0.05) synergistic effect became apparent at therapeutically relevant inhibition levels above 50% which still increased at higher inhibition levels (P ⬍ 0.01 at IC75; Figure 3a, b; CI values ⬍1). When combined with hydroxyurea equivocal results were Table 3 Experimental design of combination therapy and collection of data

Example of combination drug testing of STI571 and cytarabine in BCR-ABL-positive BV173 cells. Twenty-five combinations were tested in quadruplicate. Means of primary data from four experiments are given. A MTT assay was used to determine the affected fraction (Fa) or the level of growth inhibition (eg Fa 0.5 = 50% growth inhibition). The synergism becomes apparent when comparing STI571 alone (0.8 ␮M, Fa 0.70), cytarabine alone (0.02 ␮M, Fa 0.73) and the combination STI571 (0.4 ␮M)/cytarabine (0.01 ␮M) [50% dose, each] where the growth inhibition is higher than expected (Fa 0.92). In cases of additive activity the growth inhibition of the drug combination [50% dose, each] would be equal to the full dose of the single agents (Fa ca. 0.7) while in cases of antagonism a lower growth inhibition would be observed (Fa ⬍ 0.7). The IC50 of ST1571 was 0.32 ␮M and the IC50 for cytarabine was 0.008 ␮M or a ratio of 40:1. Using this equitoxic ratio of 40:1 the respective combinations (boxed values) were chosen for further analysis (cf. Figure 3).

Figure 2 Classic isobologram at IC75, ST1571 + cytarabine in BV173 cells. Mean combination index (CI) values of four experiments are shown with STI571/cytarabine molar ratios of 10:1 (a), 40:1 (b, equitoxic molar ratio) and 160:1 (c). CI values were calculated from a series of primary data presented in Table 4. As CI values of ⬍1, 1 and ⬎1 express synergism, mere additive effect and antagonism, respectively, this case represents an example of a synergistic combination (CI ⬍ 0.69).

obtained (Figure 3, c–e). In BV173 cells the combination STI571 + hydroxyurea was slightly synergistic at inhibition levels above 80% (CI ⬍ 1, P = 0.05), while in EM-3 cells an additive effect (CI = 1) and in K562 cells an antagonistic effect (CI ⬎ 1) was observed. STI + mafosfamide were synergistic above inhibition levels of 60% (P ⬍ 0.05, Figure 3f–h). STI571 + etoposide (Figure 3i–k) were similarly synergistic as found for cytarabine (P ⬍ 0.05 at IC50 and P ⬍ 0.01 at IC75).

Combination drug treatment, BCR-ABL-negative cells Since STI571 by itself produced no measurable effect in BCRABL-negative cells at therapeutically relevant concentrations, the median-effect method could not be applied in this case. Instead, an analysis in terms of potentiation or inhibition was carried out to assess the toxicity of combination drug therapy on BCR-ABL-negative cells. For better comparability between different cell types the results are given as relative IC50 value, a ratio of the IC50 value of a chemotherapeutic agent given in combination with STI571 to the IC50 value of the same agent given alone. No significant potentiation was observed as shown in Figure 4. Interestingly, a trend to lower sensitivity to chemotherapeutic drugs was observed in normal progenitors (relative IC50 ⬍ 1) while in leukemic cell lines the mean relative IC50 tended to be ⬎1. Discussion On the cellular level, CML is distinguished by the Philadelphia chromosome (Ph),11 an abnormally short chromosome 22, which is found in more than 90% of all cases of CML, arising from a reciprocal t(9;22) translocation.12 The product of this translocation, a 210 kDa BCR-ABL protein (p210BCR-ABL), is characterized by a constitutively enhanced tyrosine kinase activity as compared with that of the normal c-ABL protein13,14 and is known to interfere with a variety of cytoplasmatic and cytoskeletal signaling proteins and cascades,15 which eventually leads to inhibition of apoptosis. p210BCR-ABL-mediated protection of hemopoietic cells from the induction of apoptosis induced by irradiation, treatment with cytotoxic Leukemia


346

Figure 4 Effects of chemotherapeutic drugs ± 0.8 ␮m STI571 on BCR-ABL-negative cells. For better comparability between different cell types data are presented as relative IC50 of chemotherapeutic drugs (IC50 with STI571/IC50 without STI571). Values of 1, ⬎1 and ⬍1 correspond to no influence, potentiation and inhibition, respectively. Addition of STI571 to the chemotherapeutic drugs does not potentiate their toxicity on BCR-ABL negative cells. Black columns, HL-60 cell line; white columns, KG1a cell line; hatched columns, normal CD34+ cells; bars, 1 s.d. Means of four (cell lines) or three (nCD34) paired experiments are presented.

Figure 3 Combination index (CI) as a function of affected fraction (Fa) in BCR/ABL-positive cell lines. Fa values of 0.25, 0.5 or 0.75 correspond to 25%, 50% or 75% growth inhibition. CI values were calculated on the basis of a range of equitoxic drug combinations as shown in Table 4. This way CI values for different levels of growth inhibition, for different drug combinations or for different cell lines can easily be compared. It is notable that the synergistic activity of most combinations (plots a, b, f, g, h, i, j, k) increases with higher levels of growth inhibition, suggesting that when ST1571 is efficiently blocking the ATP binding site of BCR-ABL the downregulation of anti-apoptotic mechanisms allows the full activity of pro-apoptotic chemotherapeutic drugs on the CML cells to occur. Data points, means of four independent experiments; bars, 1 s.d.

drugs,16–18 Fas ligation,19 and cytokine withdrawal20 has been reported. The inherent resistance of BCR-ABL-positive cells to cytotoxic therapy is a major impediment to the management of CML. Systematic development of inhibitors targeting the ATPbinding sites of a number of tyrosine kinases lead to generation of STI571, which was shown to reversibly inhibit the tyrosine kinases of the normal c-ABL, ␣- and ␤-PDGF-R, and c-kit proteins and of the BCR-ABL and Tel-PDGF-R fusion proteins at micromolar concentrations.21–24 Inhibition of the BCRABL tyrosine kinase by STI571 leads to induction of apoptosis in BCR-ABL-positive cells25,26 and allows the selection for BCR-ABL-negative long-term culture initiating cells from leukapheresis products of CML patients.7 In agreement with the earlier published data, we found STI571 as single agent to exert selective toxicity on cells carrying BCR-ABL. Whereas the sensitivity of BCR-ABL-positive cells to chemotherapeutic drugs differed substantially, their sensitivity to STI571 showed a minor variability (IC50 values ranged from 0.17 to 0.68 ␮m, Table 1), which implies absence Leukemia

of primary resistance to STI571 in the tested BCR-ABL-positive cells and independence from existing resistance to chemotherapeutic drugs used in our study. On the other hand, the development of secondary resistance to STI571 in the course of a long-term incubation is becoming evident as reported recently.27,28 Combination drug treatment may be one way to overcome this problem. Our data imply that STI571 exhibits strong synergism with apoptosis-inducing cytarabine, mafosfamide and etoposide at higher levels of growth inhibition, which may originate from increasing inhibition of the BCR-ABL tyrosine kinase with subsequent induction of apoptotic pathways by these chemotherapeutic drugs. This points out the necessity of a high level of BCR-ABL tyrosine kinase inhibition in the course of the combination drug therapy to achieve higher levels of synergy and, thus, to fully exploit the therapeutic potential of the drug combination. Hydroxyurea, being a representative of G1/S arresting agents, is ineffective in induction of apoptosis,29,30 which explains the merely additive or even slightly antagonistic effect when combined with STI571. In comparison to drug combinations that we studied earlier for synergistic activity where both agents exerted strong toxicity on normal cells,31,32 the combination of chemotherapeutic drugs with STI571 has the advantage of increased selectivity to the leukemic clone. STI571 at therapeutically relevant concentration levels neither produced any measurable growth inhibition in BCR-ABLnegative cell lines and normal CD34+ progenitors nor potentiated the toxicity of chemotherapeutic drugs on these cells thus leading to a higher therapeutic index of the drug combination as compared to chemotherapeutic drugs alone. Whether synergistic activity in vitro will translate in better response and survival rates of CML patients receiving combination therapy vs STI571 alone must be proven in future clinical trials. In our experimental model, we used BCR-ABL-positive leukemic cell lines derived from the blast crisis of CML where STI571 by itself has only transient efficacy. Therefore, the data on synergistic effects of STI571 and cytarabine or eto-


poside are of special interest for this patient group and may aid in the design of future clinical trials.

Acknowledgements We appreciate the continuing support of Prof AD Ho (Department of Internal Medicine V, University of Heidelberg, Germany). We are grateful to S Heil, B Berkus, and HJ Engel for expert technical assistance (German Cancer Research Center, Heidelberg, Germany), to Dr E Buchdunger and B Willi (Novartis Pharma AG, Basel, Switzerland) for supplying STI571 and to Dr J Pohl (ASTA Medica, Frankfurt am Main, Germany) for supplying mafosfamide. This work was supported in part by grant DJCLS-R00/03 of the Deutsche Jose´ Carreras Leuka¨mie Stiftung eV.

References 1 Sawyers CL. Chronic myeloid leukemia. N Engl J Med 1999; 340: 1330–1340. 2 Druker BJ, Talpaz M, Resta D, Peng B, Buchdunger E, Ford J, Sawyers CL. Clinical efficacy and safety of an Abl-specific tyrosine kinase inhibitor as targeted therapy for chronic myelogenous leukemia. Blood 1999; 94 (Suppl. 1): 368a. 3 Kasper B, Fruehauf S, Schiedlmeier B, Buchdunger E, Ho AD, Zeller WJ. Favorable therapeutic index of a p210(BCR-ABL)-specific tyrosine kinase inhibitor; activity on lineage-committed and primitive chronic myelogenous leukemia progenitors. Cancer Chemother Pharmacol 1999; 44: 433–438. 4 Schiedlmeier B, Ku¨hlcke K, Eckert HG, Baum C, Zeller WJ, Fruehauf S. Quantitative assessment of retroviral transfer of the human multidrug resistance 1 gene to human mobilized peripheral blood progenitor cells engrafted in nonobese diabetic/severe combined immunodeficient mice. Blood 2000; 95: 1237–1248. 5 Silver RT, Woolf SH, Hehlmann R, Appelbaum FR, Anderson J, Bennett C, Goldman JM, Guilhot F, Kantarjian HM, Lichtin AE, Talpaz M, Tura S. An evidence-based analysis of the effect of busulfan, hydroxyurea, interferon, and allogeneic bone marrow transplantation in treating the chronic phase of chronic myeloid leukemia: developed for the American Society of Hematology. Blood 1999; 94: 1517–1536. 6 Carlo-Stella C, Mangoni L, Almici C, Caramatti C, Cottafavi L, Dotti GP, Rizzoli V. Autologous transplant for chronic myelogenous leukemia using marrow treated ex vivo with mafosfamide. Bone Marrow Transplant 1994; 14: 425–432. 7 Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assay. J Immunol Meth 1983; 65: 55–63. 8 Finlay GJ, Wilson WR, Baguley BC. Comparison of in vitro activity of cytotoxic drugs towards human carcinoma and leukemia cell lines. Eur J Cancer Clin Oncol 1986; 22: 655–662. 9 Pieters B, Huismans DR, Leyva A, Veerman AJP. Adaptation of the rapid automated tetrazolium dye based (MTT) assay for chemosensitivity testing in childhood leukemia. Cancer Letters 1988; 41: 323–332. 10 Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984; 22: 27–55. 11 Nowell PC, Hungerford DA. A minute chromosome in chronic granulocytic leukemia. Science 1960; 132: 1497–1501. 12 Rowley JD. A new consistant chromosomal abnormality in chronic myelogenous leukemia identified by quinacrine, fluorescence and Giemsa staining. Nature 1973; 243: 290–293. 13 Konopka JB, Watanabe SM, Witte ON. An alteration of the human c-abl protein in K562 leukemia cells unmasks associated tyrosine kinase activity. Cell 1984; 37: 1035–1042.

14 Sawyers CL, Denny CT, Witte ON. Leukemia and the disruption of normal hematopoiesis. Cell 1991; 64: 337–350. 15 Warmuth M, Danhauser-Riedl S, Hallek M. Molecular pathogenesis of chronic myeloid leukemia: implications for new therapeutic strategies. Ann Hematol 1999; 78: 49–64. 16 McGahon A, Bissonette R, Schmitt M, Cotter KM, Green DR, Cotter TG. BCR-ABL maintains resistance of chronic myelogenous leukemia cells to apoptotic cell death. Blood 1994; 83: 1179– 1187. 17 Bedi A, Barber JP, Bedi GC, el-Deiry WS, Sidransky D, Vala MS, Akhtar AJ, Hilton J, Jones RJ. BCR-ABL-mediated inhibition of apoptosis with delay of G2/M transition after DNA damage: a mechanism of resistance to multiple anticancer agents. Blood 1995; 86: 1148–1158. 18 Nishii K, Kabarowski JH, Gibbons DL, Griffiths SD, Titley I, Wiedemann LM, Greaves MF. ts BCR-ABL kinase activation confers increased resistance to genotoxic damage via cell cycle block. Oncogene 1996; 13: 2225–2234. 19 McGahon AJ, Nishioka WK, Martin SJ, Mahboubi A, Cotter TG, Green DR. Regulation of the Fas apoptotic cell death pathway by Abl. J Biol Chem 1995; 270: 22625–22631. 20 Bedi A, Zehnbauer BA, Barber JP, Sharkis SJ, Jones RJ. Inhibition of apoptosis by BCR-ABL in chronic myeloid leukemia. Blood 1994; 83: 2038–2044. 21 Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, Fanning S, Zimmermann J, Lydon NB. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of bcr/abl positive cells. Nat Med 1996; 2: 561–566. 22 Buchdunger E, Zimmermann J, Mett H, Meyer T, Muller M, Druker BJ, Lydon NB. Inhibition of the Abl protein-tyrosine kinase in vitro and in vivo by a 2-phenylaminopyrimidine derivative. Cancer Res 1996; 56: 100–104. 23 Carroll M, Ohno-Jones S, Tamura S, Buchdunger E, Zimmermann J, Lydon NB, Gilliland DG, Druker BJ. CGP57148, a tyrosine kinase inhibitor, inhibits the growth of cells expressing BCR-ABL, TEL-ABL and TEL-PDGFR fusion proteins. Blood 1997; 90: 4947–4952. 24 Buchdunger E, Cioffi CL, Law N, Stover D, Ohno-Jones S, Druker BJ, Lydon NB. Abl protein-tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-Kit and platelet-derived growth factor receptors. J Pharmacol Exp Ther 2000; 295: 139– 145. 25 Gambacorti-Passerini C, le Coutre P, Mologni L, Fanelli M, Bertazzoli C, Marchesi E, Di Nicola M, Biondi A, Corneo GM, Belotti D, Pogliani E, Lydon NB. Inhibition of the ABL kinase activity blocks the proliferation of BCR/ABL+ leukemic cells and induces apoptosis. Blood Cells Mol Dis 1997; 23: 380–394. 26 Horita M, Andreu EJ, Benito A, Arbona C, Sanz C, Benet I, Prosper F, Fernandez-Luna JL. Blockade of the Bcr-Abl kinase activity induces apoptosis of chronic myelogenous leukemia cells by suppressing signal transducer and activator of transcription 5-dependent expression of Bcl-xL. J Exp Med 2000; 191: 977–984. 27 le Coutre P, Tassi E, Varella-Garcia M, Barni R, Mologni L, Cabrita G, Marchesi E, Supino R, Gambacorti-Passerini C. Induction of resistance to the abelson inhibitor STI571 in human leukemic cells through gene amplification. Blood 2000; 95: 1758–1766. 28 Mahon FX, Deininger DWN, Schultheis B, Goldman JM, Melo JV. Selection and characterization of BCR-ABL positive cell lines with differential sensitivity to the signal transduction inhibitor STI571: mechanisms of resistance. Blood 1999; 94 (Suppl. 1): 656a. 29 Lin CK, Nguyen TT, Morgan TL, Mei RL, Kaptein JS, Kalunta CI, Yen CF, Park E, Zou HY, Lad PM. Apoptosis may be either suppressed or enhanced with strategic combinations of antineoplastic drugs or anti-IgM. Exp Cell Res 1998; 244: 1–13. 30 Johnson KR, Young KK, Fan W. Antagonistic interplay between antimitotic and G1-S arresting agents observed in experimental combination therapy. Clin Cancer Res 1999; 5: 2559–2565. 31 Zeller WJ, Berger M, Schma¨hl D. Synergistic action of vincristine and adriamycin in the treatment of experimental rat leukemia L5222. Cancer Res 1979; 39: 1071–1073. 32 Zeller WJ, Berger MR, Henne T, Weber E. More than additive toxicity of the combination of 1-methyl-1-nitrosourea plus 1,3-bis(2chloroethyl)-1-nitrosourea in the rat. Cancer Res 1986; 46: 1714–1716.

347

Leukemia