Schedule-Dependent Synergism and Antagonism Between Methotrexate

Home , Amsacrine, Cytarabine, Doxorubicin, Methotrexate, Raltitrexed, Vincristine

Leukemia (2002) 16, 1808–1817  2002 Nature Publishing Group All rights reserved 0887-6924/02 $25.00 www.nature.com/leu Schedule-dependent synergism and antagonism between methotrexate and cytarabine against human leukemia cell lines in vitro M Akutsu1, Y Furukawa2, S Tsunoda1, T Izumi1,3, K Ohmine1,3 and Y Kano1

1Division of Medical Oncology, Tochigi Cancer Center, Tochigi, Japan; 2Division of Stem Cell Regulation, Jichi Medical School, Tochigi, Japan; and 3Division of Hematology, Jichi Medical School, Tochigi, Japan

Methotrexate (MTX) and cytarabine have been widely used for istic effects.16,18,23 The sequential exposure to both agents also the treatment of acute leukemias and lymphomas for over 30 produced conflicting results.15,17,18,20,21,23,24 No well-designed years. However, the optimal schedule of this combination is yet to be determined and a variety of schedules of the combination clinical randomized trial of this combination has been has been used. We studied the cytotoxic effects of MTX and available. cytarabine in combination against human leukemia cell lines at In experimental combination studies, a great deal of con- various schedules in vitro. The effects of the combinations at fusion has arisen because dose–response curves or survivals the concentration of drug that produced 80% cell growth inhi- for anticancer agents are variable and different investigators 80 bition (IC ) were analyzed using the isobologram method of have disagreed as to what constitutes an ‘expected’ level of Steel and Peckham. Simultaneous exposure to MTX and cytara- 25,26 bine for 3 days produced antagonistic effects in human T cell effect when two agents are combined. Although dose- leukemia, MOLT-3 and CCRF-CEM, B cell leukemia, BALL-1, response curves and survivals of anticancer agents are Burkitt’s lymphoma, Daudi, promyelocytic leukemia, HL-60 and important factors in the analysis of the effects of drug combi- Philadelphia chromosome-positive leukemia, K-562 cells. Sim- nations, many studies lack full dose–response data for each ultaneous exposure to MTX and cytarabine for 24 h produced drug, either alone or in combination. antagonistic effects, sequential exposure to MTX for 24 h fol- In in vivo studies, synergism (therapeutic synergism)is often lowed by cytarabine for 24 h produced synergistic effects, and the reverse sequence produced additive effects in both CCRF- defined as if the combinations produce a better response than CEM and HL-60 cells. Sequential exposure to MTX for 24 h fol- either drug alone, with acceptable toxicity. This definition is lowed by cytarabine for 3 days also produced synergistic a quite different concept from true synergism, which means effects in MOLT-3 cells. Cell cycle analysis supported these that the effect of combination is more than expected, when observations. Our findings suggest that the simultaneous combined.27 In in vitro studies, the classical isobologram28 administration of MTX and cytarabine is not appropriate and and fractional product method29 have often been used, which the sequential administration of MTX followed by cytarabine may be the optimal schedule of this combination. assume linear–dose response curves for each drug and combi- Leukemia (2002) 16, 1808–1817. doi:10.1038/sj.leu.2402573 nations, which is seldom the case. The median effect Keywords: methotrexate; cytarabine; synergism; antagonism; iso- method30 requires rigid sigmoid dose–response curves, based bologram upon the Michaels-Menten and Hill model. However, cytotoxic mechanisms of anticancer agents are multifactorial and dose–response curves are often not sigmoid and regression Introduction analyses are difficult or forced. We use the isobologram method of Steel and Peckham,31 because this method can be Methotrexate (MTX)and cytarabine have been widely used used to calculate additive interaction of most combinations, for the treatment of acute leukemias and lymphomas for over irrespective of the shapes of the dose–response curves of the 30 years. However, the optimal schedule of the combination agents and whether they have independent or overlapping of MTX and cytarabine is yet to be determined. In regimens damage. including MTX and cytarabine, the schedules of administering To clarify the optimal schedule of the combination of MTX MTX and cytarabine have been variable. In general, MTX and and cytarabine, we studied the effects of simultaneous and cytarabine have been used separately.1–3 However, in some sequential exposure to MTX and cytarabine in human regimens, MTX and cytarabine have been administered simul- leukemia cell lines in vitro using the isobologram of Steel and taneously (on the same day), eg as triple combinations for cen- Peckham. The present findings clearly demonstrated marked tral nervous system (CNS)prophylaxis and CNS leukemia, 4,5 antagonistic effects in simultaneous exposure to MTX and PROMACE-cytaBOM6,7 and AMOPLACE8 for non-Hodgkin’s cytarabine and marked synergistic effects in sequential lymphoma, POG8602 and EORTC58881 for pediatric acute exposure to MTX followed by cytarabine, suggesting that the lymphoblastic leukemia (ALL)9,10 and sequentially on different optimal schedule of this combination may be the sequential days, eg Hyper-CVAD for ALL and CHOD/BVAM for administration of MTX followed by cytarabine, but not the lymphoma.11–14 simultaneous administration of these agents. Experimental findings for the combination of MTX and cytarabine, most of which were obtained more than 20 years ago, are variable (Table 1). In both in vitro and in vivo experiments, Materials and methods the simultaneous exposure to MTX and cytarabine produced conflicting results:15–24 synergistic effects15,19–22 and antagon- Cell lines

Experiments were conducted with the T cell ALL lines MOLT- 3 and CCRF-CEM, the B cell ALL line BALL-1,32 Burkitt’s lym- Correspondence: Y Kano, Division of Medical Oncology, Tochigi Cancer Center, Yonan 4-9-13, Utsunomiya, Tochigi, 320-0834, Japan; phoma line Daudi, the Philadelphia chromosome-positive Fax: 011–81–28–658–5488 leukemia line K-562, and the acute promyelocytic leukemia Received 27 November 2001; accepted 12 March 2002 line HL-60. CCRF-CEM, HL-60, K-562, MOLT-3 cells were MTX and ara-C in combination M Akutsu et al 1809 C) (68 h) ND synergism 23 C (2 h) NDC (6 h)C (6 h) fractional product concept fractional product concept antagonism synergism synergism 23 23 24 C (42 h)C (2 h)C (68 h) classical isobologram fractional product ND concept synergism antagonism 18 21 antagonism 23 C (42 h)C (68 h)C (68 h) classical isobologram ND ND additive antagonism 18 antagonism 23 23 + + + + + + + + + + M M M M M M (M M M M → → → → → → → → → → C (1 h)C (2 h) fractional product concept fractional product concept synergism antagonism 17 21 → → CCC (5 h)CC (72 h)C (72 h) classical isobologram fractional product classical concept isobologram ND ND NDC synergismC antagonismC antagonismC 21 16 18 antagonism synergism increased dCPT survival time survival 22 time 23 survival time fractional 23 product concept synergism antagonism antagonism synergism 20 15 18 19 C survival time additive 18 C fractional product concept synergism 20 + + + + + + + + + + M fractional product concept synergism 20 → → → HL-60L5178Y3924A8999R sequential and simultaneous: sequential M (2 and simultaneous: h) C (6 sequential h) and simultaneous: C sequential (4 and h) simultaneous: C (4 h) L1210ND3924A8999RL5178Y simultaneous: M L1210L5178YL1210 simultaneous: simultaneous: M 3924A M simultaneous:8999R M sequential: M (2 h) sequential: M (3 h) sequential and simultaneous: M (6 sequential h) and simultaneous: sequential M and (3 simultaneous: h) M sequential (4 and h) simultaneous: M (4 h) L1210TLX/5L1210L1210TLX/5 simultaneous: M simultaneous: M simultaneous: M simultaneous: M sequential: M L1210 sequential: M L1210 sequential: C Previous experimental studies for the combination of methotrexate (M)and cytarabine (C) (colonyformation) BF5, N1S1, 8999F 3924 sequential sequential and and simultaneous: simultaneous: M M (4 (4 h) h) SystemIn vitro (cell growthinhibition) Cell line L5178Y L1210 simultaneous: M simultaneous: M ScheduleIn vivo Evaluation Effect Ref. Table 1 ND, not described.

Leukemia MTX and ara-C in combination M Akutsu et al 1810 obtained from the Health Science Research Resources Bank MTT assay (Osaka, Japan). BALL-1 and Daudi cells were obtained from RIKEN Cell Bank (Tsukuba, Japan). All cell lines were cultured Cell growth inhibition was determined by a 3-(4,5-dimethyl- in RPMI1640 medium (GIBCO, Grand Island, NY, USA)sup- thiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay.33 plemented with 10% heat-inactivated fetal calf serum Fifty microliters of MTT (1 mg/ml)was added to each well. (GIBCO), 50 units/ml penicillin and 5 ␮g/ml streptomycin at Following incubation for 4 h at 37°C, culture plates were cen- ° 37 C in a humidiﬁed chamber with 5% CO2 in air. trifuged at low speed and the supernatant was removed. Dimethyl sulfoxide (150 ␮l)was then added and the plates were shaken to solubilize the MTT-formazan product. Drugs Absorbance at 570 nm was measured with a Titertek multi- scan. For the background control, control (no drug), each MTX and cytarabine were obtained from Lederle Japan drug, or drug combinations, the four intermediate data among (Tokyo, Japan), and Nihon Shinyaku (Tokyo, Japan), respect- the eight data were used for the analysis, and the two highest ively. Drugs were dissolved in and diluted with culture and the two lowest data were discarded. We established a medium. linear relation between the MTT assay and cell number within the range of the experiments shown.

Cell growth inhibition by the combination of MTX and cytarabine Isobologram

Eight plates were prepared for one combined drug test. Dose–response interactions between MTX and cytarabine at

the point of IC80 were evaluated using the isobologram method of Steel and Peckham.31 The theoretical basis of the Protocol 1: simultaneous exposure to MTX and cytarabine for isobologram method and the procedure for making the isobol- 3 days: All six leukemia cell lines were used for the ogram has been described in detail.34,35 Based on the dose– experiments. Logarithmically growing cells were harvested, response curves of MTX and cytarabine, three isoeffect curves resuspended in fresh medium (0.5–1 × 105 cells/ml)and were constructed (Figure 1). If the agents were acting addi- plated in 96-well plates (Falcon, Oxnard, CA, USA)(100 ␮l tively by independent mechanisms, combined data points cell suspension per well)and all plates had a control column would lie near the mode I line (hetero-addition). If the agents containing medium alone. Drug solutions of MTX and cytara- were acting additively by similar mechanisms, the combined bine at different concentrations (50 ␮l each)were then added data points would lie near the mode II lines (iso-addition). to individual wells. Drug concentrations used were those obtained clinically and required for making dose–response

curves, up to IC90, appropriately. The plates were then incubated for 3 days.

Protocol 2: simultaneous and sequential exposure to MTX and cytarabine for 24 h: CCRF-CEM and HL-60 cells were used for the experiments. For the simultaneous exposure to MTX and cytarabine, cell suspensions (100 ␮l)and solutions of MTX (50 ␮l)and cytarabine (50 ␮l)at different concentrations were added to individual wells. The plates were then incubated for 24 h. The cells were then washed twice with medium, and then fresh medium (200 ␮l)was added. The plates were incubated again under the same conditions for 3 days. For the sequential exposure to these agents, cell solutions (100 ␮l), solutions of MTX (or cytarabine) (50 ␮l)and medium (50 ␮l)were added to individual wells. The plates were then incubated for 24 h. The cells were then washed twice with medium, and then fresh medium (150 ␮l)and solutions of cytarabine (or MTX)(50 ␮l)were added. The plates were incubated again under the same conditions for 24 h. After that the cells were washed twice, fresh medium was added (200 ␮l)and the cells were incubated again for 2 days.

Protocol 3: sequential exposure to MTX for 24 h followed by cytarabine for 3 days: MOLT-3 cells were used for the Figure 1 Schematic representation of an isobologram (Steel and experiments. Cell suspensions (100 ␮l), solutions of MTX (50 Peckham). The envelope of additivity, surrounded by mode I (solid ␮l)and medium (50 ␮l)were added to individual wells. The line)and mode II (dotted lines)isobologram lines, was constructed plates were then incubated for 24 h. The cells were then from the dose–response curves of methotrexate (MTX)and cytarabine (ara-C). The concentrations that produced 80% cell growth inhibition washed twice with medium, and then fresh medium (150 ␮l) (IC80)are expressed as 1.0 in the ordinate and the abscissa of the and solutions of cytarabine (50 ␮l)were added. The plates isobologram. Combined data points Pa, Pb, Pc, and Pd show supraad- were incubated again under the same conditions for 3 days. ditive, additive, subadditive and protective effects, respectively.

Leukemia MTX and ara-C in combination M Akutsu et al 1811 When the data points of a drug combination fell within the Results area surrounded by the three lines (envelope of additivity), the combination was regarded as additive. The envelope of Figure 2 shows the dose–response curves for MTX in combi- additivity should not be considered as a reliable definition of nation with cytarabine for the MOLT-3 cells in simultaneous additivity. The expression of uncertainty is an important con- exposure to MTX and cytarabine for 3 days (a)and sequential cept of the isobologram method of Steel and Peckham.31 We exposure to MTX for 24 h followed by cytarabine for 3 days used this envelope not only to evaluate combinations in (b). Each isobologram was generated based on such dose– which cells were simultaneously exposed to MTX and cytara- response curves. bine, but also to evaluate combinations in which the cells were sequentially exposed to both agents, since the second agent under the present experimental conditions could modu- Isobologram analysis late the cytotoxicity of the first agent. When the data points fell to the left of the envelope (ie the Protocol 1: simultaneous exposure to MTX and cytarabine for combined effect was caused by lower doses of the two agents 3 days: Figure 3 shows the isobolograms for MOLT-3 (a), than predicted), we regarded the drug combination as having CCRF-CEM (b), BALL-1 (c), Daudi (d), K-562 (e), and HL-60 a supra-additive effect (synergism). When the points fell to the (f)cells, respectively. For all cell lines tested, all or most data right of the envelope (ie the combined effect was caused by points were in the areas of subadditivity and protection. The higher doses of the two agents than predicted), but within the mean values of the observed data were larger than those of square or on the line of the square, we regarded the combi- the predicted maximum additive values (Table 2). The nation as having a sub-additive effect, ie the combination was observed data and the predicted maximum values were com- superior or equal to a single agent but was less than additive. pared by Wilcoxon signed-rank test. The observed data were When the data points were outside the square, the combi- significantly larger than the predicted maximum values (Table nation was regarded as having a protective effect, ie the com- 2), indicating antagonistic effects of simultaneous exposure to bination was inferior in cytotoxic action to a single agent. Both these two agents for all these cell lines. sub-additive and protective interactions were regarded as antagonistic effects. Protocol 2: simultaneous and sequential exposure to MTX and cytarabine for 24 h: Since simultaneous exposure to MTX Statistical analysis and cytarabine for 3 days produced antagonistic effects in all six leukemia cell lines, cytotoxic effects of simultaneous and Statistical analysis was performed as described previously.36 sequential exposure to MTX and cytarabine for 24 h were If the mean value of the observed data was equal to or smaller studied using CCRF-CEM and HL-60 cells. Figure 4a–f shows than that of the predicted maximum values and equal to or the isobolograms for CCRF-CEM (a–c)and HL-60 (d–f)cells, larger than that of the predicted minimum values, the combi- respectively. In simultaneous exposure to MTX and cytarabine nation was regarded as having an additive effect. If the mean for 24 h (a, d), all or most data points fell in the areas of value of the observed data was smaller than that of the pre- subadditivity and protection. The observed data were signifi- dicted minimum values or larger than that of the predicted cantly larger than the predicted maximum values (Table 1), maximum values, the combinations were considered to have indicating antagonistic effects for CCRF-CEM and HL-60 cells. a synergistic or antagonistic effect, respectively. To determine In sequential exposure to MTX for 24 h followed by cytarabine whether the condition of synergism (or antagonism)truly for 24 h (b, e), most data points fell in the area of supraadditiv- existed, the Wilcoxon signed-rank test was performed for comparing the observed data with the predicted minimum (or maximum)values for an additive effect, which were closest to the observed data (ie the data on the boundary (mode I or mode II lines)between the additive area and supra-additive area (or sub-additive and protective areas). Probability (P) values р0.05 were considered significant. Combinations with P Ͼ 0.05 were regarded as having an additive to synergistic (or additive to antagonistic)effect. All statistical analyses were performed using the StatView 4.01 software program (Abacus Concepts, Berkeley, CA, USA).

Cell cycle analysis by ﬂow cytometry

Cell cycle analysis was performed using MOLT-3 cells. Cells were cultured in the presence of either no drug, 50 nM MTX Figure 2 Dose–response curves MTX in combination with cytarab- alone, 50 nM cytarabine alone or 50 nM MTX and 50 nM ine in MOLT-3 cells. Cells were exposed to these two drugs simul- cytarabine for 24 h. Cells were also treated with or without taneously for 3 days (a), and sequentially MTX for 24 h followed by 50 nM MTX for 24 h, washed, and cultured with or without cytarabine for 3 days (b). The cell number was measured using the 50 nM cytarabine for a further 24 h. Then, the cells were MTT assay and was plotted as a percentage of the control (cells not stained with propidium iodide in preparation for ﬂow cytome- exposed to drugs). The concentrations of MTX are shown on the abscissa. The concentrations of cytarabine were 0 (᭺),10(b),20(࡯), try with the FACScan/CellFIT system (Becton Dickinson, San 50 (̆)and 100 ( ̄)nM. Data are mean values for three independent Jose, CA,USA). DNA histogram was obtained by analyzing experiments. Each point represents the mean value for at least three 25000 cells with the ModFIT program (Becton Dickinson).37 independent experiments; s.e. were less than 25% and were omitted.

Leukemia MTX and ara-C in combination M Akutsu et al 1812

Figure 3 Isobolograms of simultaneous exposure to MTX and cytarabine for 3 days for the MOLT-3 (a), CCRF-CEM (b), BALL-1 (c), Daudi (d), K-562 (e) and HL-60 (f) cells. Data are presented as means ± s.e. for at least three independent experiments. The data points marked * indicate that s.e. were greater than 30%. The data points for the combination fell mainly in the areas of subadditivity and protection against all these cell lines, indicating antagonism.. Even if s.e. of data points in the area of protection were greater than 30%, the error bars of data points did not cross the envelope.

Table 2 Mean values of observed data, predicted minimum and predicted maximum values of methotrexate (MTX)in combination with cytarabine (ara-C)

Schedule Cell line Observed Predicted Predicted Effect dataa minb maxc

simultaneous MTX + ara-C (3 days) MOLT-3 Ͼ1.05 0.21 0.66 antagonism (P Ͻ 0.02) CCRF-CEM Ͼ1.16 0.49 0.68 antagonism (P Ͻ 0.05) B-ALL 0.93 0.25 0.81 antagonism (P Ͻ 0.05) Daudi Ͼ0.87 0.45 0.78 antagonism (P Ͻ 0.02) K-562 0.78 0.57 0.72 antagonism (P Ͻ 0.05) HL-60 0.92 0.31 0.73 antagonism (P Ͻ 0.02) simultaneous MTX (24 h) + ara-C (24 h) CCRF-CEM Ͼ1.05 0.56 0.7 antagonism (P Ͻ 0.05) HL-60 Ͼ1.01 0.51 0.71 antagonism (P Ͻ 0.05) sequential MTX (24 h) → ara-C (24 h) CCRF-CEM 0.43 0.56 0.71 synergism (P Ͻ 0.02) HL-60 0.46 0.59 0.79 synergism (P Ͻ 0.05) sequential ara-C (24 h) → MTX (24 h) CCRF-CEM 0.76 0.69 0.8 additive HL-60 0.63 0.53 0.75 additive sequential MTX (24 h) → ara-C (3 days) MOLT-3 0.41 0.65 0.75 synergism (P Ͻ 0.02)

aMean value of observed data. bMean value of the predicted minimum values for an additive effect. cMean value of predicted maximum values for an additive effect.

ity for both cell lines. The observed data were signiﬁcantly Protocol 3: sequential exposure to MTX for 24 hours followed smaller than the predicted minimum values (Table 2), indicat- by cytarabine for 72 h: As MTX is usually administered for ing synergistic effects. In sequential exposure to cytarabine 1 day and cytarabine is administered for several days, we followed by MTX (c, f), all data points fell within the envelope further studied the cytotoxic effects of sequential exposure to of additivity for both cell lines. The observed data were larger MTX for 24 h followed by cytarabine for 3 days using MOLT- than the predicted minimum values and smaller than the pre- 3 cells. Figure 4g shows the isobologram for MOLT-3 cells of dicted maximum values (Table 2), indicating additive effects. this schedule. All data points fell in the area of supraadditivity.

Leukemia MTX and ara-C in combination M Akutsu et al 1813

Figure 4 Isobolograms of simultaneous exposure to MTA and cytarabine for 24 h (a, d), sequential exposure to MTX for 24 h followed by cytarabine for 24 h (b, e)and the reverse sequence (c, f)for the CCRF-CEM (a-c),and HL-60 cells (d-f)and an isobologram of sequential exposure to MTX for 24 h followed by cytarabine for 3 days for the MOLT-3 cells (g). Data are presented as means ± s.e. for at least three independent experiments. The error bars s.e. from the mean values. The data points marked * indicate that s.e. were greater than 30%. In simultaneous exposure to MTX and cytarabine, all or most data points fell in the areas of subadditivity and protection (a, d), suggesting antagonism. Even if s.e. of data points in the area of protection were greater than 30%, the error bars of data points did not cross the envelope. In sequential exposure to MTX followed by cytarabine (b, e, g), all or most data points fell in the area of supraadditivity, suggesting synergistic effects. In the reverse sequence (c, f), all data points fell within the envelope of additivity, suggesting additive effects.

The observed data were signiﬁcantly smaller than the pre- followed by cytarabine for 24 h induced a marked increase dicted minimum values (Table 2), indicating synergistic in apoptotic cells (Figure 5h). The remaining cells showed cell effects. cycle arrest in the G1 to S phases after the sequential addition.

Cell cycle cytometric analysis Discussion

Cell cycle analysis revealed that, when used alone, MTX and In the present study, we demonstrated that cytotoxic effects of cytarabine accumulated MOLT-3 cells in the G1 to early S MTX in combination with cytarabine are deﬁnitely schedule- phases and the late S to early G2 phases, respectively (Figure dependent. Simultaneous exposure to MTX and cytarabine for 5b and c). Simultaneous exposure of MOLT-3 cells to MTX 3 days or 24 h produced antagonistic effects. Sequential and cytarabine for 24 h resulted in G1 arrest without a sig- exposure to MTX for 24 h followed by cytarabine for 24 h niﬁcant increase in the sub G1 fraction (Figure 5b and c). In produced synergistic effects, while the reverse sequence pro- contrast, sequential treatment of the cells with MTX for 24 h duced additive effects, suggesting that the optimal schedule of

Leukemia MTX and ara-C in combination M Akutsu et al 1814

Figure 5 Cell cycle analysis of MOLT-3 cells treated with MTX and cytarabine during the experiment of protocols 1 and 3. Left column: MOLT-3 cells were cultured without drugs (None)(a)or with either 50 nM MTX (b),50 nM cytarabine (c)or both drugs (d)for 24 h. Right column: MOLT- 3 cells were cultured without drugs for 24 h, washed and cultured without drugs (None)(e)or with 50 nM cytarabine (g)for 24 h. Cells were cultured with 50 nM MTX for 24 h, washed and cultured without drugs (f)or with 50 nM cytarabine (h)for a further 24 h. DNA histogram was obtained as described in Materials and methods.

this combination is MTX followed by cytarabine. Since MTX is In the present study, the data points on the isobolograms of usually administered for 1 day and cytarabine is administered simultaneous exposure to MTX and cytarabine for 3 days for several days, we further studied the cytotoxic effects of (Figure 3a–f)and for 24 h (Figure 4a, d)fell in the areas of sequential exposure to MTX for 24 h followed by cytarabine subadditivity and protection. The observed data points for the for 3 days. This schedule also showed synergistic effects. MOLT-3 and CCRF-CEM cells (Figure 3a, b)and for the CCRF- In general, the deﬁnition of synergism or antagonism of isobol- CEM and HL-60 cell (Figure 4a, d)were greater than 1.0 ogram of Steel and Peckham is stricter that that of other methods. (Table 1), indicating that each drug alone is more cytotoxic The present ﬁndings support the observations of Burchenal et than the simultaneous combination. The observed data points al,15 Grindey et al,16 Hoovis et al17 and Tanizawa et al,24 and for four other cell lines (Figure 3c–f)were also close to 1.0 suggest that the sequential administration of MTX followed by (Table 1), suggesting that the combination of both agents was cytarabine would be the optimal schedule of this combination almost equally cytotoxic to each drug alone. and simultaneous administration would be inappropriate. The mechanism underlying the antagonistic or synergistic

Leukemia MTX and ara-C in combination M Akutsu et al 1815 interaction that occurs with simultaneous exposure to MTX some patients relapse with CNS leukemia. Thus, the combi- and cytarabine is unclear. The ‘protective effects’ of the com- nation of MTX with other agents is a logical approach to bination in simultaneous exposure means that the cytotoxic improve antitumor response. The findings of our study suggest effects of MTX and cytarabine are inhibited by each other. that cytarabine appears inadequate for simultaneous adminis- Both agents are definitely cell cycle-dependent. MTX has tration with MTX. If both MTX and cytarabine are used for major cytotoxic effects on cells in the S phase and accumu- CNS leukemia and CNS prophylaxis, it may be better to lates in cells in the G1 to early S phase,38 while cytarabine administer these agents sequentially or quite separately. has major cytotoxic effects in the S phase and accumulates in MTX and cytarabine are cell cycle-specific with short-half cells in the late S and early G2 phases39 (Figure 5). Thus, one lives within the CSF.44,45 As a result, the exposure of tumor agent might reduce the cytotoxicity of the other agent by pre- cells in the CSF to the cytotoxic drug levels may be insuf- venting cells from entering the specific phase in which the ficient. Recently, a sustained-release formulating cytarabine cells are most sensitive to the other agent. capable (Depo-Cyto)of maintaining a cytotoxic concentration Sequential exposure to MTX followed by cytarabine showed in the CSF for more than 14 days after a single injection has synergistic effects. The synchronization of the population in been incorporated into clinical use.46 The present findings the early S phase by MTX would prime cells for increased suggest that the intrathecal use of MTX within 14 days after incorporation of cytarabine. Since cytarabine is a deoxycytid- Depo-Cyto may be inappropriate. If MTX is combined with ine analog, the potentiation of its activity may also have Depo-Cyto, MTX followed by Depo-Cyto (for example MTX resulted from depletion of dCTP after MTX treatment. MTX, an day 1 and Depo-Cyto day 4)would be the optimum. inhibitor of dihydrofolate reductase, simultaneously reduced The third-generation regimen for non-Hodgkin’s lymphoma deoxynucleotide pools.40 In cells treated with MTX followed ‘PROMACE-cytaBOM’ and its modification ‘intensive PRO- by cytarabine, depletion of dCTP by MTX may have increased MACE-cytaBOM’6,7 are also used for the simultaneous admin- deoxycytidine kinase activity. The sequential treatment with istration of MTX and cytarabine. This dose-intensive regimen, MTX then cytarabine has been observed to augment the gen- at first, reported an improved response rate and survival com- eration of ara-CTP and the formation of DNA strand pared with standard ‘CHOP’ therapy. However, a large ran- breaks.17,21,24,41 domized trial did not show any significant difference between It must be noted that there are a number of difficulties in them.47 The inadequate schedule of the combination of MTX translating results from in vitro to clinical therapy. First, the and cytarabine might partially contribute to the results. biochemistry of the cells may be quite different although the The POG protocol 8602 for ALL randomized treatment mechanisms of cytotoxicity are generally thought to be simi- groups for consolidation treatment was as follows: one receiv- lar. Second, standard culture medium contains high concen- ing six MTX (24 h infusion)alone (regimen A);the other six trations of folate instead of 5-methyltetrahydrofolate (major MTX overlapping cytarabine (24 h infusion began at 12 h of folate in human plasma). Third, pharmacokinetic, cell kinetics MTX infusion)(regimen C).9 During consolidation treatment, and other obvious parameters may differ significantly between regimen C had significantly more bacterial infection. There the two. Fourth, the toxic effects of the combination can not was no significant advantage of regimen C in terms of be measured in an in vitro system. improvement in event-free survival. The results of POG8602 In clinical oncology, the simultaneous administration of suggest that overlapping administration of MTX and cytarab- MTX and cytarabine (and gluco-corticosteroid)(triple ine was not the optimum. In this regimen, the administration combination)has been widely used for the treatment of CNS of MTX and cytarabine resulted in markedly lower RBC MTX leukemia and CNS prophylaxis. Although a variety of clinical levels when compared with the levels of patients with MTX studies have been performed over the past three decades, pro- alone.48 Patients whose MTX levels were extremely low had gress has been slow and the role and optimal treatment for an adverse prognosis. Very recently, the results of the EORTC CNS prophylaxis and CNS leukemia remain controversial.4,5 58881 study were reported, which was designed to test in a Since cranial radiation is associated with significant neurologi- prospective randomized trial the value of high-dose cytarab- cal disturbances, intrathecal chemotherapy has come to be ine simultaneously added to high-dose MTX after completion preferred. The intrathecal therapy of MTX and cytarabine has of the induction-consolidation phase in ALL and lymphoblas- been preferred to that of MTX alone, since many oncologists tic lymphoma in children treated with a Berlin–Frankfurt– think the combination better than monotherapy. However, no Munster (BFM)-based regimen.10 The addition of high-dose randomized studies comparing the combination of MTX and cytarabine failed to decrease the incidence of CNS relapse cytarabine and MTX alone have been reported. CNS prophy- and to improve the overall disease-free survival. The antagon- laxis using intrathecal MTX and cytarabine was observed to istic interaction between MTX and cytarabine might contrib- produce more neuro-psychogenic toxicity than cranial radi- ute to the negative results of POG8602 and EORTC58881 for ation and intrathecal MTX.42 In recent brief-duration high- this combination. intensity chemotherapy for patients with small noncleaved- Although MTX followed by cytarabine produced synergistic cell lymphoma or L3 ALL, the simultaneous intravenous effects, the clinical application of this schedule has rarely administration of high-dose methotrexate and intermediate- been employed.10–13 For pediatric patients with advanced B dose cytarabine, simultaneous intrathecal methotrexate and cell malignancies, a treatment plan which consists of four cytarabine and cranial irradiation were associated with an courses of fractionated cyclophosphamide administered with excessive rate of intolerable and sometimes irreversible neuro- doxorubicin and vincristine alternating with sequential high- logic changes.43 These clinical observations suggest the possi- dose MTX and cytarabine showed high response rates and bility that the combination produces equivalent or decreased increased survival rate.12 For adult patients with ALL, a treat- anti-leukemic activity with increased toxicity. It is important to ment which consists of eight cycles of HyperCVAD plus high- have a randomized trial comparing the effects and neurologic dose MTX (1000 mg/m2 on day 1)followed by high-dose cyta- toxicities of MTX plus cytarabine and MTX alone against CNS rabine (3 g/m2 × 4 on days 2 and 3)showed 39% and 38% leukemia and CNS prophylaxis. of the estimated 5-year survival and 5-year CR rates, respect- Although MTX is a main drug used for CNS prophylaxis, ively.13 Since the sequential administration of MTX followed

Leukemia MTX and ara-C in combination M Akutsu et al 1816 by cytarabine is only part of the clinical regimens, it appears Lamagnere JP, Otten J, Dufillot D, Pein F. High survival rate in difficult to evaluate the efficacy of this sequence. In both regi- advanced-stage B-cell lymphomas and leukemias without CNS mens, however, this sequence might contribute to the good involvement with a short intensive polychemotherapy: results from the French Pediatric Oncology Society of a randomized trial results. of 216 children. J Clin Oncol 1991; 9: 123–132. In conclusion, simultaneous exposure to MTX and cytarab- 12 Bessell EM, Graus F, Punt JA, Firth JL, Hope DT, Moloney AJ, ine produced marked antagonistic effects, while sequential Lopez Guillermo A, Villa S. Primary non-Hodgkin’s lymphoma of exposure to MTX followed by cytarabine produced synergistic the CNS treated with BVAM or CHOD/BVAM chemotherapy effects against the human leukemia cell lines tested. Although before radiotherapy. J Clin Oncol 1996; 14: 945–954. no in vitro model is absolutely predictive for clinical activity, 13 Bowman WP, Shuster JJ, Cook B, Griffin T, Behm F, Pullen J, Link M, Head D, Carroll A, Berard C, Murphy S TI. Improved survival the present findings suggest that, in the clinical setting, the for children with B-cell acute lymphoblastic leukemia and stage simultaneous administration of MTX and cytarabine is inad- IV small noncleaved-cell lymphoma: a pediatric oncology group equate and the sequential administration of MTX followed by study. J Clin Oncol 1996; 14: 1252–1261. cytarabine is appropriate for antitumor effects if MTX and 14 Kantarjian HM, O7Brien S, Smith TL, Cortes J, Giles FJ, Beran M, cytarabine are combined. Pierce S, Huh Y, Andreeff M, Koller C, Ha CS, Keating M, Murphy S, Freireich EJ. Results of treatment with Hyper-CVAD, a dose- intensive regimen, in adult acute lymphocytic leukemia. J Clin Oncol 2000; 18: 547–561. Acknowledgements 15 Burchenal JH, Dollinger MR. Cytosine in combination with 6-mer- captopurine, methotrexate, or 5-fluorouracil in L1210 mouse leu- This work was supported in part by a Grant-in-Aid for Cancer kemia. Cancer Chemother Rep 1967; 51: 435–438. Research (11–8)from the Ministry of Health and Welfare. 16 Grindey G, Nichol CA. Interaction of drugs inhibiting different steps in the synthesis of DNA. Cancer Res 1972; 32: 527–531. 17 Hoovis ML, Chu MY. Enhancement of the antiproliferative action of 1-b-D-arabinofuranosylcytosine by methotrexate in murine leu- References kemia cells. (L5178Y) Cancer Res 1973; 33: 521–525. 18 Tattersall MH, Harrap KR. Combination chemotherapy: the antag- 1 Coiffier B, Gisselbrecht C, Herbrecht R, Tilly H, Bosly A, Brousse onism of methotrexate and cytosine arabinoside. Eur J Cancer N. LNH-84 regimen: a multicenter study of intensive chemo- 1973; 9: 229–232. therapy in 737 patients with aggressive malignant lymphoma. J 19 Avery TL, Roberts D. Dose-related synergism of cytosine arabino- Clin Oncol 1989; 7: 1018–1026. side and methotrexate against murine leukemia L1210. Eur J Can- 2 Schorin MA, Blattner S, Gelber RD, Tarbell NJ, Donnelly M, cer 1974; 10: 425–429. Dalton V, Cohen HJ, Sallan SE. Treatment of childhood acute 20 Edelson M, Vietti T, Valeriote F. The enhanced cytotoxicity of lymphoblastic leukemia: results of Dana-Farber Cancer combinations of 1-b-D-arabinofuranosylcytosine and methotrex- Institute/Children’s Hospital Acute lymphoblastic leukemia Con- ate. Cancer Res 1975; 35: 1555–1558. sortium Protocol 85–01. J Clin Oncol 1994; 12: 740–747. 3 Moser AM, Adamson PC. Intraventricular concentration times time 21 Cadman E, Eiferman F. Mechanism of synergistic cell killing when (C × T)methotrexate and cytarabine for patients with recurrent methotrexate precedes cytarabine: study of L1210 and human leu- meningeal leukemia and lymphoma. Cancer 1999; 85: 511–516. kemia cells. J Clin Invest 1979; 64: 788–797. 4 Porchedly C. Prevention of meningeal leukemia. Review of 20 22 Roberts DW, Peck C, Hilliard C, Wingo W. Methotrexcate- years of research and current recommendations. Hematol Oncol induced changes in the levels of 1-b-D-arabinofuranosylcytosine Clin N Am 1990; 4: 951–969. triphosphate in L1210 cells. Cancer Res 1979; 39: 4048–4054. 5 Pinkel D, Woo S. Prevention and treatment of meningeal leukemia 23 Jackson RC, Harkrader RJ. Synergistic and antagonistic interac- in children. Blood 1994; 84: 355–366. tions of methotrexate and 1-b-D-arabinofuranosylcytosine in 6 Longo DL, DeVita VT Jr, Duffey PL, Wesley MN, Ihde DC, Hub- hepatoma cells: the modulating effect of purines. Biochem Pharm- bard SM, Gilliom M, Jaffe ES, Cossman J, Fisher RI. Superiority acol 1981; 30: 223–229. of ProMACE-CytaBOM over ProMACEMOPP in the treatment of 24 Tanizawa A, Kubota M, Takimoto T, Kito T, Akiyama Y, Kiriyama advanced diffuse aggressive lymphoma: results of a prospective Y, Mikawa H. Synergistic effect of methotrexate and 1-beta-D-ara- randomized trial. J Clin Oncol 1991; 9: 25–38. binofuranosylcytosine on the generation of DNA strand breaks in 7 Gordon LI, Young M, Weller E, Habermann TM, Winter JN, Glick a human promyelocytic leukemia cell line. Leuk Res 1989; 13: J, Ghosh C, Flynn P, Cassileth PA. A phase II trial of 200% Pro- 151–156. MACE-CytaBOM in patients with previously untreated aggressive 25 Berenbaum MC. What is synergy? Pharmacol Rev 1988; 41: lymphomas: analysis of response, toxicity, and dose intensity. 93–141. Blood 1999; 94: 3307–3314. 26 Merlim JL. Concepts of synergism and antagonism. Anticancer Res 8 Parker BA, Santarelli M, Green MR, Anderson JR, Cooper MR, 1994; 14: 2315–2320. Case D Jr, Barcos M, Peterson BA, Gottlieb AJ. AMOPLACE treat- 27 Vendette JM, Goldin A. drug synergism in antineoplastic chemo- ment of intermediate-grade and high-grade malignant lymphoma: therapy. Adv Chemother 1964; 1: 397–498. a Cancer and Leukemia Group B study. J Clin Oncol 1993; 11: 28 Loewe S. The problem of synergism and antagonism of combined 248–254. drugs. Arzneim Forsch 1953; 3: 285–290. 9 Harris MB, Shuster JJ, Jeanette Pullen D, Borowitz MJ, Carroll AJ, 29 Webb JL. Effect of more than one inhibitor. In: Enzymes and Meta- Behm FG, Land VJ. Consolidation therapy with antimetabolite- bolic Inhibitors. Academic Press: New York, 1963, pp 66–79. based therapy in standard-risk acute lymphoblastic leukemia of 30 Chou TC, Motzer RJ, Tong Y, Bosl GJ. Computerized quantitation childhood: a pediatric oncology group study. J Clin Oncol 1998; of synergism and antagonism of taxol, topotecan, and cisplatin 16: 2840–2847. against human teratocarcinoma cell growth: a rational approach 10 Millot F, Suciu S, Philippe N, Benoit Y, Mazingue F, Uyttebroeck to clinical protocol design. Nat Cancer Inst 1994; 86: 1517–1524. A, Lutz P, Mechinaud F, Robert A, Boutard P, Marguerite G, Fers- 31 Steel GG, Peckham MJ. Exploitable mechanisms in combined ter A, Plouvier E, Rialland X, Behard C, Plantaz D, Dresse MF, radiotherapy–chemotherapy: the concept of additivity. Int J Radiat Philippet P, Norton L, Thyss A, Dastugue N, Waterkeyn C, Vilmer Oncol Biol Phys 1979; 5: 8–91. E, Otten J. Value of high-dose cytarabine during interval therapy 32 Hiraki S, Miyoshi I, Masuji H, Kubonishi I, Matsuda Y. Establish- of a Berlin–Frankfurt–Munster-based protocol in increased-risk ment of an Epstein–Barr virus-determined nuclear antigen-nega- children with acute lymphoblastic leukemia and lymphoblastic tive human B-cell line from acute lymphoblastic leukemia. J Natl lymphoma: results of the European Organization for Research and Cancer Inst 1977 59: 93–94. Treatment of Cancer 58881 randomized phase III trial. J Clin 33 Kano Y, Sakamoto S, Kasahara T, Akutsu M, Inoue Y, Miura Y. In Oncol 2001; 19: 1935–1942. vitro effects of amsacrine in combination with other anticancer 11 Patte C, Philip T, Rodary C, Zucker JM, Behrendt H, Gentet JC, agents. Leukemia Res 1991; 15: 1059–1064.

Leukemia MTX and ara-C in combination M Akutsu et al 1817 34 Kano Y, Ohnuma T, Okano T, Holland JF. Effects of vincristine in 42 Giralt J, Ortega JJ, Oliwe T, Verges R, Forio I, Salvador L. Long combination with methotrexate and other antitumor agents in term neuropsychologic sequence of childhood leukemia: com- human acute lymphoblastic leukemia cells in culture. Cancer Res parison of two CNS prophylactic regimens. Int J Radiat Oncol Biol 1988; 48: 351–356. Phys 1992; 24: 49–53. 35 Kano Y, Suzuki K, Akutsu M, Suda K, Inoue Y, Yoshida M, Sakam- 43 Lee EJ, Petroni GR, Schiffer CA, Freter CE, Johnson JL, Barcos M, oto S, Miura Y. Effects of CPT-11 in combination with other Frizzera G, Bloomﬁeld CD, Peterson BA. Brief-duration high- anticancer agents in culture. Int J Cancer 1992; 50: 604–609. intensity chemotherapy for patients with small noncleaved-cell 36 Kano Y, Akutsu M, Tsunoda S, Suzuki K, Adachi K. In vitro sched- lymphoma or FAB L3 acute lymphocytic leukemia: results of can- ule-dependent interaction between paclitaxel and SN-38 (the cer and leukemia group B study 9251. J Clin Oncol 2001; 19: active metabolite of irinotecan)in human carcinoma cell lines. 4014–4022. Cancer Chemother Pharmacol 1998; 42: 91–98. 44 Zimm S, Collins JM, Miser J, Chatterji D, Poplack D. Cytosine 37 Furukawa Y, Iwase S, Kikuchi J, Nakamura M, Terui Y, Yamada arabinoside cerebrospinal ﬂuid kinetics. Clin Pharmacol Ther H, Kano Y, Matsuda M. Phosphorylation of bcl-2 protein by cdc2 1984; 35: 826–830. kinase during G2/M phases and its role in cell cycle regulation. J 45 Strother DR, Glynn-Barnhart A, Kovnar E, Gregory RE, Murphy Biol Chem 2000; 275: 21661–21667. SB. Variability in the disposition of intraventricular methotrexate: a proposal for rational dosing. J Clin Oncol 1989; 7: 1741–1747. 38 Lorico A, Boiocchi M, Rappa G, Sen S, Erba E, D’Incalci M. 46 Glantz MJ, Jaeckle KA, Chamberlain MC, Phuphanich S, Recht L, Increase in topoisomerase-II-mediated DNA breaks and cytotoxic- Swinnen LJ, Maria B, LaFollette S, Schumann GB, Cole BF, Howell ity of VP16 in human U937 lymphoma cells pretreated with low SB. A randomized controlled trial comparing intrathecal sus- doses of methotrexate. Int J Cancer 1990; 45: 156–162. tained-release cytarabine (DepoCyt)to intrathecal methotrexate in 39 Karon M, Chirakawa S. The locus of action of 1-b-D-arabinofur- patients with neoplastic meningitis from solid tumors. Clin Cancer anosylcytosine in the cell cycle. Cancer Res 1970; 29: 687–694. Res 1999; 5: 3394–3402. 40 Chen VJ, Bewley JR, Andis SL, Schultz RM, Iversen PW, Shih C, 47 Fisher RI, Gaynor ER, Dahlberg S, Oken MM, Grogan TM, Mize Mendelsohn LG, Seitz DE, Tonkinson JL. Preclinical cellular phar- EM, Glick JH, Coltman CA Jr, Miller TP. Comparison of a standard macology of LY231514 (MTA): a comparison with methotrexate, regimen (CHOP)with three intensive chemotherapy regimens for LY309887 and raltitrexed for their effects on intracellular folate advanced non-Hodgkin’s lymphoma. N Engl J Med 1993; 328: and nucleoside triphosphate pools in CCRF-CEM cells. Br J Cancer 1002–1006. 1998; 78 (Suppl. 3): 27–34. 48 Graham ML, Shuster JJ, Kamen BA, Land VJ, Borowitz MJ, Camitta 41 Newman EM, Villacorte DG, Testi AM, Krance RA, Harris MB, B, Cheo DL, Harrison MP, Leventhal BG, Pinkel DP, Pullen DJ, Ravindranath Y, Pinkel D. Biochemical interactions between Steuber P, Whitehead VM. Changes in red blood cell methotrexate methotrexate and1-beta-D-arabinofuranosylcytosine in hematopo- pharmacology and their impact on outcome when cytarabine is ietic cells of children: a Pediatric Oncology Group study. Cancer infused with methotrexate in the treatment of acute lymphoblastic Chemother Pharmacol 1990; 27: 60–66. leukemia in children. Clin Cancer Res 1996; 2: 331–337.

Leukemia