Portal de Periódicos da CAPES

Drug-interaction studies evaluating T-cell proliferation reveal distinct activity of dasatinib and imatinib in combination with cyclosporine A

2012; Elsevier BV; Volume: 40; Issue: 8 Linguagem: Inglês

10.1016/j.exphem.2012.04.003

1873-2399

Stephen J. Blake, Timothy P. Hughes, A. Bruce Lyons,

Chronic Lymphocytic Leukemia Research

Development of small molecule tyrosine kinase inhibitors for the treatment of chronic myeloid leukemia has been astonishingly successful; however, their off-target effects have generated both challenges and opportunities for extending their clinical application. Dasatinib and imatinib are two of the most commonly used tyrosine kinase inhibitors and both have been shown to impact T-cell function. Due to this activity, their use as potential immune suppressants has been proposed. In this report, we investigated drug interactions with cyclosporine A in suppressing T-cell proliferation. Dasatinib and imatinib were titrated against varying concentrations of cyclosporine in the cultures and T-cell proliferation assessed by 5-6-carboxyfluorescein diacetate, succinimidyl ester dye dilution. These proliferation data were then used to determine the combination index to evaluate additive, synergistic, or antagonistic interactions between the drugs. This analysis uncovered a number of different drug interactions affecting T-cell proliferation. Cyclosporine had an additive or synergistic effect on T-cell proliferation when combined with dasatinib and imatinib for 3 of the 4 methods of stimulating T-cell proliferation. However, when T cells were stimulated with anti-CD3 and anti-CD28 antibodies, this interaction was found to be strongly antagonistic at low dasatinib concentrations. In contrast, this strong antagonism was not observed when imatinib was used in combination with cyclosporine A. This study suggests drug interactions affecting T cells may need to be carefully taken into account when using tyrosine kinase inhibitors. Furthermore, the technique to evaluate drug interactions is novel, and applicable to study any interaction affecting proliferation. Development of small molecule tyrosine kinase inhibitors for the treatment of chronic myeloid leukemia has been astonishingly successful; however, their off-target effects have generated both challenges and opportunities for extending their clinical application. Dasatinib and imatinib are two of the most commonly used tyrosine kinase inhibitors and both have been shown to impact T-cell function. Due to this activity, their use as potential immune suppressants has been proposed. In this report, we investigated drug interactions with cyclosporine A in suppressing T-cell proliferation. Dasatinib and imatinib were titrated against varying concentrations of cyclosporine in the cultures and T-cell proliferation assessed by 5-6-carboxyfluorescein diacetate, succinimidyl ester dye dilution. These proliferation data were then used to determine the combination index to evaluate additive, synergistic, or antagonistic interactions between the drugs. This analysis uncovered a number of different drug interactions affecting T-cell proliferation. Cyclosporine had an additive or synergistic effect on T-cell proliferation when combined with dasatinib and imatinib for 3 of the 4 methods of stimulating T-cell proliferation. However, when T cells were stimulated with anti-CD3 and anti-CD28 antibodies, this interaction was found to be strongly antagonistic at low dasatinib concentrations. In contrast, this strong antagonism was not observed when imatinib was used in combination with cyclosporine A. This study suggests drug interactions affecting T cells may need to be carefully taken into account when using tyrosine kinase inhibitors. Furthermore, the technique to evaluate drug interactions is novel, and applicable to study any interaction affecting proliferation. Small-molecule targeted therapies for chronic myeloid leukemia act by inhibiting the mutant bcr-abl kinase, and have been very effective in controlling chronic phase disease [1Talpaz M. Shah N.P. Kantarjian H. et al.Dasatinib in imatinib-resistant Philadelphia chromosome-positive leukemias.N Engl J Med. 2006; 354: 2531-2541Crossref PubMed Scopus (1462) Google Scholar, 2O'Brien S.G. Guilhot F. Larson R.A. et al.Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia.N Engl J Med. 2003; 348: 994-1004Crossref PubMed Scopus (2983) Google Scholar]. However, it has been increasingly apparent that many of these drugs have off-target effects that are T-cell inhibitory [3Schade A.E. Schieven G.L. Townsend R. et al.Dasatinib, a small-molecule protein tyrosine kinase inhibitor, inhibits T-cell activation and proliferation.Blood. 2008; 111: 1366-1377Crossref PubMed Scopus (217) Google Scholar, 4Seggewiss R. Lore K. Greiner E. et al.Imatinib inhibits T-cell receptor-mediated T-cell proliferation and activation in a dose-dependent manner.Blood. 2005; 105: 2473-2479Crossref PubMed Scopus (246) Google Scholar]. Drug inhibition of T-cell activity is important clinically in organ transplantation settings, with the standard clinical T-cell inhibitor Cyclosporine A (CsA) blocking T-cell activity by inhibiting calcium influx during T-cell activation [5Fruman D.A. Klee C.B. Bierer B.E. Burakoff S.J. Calcineurin phosphatase activity in T lymphocytes is inhibited by FK 506 and cyclosporin A.Proc Natl Acad Sci U S A. 1992; 89: 3686-3690Crossref PubMed Scopus (752) Google Scholar, 6Kahan B.D. Cyclosporine.N Engl J Med. 1989; 321: 1725-1738Crossref PubMed Scopus (1633) Google Scholar]. The tyrosine kinase inhibitors (TKIs) act in a completely different manner than CsA, inhibiting T cells by blocking the tyrosine kinases involved in T-cell activation, including ABL, and src-family tyrosine kinases LCK, Yes, or Fyn [7Huang Y. Comiskey E.O. Dupree R.S. Li S. Koleske A.J. Burkhardt J.K. The c-Abl tyrosine kinase regulates actin remodeling at the immune synapse.Blood. 2008; 112: 111-119Crossref PubMed Scopus (64) Google Scholar, 8Palacios E.H. Weiss A. Function of the Src-family kinases, Lck and Fyn, in T-cell development and activation.Oncogene. 2004; 23: 7990-8000Crossref PubMed Scopus (522) Google Scholar, 9Molina T.J. Kishihara K. Siderovski D.P. et al.Profound block in thymocyte development in mice lacking p56lck.Nature. 1992; 357: 161-164Crossref PubMed Scopus (887) Google Scholar, 10Manley P.W. Cowan-Jacob S.W. Mestan J. Advances in the structural biology, design and clinical development of Bcr-Abl kinase inhibitors for the treatment of chronic myeloid leukaemia.Biochim Biophys Acta. 2005; 1754: 3-13Crossref PubMed Scopus (156) Google Scholar]. The kinase inhibitors imatinib, dasatinib, and nilotinib have previously been identified as potentially clinically useful T-cell inhibitors showing inhibition of T-cell function in vitro and mouse models [3Schade A.E. Schieven G.L. Townsend R. et al.Dasatinib, a small-molecule protein tyrosine kinase inhibitor, inhibits T-cell activation and proliferation.Blood. 2008; 111: 1366-1377Crossref PubMed Scopus (217) Google Scholar, 4Seggewiss R. Lore K. Greiner E. et al.Imatinib inhibits T-cell receptor-mediated T-cell proliferation and activation in a dose-dependent manner.Blood. 2005; 105: 2473-2479Crossref PubMed Scopus (246) Google Scholar, 11Blake S.J. Lyons A.B. Hughes T.P. Nilotinib inhibits the Src-family kinase LCK and T-cell function in vitro.J Cell Mol Med. 2009; 13: 599-601Crossref PubMed Scopus (25) Google Scholar, 12Chen J. Schmitt A. Chen B. et al.Nilotinib hampers the proliferation and function of CD8+ T lymphocytes through inhibition of T cell receptor signaling.J Cell Mol Med. 2008; 12: 2107-2118Crossref PubMed Scopus (66) Google Scholar, 13Blake S. Hughes T.P. Mayrhofer G. Lyons A.B. The Src/ABL kinase inhibitor dasatinib (BMS-354825) inhibits function of normal human T-lymphocytes in vitro.Clin Immunol. 2008; 127: 330-339Crossref PubMed Scopus (96) Google Scholar, 14Fraser C.K. Blake S.J. Diener K.R. et al.Dasatinib inhibits recombinant viral antigen-specific murine CD4(+) and CD8(+) T-cell responses and NK-cell cytolytic activity in vitro and in vivo.Exp Hematol. 2009; 37: 256-265Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 15Fei F. Yu Y. Schmitt A. et al.Dasatinib exerts an immunosuppressive effect on CD8(+) T cells specific for viral and leukemia antigens.Exp Hematol. 2008; 36: 1297-1308Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar, 16Weichsel R. Dix C. Wooldridge L. et al.Profound inhibition of antigen-specific T-cell effector functions by dasatinib.Clin Cancer Res. 2008; 14: 2484-2491Crossref PubMed Scopus (118) Google Scholar], as well as in patients [17Sillaber C. Herrmann H. Bennett K. et al.Immunosuppression and atypical infections in CML patients treated with dasatinib at 140 mg daily.Eur J Clin Invest. 2009; 39: 1098-1109Crossref PubMed Scopus (87) Google Scholar, 18Torres H.A. Chemaly R.F. Viral infection or reactivation in patients during treatment with dasatinib: a call for screening?.Leuk Lymphoma. 2007; 48: 2308-2309Crossref PubMed Scopus (7) Google Scholar]. Recent studies have indicated that T-cell proliferation is more potently blocked by a combination of dasatinib and CsA than by either drug alone, suggesting at least additivity between the two drug types [3Schade A.E. Schieven G.L. Townsend R. et al.Dasatinib, a small-molecule protein tyrosine kinase inhibitor, inhibits T-cell activation and proliferation.Blood. 2008; 111: 1366-1377Crossref PubMed Scopus (217) Google Scholar]. This could have useful clinical implications as CsA can reduce renal function at higher concentrations, thus, reducing the CsA dose may be of benefit in this setting. Although Schade et al. demonstrated a greater inhibition of T-cell proliferation when dasatinib was used in combination with CsA, only single concentrations of each agent were used, therefore, no analytical method could be applied to determine the level of synergy between the two drugs. The aim of our study was to examine in detail the interactions between the TKIs dasatinib and imatinib, in combination with CsA, using robust analysis of the effects of these agents on T-cell proliferation. Recent in vitro and murine model studies have also identified the beneficial effects of using TKIs and CsA in combination to treat BCR-ABL–positive leukemia [19Porter C.C. Gregory M.A. Zaberezhnyy V. Klawitter J. Christians U. DeGregori J. The addition of cyclosporine to dasatinib therapy of murine Bcr-Abl+ leukemia improves disease control independent of altered pharmacokinetics.ASH Annual Meeting Abstracts. 2010; 116: 600Google Scholar, 20Gregory M.A. Phang T.L. Neviani P. et al.Wnt/Ca2+/NFAT signaling maintains survival of Ph+ leukemia cells upon inhibition of Bcr-Abl.Cancer Cell. 2010; 18: 74-87Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar], which has led to the proposing of a phase 1b clinical trial using dasatinib and CsA to treat imatinib refractory chronic myeloid leukemia (i.e., ESCAPE 1b trial, NCT01456988). Thus, identifying any off-target effects of this drug combination is of strong clinical importance. First described by Chou and Talaley [21Chou T.C. Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors.Adv Enzyme Regul. 1984; 22: 27-55Crossref PubMed Scopus (5891) Google Scholar], the Combination Index (CI) has been widely used to quantify interactions between drugs and has been incorporated into the statistical program Calcusyn (Biosoft, Cambridge, UK). This program allows CI values to be generated, such that a CI of 1 indicates an additive effect between drugs, 1 indicates an antagonistic effect. Using 5-6-carboxyfluorescein diacetate, succinimidyl ester (CFSE) dye to examine T-cell proliferation [22Lyons A.B. Analysing cell division in vivo and in vitro using flow cytometric measurement of CFSE dye dilution.J Immunol Methods. 2000; 243: 147-154Crossref PubMed Scopus (566) Google Scholar, 23Lyons A.B. Parish C.R. Determination of lymphocyte division by flow cytometry.J Immunol Methods. 1994; 171: 131-137Crossref PubMed Scopus (1465) Google Scholar], we have developed a novel method to evaluate proliferation and drug-interaction analysis, generating CI values to allow interactions between the two drug classes to be analyzed in great detail. This approach could be applied broadly to analysis of drug combinations that inhibit or enhance the proliferation of cells. Peripheral blood from normal donors was collected into lithium heparin tubes by venipuncture. Experimental use of human material was approved by the Human Ethics Committee, Royal Adelaide Hospital. Peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation using Lymphoprep for 30 minutes at 300 g (Axis-Shield, Oslo, Norway) and washed twice in Hank's buffered salt solution (SAFC Biosciences, St Louis, MO, USA) before use in experiments. Dasatinib (BMS-354825) was kindly supplied by Bristol Myers Squibb as a pure powder. The powder was dissolved in dimethyl sulfoxide to a concentration of 10 mg/mL (19.76 mM) and stored at −20°C. Imatinib was purified by dissolving the powdered drug in water before the drug was alkalized to its free base form. After extraction into ethyl acetate to purify away from water-soluble excipients, solvent was removed by vacuum evaporation, and imatinib was recrystallized into the mesylate salt. Purity by high-performance liquid chromatography was >95%. Purified imatinib was dissolved in distilled water to a final concentration of 10 mM before being filter sterilized and stored at −70°C. CsA (Sigma, St Louis, MO, USA) was dissolved in 95% ethanol at 10 mg/mL and stored at −20°C. PBMCs were labeled with CFSE (Invitrogen, Carlsbad, CA, USA) as described previously [23Lyons A.B. Parish C.R. Determination of lymphocyte division by flow cytometry.J Immunol Methods. 1994; 171: 131-137Crossref PubMed Scopus (1465) Google Scholar]. After staining, cells were resuspended at 1 × 106 cells/mL in culture medium, RPMI 1640 (SAFC Biosciences) supplemented with 10% fetal calf serum (JRH Biosciences, Lenexa, KS, USA) and penicillin/streptomycin (Sigma). Stained cells were then plated onto 96-well flat-bottomed tissue culture trays and T cells stimulated by the addition of the mitogens Concanavalin A (ConA; Sigma) or phytohemagglutin (PHA) at 10 μg/mL (Sigma), or an anti-CD3 antibody used at 100 ng/mL (Mabtech, Nacka Strand, Sweden) or CD3/CD28 T-cell expander beads (Invitrogen) added at 1 bead per PBMC. Samples were placed in a humidified 37°C 5% CO2 incubator and incubated for 4–6 days. Detection of T cells was achieved through surface staining of PBMCs with an R-phycoerythrin–conjugated anti-CD3 antibody (Becton Dickinson, Pharmingen, San Jose, CA, USA). PBMCs were removed from culture and stained for 30 minutes on ice, washed once in phosphate-buffered saline with 0.1% bovine serum albumin and resuspended in phosphate-buffered saline and stored on ice before analysis by flow cytometry. Samples were analyzed on a Beckman Coulter FC500 bench top flow cytometer running CXP acquisition software (Beckman Coulter, Villepinte, France). The level of T-cell proliferation was determined for all experiments by analyzing flow data with the Modfit LT flow cytometry modeling software (Verity Software House, Topsham, ME, USA). T cells were identified by gating on CD3-positive cells with lymphocyte and activated lymphocyte light scatter properties. Using the proliferation wizard within Modfit LT, the Proliferation Index (PI) was calculated for all samples based on the following formula:PI=(npp+nG1+nG2+nG3+…nGn)/[(npp/2pp)+(nG1/2G1)+(nG2/2G2)+(nG3/2G3)+…(nGn/2Gn)]Where n is the number of cells in each fluorescence divisional peak and each peak is identified as follows: “pp” refers to the parental peak of undivided cells, “G1” is the first divisional peak, G2 the second, G3 the third, and so on, until division is indistinguishable from background fluorescence. Thus, PI is the sum of all proliferation events in the culture divided by the calculated sum of the original parental cells at the initiation of culture. The range of drug concentrations was chosen to give approximately 5%–95% inhibition of T-cell proliferation as determined from preliminary experiments. Dasatinib and cyclosporine were used at 1, 5, 10, and 25 nM for all experiments, except when an anti-CD3 antibody was used to stimulate proliferation, where 1, 2.5, 5, and 10 nM were used, as initial experiments showed both drugs to be more potent at blocking anti-CD3–driven proliferation. In addition, a single high 50-nM concentration of each drug alone was used, corresponding to the maximal inhibition of each drug on T-cell proliferation, determined in preliminary experiments. Imatinib was used at 1, 2.5, 5, and 10 μM and maximum inhibition achieved using 20 μM. Interactions between the two drug combinations were evaluated using the Calcusyn program (Biosoft, Cambridge, UK) using PI as a quantitative measures of cell proliferation. However, the values had to be modified for analysis by Calcusyn as the program requires drug effects to be represented as values between 0 and 1, where 0 indicates no drug effect and 1 indicates complete inhibition. Initially, PI values were modified by the subtraction of 1 from each point. Thus, no proliferation was represented by a value of 0 rather than 1. Then the PI values in the presence of the inhibitors were normalized by dividing each by the PIs of the untreated control, giving the PI as a fraction of the untreated control. Each value was then subtracted from 1 to give each value as the fraction of T-cell proliferation affected by the drugs. If the proliferation of drug-treated cells was greater than that of the untreated control, this process generated a value of <0. In these circumstances, this proportion was assigned an arbitrary value of 0.0001 (representing essentially no drug effect on proliferation) to accommodate the requirement of Calcusyn that all values lie between 0 and 1. The values obtained from each 25-well grid were then entered as a drug combination of nonstandard ratio into the Calcusyn drug wizard. From this, the CI was calculated for each drug combination and then plotted onto a 3D column graph. This method of graphing CI values was described by Flaig et al. [24Flaig T.W. Su L.J. Harrison G. Agarwal R. Glode L.M. Silibinin synergizes with mitoxantrone to inhibit cell growth and induce apoptosis in human prostate cancer cells.Int J Cancer. 2007; 120: 2028-2033Crossref PubMed Scopus (56) Google Scholar] and has the advantage of allowing visual inspection of drug interactions over a broad range of concentrations. Alternatively, to more clearly show differences between synergy and antagonism, the values were graphed on a log10 scale, meaning synergy was now shown by (log CI < 0) and antagonism (log CI > 0). Alternatively, if a fixed drug ratio was performed, the transformed PI values were entered into the fixed ratio combination wizard within the Calcusyn program. CFSE labeling of PBMCs using the standard protocol [22Lyons A.B. Analysing cell division in vivo and in vitro using flow cytometric measurement of CFSE dye dilution.J Immunol Methods. 2000; 243: 147-154Crossref PubMed Scopus (566) Google Scholar] gave a level of fluorescence that allowed division peaks to be clearly defined for 6–7 generations. Proliferation assays were performed in a 96-well plate in a 5 × 5-well grid, allowing the inhibitory action of each of the possible combinations of the two drugs to be tested (Fig. 1A). After incubation for 4–6 days, cells were labeled with an anti-CD3 phycoerythrin antibody to identify T cells and the samples were analyzed by flow cytometry. As shown in Figure 1B, initial results demonstrated that concentrations of dasatinib and CsA that weakly inhibited ConA-stimulated T-cell proliferation potently inhibited proliferation when added to cultures in combination. PBMCs were stimulated in the presence of the described drug combinations and the PIs of CD3+ T cells were determined using Modfit. To visually inspect the data, the PIs at various concentrations of CsA were plotted against the concentration of dasatinib (Fig. 2). Inspection of these graphs shows that at each concentration of CsA, inhibition of proliferation induced by ConA, PHA, or an anti-CD3 antibody is increased in a dose-dependent pattern by increasing concentrations of dasatinib. Interestingly, when CD3/CD28 microbeads were used to stimulate proliferation, dasatinib and CsA appear to have antagonistic effects at the lower concentrations of dasatinib. T-cell proliferation increased with increasing concentrations of dasatinib, up to a concentration of 10 nM, beyond which dasatinib then synergized with CsA. The PI values were then modified using the Calcusyn drug wizard as described to obtain the CI values at each point. These data were expressed as a log value, in order to clearly differentiate between synergy (log CI < 0) and antagonism (log CI > 0) (Fig. 3). These graphs indicate that after stimulation with PHA, ConA, or an anti-CD3 antibody, there is a mild synergistic or additive interaction between the dasatinib and cyclosporine, with most points generating a value of 0 to −0.4. Analysis using this method also clearly demonstrated the antagonistic effect between dasatinib and CsA in CD3/CD28 microbead-stimulated cells, with log CI values considerably >0 at concentrations of dasatinib ≤10 nM. However, at a concentration of 25 nM dasatinib, the interaction with CsA became synergistic, with log CI values <0. Log graphs were used to better reveal the degree of synergy that would otherwise be truncated in a linear graph. In addition, log graphs could better accommodate the variability seen in raw CI values. For comparison, plots of representative CI values are presented in Supplementary Figure E1 (online only, available at www.exphem.org) and mean CI values and standard error of mean are presented in Supplementary Table E1 (online only, available at www.exphem.org). Initial results also demonstrated similar antagonistic effects on both CD4 and CD8 T cells, suggesting that the interaction between dasatinib and CsA affects T cells globally and is not a disproportionate effect on one of the major subsets (Supplementary Figure E2; online only, available at www.exphem.org). Dasatinib alone had an interesting effect on T-cell proliferation because although the numbers of cells entering division were impacted, the PI of cells already in division was unaltered by the drug (Supplementary Figure E3; online only, available at www.exphem.org). Additionally, CD3/CD28-induced proliferation was higher in the dividing fraction of cells treated with dasatinib, suggesting that the drug alone at levels 2.5 μM, log CI values were generally ≤0, demonstrating an overall synergistic or additive response. These data suggests that the synergy between CsA and either dasatinib or imatinib is related to the common TKI activities of the two drugs. However, the degree of antagonism observed between dasatinib and CsA when tested on T cells stimulated with anti-CD3/CD28 microbeads is not evident with imatinib and might be due to dasatinib targeting additional kinases.Figure 5Determination of CI to examine interactions between imatinib and CsA. The PI was calculated from stimulated T cells in the presence of different concentrations of imatinib and CsA and used to generate CI values. CI values were then plotted onto a log10 3D column graph to allow visualization of drug interactions with each point averaged from 2–4 independent experiments with mean CI values and standard error of mean shown in Supplementary Table E2 (online only, available at www.exphem.org). Plots represent interactions following stimulation with (A) ConA, (B) PHA, (C) anti-CD3 antibody, and (D) anti-CD3/CD28 microbeads.View Large Image Figure ViewerDownload Hi-res image Download (PPT) For the majority of experiments, nonstandard drug ratios were used for simplicity of setup and to give a large range of concentrations. However, analysis of synergy using the CI method is often performed using fixed drug ratios (e.g., 5:1 2:1, 1:1). Use of fixed ratios allows simple visual analysis of synergy in the form of isobolograms, as well as by generation of CI values at effective doses (ED) causing 50%, 75%, and 90% inhibition of proliferation (ED50, ED75, and ED90). To investigate if standard ratio analysis improved data analysis, 5:1 and 2:1 ratios of dasatinib to CsA were set up using dasatinib concentrations of 1, 5, 10, and 25 nM and tested on CFSE-labeled T cells stimulated with ConA and the resulting PI values transformed and entered into Calcusyn. As outlined in Supplementary Figure E4A, B (online only, available at www.exphem.org), the isobologram analysis of the 2:1 and 5:1 fixed drug ratios suggested moderate synergy between dasatinib and CsA, as points of interaction fell below the line connecting the ED values of each drug alone. Analysis of the data using a fixed drug ratio also allows generation of CI values at the ED50, ED75, and ED90 of the drugs under examination (Supplementary Figure E4C; online only, available at www.exphem.org). At each of these fractions of the ED, and at both drug ratios, the CI was <1. Data analysis at fixed drug ratios confirmed our finding that dasatinib and CsA act in synergy to block ConA-induced T-cell proliferation and presented the data in a simplified manner. The experiments of nonfixed drug ratio were reanalyzed using a fixed drug ratio to confirm this method was able to provide a similar result to the titration analysis. Within the original experimental design, dasatinib and CsA were used at identical concentrations, meaning a 1:1 ratio was present within the nonstandard ratio setup (e.g., 1:1 nM, 5:5 nM, etc.). Thus, all experiments were able to be reanalyzed using a 1:1 fixed ratio and CI values calculated at the ED50, 75, and 90. As shown in Table 1, mean CI values also indicate that synergistic interactions occur between dasatinib and CsA, and that this synergy is strongest when either ConA or anti-CD3 antibody is used to stimulate proliferation. However, when analyzed by a one-sample t test, only one point, the ED90 for ConA, was significantly different from 1 and showed significant synergism. The opposite was seen with CD3/CD28 microbeads, where increased CI values at increasing ED points were observed, confirming the antagonism between the drugs under these conditions. However, due to individual donor variation, the pooled data did not reach statistical significance, although the effect of the drug was consistent in each experiment. Overall, reanalysis of the experimental data using a fixed drug ratio has confirmed previous findings and, by averaging values for all donors, allows robust conclusions.Table 1Assessment of drug interactions between dasatinib and CsA at a 1:1 ratio at varying ED pointsConAAnti-CD3 antibodyPHACD3/CD28 microbeadsCI atCI atCI atCI atED50ED75ED90ED50ED75ED90ED50ED75ED90ED50ED75ED90Mean ± SD0.91 ± 0.270.82 ± 0.160.76 ± 0.150.96 ± 0.080.87 ± 0.110.80 ± 0.131.14 ± 0.350.95 ± 0.320.83 ± 0.2715.5 ± 21.7320.88 ± 31.331.46 ± 49.85p Value0.54600.09880.04970.48130.18010.12750.47520.78830.28980.33320.35280.3683PI of cells cultured with dasatinib/CsA in combination was assessed by flow cytometry. The points when dasatinib and CsA were at a 1:1 ratio, i.e., 5:5 nM as well as points where drug was used alone allowed analysis of drug interactions at a fixed 1:1 ratio. Using the fixed drug ratio function of the statistical program Calcusyn, CI values at points where proliferation was blocked 50%, 75%, and 90% were generated and then averaged between all experiments performed. Data presented as mean, standard deviation, and significant difference from baseline value from 3–4 different experiments each using different normal donors. Open table in a new tab PI of cells cultured with dasatinib/CsA in combination was assessed by flow cytometry. The points when dasatinib and CsA were at a 1:1 ratio, i.e., 5:5 nM as well as points where drug was used alone allowed analysis of drug interactions at a fixed 1:1 ratio. Using the fixed drug ratio function of the statistical program Calcusyn, CI values at points where proliferation was blocked 50%, 75%, and 90% were generated and then averaged between all experiments performed. Data presented as mean, standard deviation, and significant difference from baseline value from 3–4 different experiments each using different normal donors. With the development and use of TKIs rapidly expanding, their interaction with other drugs and potential side effects are of considerable clinical importance. We set out to develop a simple approach that employs CFSE to track cell division and uses this information to accurately determine interactions between the TKIs dasatinib and imatinib and CsA, all known to block T-cell proliferation [3Schade A.E. Schieven G.L. Townsend R. et al.Dasatinib, a small-molecule protein tyrosine kinase inhibitor, inhibits T-cell activation and proliferation.Blood. 2008; 111: 1366-1377Crossref PubMed Scopus (217) Google Scholar, 4Seggewiss R. Lore K. Greiner E. et al.Imatinib inhibits T-cell receptor-mediated T-cell proliferation and activation in a dose-dependent manner.Blood. 2005; 105: 2473-2479Crossref PubMed Scopus (246) Google Scholar, 6Kahan B.D. Cyclosporine.N Engl J Med. 1989; 321: 1725-1738Crossref PubMed Scopus (1633) Google Scholar]. The use of CFSE to track cell division provides more detailed information on cell division than the less direct methods that measure DNA synthesis or changes in cell metabolism. Using the Modfit LT software, the total proliferation occurring in the cultures could be accurately quantified. Determining drug interactions by this method was simple and required only subtle data manipulation to enter the data directly into Calcusyn, with most calculations able to be performed automatically in spreadsheet programs. Although Modfit was used to generate PI values, they can also be generated manually using simple equations described previously [22Lyons A.B. Analysing cell division in vivo and in vitro using flow cytometric measurement of CFSE dye dilution.J Immunol Methods. 2000; 243: 147-154Crossref PubMed Scopus (566) Google Scholar]. CI values and isobolograms can also be generated manually [21Chou T.C. Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors.Adv Enzyme Regul. 1984; 22: 27-55Crossref PubMed Scopus (5891) Google Scholar, 25Tallarida R.J. Drug synergism: its detection and applications.J Pharmacol Exp Ther. 2001; 298: 865-872PubMed Google Scholar]; however, a good understanding of mathematics and statistics is essential for generating meaningful results. Using this approach, we have investigated in detail the interactions between dasatinib and cyclosporine in blocking T-cell proliferation. We have demonstrated that after most polyclonal stimuli, dasatinib and CsA interact to increase the inhibition of T-cell proliferation. These interactions were generally either synergistic or additive, with stronger interactions observed after stimulation with ConA or an anti-CD3 antibody compared with PHA stimulation. Due to the intrinsic differences between donors and the limited number of replicates, only a single drug combination, the ED90 after ConA stimulation, was shown to be statistically significant. Therefore, we conclude that dasatinib and CsA act in an additive manner under these conditions. Likewise, the combination of imatinib with CsA was additive or slightly synergistic; however, imatinib showed only a weakly inhibitory effect on T-cell proliferation stimulated by anti-CD3 or anti-CD3/28 microbeads. Somewhat surprisingly, we have uncovered a strong antagonistic effect between dasatinib and CsA when proliferation was stimulated with CD3/CD28-coated microbeads, which was evident at dasatinib concentrations of 1–10 nM. The addition of dasatinib reversed inhibition of T cells by CsA and increased proliferation in the sample beyond that of untreated cells. Although dasatinib alone sometimes caused a small increase in T-cell proliferation, it was markedly enhanced by the addition of CsA. This potent antagonism was not observed with imatinib. The dasatinib/CsA antagonism appears to be specific for co-stimulation of the CD28 receptor, as an additive effect between the drugs was observed when anti-CD3 antibody alone was used to stimulate T cells. A possible cause for this antagonism may be dasatinib blockade of an inhibitory regulatory kinase that is activated through engagement of CD28. Another possibility is that the combination of dasatinib and CsA inhibits regulatory T cells more potently than effecter T cells [26Fei F. Yu Y. Schmitt A. et al.Dasatinib inhibits the proliferation and function of CD4+CD25+ regulatory T cells.Br J Haematol. 2009; 144: 195-205Crossref PubMed Scopus (59) Google Scholar], thus releasing effecter T cells from regulatory T-cell suppression, to potentially enhance overall T-cell proliferation [27Chen C. Lee W.H. Yun P. Snow P. Liu C.P. Induction of autoantigen-specific Th2 and Tr1 regulatory T cells and modulation of autoimmune diabetes.J Immunol. 2003; 171: 733-744PubMed Google Scholar]. Further investigation of the effects of dasatinib/CsA combinations on specific T-cell subsets, including regulatory T cells, would be of great interest. Alternatively, it is known that CsA inhibits the adenosine triphosphate binding cassette transporter ABCB1, which is important in efflux of dasatinib from cells [28Hiwase D.K. Saunders V. Hewett D. et al.Dasatinib cellular uptake and efflux in chronic myeloid leukemia cells: therapeutic implications.Clin Cancer Res. 2008; 14: 3881-3888Crossref PubMed Scopus (156) Google Scholar], thus the dynamics of dasatinib retention in cells may be altered. It is interesting to note that while numerous articles have shown dasatinib to be a potent inhibitor of T cells in vitro, clinical studies have shown that a small subset (20%) of patients taking dasatinib develop plural effusions, with the cause now linked to a clonal expansion of T cells or natural killer cells [29Ottmann O. Dombret H. Martinelli G. et al.Dasatinib induces rapid hematologic and cytogenetic responses in adult patients with Philadelphia chromosome positive acute lymphoblastic leukemia with resistance or intolerance to imatinib: interim results of a phase 2 study.Blood. 2007; 110: 2309-2315Crossref PubMed Scopus (288) Google Scholar, 30Mustjoki S. Ekblom M. Arstila T.P. et al.Clonal expansion of T/NK-cells during tyrosine kinase inhibitor dasatinib therapy.Leukemia. 2009; 23: 1398-1405Crossref PubMed Scopus (269) Google Scholar, 31de Lavallade H. Punnialingam S. Milojkovic D. et al.Pleural effusions in patients with chronic myeloid leukaemia treated with dasatinib may have an immune-mediated pathogenesis.Br J Haematol. 2008; 141: 745-747Crossref PubMed Scopus (114) Google Scholar, 32Kreutzman A. Juvonen V. Kairisto V. et al.Mono/oligoclonal T and NK cells are common in chronic myeloid leukemia patients at diagnosis and expand during dasatinib therapy.Blood. 2010; 116: 772-782Crossref PubMed Scopus (148) Google Scholar]. Despite the complications caused by T or natural killer cell expansion in patients taking dasatinib who develop plural effusions, those same patients had a better response during therapy, especially in the more advanced stages of disease [30Mustjoki S. Ekblom M. Arstila T.P. et al.Clonal expansion of T/NK-cells during tyrosine kinase inhibitor dasatinib therapy.Leukemia. 2009; 23: 1398-1405Crossref PubMed Scopus (269) Google Scholar], with beneficial effects linked to increased immune responses against the leukemic cells. Understanding how dasatinib can increase immune responses in patients is of major interest and is also cause for caution should other immune-suppressive drugs be used. Our work has demonstrated that the combination of dasatinib and CsA has the potential to more potently inhibit T-cell proliferation and enhance T-cell proliferation, depending on the stimuli used. Although only polyclonal T-cell stimuli were used in this study, if similar results are seen in patients, they can have implications for therapy success. If the drug combination suppresses this dasatinib-induced T-cell expansion, it might remove this potentially beneficial effect. Alternatively, if T-cell expansion is increased it might improve antileukemic immune responses, and also potentially increase other autoimmune effects, such as pleural effusions. Although we can only speculate about the effects on patient immune function, our limited results suggest that monitoring patient immune cell function during combination therapy would be important. This has recently become more clinically relevant as dasatinib and CsA combinations were shown to improve therapy against BCR-ABL–positive leukemic cells [19Porter C.C. Gregory M.A. Zaberezhnyy V. Klawitter J. Christians U. DeGregori J. The addition of cyclosporine to dasatinib therapy of murine Bcr-Abl+ leukemia improves disease control independent of altered pharmacokinetics.ASH Annual Meeting Abstracts. 2010; 116: 600Google Scholar, 20Gregory M.A. Phang T.L. Neviani P. et al.Wnt/Ca2+/NFAT signaling maintains survival of Ph+ leukemia cells upon inhibition of Bcr-Abl.Cancer Cell. 2010; 18: 74-87Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar], and a phase 1b clinical trial has been proposed to test the combination for efficacy. Overall, the use of quantitative proliferation data obtained by CFSE dye dilution to generate CI values was straightforward and applicable to any study of combining drugs that affect cell proliferation. Two different experimental layouts were investigated in this study. Analysis using the non-fixed ratio method was able to show a clear difference between drug interactions, but donor variability made statistical analysis difficult and required data to be plotted on a log scale when donor samples were pooled. The analysis by a fixed drug ratio resulted in fewer data points, which revealed the different interactions less clearly, however, it was more easily interpreted and pooled. Our data are consistent with a previous study suggesting synergy between dasatinib and CsA at blocking T-cell proliferation after stimulation with an anti-CD3 antibody [3Schade A.E. Schieven G.L. Townsend R. et al.Dasatinib, a small-molecule protein tyrosine kinase inhibitor, inhibits T-cell activation and proliferation.Blood. 2008; 111: 1366-1377Crossref PubMed Scopus (217) Google Scholar]. However, our study extends significantly beyond those findings by applying robust mathematical analysis of interactions over a variety of conditions and also direct comparison with the prototypic TKI imatinib. Additionally, we have shown that the method of stimulating T cells can profoundly impact the outcome of drug combinations. The approach described here has the major advantage of being able to tease out subtle interaction differences that might not be apparent when cruder readouts of cell behavior are used. The use of CFSE dilution analysis is applicable to any study using combinations of drugs that affect cell proliferation for a wide variety of cell types. In combination with immunofluorescence labeling of cell surface markers, the method can follow the effects of drugs on individual cell types, such as T cells within a mixed PBMC population and is also compatible with methods to identify apoptosis (Annexin V) and cell death (DNA dyes such as propidium iodide). This study was funded in part by a Cancer Council of Australia grant to Prof. T.P. Hughes and Dr. A.B Lyons. S.J. Blake was a recipient of an Adelaide University Molecular and Biomedical Sciences Postgraduate Scholarship.