Abstract
Background: Cholangiocarcinoma (CC) is highly resistant to chemotherapy and radiation, and is, therefore, difficult to cure. Polo-like kinases (Plks) are increasingly recognized as key regulators of mitosis, meiosis and cytokinesis. Alterations in PLK1- expression have been brought into relation with tumorigenesis, thus rendering PLK1 suppression an interesting target for tumor therapy. BI 2536, the first compound of the chemical class of dihydropteridinones, is a highly selective and potent inhibitor of PLK1. Materials and Methods: Retardation of cell proliferation by BI 2536 was tested in 14 CC cell lines by cell viability assay. Moreover, molecular activity of BI 2536 was investigated by Western blot, flow cytometry and real time- polymerase chain reaction (RT-PCR). Apposition of gemcitabine, 5-fluorouracil (5-FU) and insulin-like growth factor-1 receptor (IGF-1R) retardant NVP-AEW541 was also examined. Results: BI 2536 subdued proliferation in all CC cell lines, however, reaction was stronger in gallbladder carcinoma. Therapy with BI 2536 did not result in a significant change in phosphorylation of histone H3, AKT, and p42/44. However, exposure of cells to this compound caused arrest at the G2/M-checkpoint and a surge in apoptosis. Moreover, PLK1 and FOXM1 were concurrently present in all cell lines, proposing a role for their involvement. Use of a mixture of BI 2536 with 5-FU or NVP-AEW541 resulted in synergism, while a mixture with gemcitabine resulted in additive activity. Conclusion: These experiments indicate that BI 2536 is effective against CC and increases the potency of 5-FU and NVP-AEW541.
Abbreviations: APC/C: Anaphase-promoting complex/cyclosome; BTC: Biliary tract cancer; CC: Cholangiocarcinoma; DMSO: Dimethyl sulfoxide; ECC: Extrahepatic cholangiocarcinoma; EGFR: Epidermal derived growth factor receptor; 5-FU: 5-Fluorouracil; GBC: Gallbladder carcinoma; HRPC: Hormone refractory prostate cancer; IGF-1R: Insulin-like growth factor-1 receptor; ICC: Intrahepatic cholangiocarcinoma; NCSLC: Non-small cell lung cancer; OS: Overall survival; PCR: Polymerase chain reaction; PFS: Progression free survival; Plks: Polo-like kinases; SD: Stable disease; TTP:Time to tumor progression.
Polo-like kinase 1 (PLK1) is the best characterized protein of the four human polo-like kinases, a family of serine/threonine kinases involved in multiple steps during mitosis, meiosis and cytokinesis (1, 2). PLK1 is a key regulator of the eukaryotic cell cycle, performing functions such as G2/M transition by activation of Cyclin dependent kinase 1 (Cdk1)/Cyclin B1 complex, centrosome maturation, direct activation of anaphase-promoting complex/cyclosome (APC/C) or indirectly via degradation of APC/C-inhibitor Emi1, bipolar spindle assembly, chromosome segregation, and cytokinesis. Its role in proliferating tissues brought PLK1 into the focus of oncological research showing that it is highly expressed in a broad spectrum of human malignancies, such as colorectal, esophageal and gastric cancer as compared to surrounding healthy tissues. Moreover, PLK1 overexpression is associated with poor prognosis in several tumor types and a lower overall survival (3). This context renders PLK1 an attractive and selective target for anticancer drug development. BI 2536, the first compound of the chemical class of dihydropteridinones, is a highly selective and potent inhibitor of PLK1 (4). Preclinical studies showed its efficacy in vitro and in xenograft models in mice with acceptable toxicity (4). However, the role of PLK1 and its inhibition has not been examined in biliary tract cancer (BTC), a rare tumor with a grim prognosis due to a lack of early diagnosis and effective treatment paradigms. Therefore, the objectives of this study were to investigate PLK1 expression in BTC cell lines and to evaluate the efficacy of treatment with BI 2536 as monotherapy and in combination with 5-fluorouracil (5-FU), gemcitabine or insulin-like growth factor-1 receptor (IGF-1R) inhibitor NVP-AEW541, which was previously investigated by our group (5).
Materials and Methods
Cell lines and drugs. Fourteen human BTC cell lines were examined: 5 gallbladder carcinoma (GBC) (GB-CL-1, Mz-ChA-1, Mz-ChA-2, TGBC1-TKB, and Wittier) (6-8), 4 intrahepatic cholangiocarcinoma (ICC) (H-1, HuH-28, NEC, and RBE) (9, 10) and 5 extrahepatic cholangiocarcinoma (ECC) (CC-LP-1, CC-SW-1, EGI-1, Sk-ChA-1, and TFK-1) (7, 11-13). All cell lines were cultured in an incubator at 37°C with 5-10% CO2 in appropriate media, which were replaced every three days. Cells were tested for mycoplasma regularly by polymerase chain reaction (PCR) (Venor GeM, Minerva Biolabs, Berlin, Germany) according to manufacturer's instructions. BI 2536 was kindly provided by Boehringer (Ingelheim, Germany) and NVP-AEW541 by Novartis (Basel, Switzerland). Both were dissolved in dimethyl sulfoxide (DMSO) (as 10 mmol/l stock), divided into aliquots and stored at −20°C as specified by the manufacturer. Gemcitabine and 5-FU (diluted in 0.9% NaCl) were provided by our hospital pharmacy.
Inhibition of cell growth (MTT assay). Cytotoxic effects of drugs were determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich Chemie GmbH Munich, Germany) assay. A total of 1–50×103 cells were seeded in triplicates in 96-well plates (100 μl/well) and allowed to attach overnight. The medium was then replaced by medium (100 μl) containing the designated drug or vehicle control followed by an incubation for 3 or 6 d. For the 6 d experiment, the medium with/without drug was changed after 3 d. Two hours before the end of the incubation period, 10 μl of phoshate-buffered saline (PBS) containing MTT (5 g/l) was added to each well. Following this, the medium was removed. The precipitate was then resuspended in 100 μl of lysis buffer (DMSO, 10% SDS, 0.6% acetic acid). The absorbance was measured on an Infinite M200 microplate reader (Tecan, Grödig, Austria) at 570 nm and a reference wavelength of 690 nm. Each experiment was performed in triplicate.
Real time-quantitative polyperase chain reaction (RT-qPCR) for PLK1. After treatment of cells for 1 to 3 d, total cell RNA was extracted using the RNeasy Mini Kit (Qiagen GmbH, Hilden, Germany) after homogenization with a QIAshredder (Qiagen) according to the manufacturer's instructions and stored at −80°C. RNA was dissolved in water and quantified at 260 nm with a biophotometer (Eppendorf, Hamburg, Germany); purity was verified by optical density (OD) with an absorption ratio OD260 nm/OD280 nm of between 1.93 and 2.06. Single-step quantitative RT-PCR analysis was carried out in a LightCycler system (Roche, Mannheim, Germany), with primers and fluorochromes obtained from Qiagen (QuantiTect Primer Assay Hs_PLK1_1_SG, Hs_GAPDH_2_SG and SYBR-Green RT-PCR kit) and used according to the manufacturer's instructions. Products of RT-PCR were separated by gel electrophoresis to confirm correct amplification and size. Water was used to detect primer interactions and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the reference gene to ensure equal loading. Relative gene expression (treated to untreated) was calculated with REST software tool by Pfaffl, Horgan and Dempfle (14).
Cell cycle analysis. Cells were seeded in T-25 flasks (3.5×105), treated with different concentrations of BI 2536 or vehicle control for 48 h, washed with PBS, trypsinized, centrifuged, and fixed in ice-cold ethanol with PBS containing 1 mmol/l EDTA. DNA was labelled with 1:100 diluted propidium iodide after digestion of RNA by RNAse A. Cells were analysed by flow cytometry with a FACSCalibur system (Becton Dickinson, San Diego, USA) and cell cycle profiles were determined using WinMDI 2.9 and Cylchred software (freeware). Doublets were excluded by gating for width of fluorescence signal (FL2-W). Each experiment was performed at least in triplicates.
Immunoblotting. Cell culture monolayers were washed with ice-cold PBS and lysed in flask with a buffer containing Tris-HCl (50 mmol/l, pH 7.4), NP-40 (1%), NaCl (200 mmol/l), sodium-orthovanadate (1 mmol/l), 2-glycerophosphate (1 mmol/l), sodium fluoride (20 mmol/l), dithiothreitol (DTT; 10 mmol/l), phenylmethylsulfonylfluoride (PMSF; 1 mmol/l) and 0.2% proteinase inhibitor cocktail (Sigma-Aldrich Chemie) on ice for 30 min. The lysate was then centrifuged at 13,000 rpm for 15 min. Histone H3 and p(hospho)-histone H3 samples were treated additionally for 10 s with a sonicator (Sonoplus HD 2070; Bandelin, Berlin, Germany) prior to centrifugation. All supernatants were divided into aliquots, rapidly frozen in liquid nitrogen and stored at −80°C after protein quantification by Bradford protein assay (Bio-Rad, Munich, Germany). Fifty to 75 μg of cell lysates were separated on SDS-polyacrylamide gels and electroblotted onto polyvinylidene difluoride membranes (Carl Roth, Karlsruhe, Germany). Membranes were then incubated in blocking solution [5% BSA in TBS-T (10 mmol/l Tris-HCl, 140 mmol/l NaCl, 0.1% Tween-20)], followed by incubation with the primary antibody at 4°C overnight [5% bovine serum albumine (BSA) in TBS-T]. The membranes were then washed in TBS-T and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at room temperature. Antibody detection was performed with an enhanced chemoluminescence reaction (SuperSignal West Dura or SuperSignal West Femto, Pierce, Rockford, USA) and measured by LAS 1000 plus (Fujifilm, Düsseldorf, Germany). The data were analyzed by densitometry on GeneTools Software 4 (SynGene, Cambridge, UK). Monoclonal (mc) β-Actin antibody was purchased from Sigma (Sigma-Aldrich Chemie, Munich, Germany), polyclonal (pc) PLK1, pc FOXM1, pc p-histone H3, pc histone H3, mc p-p42/44, mc p42/44, mc p-AKT, and mc AKT antibodies were all purchased from Cell Signaling (Cell Signaling Technology, Beverly, USA).
Statistical analysis. Numeric data arepresented as the mean value with standard deviation (SD). Inter-group comparisons were performed with Student's t-test. P-values less than 0.05 were considered significant. Correlations were calculated based on Pearson correlation coefficient and tested for significance using Fisher transformation.
Results
Inhibition of cell growth. After incubation for 3 d, all BTC cell lines except TGBC1-TKB, were affected by BI 2536 (mean IC50=9.81±4.39 nmol/l; IC50 values were calculated with linear regression model, which was restricted to IC50 area). Wittier, Mz-ChA-2, Mz-ChA-1 and CC-LP-1 were the most sensitive cell lines, whereas Sk-ChA-1, GB-CL-1, RBE and TFK-1 were less sensitive (Figure 1a and Table I). In most cell lines, concentrations of BI 2536 above 20 nmol/l did not reveal additional effects. Response differed between the group of the GBC cell lines (mean IC50=6.64±5.50 nmol/l; TGBC1-TKB not included) and the group of ICC+ECC cell lines (mean IC50=11.22±3.21 nmol/l), although not statistically significantly. Treatment duration of 6 d increased the effects (mean IC50=5.29±2.23nmol/l). Response in the 6 d experiment was again slightly better for the GBC cell lines (mean IC50=3.95±2.21 nmol/l) in comparison to the group of ICC+ECC cell lines (mean IC50=6.03±1.98 nmol/l) (Figure 1b and Table I) (p>0.05). DMSO alone, the solvent for BI 2536 showed no effects on the viability at the applied concentrations (data not shown).
Inhibition of cell growth by in vitro treatment with BI 2536.
For the subsequent experiments, three cell lines were selected as being representative of each tumor entity and different in response to BI 2536: GB-CL-1, RBE and CC-LP-1. Tests using different lengths of treatment showed maximum effects being reached after 4 d. Cells treated for 2 d were significanty less viable than cells treated for only 1 d. There was no significant difference between the effects after 2 and 3 d, likewise between 4, 5, and 6 d (Figure 1c). Using data of this experiment, replication times for the three cell lines were calculated, which were similar.
RT-qPCR for PLK1. In cell lines, GB-CL-1, RBE and CCLP-1, mRNA expression of PLK1 was detectable. Relative mRNA expression of PLK1 in the cell lines was not significantly influenced by 1 d or 3 d treatment with IC50-concentrations of BI 2536 (Figure 2).
Cell cycle analysis. There was a significant negative correlation between the concentration of BI 2536 and the proportion of cells in the G2/M- fraction for GB-CL-1 and CC-LP-1 cells (correlation coefficient −0.97 and −0.99, p<0.01). Furthermore, the proportion of G 0/G 1- fraction cells correlated significantly positive with increasing BI 2536 concentrations in these cell lines (correlation coefficient 0.88 and 0.99, p<0.05). The RBE cell line exhibited no significant changes (Figure 3a-c). The sub G1/G0- fraction (marker of apoptosis) increased significantly for CC-LP-1 and there was a change in this direction in GB-CL-1 cell line (correlation coefficient 0.96 and 0.91, p<0.05). For the RBE cell line, there was no significant increase due to a relatively high sub G1/G0- fraction in untreated samples (Figure 3d).
Immunoblotting. In all 14 cell lines, immunoblot analysis revealed protein expression of PLK1 and forkhead box M1 transcription factor (FOXM1), a putative upstream regulation of PLK1 gene (Figure 4a). The level of expression of both proteins was not affected by treatment with 10 nmol/l BI 2536 for 2 d (mean expression on treatment relative to control was 1.42±0.84 and 1.06±0.42, respectively). There was no clear correlation between protein expression levels and response to BI 2536 treatment (correlation coefficients were between −0.05 and −0.53). Treatment of cell lines GB-CL-1, RBE and CC-LP-1 with different BI 2536 concentrations (0 nmol/l, IC10, IC30, IC50, and 20 nmol/l) did not result in obvious changes of p-histone H3, histone H3, p-p42/44, p42/44, p-AKT and AKT levels (Figure 4b). Nevertheless, there was a tendency for lower p-AKT and AKT expression after treatment (except AKT level for GB-CL-1, which slightly increased). The p-p42/44 and p42/44 expression level was slightly diminished for CC-LP-1 after treatment. Staining with β-Actin antibody served as control for protein loading in all immunoblots.
In vitro combination of BI 2536 with 5-FU, gemcitabine, or NVP-AEW541. To examine the impact of BI 2536 in combination with commonly used chemotherapeutic agents in BTC (5-FU and gemcitabine) and small molecule inhibitor of insulin-like growth factor-1 receptor (NVP-AEW 541), further trials were performed with cell lines GB-CL-1, RBE and CC-LP-1. Cells were treated with increasing concentrations of the selected combination partner with or without BI 2536 at a constant IC20 dose for 3 d. Assuming that the drugs do not directly interact and using the model of effect multiplication postulated by Berenbaum (15), values equal to those calculated represent additive effects. Measured values for viability lower than those calculated indicate synergistic effects, values higher than those calculated indicate antagonism. Combination of BI 2536 with 5-FU (Figure 5 a-c) and NVP-AEW541 (Figure 5 g-i) resulted in synergism. In contrast, BI 2536 in combination with gemcitabine resulted only in additive effects (Figure 5 d-f).
In vitro cell growth inhibition. a): Treatment of 14 human biliary tract cancer cell lines with BI 2536 for 3 d (n=9); b): 6 d incubation (n=9); c): incubation of selected cell lines GB-CL-1, RBE and CC-LP-1 with calculated doses of BI 2536 for 24-144 h (n=9).
Discussion
Biliary cancer as a rare tumor of the gastrointestinal tract is subdivided into GBC, ICC and ECC. Although complete surgical resection is the only curative approach, this can only be accomplished in a minority of patients, since most of them present with advanced disease. In addition, those patients who have undergone complete surgical resection experience a high tumor recurrence rate. Non-resectable BTC is associated with a poor prognosis due to wide resistance to chemotherapeutic agents and radiotherapy. In addition, there is a lack of prospectively randomized phase III trials testing modern chemotherapy regimens. The only exceptions are a recent UK trial (ABC-02), which detected a significantly increased median time to tumor progression (TTP) and median overall survival (OS) with a combination of gemcitabine and cisplatin in BTC patients (16) and a recent Indian trial which detected a significantly increased median progression-free survival (PFS) and median OS with a combination of gemcitabine and oxaliplatin in GBC patients (17). It is, therefore, essential to search for new therapeutical approaches.
Influence of BI 2536 (1 d and 3 d treatment with calculated IC50) on mRNA expression of PLK1 in three of the tested cell lines. The ratio of PLK1 mRNA expression in treated vs. untreated cell lines is shown (n=3).
After several years of preclinical research, the first clinical study data are now available for this tumor entity. Inhibitors of the epidermal-derived growth factor receptor (EGFR) family, such as erlotinib, cetuximab, and lapatinib were recently investigated. Furthermore, bortezomib, an inbibitor of the proteasome, imatinib, bevacizumab, and sorafenib, were studied, as well. Although early evidence of antitumor activity was seen, the results are still preliminary and require further investigation (for review see (18)).
In the current preclinical study another approach was investigated: inhibition of polo-like kinase 1 (PLK1) by BI 2536. This may be interesting, since it has been shown that PLK1 inhibition can be used even to treat tumors with p53 or RAS mutations (3, 19, 20). In vitro investigations of BI 2536 on human cancer cell lines from different entities revealed an IC50 of 1.4 to 26 nmol/l after 3 d of treatment (4, 21-23). These results are well comparable to ours for BTC cell lines. In addition, we tested different lengths of treatment and demonstrated maximum effects to be reached after 4 d. In a next step, we tried to identify possible mechanisms of action of BI 2536 in a selection of our BTC cell lines. Since PLK1 expression has been shown to correlate strongly with the mitotic activity of cells and is a prerequisite as BI 2536 target, we performed qRT-PCR and Western blot analysis for PLK1. As has been shown for tumors of various origin (lung, colon, stomach, smooth muscle, and esophageal tumors as well as non-Hodgkin lymphomas), which express high levels of PLK1 transcripts in about 80% (24), we detected PLK1 expression in all BTC cell lines. Moreover, BI 2536 treatment did not significantly influence the level of PLK1 expression. Transcriptional factors regulating PLK1 include BReast CAncer 1 (BRCA1), the pRb family members, and FOXM1 (25-27). FOXM1 regulates the expression of many G2-specific genes and is essential for chromosome stability. For hepatoblasts, FOXM1 is essential for differentiation towards biliary epithelial cell lineage (28). This function in embryogenesis may be indicative of a role in transformed biliary tract cells. It has been shown that a positive feedback mechanism between PLK1 and FOXM1 is essential at the G2/M checkpoint and for expression of many key regulators (29). The expression of PLK1 is driven by Ha-Ras via FOXM1 (30). Thus, it is uncertain if inhibition of PLK1 causes changes in FOXM1 expression. In our study, we did not detect a significant change in FOXM1 expression after BI 2536 treatment. Since it has been shown in preclinical studies that inhibition of PLK1 causes G2/M arrest, and consequently increases the level of mitosis marker p-histone H3 (31), we studied these effects in our BTC cell lines, too. Whereas the first mechanism was confirmed in our cell cycle analysis, it is unclear why we were unable to detect a significant increase in histone H3 phosphorylation. Since a relevant rate of apoptosis occurred after BI 2536 treatment, as shown by cell cycle analysis, we were interested to determine whether the level of p-AKT, an important molecule of downstream cell signaling (PI3K/AKT cascade) and an important inhibitor of apoptosis, was diminished. Unfortunately, this was not the case. In addition, there was no proof that the cell proliferation rate was diminished by dephosphorylation of p-p42/44, an important member of the rat sarcoma (Ras)/Raf/mitogen-activated protein kinase kinase (MEK)/mitogen-activated protein kinase (MAPK) signaling cascade. Based on preclinical data, the small compound BI 2536 has been studied in patients with solid tumors or hematologic malignancies. During phase I trials for solid tumors, intravenously administered BI 2536 showed dose proportional pharmacokinetics when administered using various dose schedules. Hematologic toxicity was dose- limiting, establishing maximum tolerated doses of 200 mg or 300 mg when administered every 21 d to patients with advanced solid tumors (32-34). In four phase II trials, BI 2536 was then investigated in patients with metastatic hormone-refractory prostate cancer (HRPC), unresectable advanced pancreatic cancer, stage IIB/IV non-small cell lung cancer (NCSLC) (second-line), and mixed solid tumors (advanced head and neck cancer, breast cancer, ovarian cancer, soft tissue sarcoma and melanoma) (35-38). In the first study, BI 2536 was very well tolerated in metastatic HRPC patients (n=15), with some indications of antitumor activity by PSA reduction and radiologically stable disease. However, the stage I study objectives were not reached. In the second study, no objective response was observed in either treatment arm and achievement of stable disease for at least 12 weeks did not exceed 20% of patients with pancreatic cancer (n=86). A median OS of 5 months and a one-year survival of 18% were no better than commonly reported for gemcitabine in earlier trials. In the third trial, four NCSLC patients (4.2%) had a partial response; two were confirmed by independent review. The median PFS was 8.3 weeks and 7 weeks assessed by investigator and independent review, respectively. The median OS was 28.7 weeks. In the fourth trial, 76 patients with mixed solid tumors were included, 71 started treatment and received a median number of two cycles (four in ovarian cancer). Confirmed objective responses were not observed. All cohorts were closed after the entry of 14-15 eligible non-responding patients.
Cell cycle analysis. In vitro treatment of selected cell lines GB-CL-1 (a), RBE (b), and CC-LP-1 (c) with BI 2536 for 48 h (n=3) (**p<0.01 for G 2/M fraction, treated cells vs. control). Cells having less than single DNA content (subG 1-fraction) were presumed to be apoptotic (n=3) (*p<0.05 treated cells vs. control) (d).
Mechanism of BI 2536 drug action. Protein expression of PLK1 and FOXM1 in tested human biliary tract cancer cell lines was determined by immunoblot after treatment with 0 or 10 nmol/l BI 2536 (a); p-histone H3, histone H3, p-p42/44, p42/44, p-AKT and AKT protein levels were examined by immunoblot for selected cell lines GB-CL-1, RBE and CC-LP-1 (b). β-Actin served as loading control in all experiments. Densitometry was performed to analyze results.
In vitro treatment with drug combinations of BI 2536 and 5-fluorouracil (5-FU), NVP-AEW541, or gemcitabine. Selected cell lines GB-CL-1, RBE and CC-LP-1 were incubated with rising concentrations of 5-FU (a-c), gemcitabine (d-f), or NVP-AEW541 (g-i) alone and in combination with fixed IC20 concentration of BI 2536. The shaded area represents possible drug synergism, error bars represent SD (n=3).
These phase II study data sparked our interest in investigation of in vitro combination therapy. The combined drugs should ideally tackle the cancer cells in different cell phases, using different modes of action. Gemcitabine and 5-FU are currently used as chemotherapy for BTC (39). Gemcitabine is a nucleoside analoge that is utilized instead of cytidine during DNA replication, leading to premature chain termination and subsequent apoptosis. While 5-FU is principally also a nucleoside analoge, its main effects are exerted through inhibition of thymidylate synthase and hence reduction of thymidine necessary for DNA replication. Both drugs affect cells mainly during the S phase, while our flow cytometric experiments showed that treatment with BI 2536 caused G2/M phase arrest leading to synergistic effects in combination with 5-FU and additive effects in combination with gemcitabine. Additionally, we tested the combination of BI 2536 with IGF-1R inhibitor NVP-AEW541 mediating G1 arrest. The IGF-1R system has emerged as an interesting target for cancer therapy, as it represents an important promoter of tumor transformation and survival of malignant cells, but is only partially involved in normal cell growth (40-42). This is in part attributed to interactions with oncogenes. Moreover, activation of IGF-1R may contribute to tumor angiogenesis by up-regulating of vascular endothelial growth factor expression in certain cancer entities (43-45). Thus, targeting the IGF-1R system with small molecule tyrosine kinase inhibitors, such as NVP-AEW541, a novel compound which is more selective for IGF-1R than the insulin receptor by a factor of 27 on a cellular level, may be a new strategy of cancer growth inhibition (46, 47). As a result, combination of BI 2536 with NVP-AEW541 resulted in synergistic effects. However, since this was observed mainly at high micromolar concentrations one cannot exclude a possible off-target effect.
In summary, our findings suggest that BI 2536 is active against BTC in vitro. In addition, the compound potentiated the efficacy of 5-FU and NVP-AEW541. Based on these data and data from phase II studies, further clinical evaluation of this new drug for combination treatment of BTC is recommended.
Acknowledgements
The Authors thank Boehringer-Ingelheim for the provision of BI 2536 and Novartis Pharma for the provision of NVP-AEW541. Furthermore, we thank the Institute for Medical Informatics, Statistics and Epidemiology for statistical guidance.
- Received July 3, 2011.
- Revision received August 17, 2011.
- Accepted August 18, 2011.
- Copyright© 2011 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.