Tozasertib

In vitro evaluation of the Aurora kinase inhibitor VX-680 for Hepatoblastoma

Abstract

Purpose
Hepatoblastoma (HB) has a poor prognosis in advanced stages. The aim of this study was to enhance the effectiveness of chemotherapy with antineoplastic kinase inhibitors.

Methods
Viability was monitored in HB cells (HUH6, HepT1) in monolayer and spheroid cultures treated with kinase inhibitors VX-680, Wee1-Inhibitor II, and SU11274 alone or in combination with cisplatin (CDDP) using MTT assays. Apoptosis was revealed by Caspase-3 assay. Western blot and immunohistochemical analyses were performed to determine histone H3 phosphorylation.

Results
Among the kinase inhibitors, the strongest anti-proliferative effect on HB cells was documented for VX-680. HUH6 cells responded more sensitively to the Aurora kinase inhibitor than HepT1 cells (IC50 8 and 16.6 μM, respectively). While VX-680 and CDDP showed no additive effects, the combination of VX-680 and histone deacetylase inhibitor SAHA had a synergistic effect on the proliferation of HUH6 cells. The inhibition with VX-680 led to reduced histone H3 phosphorylation, to an increase of apoptotic cells, and to morphological changes such as vacuolization and swelling of the cells and nuclei.

Conclusion
The data provide evidence that VX-680 might improve treatment results in HB with increased Aurora kinase activity by inhibiting cell proliferation and induction of apoptosis.

Keywords
Kinase, Aurora Kinase A, Wee kinase, Met, Histone deacetylation, Hepatoblastoma, Multidrug resistance

Introduction

Hepatoblastoma (HB), accounting for 1% of all infantile cancer diagnoses, is the most common malignant liver tumor in childhood. Although there have been great achievements in treating patients with standard risk HB in the past decades, treatment results for those with high-risk HB still remain poor. In the latter case, a developing multidrug resistance during multiple chemotherapy cycles often appears to be the major challenge. One option to overcome this problem is the recently, widely discussed and explored molecular targeted therapy. Recent advances have been made in identifying molecules which are crucial for the malignant phenotype. Substances targeting those specific molecules are, for example, small molecule kinase inhibitors such as the Aurora kinase inhibitor VX-680. Aurora kinases A, B, and C are serine/threonine kinases which are essential for cell proliferation and play a key role during mitosis. They are overexpressed in several tumors suggesting an important role in tumorigenicity. While Aurora kinase A participates in centrosome function, mitotic entry, and spindle assembly, Aurora kinase B is involved in chromatin modification, microtubule-kinetochore attachment, the spindle checkpoint, and cytokinesis. The Aurora kinase inhibitor VX-680 is a pan-inhibitor that blocks cell progression through inhibition of histone H3 phosphorylation, thus leading to polyploidy, inhibition of cell proliferation, and apoptosis. Its action has shown promising efficiency in many human tumor types.

Other molecules that interfere with the cell cycle progression are Wee1-inhibitors and histone deacetylase (HDAC) inhibitors (HDACi). The tyrosine kinase Wee1, important for the G2/M transition during the cell cycle, is a critical regulator with respect to DNA-related double-strand breaks that acts through phosphorylation of CDC2 thus leading to a stop in G2, allowing the cell to repair the damage. Inhibition of Wee1 has been shown to abrogate the G2/M checkpoint, forcing cells with DNA damage, for example through cytotoxic agents such as cisplatin (CDDP), to enter mitosis which consequently leads to cell death.

The impact of kinase inhibitors depends on active signal transduction within the malignant cell. Epigenetic modulators may restore control mechanisms of cell death and support the effect of kinase inhibitors. HDAC inhibitors such as vorinostat (suberoylanilide hydroxamic acid, SAHA) unfold their potential at the epigenetic level. The balance between acetylation through acetyltransferases (HATs) and deacetylation through HDACs on histone proteins is crucial in regulating gene expression. While HATs catalyze histone acetylation and relax the chromatin to increase accessibility for transcription factors, HDACs repress gene transcription. In several tumor types, including gastric and prostate cancer, an overexpression of HDACs could be shown tilting the balance in favor of a hypoacetylated state. Consequently, tumor cells downregulate gene expression while focusing on their cell division. Inhibition of HDACs leads to altered gene expression thus in turn leading to apoptosis and autophagy.

Hence, inhibition of cell cycle-modulating kinases might provide a promising tool to increase the effectiveness of chemotherapy in patients with high-risk or recurrent HB. In this study, we describe the effects of various kinase inhibitors either alone or in combination with epigenetic modulators and cytotoxic drugs commonly used in the treatment protocols for HB.

Methods

Drugs
VX-680, Wee1-Inhibitor II, Met-Inhibitor SU11274, and SAHA were purchased from Selleck Chemicals (Houston, TX, USA), solubilized in DMSO and diluted with medium to final concentrations in the cell culture between 1 and 100 μM. The highest DMSO concentration in cultures was 0.1 μg/ml. Cisplatin (CDDP, Neocorp AG, Weilheim, Germany) was commercially available as drug formulation.

Cells and Culture Conditions
The HB cell lines HepT1 and HUH6 were used for all experiments. Tumor cells were grown as monolayer in Dulbecco’s MEM (Biochrom, Berlin, Germany) supplemented with 10% fetal calf serum and 1% glutamine. The cells were grown at 37 °C in a humidified atmosphere containing 5% carbon dioxide. All used cells were mycoplasma negative. For spheroid cultures, low attachment plates were used (Corning Inc, Corning, NY, USA) as previously described.

Cell Viability Assay
HB cells (10,000 cells/100 μl) were seeded out in 96-well plates (Becton–Dickinson GmbH, Heidelberg, Germany) and cultured as described above. At day two, drugs were added to the cells at different concentrations of up to 100 μM. Experiments were repeated with kinase inhibitors alone or in combination with cytotoxic drugs like CDDP and SAHA. Drug solutions were prepared shortly before administration. All assays were performed three times in triplicates.

Cell viability was assessed by MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide] assay (Sigma-Aldrich, Munich, Germany). 25 μl MTT (5 mg/ml) dissolved in PBS was added to each well. After incubation for 6 hours 100 μl/well lysis solution (10% SDS in acid water; Merck, Darmstadt, Germany) was added and further incubated overnight in the dark at room temperature. Cell viability was assessed by measuring absorption at 570 nm using a Milena Kinetic Analyzer (DPC Bierman, Bad Nauheim, Germany). Percentages of cell viability were calculated by normalization between background of cultures without cells and untreated cultures as control. Dose-dependent viability curves were computed by sigmoidal curves with variable slope to determine IC50.

Apoptosis Assay
Cells were seeded at a density of 1 x 10^4 cells/100 μl and incubated overnight. Later, cells were treated with increasing concentrations of VX-680 (6 and 12.5 μM) for 24 hours. NucView™ 488 Caspase-3 Assay Kit for Live Cells was used to detect apoptosis (Biotium, Inc., Hayward CA, USA). Five microliters of NucView™ 488 Caspase-3 substrate was added to the cells and thereafter incubated for 30 minutes. Cells positive for apoptosis showed a green fluorescent signal. Immunofluorescence microscopy was carried out on a Zeiss Axio Scope epifluorescence microscope (Carl Zeiss, Oberkochen, Germany) with a MRC5 camera. Images were processed using AxioVision 4.8.1 software.

Western Blot Analysis
Western blot analysis was performed as previously described. For detection of histone H3 phosphorylation, rabbit anti-phospho-Histone H3 (Ser 10) antibody was used. Horseradish peroxidase (HRP)-goat-anti-rabbit antibody (both antibodies purchased from Cell Signaling Technology, Danvers, MA, USA) served as secondary antibody. Visualization of antibody binding was carried out by chemiluminescence (Roti Lumin, Carl Roth, Karlsruhe, Germany). Densitometric analysis was done with AlphaDigiDoc software (Biozym Scientific, Oldendorf, Germany).

Immunohistological Staining
Tumor tissue from xenografts of HUH6 and HepT1 were from a previously described experiment. Primary HB samples were collected for expression analysis after obtaining informed consent from the patient’s parents. For cryosections, tumor specimens were embedded in Tissue Tec OCT™ (Sakura Finetek, Alphen aan den Rijn, The Netherlands) and frozen in liquid nitrogen. Frozen specimens were sectioned into 10 μm sections using a Leica Cryotom. Before staining, sections were fixed with methanol/acetone 1:1 at -20 °C for 10 minutes and then dried at room temperature. Slides were incubated in goat-serum (1%, DAKO, Hamburg, Germany) for 30 minutes to block unspecific binding areas. For detection of phosphorylated histone H3 proteins, a rabbit anti-phospho-Histone H3 (Ser 10) antibody (Cell Signaling Technology, Danvers, MA, USA) was used. After overnight incubation, cells were washed in PBS. Peroxidase-labeled goat anti-rabbit antibody served as secondary antibody (Biozol, Eching, Germany). Cells were washed and sections were then incubated with the ABC-Kit (Vector SK-6100; Vector Laboratories, Burlingame, CA, USA) for 30 minutes, stained with DAB (Biocompare, San Francisco, USA) for 5 minutes, and counterstained with hematoxylin-eosin. Microscopy was carried out on a Zeiss Axiovert 135 microscope (Axiovert 135, Carl Zeiss, Oberkochen, Germany).

Statistics
Statistical analysis was carried out by one-way ANOVA on ranks test using GraphPad Prism 4.00 (GraphPad Software, San Diego, California, USA, www.graphpad.com). Viability curves were fitted with a sigmoidal dose response function with variable slope. F Test was used to compare curve parameters of treatment. All numeric data are expressed as means. Data plotted on graphs are mean and SD. Significance was assumed for all p < 0.05. Results Effects of Kinase Inhibitors on Hepatoblastoma Cell Lines Since there is great evidence that overexpression of Aurora kinase A, the tyrosine kinase receptor MET as well as the Wee1 kinase are linked to the development and poor prognosis of hepatocellular carcinoma, we analyzed the gene expression of these kinases based on microarray data retrieved from 25 primary HB tissues and normal liver. When the HB tissues of those patients who had an event-free survival in the following 3 years after chemotherapy/resection were compared with those of patients who had died during this period of time, a significant increase for Aurora kinase A gene expression (p = 0.01) could be observed in the latter set. For Met expression, there were no detectable changes; both HB tissues and normal liver tissues expressed Met at the same level. In contrast, Wee1 expression levels in HB tissues differed from those of normal liver tissues. There was a significant decrease in Wee1 gene expression (p = 0.01) which was about two times lower compared with normal liver. To evaluate whether these kinases play a role in the proliferation of HB cells, kinase inhibitors were used to assess cell viability in an MTT assay. Cells were incubated with increasing concentrations (0–100 μM) of either the Aurora kinase inhibitor VX-680, the MET-inhibitor SU11274, or the Wee1-inhibitor Wee I. For both cell lines, HepT1 and HUH6, a dose-dependent decrease in cell viability was observed for VX-680 and Wee-inhibitor I. For VX-680, 50% growth inhibition (IC50) was assessed at 16.6 ± 1.2 μM for HepT1 and 8.0 ± 1.1 μM for HUH6. For HUH6, even low concentrations of the Aurora kinase inhibitor led to cell inhibition of approximately 10%. In comparison with the Wee-inhibitor I, 50% growth inhibition was assessed at higher concentrations with approximately 29 μM for both HepT1 and HUH6. In contrast, treatment with the MET-inhibitor SU11274 did not lead to a decrease in cell viability of more than 36.6 ± 4.42% regardless of the concentration (for HUH6). Taken together, of the three tested kinase inhibitors VX-680 showed the strongest effect on HB cell growth. However, 50% cell inhibition for HepT1 was observed with almost twice the concentration of that for HUH6. The observed cell death under treatment with VX-680 could be due to apoptosis, necrosis, or autophagy. To further evaluate whether the Aurora kinase inhibitor triggers apoptosis in HB cell lines, both HepT1 and HUH6 were treated with VX-680 (6, 12.5 μM) for 24 hours and a Caspase-3 assay was performed. Treatment with VX-680 increased the activity of Caspase-3 in a dose-dependent manner. In HUH6 cells, more cells with active Caspase-3 were observed at the same cell density when compared with HepT1 cells. Hence, HUH6 cells show a more sensitive response to the Aurora kinase inhibitor VX-680. Effects of VX-680 in Combination with CDDP and SAHA The Aurora kinase inhibitor VX-680 was shown to increase apoptosis when combined with the DNA-damaging agent CDDP and the histone deacetylase inhibitor SAHA, respectively. Therefore, we assessed whether an additional effect in growth inhibition could be attained by combining these two agents with VX-680. HepT1 and HUH6 were incubated with either increasing concentrations of VX-680 (1–50 μM) alone or with CDDP (2.5 μg/ml) and SAHA (0.5 μM). While treatment with only the Aurora kinase inhibitor led to an IC50 of 16.6 ± 1.2 μM for HepT1 and 9.9 ± 1.0 μM for HUH6, the combination with CDDP led to a slight decrease in HepT1 with a growth inhibition of 50% at concentrations of 12 ± 1.8 μM. In HUH6, no IC50 for the combination with CDDP could be assessed since even at the lowest VX-680 concentration growth inhibition was determined to be less than 50%. It was also observed that treatment with CDDP alone showed a decrease in cell viability of fewer than 58% in HUH6 and only 75% in HepT1 cultures. However, no synergistic effect in both cell lines could be observed. The combination with SAHA also led to a shift of the IC50 towards lower concentrations of VX-680. For HUH6, 50% growth inhibition was observed at 5.2 ± 1.1 μM, for HepT1 at 15.4 ± 1.1 μM. Hence, there is little sensitivity for HepT1 towards a combination of VX-680 and SAHA. In contrast, in HUH6 cultures only half of the concentration of VX-680 was needed to reach 50% cell inhibition when 0.5 μM SAHA was added. Taken together, there is evidence for a synergistic effect between the Aurora kinase inhibitor and SAHA in HUH6 cells. Treatment with VX-680 and SAHA also induced prominent morphological changes in both cell lines. Nuclei and cell diameter were determined via DAPI staining and microscopy in viable cells after 72 hours of drug exposure. Compared with controls, an increase of vacuoles and dead cells was observed in both cell lines under combined treatment. The effect of VX-680 and/or SAHA on cell morphology was similar in both HepT1 and HUH6 cells. There was also a significant increase in both the nuclei and cell diameter in HepT1 cells when treated with VX-680 alone or in combination. The same applied to HUH6 cells with an additional significant increase when treated with SAHA alone. In HUH6 cells, nuclei diameters increased during the combined treatment from 10.5 ± 1.5 to 18.8 ± 1.8 μm, which is about 1.8-fold higher. Compared with HepT1, nuclei diameters increased approximately 1.5-fold. Effect of VX-680 and SAHA on the Phosphorylation of Histone H3 Proteins in HB Cells Since both VX-680 and SAHA inhibit phosphorylation of the histone H3 protein, we further evaluated the effects of these substances with respect to the phosphorylation extent of H3 proteins in HB cells. HUH6 and HepT1 cells were therefore treated with VX-680 (10 μM), SAHA (0.2 μM), or a combination of both for 24 hours and Western blot analysis was performed. A very weak histone H3 phosphorylation in HepT1 could be observed when compared with HUH6, indicating that in HepT1 cells most of the histone H3 proteins are already dephosphorylated. In both HB cell lines, treatment with VX-680 and the combination with SAHA led to a decrease in histone H3 phosphorylation. No phosphorylated histone H3 protein could be detected. The decrease after combined treatment is probably due to the action of VX-680 alone since exclusive treatment with SAHA could not dephosphorylate histone H3 proteins. Taken together, a decrease in histone H3 phosphorylation could be observed after treatment with the Aurora kinase inhibitor VX-680. Treatment with SAHA alone or in combination with VX-680 could not contribute to this effect. Phosphorylation Status in HB Xenografts and Primary HB Tissues To further evaluate phosphorylation of histone H3 proteins in HB primary tissues and xenografts, immunohistochemical staining was performed using an antibody which recognizes Phospho-10S in histone H3 proteins. For all tumor specimens, negative controls were carried along by omitting incubation with the primary antibody. No phosphorylated histone H3 proteins could be detected in these tumor samples. In HB tumors induced by subcutaneous injection of HepT1 and HUH6 cells, sporadic cells with phosphorylated histone H3 proteins could be observed. In addition, phosphorylation density was the same in HepT1 and HUH6 xenografts. Similar results were obtained from primary HB tissues. However, the extent of histone H3 phosphorylation was heterogeneous depending on the sample. Compared with xenotransplants, more cells with phosphorylated histone H3 proteins were found in primary HB. Effects of VX-680 and SAHA on Spheroids The observed effects of VX-680 and SAHA on cells cultured as monolayer were also evaluated under chemoresistant conditions. Therefore, HB cells were cultured on low binding plates to form 3D cultures, also known as spheroids, which represent a model with multidrug resistance character. Cells grown as spheroids displayed a less sensitive effect towards VX-680 (0–100 μM) and the combination with 0.5 μM SAHA. Compared with 2D cultures, growth inhibition of 50% could not be attained for both cell lines. For instance, in HUH6 cells, cell viability in 2D cultures decreased at a concentration of 25 μM VX-680 to 5.37 ± 1.2%, whereas cells grown as spheroids showed almost 100% viability. However, the combination with SAHA in HUH6 3D cultures led to a moderate synergistic effect as observed for monolayer cultures. This synergistic effect could also be observed in HepT1 spheroids, but not in the monolayer culture. In addition, the response of HepT1 3D cultures to VX-680 and SAHA was more prominent compared with HUH6 spheroids; as for 25 μM VX-680, the cell viability decreased to 75% of control cultures. Here we showed that VX-680 and the combination with SAHA could not overcome the multidrug resistance caused by spheroid cultures. Discussion An acquired multidrug resistance resulting from multiple chemotherapy courses is a major problem in the treatment of patients facing HB and various other tumor entities. For patients with high-risk HB or those who suffer from relapse, it is therefore of great interest to increase the outcome by developing new therapy strategies since, with 69%, the 3-year survival still remains poor. To overcome this issue, recent advances have been made through the development of small molecule inhibitors in order to improve current therapy results. These inhibitors include, among others, kinase inhibitors such as the Aurora kinase inhibitor VX-680 as well as the histone deacetylase inhibitor vorinostat (SAHA), which recently have been shown to block tumor cell proliferation, induction of apoptosis, and tumor regression. Therefore, combination of these drugs with cytotoxic agents such as CDDP, which is currently used in the standard chemotherapy regimen for high-risk HB, may be a promising strategy. In this study, we evaluated the effect of several kinase inhibitors either alone or in combination with CDDP or SAHA. Incubating HB cells with Wee1-inhibitor or the Met-inhibitor SU11274 showed none or only moderate reduction of cell viability, whereas the Aurora kinase inhibitor VX-680 proved to be the most promising substance tested. SU11274 represents a specific inhibitor of the receptor tyrosine kinase Met. The kinase activity is usually triggered by binding of hepatocyte growth factor (HGF), which leads to an autophosphorylation of the kinase and thus in turn triggers various cellular events, e.g., cell growth and motility, angiogenesis, and tissue regeneration. Met is highly overexpressed in a number of solid tumors, including hepatocellular carcinoma, and there is evidence that overexpression correlates with an aggressive phenotype and poor prognosis. For SU11274, a moderate effect on cell viability was observed even though the highest concentrations were about tenfold of what is regularly used for in vitro assays. This result may be due to low levels of phosphorylated Met in HB cells since there is evidence that the effectiveness of SU11274 also depends on the phosphorylation status of Met. The inhibitor may also counteract cell proliferation when exogenous HGF is added; however, in cultures with serum stimulation, proliferation of HB cells seems not to be driven by Met. Taken together, these results suggest a less-pronounced dependency of HB on HGF and Met due to the unchanged expression of Met in HB primary tissues when compared with normal liver tissue as gene expression analysis revealed. The observed weak effectiveness of Wee1 inhibition may be due to the low expression levels of Wee1 kinase in HB cell lines as emphasized by gene expression analysis. Consequently, the action of this substance may rely on an elevated expression of Wee1 thus leading to an abrogation of the G2/M checkpoint followed by cell death. This is also consistent with data from other studies showing selective and correlating sensitivity of the Wee1 inhibitor towards other cell lines depending on the expression and gene copy number of Wee1 kinase. For instance, breast cancer and prostate carcinoma cell lines with weak Wee1 expression were not sensitive towards Wee1 inhibition. In contrast, other cell lines which showed elevated expression levels of this kinase were sensitive. The expression analysis of Aurora kinase A, which shows a significant upregulation in patients who have died from HB, suggests that targeting this enzyme with VX-680 may be a promising tool in the treatment of HB. VX-680 exerts its action mainly through inhibition of histone H3 phosphorylation, thus in turn leading to inhibition of cell proliferation and apoptosis. Comparing the two cell lines, reduction in cell viability and thus a lower IC50 was found for HUH6 cells. This is consistent with previous reports that also showed a greater sensitivity of HUH6 towards commonly used drugs. VX-680 also exhibited apoptosis as confirmed by Caspase-3 assay. This is presumably due to subsequent cell arrest in a pseudo G1-state with 4N DNA content or endoreduplication in that the cell exits mitosis and proceeds through S-phase in the absence of cell division, resulting in a DNA content of more than 4N. Consequently, this leads to apoptosis due to proliferation in the presence of aberrant mitosis and failed cytokinesis. Combined treatment with either CDDP or the histone deacetylase inhibitor SAHA did not lead to synergistic effects in both cell lines. For medulloblastoma cell lines, it is reported that the synergistic effect between CDDP and an Aurora kinase A inhibitor depends on the chronological order of these drugs. If given the chemotherapeutic agent prior to Aurora kinase inhibition, the agents were antagonistic. Cycling cells are needed for both drugs; however, an arrest in G2 by VX-680 may not contribute to the activity of CDDP whose mode of action results in a transient S-phase and subsequent G2/M arrest for as long as 96 hours. Hence, there is a stop in cell progression before VX-680 can carry out its effect. This in turn implies that simultaneous application may have no synergistic effect as observed for both HUH6 and HepT1 cells. However, CDDP is not the only agent administered to patients with high-risk HB; for example, Doxorubicin and other drugs are also frequently used in the therapy regimen PLADO. Therefore, future studies need to reveal an additional and/or more effective combination with VX-680. No or little synergism was seen for the combination with SAHA. The idea of synergistic coupling through enhanced histone H3 phosphorylation driven by increased acetylation caused by SAHA was proven to be less effective. Even though there was a shift to lower concentrations of VX-680 observed in HUH6, HepT1 cells failed to respond to the combined treatment. There is evidence that treatment with SAHA also results in a decrease of Aurora A and B thus reinforcing the effect of VX-680. Western blot analysis also confirmed that increase as well as the combination with VX-680 compared with exclusive VX-680 treatment. This gives rise to the assumption that the increase in nuclei and cell diameter for the respective treatments may rely on different mechanisms as well. While the Aurora kinase inhibitor VX-680 is accompanied by endoreduplication causing a DNA content equal to or higher than 4N thus in turn causing an increase of the nuclei and cell diameter, HDAC inhibitors lead to histone acetylation thereby increasing transcriptional activity as well as aberrant cytokinesis with binuclear cells. Taken together, the combination of VX-680 and SAHA may have indeed a synergistic effect leading to a further increase in nuclei and cell diameter due to the interplay of both mechanisms described, but these two substances may interfere with each other on the level of Aurora kinase inhibition. It is also reported that the combination of VX-680 and SAHA leads to a greater induction and upregulation of the proapoptotic protein Bim, thus sensitizing the cells to a greater degree for apoptosis. Hence, upregulation of proapoptotic proteins and failed cytokinesis together with an aberrant DNA content increase the probability to undergo apoptosis. This is consistent with the results in the HB cell line HUH6. The lack of additional effects observed for the combined treatment in HepT1 cells could be due to the fact that these cells exhibit an even slower doubling time, that is a slower cell cycle progression. However, the effects of both substances depend on cycling cells suggesting that with a decreasing rate of growth a decrease in the efficiency of the agents is accompanied. A possible therapeutic potential of VX-680 requires high phosphorylation of histone H3 proteins in primary HB tumors. Since few HB cells are in the G2 phase and the observed histone H3 phosphorylation is even low, a long-term administration of VX-680 can be emphasized. However, one should also consider off-target as well as side effects of the Aurora kinase inhibitor which add up to successfully combat cancer. For instance, VX-680 has been shown to target the FLT3 kinase in primary acute myelogenous leukemia cells thereby ablating colony formation. It also has an anti-proliferative effect on tumor-associated endothelial cells in clear cell renal cell carcinoma. Hence, VX-680 is still under intensive investigation and all targets and effects have presumably not yet been identified. 3D cultures, also known as spheroids, have been previously shown to be a reliable model for investigating substances with respect to their ability to overcome drug resistance. A weak synergistic effect of VX-680 and SAHA could be observed between the two treatment conditions. This might be due to an existing drug concentration gradient, that is, a decrease of the amount of drugs along the spheroid with the lowest concentration representative in the core. Hence, cells that are farther away from the surface are less vulnerable to the substances applied and this might explain the high cell viability observed. Additionally, the slow doubling time of HB cells is to be considered, especially for HepT1 cells. The effects of these substances rely on cycling cells, causing a stop in cell progression and thus leading to apoptosis. Hence, slow cycling cells may exert this effect with even longer exposure times. Taken together, we evaluated for HB several cell cycle-modulating kinases which are currently under intensive investigation. So far, inhibition of Aurora kinase A seems promising as an additive Tozasertib or an alternative in the treatment of selected HB with increased kinase activity.