WO2018083483A1

WO2018083483A1 - Therapy

Info

Publication number: WO2018083483A1
Application number: PCT/GB2017/053309
Authority: WO
Inventors: Linda O'FLAHERTY; Jeremy Myles TAVARÉ; Michael Julian Seckl; Olivier Emmanuel Rafi Pardo
Original assignee: The University Of Bristol; Imperial Innovations Limited; Cancer Research Technology Limited
Priority date: 2016-11-02
Filing date: 2017-11-02
Publication date: 2018-05-11
Also published as: GB201618477D0

Abstract

The present invention provides a GSK3 inhibitor for use in treating cancer, wherein the GSK3 inhibitor is to be administered in combination with a chemotherapy agent, methods of treatment comprising the same and methods for identifying cancers sensitive to such treatment.

Description

Therapy

Field of Invention

The present invention relates to a novel therapy for the treatment of cancer and methods for identifying cancers sensitive to such therapy.

Background to the Invention

Since its discovery in the 1980's, glycogen synthase kinase 3 (GSK3) has been shown capable of regulating numerous protein substrates and transcriptional factors thus involving the enzyme in a plethora of cellular processes, including cell proliferation, differentiation and apoptosis. Not surprisingly, the multi-functionality of GSK3 has resulted in the kinase's implication in a diverse range of disease states, ranging from schizophrenia to cancer. The role of GSK3 in cancer remains to be fully understood and is somewhat paradoxical, with some evidence advocating it to be a tumour suppressor while others suggest it to be a tumour promoter.

Non-small cell lung cancer (NSCLC) comprises 80% of lung cancer diagnoses and despite advances in detection and therapies, the five year survival rate remains unacceptably low (-5-15%). However, with an estimated 50% of genetic aberrations found in NSCLC believed to be targetable, there is a renewed energy and lot of focus on developing and improving targeted therapies. NSCLC patients harbouring activating EGFR mutations are usually stratified for treatment with tyrosine kinase inhibitors (TKIs), which include erlotinib, gefitinib and afatinib. Despite good initial responses to TKI treatment, patients often progress within the first 2 years due to the emergence of secondary EFGR mutations, the most common of which is T790M. KRAS mutations also render patients resistant to Crizotinib, a TKI used for treating NSCLC patients with ALK rearrangements. In order to keep one step ahead of emerging resistance, the identification of new therapeutic targets (biomarkers) and/or consideration of new drug combinations which are capable of 'hitting' more than one target simultaneously, is of high priority. Summary of the Invention

The present inventors have demonstrated a novel combination of a GSK3 inhibitor and a chemotherapy agent which significantly and synergistically inhibits tumour cell growth in vitro and in vivo. Accordingly, in a first aspect the present invention provides a GSK3 inhibitor for use in treating cancer, wherein the GSK3 inhibitor is to be administered in combination with a chemotherapy agent. Methods of treating cancer comprising administering a GSK3 inhibitor and a chemotherapy agent to a patient are also provided.

In a further aspect the present invention provides methods for identifying cancers sensitive to treatment with a combination of a GSK3 inhibitor and a chemotherapy agent.

Description

GSK3 inhibitors refer to compounds that bind to GSK3 and modulate its activity such that net enzyme activity is decreased, i.e. the compounds have the same net effect as an inhibitor. The GSK-3 inhibitors may bind reversibly or irreversibly to the enzyme and may be ATP competitive inhibitors or non-ATP competitive inhibitors. In more detail, the GSK3 inhibitor may block the GSK-3 substrate binding site (non-ATP competitive) or the ATP binding pocket (ATP-competitive). Preferably the GSK3 inhibitor acts on either the GSK3a isoform or the GSK3P isoform, or both. The GSK3 inhibitor may have IC50 values of 50 nM or less, or 20 nM or less, or 10 nM or less for GSK-3 β and/or GSK-3 a.

Suitable GSK3 inhibitors for use in the present invention may include metal cations, such as beryllium, copper, lithium, mercury or tungsten; thiadiazolidindiones, such as tideglusib or TDZD-8; indirubins; CHIR99021; CHIR98014; SB216763; SB415286; AZD-1080; AZD-2858; AR-A014418; and LY2090314. In preferred embodiments of the invention the GSK3 inhibitor is CHIR99021. Other GSK3 inhibitors that may be suitable for use in the present invention are described in Pandey and DeGrado 2016 (incorporated herein by reference) and include pyrazolopyrimidines, benzimidazoles, pyridinones, pyrimidines, indolylmaleimide, imidazopyridines, oxadiazoles, pyrazines and 5-imino-l,2,4-Thiadiazoles.

GSK3 is known to act on the mitotic apparatus, playing a role in microtubule organisation and chromatin segregation. Without being bound by theory, the present inventors believe that GSK3 inhibitors act to destabilise mitotic microtubules. When administered in combination with a chemotherapy agent the effect of the chemotherapy agent can be enhanced, providing a treatment that may be more effective than a conventional monotherapy. Such enhancement may be particularly noticeable when the chemotherapy agents acts via a similar mechanism to the GSK3 inhibitor. Accordingly, in embodiments of the invention the chemotherapy agent may act to induce apoptosis by inhibiting formation or function of the mitotic apparatus, for example by destabilising microtubules. In embodiments of the invention, the chemotherapy agent may act to induce abnormalities in metaphase microtubule structures and/or to induce de-linearisation of chromatin on the metaphase plate. The combination therapy of the present invention may therefore result in dividing cells being trapped in the G1/G0 phase, thus prolonging the time spent at the cell cycle checkpoint and increasing the possibility of initiating an apoptotic response.

In embodiments of the invention the chemotherapy agent may be selected from one or more of an antimitotic agent, such as paclitaxel, docetaxel or nabpaclitaxel; a topisomerase inhibitor, such as irinotecan, topotecan, camptothecin and lamellarin D, etoposide (VP- 16), teniposide, doxorubicin, daunorubicin, mitoxantrone, amsacrine, ellipticines, aurintricarboxylic acid, or HU-331; a monoclonal antibody such as cetuximab, trastuzumab, nivolumab zalutumumab, nimotuzumab, matuzumab or pembrolizumab; an antimetabolite, such as folic acid, pyrimidine, purine or analogs thereof; and an EGFR kinase inhibitor, such as gefitinib, erlotinib, afatinib, crizotinib or lapatinib. Preferably the chemotherapy agent is paclitaxel.

In embodiments of the invention the chemotherapy agent is not a platinum-based chemotherapy agent, such as carboplatin. The GSK3 inhibitor and the chemotherapy agent may be administered simultaneously or sequentially. When administered simultaneously the GSK3 inhibitor and the chemotherapy agent may be administered as separate compositions or in a single composition. For example, the GSK3 inhibitor and chemotherapy agent may be supplied in an encapsulated formulation such as a liposome, a micelle or a nanoparticle in which either the GSK3 inhibitor or the chemotherapy agent or both are encapsulated.

When administered sequentially the GSK3 inhibitor may be administered before the chemotherapy agent. Alternatively the chemotherapy agent may be administered before the GSK3 inhibitor.

When the GSK3 inhibitor and the chemotherapy agent are administered sequentially the interval between administering each component is preferably not more than 12 hours, or 24 hours, or 48 hours. Preferably the interval between administering each component is 24 hours or less, or 12 hours or less.

The GSK3 inhibitor may be administered orally or parenterally. Parenteral administration may include intravenous, intramuscular or intraperitoneal administration. For example, the GSK3 inhibitor may be administered orally via tablets, capsules, powders, granules, drops, liquid medications or syrups.

The chemotherapy agent may be administered orally or parenterally. Parenteral administration may include intravenous, intramuscular or intraperitoneal administration. For example, the chemotherapy agent may be administered via injection or infusion.

When administered sequentially the GSK3 inhibitor and the chemotherapy agent may be delivered via the same route of administration or via different routes of administration. When administered simultaneously the GSK3 inhibitor and the chemotherapy agent are preferably administered via the same route of administration. In embodiments of the invention the cancer may be characterised by GSK3 dependency. For example, the cancer may be characterised by GSK3 hyperactivity. In other words, GSK3 activity may be higher in tumour samples than in normal tissue, said normal tissue may be patient-matched tissue and may be obtained from a region surrounding the tumour area. Methods for determining GSK3 activity will be familiar to the skilled person. For example, GSK3 activity may be measured by determining the activity of downstream substrate proteins that are phosphorylated by GSK3 (e.g. CRMP-2, glycogen synthase and/or F-kappaB) or by determining the extent of phosphorylation of the N-terminal serine residue on GSK3 (i.e. serine 21 on GSK3a and serine 9 on GSK3P). Additionally or alternatively GSK3 activity may be determined by immunohistochemistry, ELISA of biopsy lysates, western blotting of tissue lysates, by in vitro protein kinase assay or by proximity ligation assay.

The cancer may be one or more of Non-Small Cell Lung Carcinoma (NSCLC), pancreatic cancer, prostate cancer, colon cancer, leukaemia, ovarian cancer, liver cancer or glioblastoma. Preferably the cancer is NSCLC.

The present invention also provides a method for identifying a cancer sensitive to the treatment described herein, the method comprising detecting the activity of GSK3 in a sample of cancer cells obtained from a patient and comparing the activity of GSK3 in the sample to a threshold, wherein if the activity of GSK3 in the sample is higher than the threshold the cancer is sensitive to the treatment.

The patient is preferably a mammal, especially a primate. In one embodiment, the patient is a human.

The sample may be a biopsy obtained from a patient, such as a tumour biopsy. In embodiments of the invention the sample may comprise tumour cells, such as tumour cells obtained from or isolated from a blood sample obtained from a patient.

The activity of GSK3 may be detected by immunohistochemistry, ELISA of biopsy lysates or western blotting of tissue lysates. Additionally or alternatively, GSK3 activity may be determined by measuring the activity of downstream substrate proteins that are phosphorylated by GSK3 (e.g. CRMP-2, glycogen synthase and/or F-kappaB) or by measuring the extent of phosphorylation of the N-terminal serine residue on GSK3 (i.e. serine 21 on GSK3a and serine 9 on GSK3P).

For example, when GSK3 activity is determined by immunohistochemistry immunoreactivity of GSK3 in a tissue section may be scored using a refined 0-300 scoring system to drive a cutpoint analysis for a subsequent 1-3 or 1-10 grading system, where the maximum number represents intense staining of most or all of the tumor tissue and the lowest number is no staining whatsoever.

The activity of GSK3 in the sample may be compared with a threshold determined by a standard activity of GSK3 appropriate for the sample type and the subject from whom the sample is collected. Thresholds may be established by testing a large number of comparable samples from a number of healthy subjects. Comparable samples may be taken from subjects of the same species, gender and age group. Preferably, the threshold is determined by comparing GSK3 activity in the sample of cancer cells with patient-matched normal tissue, which may be obtained from region surrounding the tumour area.

The increase in GSK3 activity is preferably statistically significant. For example, the increase may be at least 5%, preferably at least 10%, more preferably at least 15%, more preferably at least 20%, more preferably at least 25%, more preferably at least 30%), more preferably at least 40%. In embodiments of the invention the increase may be about 50%.

The above method for identifying a cancer sensitive to the treatment of the invention may further comprising treating the cancer by methods described herein.

Brief Description of the Drawings The invention will now be described in detail, by way of example only, with reference to the figures.

Figure 1 shows that NSCLC cell lines treated with CHIR99021 and Paclitaxel, as mono-treatments or in combination, show varying degrees of sensitivity to drug- mediated attenuation of cell growth. Cell growth was measured in the following NSCLC cell lines: (i) H1975, (ii) H1299, (iii) Hccl93, (iv) Hcc95, (v) PC9 and (vi) Hcc827. Cells were treated for 96 hours with 2-fold increasing concentrations of CHIR99021 (μΜ), Paclitaxel (nM) or CHIR99021 with Paclitaxel at a constant ratio of 1 :800 paclitaxel to CHIR99021, respectively. In comparison to the effects of the drugs applied as mono-therapy, the combination of both drug induced a considerable reduction in NSCLC cell growth in HI 975 and cell lines. A moderate, but nonsignificant, amelioration of Heel 93 cell growth was observed for 1.6 μΜ CHIR99021 + 2 nM paclitaxel and 3.2 μΜ CHIR99021+ 4 nM paclitaxel. At higher concentrations of both drugs, a saturation of the dose response curves was observed in H1975 and Hccl93 cells. While Hcc95 and PC9 cells showed no response to CHIR99021 treatment, a reduction in cell growth occurred in a dose-dependent manner following Paclitaxel treatment. The growth trend for the combination treatment in Hcc95 and PC9 mimicked the paclitaxel response. No response to any of the three treatments was observed in the Hcc827 cells.

Figure 2 shows that targeted knockdown of GSK3 by shRNA corroborates the CHIR99021 phenotype. A(i) HI 975 cells stably transduced with shRNA against GSK3a, GSK3p and pooled anti-GSK3a and -GSK3p shRNA (GSK3a+GSKp) showed a 36.6% (*, p<0.05), 31.4% (ns) and 36.4% (p=0.05) decrease in cell growth, respectively. Error Mean ± St Dev, n=3. A(ii) Western blot confirmation of shRNA- mediated reduction in protein expression of the GSK-3a and GSK-3P isoforms in H1975 cells transduced with the GSK3a+GSKp pooled shRNA, compared to untransduced HI 975 cells and HI 975 cells transduced with a non-targeting control (NTC) shRNA. Confirming that the CHIR99021 and Paclitaxel co-treatment phenotype occurs in a GSK-3 dependent manner, treatment of GSK3a+GSK3p transduced cells with Paclitaxel reduces cell growth to a greater degree than in NTC transduced cells. HI 975 cells transduced with either NTC shRNA B(i) or GSK3a+GSK3p pooled shRNA B(ii) were treated with 2-fold increasing concentrations of Paclitaxel (nM) for 72 hours, following which cell colony growth was measured by crystal violet staining. GSK3 in combination with Paclitaxel treatment reduced NSCLC cell growth in vitro to a greater extent than the sum of the effects of the individual treatments B(iii). Error is Mean ± St Dev, n=3.

Figure 3 shows reduced NSCLC cell growth in vivo following combined treatment with CHIR99021 and Paclitaxel. Athymic nude mice harbouring H1975-derived tumours, were treated with CHIR99021 for 5 days every week over 3 weeks via peroral gavage (CHIR99021) or with a single administration of paclitaxel by IP injection at the start of the study (n=8 per group). A Reduction in tumour cell growth was observed with both concentrations of Paclitaxel, most notably at the 20mg/kg, while no effect was observed for either concentration of CHIR99021. However a further, more substantial, attenuation of cell growth was observed when 37.5 mg/kg CHIR99021 was combined with 10 mg/kg Paclitaxel. B (i) Tumour size as observed on day 20 (study end point). Vehicle-only control (i), combination treatment [37.5 mg/kg CHIR99021 + 10 mg/kg Paclitaxel] (ii), 20 mg/kg Paclitaxel (iii), 10 mg/kg Paclitaxel (iv), 75 mg/kg CHIR99021 (v) and 37.5 mg/kg CHIR99021 (vi). B (ii) Data are shown as scatter plot for end point values (n=4-7 per group) with the line in the middle representing median tumor size. Significant reductions in tumor growth were found between the following cohorts: drug combination vs. control group (p<0.0001), 37.5 mg/kg CHIR99021 vs. the combination group (p=0.002) and the 10 mg/kg Paclitaxel vs. the combination group (p=0.004). The line represents the median. C In vivo inhibition of GSK-3 activity by CHIR99021 was achieved at 75 mg/kg but not with 37.5 mg/kg CHIR99021. This was confirmed by western blot analysis of the phosphorylation status of the GSK-3 substrate, CRMP2, in a representative panel of day 20 tumor tissues. Arrow indicates pCRMP2 band (i). ImageJ quantification showed a significant decrease (p<0.05) in pCRMP2 expression in 75 mg/kg CHIR99021 vs. control groups (levels normalized to ATP5B loading control) (ii). For all experiments error was represented as Mean ± SD. Figure 4 shows that combined treatment with CHIR99021 and Paclitaxel results in disorganisation of mitotic microtubule structures and chromosomal delinearization. A Subsequent to 72 hours cultured in the presence of 1 μΜ CHIR99021, 2 nM paclitaxel or both drugs in combination, HI 975 cells were fixed and stained for a chromosomal (γ-tubulin, red), microtubule structure (a-tubulin, green) and DNA (DAPI, blue) markers. In support of previous findings, both CHIR99021 (1 μΜ) and Paclitaxel (2nM) mono-treatments disrupted normal microtubule organisation and structure, although this phenotype is more striking in the latter. Moreover, HI 975 cells treated with both CHIR99021 and Paclitaxel in combination displayed more prominent abnormalities in DNA linearization and spindle morphology. In comparison to single- compound treatments, cells treated with both drugs had higher instances of mono- nucleation of the DNA, lengthening of the axial microtubules in the spindle complex and greater pole-pole distances. Images shown are representative of a minimum of 17 images, from 3 independent experiments, that were analysed. Scale bar represents 200 pixels. B Cells were treated for 72 hours as follows: control (i), 1 μΜ CHIR99021 + 2 nM Paclitaxel (ii), 2 nM Paclitaxel (iii) and 1 μΜ CHIR99021 (iv). Cells were stained with PI and apoptosis measured by flow cytometry. The apoptotic population is indicated by the blue peaks. Gl (left peak) and G2 (right peak) phases are in red with cellular debris shown in purple. Control cells (i) showed a large of cell in Gl compared to G2 with only a small proportion of cell undergoing apoptosis. In contrast to the reduced cell number described in Figure l(i), there was no observed increase in the apoptotic peak in CHIR99021 treated cells compared to the control (4.3% vs 4.0% apoptosis, respectively). Conversely a substantial increase in apoptosis was evident in paclitaxel treated cell (47.9% apoptosis) (iii). HI 975 cells exposed to the CHIR99021 and paclitaxel combination did not show as great an induction of apoptosis (44.6%) as was expected, based on the growth assay data of Figure l(i) and compared to paclitaxel alone. However, there was still a considerably higher proportion of apoptotic events when compared to CHIR99021 alone.

Examples

Methods and Materials Cell Culture

H1975, H1299, Hccl93, Hcc95, Hcc827 and PC9 cells were a kind gift from Prof. Michael Seckl (Imperial College London). All cells were grown in RPMI-1640 supplemented with 10% fetal bovine serum (FBS), 2 mM L-Glutamine and 100 U Penicillin/0.1 mg Streptomycin and maintained at 37°C in a humidified atmosphere with 5% C02. Cells were passaged approx. twice weekly, when a confluency of 80- 90% was reached. All cell culture reagents were purchased from Sigma.

Drugs and inhibitors

CHIR99021 was purchased from A Chemtek (USA) and Paclitaxel from Sigma. Drugs were dissolved in dimethyl sulfoxide (DMSO), aliquoted and stored at -20°C. For working concentrations, drugs were further diluted in DMSO and/or RPMI medium. The maximum volume of DMSO in cell culture was <0.1%.

Cell Growth Assays

Cells were passaged and seeded at 1000 cells/per well of a black coated, clear- bottomed 96-well plate (Greiner) and allowed to settle overnight. Drugs were diluted to working concentrations in a v-bottomed clear 96-well plate (Greiner). Medium from cell-containing plates was gently aspirated and replaced with 100 μΐ_^ of drug- containing medium. For the control, cells were exposed to medium containing DMSO only. Cells were grown in the presence of the drug(s) for 96 hours, under standard cell culture conditions. At 96 hours, cells were gently washed in PBS and fixed in 4% paraformaldehyde for 30 minutes. After two further PBS washes, ΙΟΟμΙ Crystal Violet solution (0.05%) Crystal Violet (Sigma) in 20% ethanol) was added to each well and incubated at room temperature for 30 minutes. Following 3 x PBS washes, the crystal violet stain was solubilized on plate shaker for 1 hour in ΙΟΟμΙ of 1%> sodium dodecyl sulphate (SDS) (Sigma) solution (in PBS). Absorbance was measured at 595nm on a Perkin Elmer Fusion a plate reader and raw data analysed with Microsoft Excel. shRNA knockdown of GSK3

SMARTvector lentiviral shRNAs against GSK3aand GSK3P, including a pooled non- targeting shRNA control, were purchased from Dharmacon (GE Lifesciences). shRNA transductions were carried out according to manufacturer's instructions. Briefly, HI 975 cells were seeded at 1,000 cells per well of a 96 well plate 24 hours prior to transduction. Cells were serum starved for 2 hours after which time shRNAs were added at an MOI of 20 in the presence of 8 μg/ml polybrene. For combined knockout of GSK3a and GSK3P isoforms, the respective shRNA sequences were pooled and added to the cells. All knockdowns were carried out in triplicate. After 72 hours transduction, stable shRNA cell lines were selected for in 2 μg/ml puromycin. After 96 hours, 10% v/v AlamarBlue™ (Thermo Fisher Scientific) was added to each well. Cell viability was taken as a measurement of fluorescence at 535nm excitation and 585nm emission on a Perkin Elmer Fusion a plate reader and data analysed with Microsoft Excel. Western blot analysis was done by aspirating AlamarBlue™- containing medium, washing cells with PBS and lysing cells in the well and pooling samples together. For generation of stable shRNA cell lines, H1975 cells were seeded at 10,000 cells per well of a 6-well plate and transduced 24 hours later at an MOI of 20 for 72 hours. After this time, cells were selected with 2 μg/ml puromycin, as described above. These cells were kept at a low passage number and used for paclitaxel treatment assays (as described previously).

Western Blotting

Cells were lysed and western blots run as previously described (Vincent et al, 2014). Briefly, cells were lysed in IX Lysis Buffer (Cell Signaling) and protein concentration determined with a Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific). Primary antibodies used were: pCRMP2 (Abeam), ATP5ase (Santa Cruz), a- Tubulin (Sigma) and total GSK3a/GSK3p (Cell Signaling). All primary antibodies were incubated overnight at 4°C. For standard western blot, HRP conjugated secondary anti-mouse (Jackson Laboratories) or anti-rabbit (Cell Signaling) antibodies were used. For fluorescent quantification of blots, anti-mouse and anti-rabbit secondary fluorophores were used. Blots were either developed by film or fluorescent signal measured on a Licor Odyssey machine.

Xenograft A maximum tolerated dose (MTD) study was carried out to determine the maximum and half maximum doses of CHIR99021 which were subsequently determined to be 75 mg/kg and 37.5 mg/kg respectively. Due to the relative insolubility of CHIR99021 in saline, the drug needed to be administered in a DMSO/Arachis Oil solution. For the xenografts, donor tumours were established in athymic nude mice by subcutaneous injection of HI 975 cells, followed by subcutaneous implantation of donor tumour fragments into the right flank of athymic nude mice. Treatment commenced as soon as the H1975-derived tumours reached a diameter of ~4x4mm³. The animals were separated into groups of 8 mice per one of the following six cohorts: vehicle control, 75 mg/kg CHIR99021, 37.5 mg/kg CHIR99021, 20 mg/kg paclitaxel, 10 mg/kg paclitaxel or 37.5 mg/kg CHIR99021 co-administered with 10 mg/kg paclitaxel. Multiple dose administration of the two concentrations of CHIR99021 was given via peroral gavage (po) 2x daily every 5 days over 20 days. A single administration of paclitaxel was delivered by intraperitoneal (IP) injection at commencement of the study (day 1). Animal observations and tumour measurements were taken every 5 days over the 20 day treatment period. Due to the short half-life of CHIR99021 in vivo (~2 hours), animals were culled 90 minutes after administration of the final CHIR99021 concentrations; images were taken and tumour tissues snap frozen.

Immunofluorescence

HI 975 cells were seeded at 1000 cells/ well of an 8 well Nunc® Lab-Tek® Chamber slide™ (Sigma), allowed to settle overnight before being treated with 2 μΜ CHIR99021, 1 nM paclitaxel or 2 μΜ CHIR99021 in combination with 1 nM paclitaxel for 72 hours. Cell medium was then aspirated, cells washed in PBS and fixed with ice-cold methanol for 5 minutes at -20°C. Following aspiration of the methanol, cells were brought to room temperature in PBS before permeabilization in 0.1% Triton X100 (in PBS). A blocking step in 0.1% Triton X100 + 5% bovine serum albumin (BSA) (Sigma) preceded incubation with primary antibodies at room temperature. Primary antibodies used were rabbit anti-y-Tubulin (Sigma) and DM1 A mouse anti-a- Tubulin (Sigma). Cells were then washed prior to incubation with secondary antibodies (in block): goat anti-mouse Alexa Fluor® 480 and goat anti- rabbit Alexa Fluor® 594 (both Thermo Fisher Scientific). After final washes in PBS, the plastic chamber was carefully removed and the slide mounted with DAPI- containing VECTASHIELD® Mounting Medium (Vectorlabs). Images were taken at 63X magnification on a Leica SP5 confocal microscope.

Flow Cytometry

HI 975 cells were grown overnight in 10 cm dishes until -80% confluency was reached. Cells were then treated with 2 μΜ CHIR99021, 1 nM paclitaxel or 2 μΜ CHIR99021 in combination with 1 nM paclitaxel for 72 hours. Cells were then harvested and 0.5-1x106 cells resuspended in PBS, spun at 1200 rpm for 5 minutes. Supernatant was discarded and 1 ml cold 70% ethanol added drop wise to the cell pellet while gently vortexing. Cells were fixed for 30 minutes at 4°C and spun for 5 minutes at 2000 rpm. Supernatant was removed, the pellet was resuspended in 1 ml PBS and spun at 2000 rpm for 5 minutes. Following a further two wash steps, 100 μΐ ribonuclease (100 μg/ml Dnase free, Sigma) was added to remove RNA and left at room temperature for 5 minutes. Finally, 400μ1 propidium iodide (50μg/ml in PBS) was added and stored at 4°C, protected from light, until ready to analyse. Samples were run on a BD Influx fluorescence activated cell sorter (FACS ) machine and data analysed with FACSDiva™ software (BD Biosciences).

Statistical Analysis

Statistical analyses were performed using GraphPad Prism. For in vitro viability assays, Two-Way ANOVA with a Tukey's post-hoc multiple comparison test performed. Standard two-tailed student t-tests were performed for analysis of the xenograft end point data. Unless otherwise stated, all experiments were carried out as a minimum of n=3. Error is represented as Mean ± Standard Deviation (St Dev).

Results

Further to the findings reported by Vincent et al, we aimed to address whether inhibition of GSK3 in combination with another chemotherapeutic agent is capable of having an additive, negative impact on tumour cell growth. We approached this by investigating the effect of GSK3 inhibition, using a highly specific GSK3 inhibitor, CHIR99021, in combination with paclitaxel in a small panel of NSCLC cell lines. Similarly to what is observed between patients, we saw a heterogeneous inter-cell line response to the combined treatments. HI 975 cells showed the most sensitive response, not only to CHIR99021 (as previously published in (Vincent, 2014)) and paclitaxel alone but also when the two drugs are combined. From as low as 2 nM paclitaxel combined with 1.6 μΜ CHTR99021 we see a significantly (p<0.0001) greater reduction in cell growth compared to the control or either compound administered as a mono-treatment (Figure 1).

To determine whether the additive effect we observe by combining CHR99021 with paclitaxel is attributed to the inhibition of GSK3 rather than an mere off target effect, we used an shRNA approach to stably knockdown the GSK3a and GSK3P isoforms, both individually and together, in HI 975 cells (Figure 2A(i)). Compared to the non- targeting shRNA control (NTC) cells, knockdown of GSK3a (p<0.05), GSK3p or GSK3a+GSK3p ameliorated cell growth. Knockdown of the isoforms was confirmed by western blot (Figure 2A(ii)). We then measured cell growth 96 hours after exposure to increasing doses paclitaxel (0.5- 4 nM). Comparison of the control cells to the combined GSK3a and GSK3P knockdown cells showed a greater reduction in growth in the cell line depleted of both GSK3 isoforms (Figure 2B(i)-(iii)). We showed this using the following equation, A+B<C. Whereby A is the growth difference between the control and GSK3 knockdown cells at baseline, B is the response of the control to paclitaxel and C is the response of the GSK knockdown cells to paclitaxel. Despite not being a significant difference, by showing that the value of C is greater than the combined values of A and B, we demonstrated that the loss of GSK3 expression mimics the effects of CHIR99021, when exposed to paclitaxel, and thus that the combined effect of the two drugs occurs in a GSK3- dependent manner.

To investigate whether our in vitro findings would be corroborated in an in vivo system, we selected a representative cell line, HI 975 to establish a xenograft model as this cell line gave the best response to CHIR99021, both alone and in combination with paclitaxel. Treatment commenced as soon as H1975-derived tumours reached a diameter of ~4x4mm³ and no adverse effect on animal weight was observed throughout treatment period (data not shown). While each cohort consisted of 8 mice at baseline, the final numbers per group at the end point varied between groups. Where no response to treatment occurred, animals were culled as a result of tumour size reaching the ethical limit.

In contrast to our in vitro findings, we observed no notable reduction in tumour volume from either dose of CHIR99021, administered as mono-therapy (Figure 3 A). This discrepancy may be accounted for by differences in the complex regulation of tumour growth in three-dimensions in the animal model as compared to two- dimensions in cell culture. Reduced tumour growth in both mono-therapy paclitaxel groups was clearly noticeably and occurred in a dose-dependent manner. Interestingly, in the cohort treated with the combined therapy of 37.5mg/kg CHIR99021 and lOmg/kg paclitaxel (concentrations which were determined to be at half-maximally tolerated dose of each individual compound) we see a substantially greater loss in tumour size compared to the 20mg/kg paclitaxel group (which represents a maximally tolerated dose), which gave the best response of the mono-therapy groups. When we compared tumour volume across the six groups at Day 20 (end point), we saw a significant difference between the control, 37.5mg/kg CHIR99021 and lOmg/kg paclitaxel to the combination group (Figure 3B). While no significant difference was found between the 20mg/kg paclitaxel and the combination group (data not shown), we found these results to be promising, as the overall effect on tumour growth with the two drugs at half their maximum tolerated doses was still greater than the maximum dose of paclitaxel, although not significantly so (data not shown).

To confirm if CHIR99021 was effectively inactivating GSK3 within the tumour tissues, we ran a western blot analysis of the phosphorylated GSK3 substrate, pCRMP2, as a measure of in vivo GSK3 activity. A representative blot from four animals within the three groups is shown in Figure 3C(i). Compared to the control animals, a significant loss of pCRMP2 phosphorylation is apparent in animals treated with 75mg/kg CHIR99021 (Figure 3C(ii)). Paclitaxel induces apoptosis via destabilization of microtubule structures and thus disrupting chromatin segregation. As GSK3 is also reported to play a role in microtubule organization and chromatin segregation, we aimed to investigate whether inhibition of GSK3 via CHIR99021 is capable of enhancing the known effect of paclitaxel. To assess whether CHIR99021- and paclitaxel-induced abnormalities in metaphase microtubule structures may contribute to apoptosis, we used immunofluorescent (IF) staining of γ-tubulin and a-tubulin, which are spindle pole and microtubule markers, respectively. HI 975 cells were treated with 1 μΜ CHIR99021, 2 nM Paclitaxel and 1 μΜ CHIR99021 in combination with 2 nM Paclitaxel for 72 hours (n=3). Qualitative analysis of the resulting images (-17-22 images per treatment group) showed distinct microtubule abnormalities and de- linearization of chromatin on the metaphase plate in a high proportion of cells treated with the CHTR99021 and paclitaxel combination, when compared to control and CHIR99021 treated cells (Figure 4A). Microtubule destabilization is also evident in cells treated with paclitaxel-only, however the phenotype is clearly enhanced when CHIR99021 is added in tandem with paclitaxel. Furthermore, in the three independent experiments analysed, instances of spindle tripolarity and monopolarity were also found in a proportion of HI 975 cells subject to the drug combination but were not observed in any of the control images (Data not shown). A single instance of multipolarity was noted in the 2 nM paclitaxel group and one tripolar cell in the 1 μΜ CHIR99021 group.

In light of these findings, one possible explanation for reduced cell growth in the presence of the CHIR99021 and paclitaxel treatments was that destabilization of the spindle complex during metaphase, resulting from this drug combination, trapped dividing cells in the G0/G1 phase, thus prolonging the time spent at cell cycle checkpoint and increasing the possibility of initiating an apoptotic response. To address this, we used flow cytometry to analyse cell cycle status in HI 975 cells, treated under the same conditions as outline for Figure 4A. These cells were stained with propidium iodide (PI), a fluorescent molecule that is impermeable to intact cell membranes of viable cells, to determine whether cells treated with the CHIR99021 and paclitaxel combination accumulate in the pre-GO/Gl phase using flow cytometry (Figure 4B). Compared to the control and 1 μΜ CHIR99021 conditions, a higher degree of apoptosis was found in the cells treated with the two compounds in combination (Figure 4B(ii)). However, a similar apoptotic induction to the combination treatment was noted in the 2 nM paclitaxel samples (Figure 4B(iii)).

Discussion

In this study we show a promising additive response in reducing tumour growth in vivo using a combination of the GSK3 inhibitor with Paclitaxel.

Despite -50% of NSCLC patients presenting with targetable genetic aberrations, emergence of resistance to target therapeutics, usually within the first year or two of treatment, is a concerning set-back in personalized medicine. In light of this, new approaches and drug combinations need to be considered in order to have the biggest impact on reducing tumour progression, ideally at the stage of first line treatment. Due to its duplicitous role as both a tumour promoter and suppressor in different cancer types, which would need careful consideration when treating metastases at different organ sites, there may be concerns targeting GSK3 for inhibition as part of a cancer regime. However, all cancer treatments, with the exception of surgical intervention, carry a certain risk of promoting tumour progression, especially radiation therapy. Of course, these risk are often small and are monitored carefully by medical professionals so that the potential benefits of such treatments in prolonging overall survival often outweigh the risks.

Our study offers a further insight into the potential for GSK3 inhibition to be incorporated into therapeutic regimens. To the best of our knowledge, we show a novel combination of the GSK3 inhibitor, CHIR99021, and paclitaxel which administered together in HI 975 NSCLC cells significantly and synergistically attenuate cell growth, both in vitro and in vivo. Preliminary investigations into the mechanism behind this synergistic effect suggest a link between a paclitaxel- dependent destabilization of mitotic microtubules and the role GSK3 in controlling microtubule dynamics. It may be that CHIR99021 -mediated inhibition of GSK3 further enhances the paclitaxel phenotype. Patient-derived tumour explants are an increasingly popular model for analysing not only molecular mechanisms of disease but also for testing the efficacy of novel therapeutic compounds. Cell lines derived from these explants have the potential to give a better insight into intra-tumour heterogeneity and thus it would be interesting to further address the utility of administering CHIR99021, in combination with a broader spectrum of clinically available therapeutics, in such a system. Inter- and intra-patient heterogeneity is a complex issue which needs urgent addressing for improving cancer treatment. We hypothesize that disruption to mitotic spindle structures during metaphase is likely be the mechanism of action for the additive effect of the CHIR99021 and paclitaxel combination on aberration of NSCLC cell growth. Furthermore, we believe from our initial findings that this drug combination has promising potential for further development as a novel therapeutic adjuvant.

References

Pandey MK, and DeGrado, TR (2016). Glycogen Synthase Kinase-3 (GSK-3)- Targeted Therapy and Imaging. Theranostics 6 (4):571-583

Vincent, E.E., et al. (2014). Glycogen Synthase 3 protein kinase activity is frequently elevated in human non-small cell lung carcinoma and supports tumour cell proliferation. PLoS One. 2014; 9(12): el 14725

Claims

1. A GSK3 inhibitor for use in treating cancer, wherein the GSK3 inhibitor is to be administered in combination with a chemotherapy agent.

2. A GSK3 inhibitor for use according to claim 1, wherein the GSK3 inhibitor is selected from one or more of CHIR99021, CHIR98014, SB216763, SB415286, AZD- 1080, AZD-2858, AR-A014418, LY2090314, Tideglusib, TDZD-8, indirubin, and lithium.

3. A GSK3 inhibitor for use according to claim 1 or 2, wherein the chemotherapy agent is selected from one or more of an antimitotic agent, a topisomerase inhibitor, a monoclonal antibody, an antimetabolite and an EGFR kinase inhibitor.

4. A GSK3 inhibitor for use according to any of claims 1 to 3, wherein the chemotherapy agent is not a platinum-based chemotherapy agent.

5. A GSK3 inhibitor for use according to any of claims 1 to 4, wherein the GSK3 inhibitor and the chemotherapy agent are to be administered simultaneously or sequentially.

6. A GSK3 inhibitor for use according to claim 5, wherein the GSK3 inhibitor and the chemotherapy agent are to be administered sequentially at an interval of not more than 24 hours.

7. A GSK3 inhibitor for use according to any of claims 1 to 6, wherein the GSK3 inhibitor is to be administered orally, intravenously, intraperitoneally, or

intramuscularly.

8. A GSK3 inhibitor for use according to any of claims 1 to 7, wherein the chemotherapy agent is to be administered orally, intravenously, intraperitoneally, or intramuscularly

9. A GSK3 inhibitor for use according to any of claims 1 to 8, wherein the cancer is characterised by GSK3 hyperactivity.

10. A GSK3 inhibitor for use according to any of claims 1 to 9, wherein the cancer is one or more of Non-Small Cell Lung Carcinoma (NSCLC), pancreatic cancer, prostate cancer, colon cancer, leukaemia, ovarian cancer, liver cancer or glioblastoma.

11. A method for treating cancer, the method comprising administering a GSK3 inhibitor and a chemotherapy agent to a patient.

12. The method of claim 11, wherein the GSK3 inhibitor is selected from one or more of CHIR99021, CHIR98014, SB216763, SB415286, AZD-1080, AZD-2858, AR-A014418, LY2090314, Tideglusib, TDZD-8, indirubin, and lithium.

13. The method of claim 11 or 12, wherein the chemotherapy agent is selected from one or more of an antimitotic agent, a topisomerase inhibitor, a monoclonal antibody, an antimetabolite and an EGFR kinase inhibitor.

14. The method of any of claims 11 to 13, wherein the chemotherapy agent is not a platinum-based chemotherapy agent.

15. The method of any of claims 11 to 14, wherein the GSK3 inhibitor and the chemotherapy agent are to be administered simultaneously or sequentially.

16. The method of claim 15, wherein the GSK3 inhibitor and the chemotherapy agent are to be administered sequentially at an interval of not more than 24 hours.

17. The method of any of claims 11 to 16, wherein the GSK3 inhibitor is to be administered via orally, intravenously, intraperitoneally, or intramuscularly.

18. The method of any of claims 11 to 17, wherein the chemotherapy agent is to be administered orally, intravenously, intraperitoneally, or intramuscularly.

19. The method of any of claims 11 to 18, wherein the cancer is characterised by GSK3 hyperactivity.

20. The method of any of claims 11 to 19, wherein the cancer is one or more of Non-Small Cell Lung Carcinoma (NSCLC), pancreatic cancer, prostate cancer, colon cancer, leukaemia, ovarian cancer, liver cancer or glioblastoma.

21. A method for identifying a cancer sensitive to the treatment of any of claims 11 to 20, the method comprising:

detecting the activity of GSK3 in a sample of cancer cells obtained from a patient; and

comparing the activity of GSK3 in the sample to a threshold;

wherein if the activity of GSK3 in the sample is higher than the threshold the cancer is sensitive to the treatment.

22. The method of claim 21, further comprising treating the cancer according to any of claims 11 to 20.