ALK INHIBITORS FOR TREATMENT OF ALK-NEGATIVE CANCER AND PLASMA CELL-MEDIATED DISEASES

Abstract

The present invention provides a method of treating ALK-negative/LTK-positive cancer in a subject, comprising administering to the subject a pharmaceutically-effective dose of a linear inhibitor of ALK. The invention is of particular utility in treating multiple myeloma, including proteasome inhibitor-resistant multiple myeloma.

Claims

1. A linear inhibitor of anaplastic lymphoma kinase (ALK) for use in the treatment of cancer or a plasma cell-mediated disease in a subject, wherein said cancer or plasma cell-mediated disease is characterised by cells which are ALK-negative and leukocyte tyrosine kinase (LTK)-positive.

2. The inhibitor of ALK for use according to claim 1, wherein said use is the treatment of cancer.

3. The inhibitor of ALK for use according to claim 2, wherein said cancer is multiple myeloma.

4. The inhibitor of ALK for use according to claim 2, wherein said cancer is chronic lymphocytic leukaemia or hepatic cancer.

5. The inhibitor of ALK for use according to any one of claims 2 to 4, wherein said cancer is resistant to treatment with a proteasome inhibitor.

6. The inhibitor of ALK for use according to claim 5, wherein said proteasome inhibitor is bortezomib.

7. The inhibitor of ALK for use according to claim 1, wherein said plasma cell-mediated disease is an autoimmune disease.

8. The inhibitor of ALK for use according to claim 7, wherein said autoimmune disease is lupus or immune thrombocytopenia.

9. The inhibitor of ALK for use according to claim 7, wherein said plasma cell-mediated disease is graft-versus-host disease.

10. The inhibitor of ALK for use according to any one of claims 1 to 9, wherein said inhibitor has a structure as set forth in Formula XI ##STR00008## wherein: R.sub.1 is CH.sub.3 or iPr; R.sub.2 is H or CH.sub.3; R.sub.3 is ##STR00009## and R.sub.4 is ##STR00010##

11. The inhibitor of ALK for use according to claim 10, wherein said inhibitor is ceritinib.

12. The inhibitor of ALK for use according to claim 11, wherein said treatment comprises daily administration of a dose comprising 450 mg ceritinib, and said dose is administered orally.

13. The inhibitor of ALK for use according to claim 11, wherein said treatment comprises daily administration of a dose comprising 200 mg to 400 mg ceritinib, and said dose is administered orally.

14. The inhibitor of ALK for use according to claim 11, wherein said treatment is for multiple myeloma, and wherein said treatment comprises daily administration of a dose comprising 400 mg to 500 mg ceritinib, and said dose is administered orally.

15. The inhibitor of ALK for use according to claim 10, wherein said inhibitor is brigatinib.

16. The inhibitor of ALK for use according to any one of claims 1 to 9, wherein said inhibitor is selected from crizotinib, ensartinib, alectinib and entrectinib.

17. The inhibitor of ALK for use according to any one of claims 1 to 16, wherein said subject is a human.

18. A method of treating cancer or a plasma cell-mediated disease in a subject, wherein said cancer or plasma cell-mediated disease is characterised by cells which are ALK-negative and LTK-positive, said method comprising administering to said subject a pharmaceutically-effective dose of a linear inhibitor of ALK.

19. A method of diagnosing and treating an ALK-negative cancer in a subject, comprising: (i) diagnosing cancer in said subject; (ii) testing said cancer for expression of ALK and LTK; and (iii) when said cancer is found to be ALK-negative and LTK-positive, administering a pharmaceutically-effective dose of a linear inhibitor of ALK to said subject.

20. A method of diagnosing and treating an ALK-negative cancer in a subject, comprising: (i) diagnosing cancer in said subject; (ii) determining that said cancer is ALK-negative and LTK-positive; and administering to said subject a linear inhibitor of ALK in an amount effective to treat said cancer.

21. The method of any one of claims 17 to 19, wherein said cancer, plasma cell-mediated disease, inhibitor of ALK, subject and/or dose is as defined in any one of claims 2 to 17.

22. Use of a linear inhibitor of ALK in the manufacture of a medicament for treating cancer or a plasma cell-mediated disease in a subject, wherein said cancer or plasma cell-mediated disease is characterised by cells which are ALK-negative and LTK-positive.

23. The use of claim 22, wherein said cancer, plasma cell-mediated disease, inhibitor of ALK, treating or subject is as defined in any one of claims 2 to 17.

Description

FIGURE LEGENDS

[0126] FIG. 1: LTK knockdown results in ER stress.

[0127] The figure shows the effect of LTK knockdown on ER stress caused by protein hypersecretion. HeLa cells which encode inducibly expressible IgM (heavy and light chain) were transfected with control or LTK siRNA. After 72 h, cells were treated with mifepristone to induce IgM expression, followed by lysis (at the indicated times following induction) and immunoblotting against the indicated proteins.

[0128] FIG. 2: LTK allows cells to cope with high secretory load.

[0129] A, L363 cells were treated with 5 μM crizotinib and the expression of spliced vs non-spliced XBP1 and ATF4 was determined 4, 8 and 12 h post-treatment. Splicing of XBP1 and accumulation of ATF4 are signs of the induction of ER stress. B, Bortezomib-resistant L363aBTZ cells were treated with 5 μM crizotinib and the expression of spliced vs non-spliced XBP1 and ATF4 was determined at 4, 8 and 12 h post-treatment. C, HeLa cells expressing inducible IgM (heavy and light chain) were treated with mifepristone to induce IgM expression over a period of 2 weeks. Afterwards, cells were treated with 1 μM crizotinib for 24 h followed by lysis and immunoblotting against spliced XBP1 to measure ER stress. + and − indicate cells with or without mifepristone treatment, respectively. Induced cells are considered hypersecretory and non-induced cells have a normal level of secretion. Numbers above the blot show the quantification of densitometric measurements from three independent experiments ±SD. Note that hypersecretory cells exhibit a higher level of ER stress upon treatment with crizotinib. D, Comparison of LTK mRNA expression by quantitative PCR in HeLa cells that do not overexpress IgM versus cells with prolonged (14 days) IgM overexpression.

[0130] FIG. 3: ALK.sup.NegLTK.sup.Pos cells are susceptible to crizotinib treatment and undergo apoptosis.

[0131] The figure shows that inhibition of LTK in hypersecretory cells reduces their viability. In A, HeLa cells expressing inducible IgM (heavy and light chain) were treated with mifepristone to induce IgM expression over a period of 2 weeks. Afterwards, cells were treated with 1 μM crizotinib for 24 h followed by lysis and immunoblotting for caspase 3. The upper blot shows non-cleaved caspase 3 and the lower blot shows cleaved caspase 3. Cleavage of caspase 3 is indicative of apoptosis induction. + and − indicate cells with or without mifepristone treatment, respectively. B, Concentration-response curves for the effect of crizotinib treatment (24 h) on the viability of three different myeloma cell lines. Results are the means of three independent experiments. C, Expression levels of LTK and ALK mRNA in the indicated myeloma cell lines determined by qPCR.

[0132] FIG. 4: ALK.sup.NegLTK.sup.Pos cell lines are susceptible to ALK inhibitor treatment.

[0133] The figure shows concentration-response curves for the effects of ceritinib, alectinib, ensartinib, lorlatinib, entrectinib and crizotinib (24 hr treatment) on the viability of the myeloma cell lines L363 and AMO-1 and their bortezomib-resistant clones (L363-BTZ & AMO-1-BTZ). The Results are the means of 3 independent experiments. Note the concentration-dependent reduction of viability of myeloma cells.

[0134] FIG. 5: LTK regulates secretion; a linear ALK inhibitor reduces secretion.

[0135] The figure shows the effect of LTK inhibition by ceritinib on protein secretion. In A, HeLa cells expressing flag-tagged LTK were treated with 1 μM of crizotinib or ceritinib for 30 minutes. Afterwards, cells were lysed and immunoblotted against phosphorylated LTK, which is used as a surrogate for the active form of LTK. The lower blot shows the levels of total LTK in cells. B, HeLa cells were treated with 1 μM ceritinib for 30 minutes followed by fixation and staining for Sec31A to label ERES. Cells were imaged by confocal microscopy and ERES were counted using ImageJ. C, HeLa cells stably expressing mCherry-tagged mannosidase-II RUSH reporter. Cells were treated with biotin to release the reporter from the ER. After 20 min, a time point where most of the reporter has reached the Golgi, cells were fixed and immunostained for GM130 to label the Golgi apparatus. The images show representative examples of cells after 20 min of biotin addition, where crizotinib-treated cells exhibit less of the reporter in the Golgi region. The graph on the right shows the quantification with the ratio of reporter fluorescence (red) in the Golgi (green) area. D, IgA secreting AMO-1 cells were treated for 3 h with the indicated doses of crizotinib and the amount of IgA in cell supernatant was determined by ELISA.

[0136] FIG. 6: Multiple myeloma cells from patients are susceptible to crizotinib treatment.

[0137] The figure shows dose-response curves for the effect of crizotinib treatment (72 hr) on the viability of purified CD138.sup.+ multiple myeloma cells isolated from seven patients as indicated. Viability was measured by CellTiterGlo.

[0138] FIG. 7: LTK and ALK expression in multiple myeloma cells from myeloma patients.

[0139] A and B show Omics data derived from the CoMMpass study, IA-13 build. Data comprised 767 multiple myeloma samples (www.themmrf.org) from the Multiple Myeloma Research Foundation Personalized Medicine Initiatives (https://research.themmrf.org). RNA-seq data are represented as Fragments Per Kilobase of transcript per Million mapped reads, FPKM.

[0140] A. RNA-seq data from multiple myeloma cells of 767 patients in the CoMMpass study. Expression of LTK and secretory pathway genes such as ERGIC-53 (LMAN1), VIP36 (LMAN2), SURF4, BiP (HSP5A), the ALK gene and the negative control IL-2 are shown. IL-2 is not expressed in multiple myeloma cells (expression lies below the cut-off of FPKM=1, dotted line). Violin plots include medians (dashed line) and quartiles (dotted line).

[0141] B. Dotplot showing LTK vs. ALK expression of 767 patients. Dotted lines show expression cut-offs for LTK and ALK (FPKM=1).

[0142] FIG. 8: Xenografted human multiple myeloma cells in NSG mice are susceptible to crizotinib.

[0143] In this figure, immunocompromised NSG mice were intrafemorally injected with human L363-BTZ (bortezomib-resistant) myeloma cells. The cells were equipped with luciferase for in vivo monitoring of disease progression. After 10 days, mice were treated with bortezomib i.v. twice over a 7 day period and with crizotinib orally every day over the 7 day period. The images on the left show examples of tumour load in the different groups at days 4 and 7 of treatment (days 14 and 17 post tumour inoculation), and the graph on the right depicts the average luciferase signal for each mouse in each cohort (n=4) normalised to day 0 of the treatment (day 10 post tumour inoculation). Statistical significance between the groups was assessed with two-way ANOVA and between the time-points using one-way ANOVA with Tukey post-test.

[0144] FIG. 9: Expression of LTK and ALK in Chronic lymphocytic leukemia (CLL) cells.

[0145] Gene expression data from CLL cell samples of 107 patients were analysed for expression of LTK and ALK. 94 of 107 patients (88%) of CLL samples expressed LTK over the cut-off (dotted line), while none of the samples expressed ALK (right histogram), as shown.

[0146] FIG. 10: CLL cells from patients are susceptible to the linear ALK inhibitor crizotinib.

[0147] The Figure shows the response of activated proliferating CLL blasts in response to a linear ALK inhibitor.

[0148] Left: Dose response curves of CLL cells from 21 patients.

[0149] Middle: IC.sub.50 values are displayed in a violin plot that shows 17 samples that are sensitive to Crizotinib. These are indicated as super-sensitive (CLL cells from 3 patients), highly sensitive (3 patients), and sensitive (11 patients).

[0150] Right: Drug sensitivity scores (DSS, Yadav et al. Scientific Reports 4: 5193, 2014) integrate multiple dose-response relationships in high-throughput compound testing studies. DSS of CLL samples demonstrate that a third of the patients have CLL cells that are highly sensitive to crizotinib.

[0151] FIG. 11: CLL cells from IgVH.sup.Unmut patients are susceptible to crizotinib.

[0152] CLL samples from FIG. 10 were divided in terms of mutational status of immunoglobulin variable heavy chain (IgVH) genes.

[0153] Left: Four of six patient samples that have IgVH.sup.Unmut CLL are highly sensitive to crizotinib. None of the IgVH.sup.Unmut samples showed low sensitivity.

[0154] Right: Three out of 15 IgVH.sup.Mut CLL samples are highly sensitive to crizotinib. One of these barely qualifies as mutated (CLL168 that has 97.9% homology). Four of the 15 IgVH.sup.Mut samples showed low sensitivity.

[0155] FIG. 12: CLL cells from patients are susceptible to ALK inhibitor treatment CLL cells from 12 patients were tested for responses to ALK inhibitor treatment with brigatinib, ceritinib, ensartinib, entrectinib and lorlatinib. Drug sensitivity scores (DSS, Yadav et al. Scientific Reports 4: 5193, 2014) are shown. A cut-off line is set at DSS=10, dotted line.

[0156] FIG. 13: Expression of LTK and ALK in B-cell differentiation.

[0157] This figure analyses the expression of ALK and LTK in B-cell subsets in humans. In A, expression signals are shown for B-cell subsets: LTK expression (left plot) and ALK expression (right). LTK is expressed over cut-off in memory cells, plasmablasts and plasma cells. ALK is not expressed in the B-cell subsets.

[0158] B. RNA-seq data was downloaded from deposited RNA-seq and flow cytometry data for 29 immune cell types within the peripheral blood mononuclear cell (PBMC) fraction of healthy donors. Data showed naïve B-cells have low expression of LTK (18 of 105 were positive) and were negative for ALK. Memory B-cells and plasma cells were positive for LTK, and all negative for ALK. Memory B-cells and plasma cells were ALK.sup.NegLTK.sup.Pos.

[0159] FIG. 14: In vitro-generated plasma cells were sensitive to crizotinib.

[0160] CD19.sup.+ B-cells from 5 blood bank donors were stimulated by adherent murine L cells expressing CD40L, BAFF and April for 3 days to make B-cell blasts (plasma blasts/cells). plasma blasts/cells were tested for sensitivity to crizotinib for 72 h (CellTiterGlo).

[0161] Left: Relative percentage cell viability in drug-sensitivity profiles of 5 normal plasma blasts/cells treated with crizotinib.

[0162] Right: IC.sub.50 of 5 normal plasma blasts/cells treated with crizotinib.

[0163] FIG. 15: In vitro-generated plasma cells are sensitive to the linear ALK inhibitors crizotinib and ceritinib.

[0164] Negatively selected normal B-cells from blood bank donors were stimulated with sCD40L, IL-21, IL-4 for 5 days. This protocol allows generation of 10-20% end differentiated plasma cells that express IRF4 and BLIMP1. From day 5-7, cells were exposed to titred crizotinib in A, or ceritinib in B.

[0165] In A and B, IRF4 vs BLIMP1 in gated CD19.sup.+ B cells is shown. IRF4 and BLIMP1 double positive cells represent end stage plasma cells. Such plasma cells are gated (blue elliptic region) and percent plasma cells are shown. The conditions without inhibitor (0) and with linear Alk-inhibitor (crizotinib, A; ceritinib, B) are shown for concentrations 0.01 μM, 0.1 μM, 1 μM, 10 μM and 100 μM.

[0166] Panel C shows calculated percentage plasma cell viabilities in response to crizotinib and ceritinib doses as indicated.

[0167] FIG. 16: Plasma cells are sensitive to crizotinib in vivo and reduced the secretion of Ig in vitro.

[0168] Luciferase.sup.+ L363wt plasma cell lines were injected intrafemorally into immunocompromised NSG mice. After 10 days, mice were treated with crizotinib (50 mg/kg, orally, daily) every day for 7 days.

[0169] Left: Plasma cell line is killed by crizotinib in xenografted mice. Mice were imaged on an IVIS optical imager on days 10, 14 and 17. Ventral aspects of 3 (of 4) mice are shown. Mice received vehicle (top) or crizotinib (bottom). Crizotinib dramatically reduced the signals from plasma cells in 3 of 4 mice.

[0170] Right: Secretion of Ig is reduced by ceritinib. The plasma cell line AMO-1 was washed and added to wells (10.sup.6 cells/well) in the presence of titred ceritinib for 3 h. Supernatants were assayed for IgA secreted by the AMO-1 cells by use of the Human IgA ELISA Kit (Abcam, Cambridge, UK).

[0171] FIG. 17: Liver hepatocellular carcinoma are ALK.sup.NegLTK.sup.Pos and are sensitised to apoptosis by crizotinib.

[0172] RNA-seq data from 365 hepatocellular carcinomas (HCCs) were generated by The Cancer Genome Atlas (TCGA) and reported as FPKM (number of Fragments Per Kilobase of exon per Million reads). 365 patient samples from hepatocellular carcinomas were investigated for expression of LTK and ALK. In A, the left plot shows LTK vs ALK with cut-off as indicated by dotted line. All patients have samples that are ALK.sup.Neg. About a quarter of the samples are LTK.sup.Pos and thereby ALK.sup.NegLTK.sup.Pos.

[0173] In B, Hepatocellular carcinoma cells (HepG2) were treated with DMSO (untreated) or with increasing concentrations (1,2 and 5 μM) of doxorubicin for 24 h and apoptosis was assessed by immunoblotting for cleaved caspase 3. Note that 5 μM doxorubicin only weakly induces cell death. Cells were also treated with the same concentrations of doxorubicin in the presence of 1 μM crizotinib for 24 h. Note that the presence of crizotinib renders HepG2 cells more sensitive to the apoptosis inducing drug doxorubicin indicating that providing a linear ALK inhibitor sensitises HCC cells to apoptosis.

EXAMPLES

[0174] Myeloma Experiments, Materials and Methods

[0175] Patient Samples and Primary Multiple Myeloma Cell Sample Processing

[0176] Multiple myeloma patients were recruited from the Oslo Myeloma Centre at Oslo University Hospital. The study was approved by the regional Committee for Medical and Health Research Ethics of South-East Norway (REC#2016/947 and 2012/174); Bone marrow aspirates were obtained from multiple myeloma patients following signed informed consent in compliance with the Declaration of Helsinki.

[0177] Seven multiple myeloma patient samples were included. Patients were due for treatment after second relapse (patient number 1701), third relapse (2101, 2021, 1802, 9041, 2504) or fourth relapse (2201). All patients had received proteasome inhibitors (bortezomib or carfilzomib) in at least one line of treatment.

[0178] Bone marrow mononuclear cells (BMMCs) were prepared from patient bone marrow aspirates by Lymphoprep density gradient centrifugation. After CD8 depletion by Dynabeads (Life Technologies) BMMCs were subsequently stimulated by expanding Th cells in the presence of Human T-Activator CD3/CD28 Dynabeads (Life Technologies) and 100 U/ml human interleukin-2 (hIL-2, Roche, Mannheim, Germany). After 48 h, BMMCs were subjected to CD138+ enrichment to isolate multiple myeloma plasma cells using MACS CD138+ microbeads (Milteny Biotec, Bergisch Gladbach, Germany).

[0179] Drug Treatment and Cell Viability Assay (Multiple Myeloma Patient Cells)

[0180] CD138+ multiple myeloma cells (5,000-10,000 cells/well) from activation assays were tested in 384-well plates for their response to crizotinib. 6 crizotinib concentrations were tested, in 10-fold dilutions covering a 0.1-10,000 nM concentration range (plates dispensed by an Echo acoustic dispenser, LabCyte Inc., San Jose, Calif., USA).

[0181] The plate was incubated in a humidified environment at 37° C. and 5% CO.sub.2. Cell viability was determined after 72 hr using the CellTiterGlo (Promega, Madison, Wis., USA) ATP assay according to the manufacturer's instructions and using an Envision Xcite plate reader (Perkin Elmer, Shelton, Conn., USA). Relative percentage (%) cell viability was calculated by normalising luminescence units for each well to negative controls (DMSO 0.1%) and positive control wells (100 μM BzCl) and curve-fitting conducted to obtain IC.sub.50 values.

[0182] Cell Culture and Transfections

[0183] HeLa cells were cultured in DMEM (GIBCO) supplemented with 10% FCS and 1% Penicillin/Streptomycin (GIBCO). For overexpression of plasmids, cells were transfected using either Fugene 6 or with TranslT-LT1 (Mirus). For knockdown experiments, cells were reverse transfected with 10 nM siRNA (final concentration) using HiPerfect (Qiagen) according to the manufacturer's instructions.

[0184] Cell Lysis and Immunoblotting

[0185] Cells were washed twice with PBS and collected in lysis buffer (50 mM Tris-HCl, pH7.4; 1 mM EDTA, 100 mM NaCl, 0.1% SDS and 1% NP-40) supplemented with proteinase and phosphatase inhibitor (Pierce Protease and Phosphatase Inhibitor Mini Tablets, EDTA-free). Lysates were incubated on ice for 10 minutes followed by clearing centrifugation at 20,000×g at 4° C. for 10 min. Supernatants were transferred into fresh tubes and reducing loading buffer was added. Lysates were subjected to SDS-PAGE and transferred onto a nitrocellulose membrane using semi-dry transfer. The membrane was blocked (in ROTI buffer (Roth) or 5% milk in PBS with 0.1% tween) and probed with the appropriate primary antibodies. Subsequently membranes were incubated with horseradish peroxidase-conjugated secondary antibody. Immunoblots were developed using a chemiluminescence reagent (ECL clarity, BioRad) and imaged using ChemiDoc (BioRad).

[0186] Multiple Myeloma Cell Lines, Cell Culture and Viability Assays

[0187] The cell lines used were the human multiple myeloma cell lines L363 (DSMZ catalogue number ACC 49), AMO-1 (DSMZ catalogue number ACC 538) and ARH-77 (ATCC CRL-1621).

[0188] Multiple myeloma cell lines were maintained in RPMI-1640 culture medium (Sigma-Aldrich, Buchs, Switzerland) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 μg/ml streptomycin and 100 U/ml penicillin/streptomycin (Sigma-Aldrich, Buchs, Switzerland). The bortezomib-resistant cell lines were established and maintained from their parental cell line by continuous drug exposure>12 months to their parental cell lines (Soriano et al,. Leukemia 30: 2198-2207, 2016).

[0189] The viability of the cells was determined 24 hr post treatment using Cell Counting Kit-8 (CCK-8; MedChemExpress, NJ, USA) according to the manufacturer's instructions.

[0190] Gene Expression Analysis

[0191] Omics data were downloaded from the CoMMpass℠ study, IA-13 build and comprised 767 multiple myeloma samples (www.themmrf.org). These data were generated as part of the Multiple Myeloma Research Foundation Personalized Medicine Initiatives (https://research.themmrf.org). RNA-seq data were represented as Fragments Per Kilobase of transcript per Million mapped reads, FPKM. FKPM<1 was considered negative.

[0192] RNA Isolation and RT-PCR (s/u XBP1 and ATF4 Induction)

[0193] Total RNA was isolated from cell lines using Trizol (Ambion/Thermo Fisher Scientific, MA, USA) and Direct-zol RNA MiniPrep (Zymo Research, CA, USA). 500 ng of total RNA was reverse transcribed using High Capacity cDNA Reverse Transcription kit (Applied Biosystems/Thermo Fisher Scientific, MA, USA) according to the manufacturer's instructions. Subsequently, 10 ng of cDNA was used per qPCR reaction with 2× PowerUp SYBR Green Master Mix (Applied Biosystems/Thermo Fisher Scientific, MA, USA) and the following primers: spliced, unspliced and total XBP1 and ATF4 (Oslowski & Urano, Methods in Enzymology 490: 71-92, 2011), and GAPDH as a housekeeping gene control on a QuantStudio5 Real-Time PCR system (Applied Biosystems/Thermo Fisher Scientific, MA, USA).

ELISA

[0194] 1×10.sup.6/ml IgA-secreting AMO-1 cells were seeded in fresh medium with increasing doses of compounds of interest for 3 hr. Subsequently, the medium was collected, diluted 20-fold and a Human IgA ELISA Kit (Abcam, UK; product number ab137980) was used to determine the level of IgA antibody in the cell culture supernatant, according to the manufacturer's instructions.

[0195] In Vivo Experiments

[0196] NSG (NOD-scid I L2Rgamma.sup.null) mice were obtained from The Jackson Laboratory (No. 005557, CA, USA) and kept in isolated ventilated cages with food ad libitum. Age matched (6-8 weeks old) mice were injected with 0.5×10.sup.5 L363-BTZ_Luc+_TdTomato+ multiple myeloma cells (equipped with luciferase vector for monitoring, obtained from Addgene: #72486) into the right femur. Tumour growth was monitored twice over the week following luciferin injection (150 mg/kg, obtained from BioVision, CA, USA). 10 days post tumour inoculation, treatment was initiated with crizotinib (50 mg/kg, orally, daily), bortezomib (1 mg/kg, i.v. twice over 7 days) or with vehicle, for 7 days. The study was carried out in accordance with the 3Rs principle. Experiments were approved by the Committee for Animal Experiments (St Gallen, Switzerland), application no. 30177. Bortezomib and crizotinib were purchased from MedChem Express, NJ, USA.

[0197] CLL Experiments, Materials and Methods

[0198] Gene Expression Data

[0199] Gene expression data for CLL cells were obtained by analysis of previous data sets obtainable at http://www.genomicscape.com/microarray platform. Data from CLL cells of 107 CLL patients were analysed for the expression of LTK and ALK genes. Data were from the Herold, chronic lymphocytic leukemia, HGU133P (107 samples) dataset from the Affymetrix Human Genome U133 Plus 2.0 Array (Herold et al., Leukemia 25(10): 1639-45, 2011), normalised with Robust Multiarray Averaging (RMA).

[0200] Dose Response Experiments for CLL Samples

[0201] CLL cells from patients were stimulated by adherent murine L cells expressing CD40L, BAFF and April for 24 h. The L cells were then removed by immuno-magnetic separation. Washed CLL cells were plated into 384-well plate format (10,000 cells/well) and tested against crizotinib at 5 concentrations in 10-fold dilutions (plates dispensed by an Echo acoustic dispenser, LabCyte Inc., San Jose, Calif., USA). The plate was incubated in a humidified environment at 37° C. and 5% CO.sub.2. Cell viability was determined after 72 h using the CellTiter-Glo (Promega, Madison, Wis., USA) ATP assay according to the manufacturer's instructions and using an Envision 2102 Multilabel reader (PerkinElmer, Shelton, Conn., USA). Relative percentage cell viability was calculated by normalising luminescence units for each well to negative (DMSO 0.1%) and positive (100 μM BzCl) control wells. Curve fitting was conducted to obtain IC.sub.50 values.

[0202] Quantitative scoring of differential drug sensitivity for individual drugs was calculated as drug sensitivity scores (DSS values, Yadav et al., Scientific Reports 4: 5193, 2014). DSS values were calculated for integrated multiple dose-response relationships in high-throughput compound testing as shown.

[0203] Gene Expression Analysis for Human B-Cells

[0204] Gene expression data for B-cell differentiation in humans were obtained by analysis of data sets obtainable at http://www.genomicscape.com/microarray platform. The GEP Dataset was derived from human plasma cell differentiation, normalised by MAS5 on the Affymetrix Human Genome U133 Plus 2.0 Array (Jourdan et al., Leukemia 28(8): 1647-56, 2014; Jourdan et al., Journal of Immunology 187(8): 3931-41, 2011; Jourdan et al., Blood 114(25): 5173-81, 2009; Caron et al., Journal of Immunology 182(12): 7595-602, 2009).

[0205] Data from donors were analysed for the expression of LTK and ALK genes in B lymphopoiesis of humans. B-cells were sorted into subsets as follows: 1. naive B-cells (n=5); 2. centroblasts (n=4); 3. centrocytes (n=4); 4. memory B-cells (n=5); 5. preplasmablasts (n=5); 6. plasmablasts (n=5); 7. early plasma cells (n=5); 8. bone marrow plasma cells (n=5).

[0206] RNA-seq data was downloaded from deposited data (Monaco et al., Cell Reports 26(6): 1627-40, 2019). Data were analysed for 29 immune cell types within the peripheral blood mononuclear cell (PBMC) fraction of healthy donors using RNA-seq and flow cytometry. Data from naïve B-cells were from Schmiedel et al., Cell 175(6): 1701-15, 2018).

[0207] Drug Sensitivity Screens Activated B-Cell Blasts/Plasma Cells

[0208] CD19.sup.+ B-cells from 5 blood bank donors were stimulated by adherent murine L cells expressing CD40L, BAFF and April for 3 days to make B-cell blasts (plasma blasts/cells). Washed plasma blasts/cells were detached and plated into 384 well format (10,000 cells/well) and tested against crizotinib at 6 concentrations in 10-fold dilutions covering a 10.sup.−3-10 μM concentration range. Cell viability was determined after 72 h using the CellTiterGlo. Relative percentage cell viability was calculated by normalising luminescence units for each well to negative controls (DMSO 0.1%) and positive control wells (100 μM BzCl) and curve-fitting conducted to obtain the IC.sub.50.

[0209] Results

[0210] LTK is a Regulatory Node in the Proteostasis Network

[0211] Mathematical modeling suggested that LTK is a positive regulator of ER export that is activated by secretory flux. To validate this prediction experimentally, HeLa cells which inducibly express IgM were used (Bakunts et al., supra). Induction of IgM expression for 24 hours resulted in a mild activation of the UPR (which is a marker of ER stress), as indicated by the rise of XBP1s levels (FIG. 1). When LTK expression was silenced, cells responded with higher levels of ER stress upon induction of IgM expression (FIG. 1). This confirms that LTK is a regulatory node of the proteostasis network.

[0212] LTK Helps Multiple Myeloma Cells Cope with Elevated Secretory Load

[0213] Multiple myeloma is characterised by secretion of excessive amounts of immunoglobulins. To determine whether LTK inhibition induces ER stress in multiple myeloma cell lines, L363 cells were treated with crizotinib, a compound which has previously been shown to inhibit LTK and its effects on secretion (Centonze et al., supra). Crizotinib treatment induced ER stress as evident by increased splicing of XBP1 and induction of ATF4 (FIG. 2A). A similar observation was made in L363-BTZ cells that are resistant to bortezomib (FIG. 2B). Of note, L363 cells as well as their bortezomib-resistant clones are positive for LTK, but negative for ALK (see below), minimising the possibility of an off-target effect of crizotinib.

[0214] To test whether LTK helps cells adapt to high secretory cargo load a system was used that allows direct comparison of the same cells with or without excessive secretion. Rather than using pulsed induction of IgM production and secretion for only short periods of time, IgM production was induced for 14 days, in order to take into account potential adaptive cellular mechanisms to chronic overexpression of secretory proteins. HeLa cells induced to overexpress IgM for 14 days were compared to non-induced HeLa cells. Treatment of non-induced cells with the LTK inhibitor crizotinib did not result in any appreciable increase in ER stress (FIG. 2C), whereas cells that experienced increased secretory burden for prolonged periods of time exhibited a 2.5-fold higher level of ER stress (FIG. 2C). The heightened dependency of hypersecretory cells on LTK is in line with the observation that hypersecretory cells exhibit a 4-fold increase in LTK expression (FIG. 2D).

[0215] Targeting LTK Reduces the Viability of Multiple Myeloma Cells

[0216] The observation that LTK inhibition induces ER stress in hypersecretory cells prompted the inventor to test whether it would also cause cell death. The effects of crizotinib on caspase 3 cleavage in HeLa cells with or without IgM hypersecretion were investigated. Non-secretory cells did not show any apoptotic response to crizotinib treatment (FIG. 3A), which is in line with the observation that these cells did not respond with ER stress. However, cells with chronic increase of secretory burden showed a marked increase in apoptosis upon crizotinib treatment (FIG. 3A). Two multiple myeloma cell lines (L363 and ARH77) were then treated with crizotinib, and clear reductions in cell viability were observed at pharmacologically relevant drug concentrations. The same was found to be the case with a bortezomib-resistant clone of L363 (FIG. 3B). Of note, these cells are positive for LTK but are negative for ALK (FIG. 3C), as were multiple myeloma cells obtained from patients (see below).

[0217] Similarly to crizotinib, other linear ALK inhibitors, such as ceritinib, alectinib, ensartinib and entrectinib, were found to display cytotoxic activity against multiple myeloma cells (FIG. 4, left panels). The cyclic ALK inhibitor lorlatinib did not efficiently kill myeloma cells. The following IC.sub.50 values (in μM) were obtained for ceritinib in the tested cell lines: 1.03 in L363, 0.79 in L363-BTZ, 2.09 in AMO-1 and 2.77 in AMO-1-BTZ. The following IC.sub.50 values (in μM) were obtained for crizotinib in the tested cell lines: 3.323 in L363, 4.15 in L363-BTZ, 3.92 in AMO-1 and 6.477 in AMO-1-BTZ. The following IC.sub.50 values (in μM) were obtained for alectinib in the tested cell lines: 6.12 in L363, 6.22 in L363-BTZ, 13.31 in AMO-1 and 17.6 in AMO-1-BTZ. A similar sensitivity to LTK inhibition was observed with the bortezomib-resistant versions of L363 and AMO-1 (FIG. 4, right). This suggests that LTK acts on a proteasome-independent part of the proteostasis network, and that proteasome inhibitor resistance in multiple myeloma may be overcome by inhibiting LTK. This can be achieved by using approved ALK-inhibiting drugs.

[0218] In terms of mechanism, linear ALK inhibitors such as ceritinib inhibit LTK (FIG. 5A), which inhibits ER-to-Golgi trafficking (FIG. 5B, C) and consequently also inhibits immunoglobulin secretion (FIG. 5D).

[0219] The effect of crizotinib was then tested on primary myeloma cells obtained from relapsed patients who were at the 3.sup.rd-5.sup.th line of therapy. CD8-depleted bone marrow-derived mononuclear cells were stimulated by anti-CD3/CD28 beads in IL-2. This strategy stimulated CD4+ helper T cells (Th cells) that provide support for multiple myeloma cell activation. After 48 hr, CD138+ multiple myeloma cells from each of 15 patients were transferred to wells in the presence of titrated crizotinib, and concentration-response curves were calculated (FIG. 6). LTK inhibition had a striking effect on all patient-derived multiple myeloma cells, and IC.sub.50 levels were between 0.8-5 μM. Notably, analysis of a cohort of 767 patients with multiple myeloma showed that the mean expression level of LTK mRNA was 8.18, while that of ALK was 0.02 (FIG. 7A), below the cut-off (10.sup.0). The patients were divided into the following categories: myeloma cells that were Alk.sup.NegLTK.sup.Pos were found in 636 of 767 patients (83%); myeloma cells that were Alk.sup.NegLTK.sup.Neg in 130 of 767 patients (17%); and Alk.sup.PosLTK.sup.Pos myeloma in only 1 of 767 patients (0.001%) (FIG. 7B). Since the ALK inhibitors lack ALK as a target, the killing of primary myeloma cells with crizotinib is likely a result of LTK inhibition.

[0220] The in vivo effect of crizotinib on multiple myeloma cells was then tested. To this end, bortezomib-resistant L363 cells engineered to express luciferase were injected into the femurs of immunocompromised mice. To facilitate monitoring of tumour growth in vivo, cells were engineered to express luciferase. This model takes into account both the special biology of proteasome inhibitor-resistant multiple myeloma cells as well as the protective and drug resistance-facilitating features of the bone marrow microenvironment that are critical for multiple myeloma biology in vivo. Treatment with crizotinib was commenced at a stage where tumours were readily detectable (at 10 days after tumour inoculation), mimicking the clinical situation. Mice were also treated with bortezomib, which was not expected to have any effect given that bortezomib-resistant L363 cells were used, but confirmed proteasome inhibitor resistance. Mice were administered crizotinib orally daily for 7 days (50 mg/kg). Bortezomib was administered intravenously twice (1 mg/kg) over the period. Treatment with crizotinib caused a reduction in tumour burden relative to the baseline, as well as to the untreated control and bortezomib-treated mice (FIG. 8). This shows that treatment with an inhibitor of LTK can overcome proteasome inhibitor resistance in multiple myeloma.

[0221] Primary CLL Cells are ALK.sup.NegLTK.sup.Pos

[0222] Turning to chronic lymphocytic leukemia (CLL), we investigated the transcription of ALK and LTK and found that 94 out of 107 CLL cell samples (88%) were LTK.sup.Pos and that all 107 samples were ALK.sup.Neg (FIG. 9). CLL is therefore a cancer where a majority of patients have ALK.sup.NegLK.sup.Pos CLL cells. CLL is a cancer where leukaemic cells in the blood are quiescent or pre-apoptotic. CLL cell divisions occur in a specialised microenvironment in the lymph node, spleen or in the bone marrow. Here dividing CLL blast cells are juxtaposed to stimulatory cells in the microenvironment (so-called pseudofollicles) that support blastogenesis and mitosis. In this stage, CLL cells are hypersecretory, secreting monoclonal antibodies before undergoing cell division (Darwiche et al., supra).

[0223] Primary CLL Cells are Sensitive to Linear ALK Inhibitors

[0224] Microenvirionmental factors that are important for CLL cell activation (Patten et al., Blood 111: 5173-5181, 2008; Bagnara et al., Blood 117: 5463-5472, 2011; Hall et al., Blood 105: 2007-2015, 2005; Os et al., Cell Reports 4(3): 566-577, 2013) were provided and proliferating CLL blasts were tested for sensitivity to linear ALK inhibitors. We found that 17 of 21 patient samples (81%) had CLL cells that were sensitive to the linear ALK inhibitor crizotinib (FIG. 10). Of these, 11 were normally sensitive (IC.sub.50<2 μM,), 3 were highly sensitive (IC.sub.50<2×10.sup.−1 μM) and 3 were supersensitive (IC.sub.50<10.sup.−2 μM). Further integrating multiple dose-response relationships in our high-throughput compound testing, we found that drug sensitivity scores (DSS) demonstrated that a third of the patients have CLL cells that were highly sensitive to crizotinib.

[0225] Poor Prognosis Patients More Often have CLL Cells Sensitive to Linear ALK Inhibitors

[0226] CLL is divided into two subtypes depending upon mutation load of the V regions of the immunoglobulin variable genes of the heavy chain (IgVH) as detailed above. The IgVH unmutated subtype has poor prognosis, while the IgVH mutated (<98% homology to germline) is found in indolent CLL with better prognosis. The results showed that the poor prognosis subtype had a higher response rate and sensitivity to linear ALK inhibitors (FIG. 11). This suggests that linear ALK inhibitors are especially relevant for treatment of the IgVH unmutated, poor prognosis subtype.

[0227] CLL cells from 12 new patients were tested for susceptibility to ALK inhibitors. Nearly all patients had CLL cells that showed susceptibility to ceritinib, crizotinib, ensartinib and entrectinib (when using a cut-off of DSS=10, the results are 11/12, 10/12, 11/12, 12/12), as shown in FIG. 12. Brigatinib inhibited CLL cells from 4/12 patients, lorlatinib inhibited CLL cells from 0/12 patients.

[0228] Normal Memory B-Cells, Plasmablasts and Plasma Cells are ALK.sup.NegLTK.sup.Pos

[0229] We next turned to normal B-cells and the B-cell differentiation lineage that includes naïve mature B-cells, germinal centre centroblasts, germinal centre centrocytes, memory B-cells, pre-plasmablasts, plasmablasts, early plasma cells and bone marrow plasma cells. Gene expression (array based) and RNA sequencing demonstrated that memory B-cells plasmablasts and plasma cells all expressed LTK, but not ALK (FIG. 13). Memory B-cells, plasmablasts and plasma cells are therefore ALK.sup.NegLTK.sup.Pos.

[0230] Normal Memory B-Cells, Plasmablasts and Plasma Cells are Sensitive to Linear ALK Inhibitors

[0231] Activated B-cells become B-cell blasts after 3 days. Such enlarged cells secrete high levels of antibody and are therefore hypersecretory. Stimulated day 3 B-cells were cultured for 3 further days with linear ALK inhibitor. All samples had a similar sensitivity with an IC.sub.50 between 1 and 1.5 μM for crizotinib (FIG. 14). Results suggest that at least half of blastoid B-cells (memory B-cell blasts, plasmablasts and plasma cells) are killed within 3 days of treatment. To further investigate the sensitivity of cell subtypes, we stimulated B-cells for 5 days to allow plasma cells to develop. End stage plasma cells typically express IRF4 and BLIMP1 (FIG. 15) and secrete high levels of antibody. Exposure to linear ALK inhibitor (crizotinib or ceritinib) efficiently killed plasma cells with high sensitivity. Plasma cells responded with an IC.sub.50 of 0.8 μM (crizotinib) and <0.001 μM (ceritinib) demonstrating that plasma cells were very sensitive to linear ALK inhibitors. A reduction of all other activated B-cell subsets and plasmablasts were seen, but plasma cells were most sensitive to linear ALK inhibitors and plasma cells were supersensitive to ceritinib.

[0232] Crizotinib Kills Plasma Cells in Xenografted NSG Mice

[0233] We injected ALK.sup.NegLTK.sup.Pos plasma cell cell lines intrafemorally into NSG mice. Treatment of mice with crizotinib reduced plasma cell density (FIG. 16, luciferase dependent signals, in optical IVIS imaging) demonstrating that plasma cell cell lines also could be inhibited in vivo.

[0234] Ceritinib Inhibits Secretion of Antibodies by Plasma Cells In Vitro

[0235] We tested if plasma cell secretion was inhibited by linear ALK inhibitor (FIG. 16). Three hour exposure reduced the levels of secreted immunoglobulin from plasma cell cell lines suggesting blockade of ER.fwdarw.Golgi trafficking as demonstrated in FIG. 5A-C.

[0236] Hepatocellular Carcinoma (HCC) Cells were ALK.sup.NegLTK.sup.Pos and Sensitive to Crizotinib

[0237] RNA-seq data from 365 HCCs revealed that all patients had HCC samples that are ALK.sup.Neg. About a quarter of the samples were LTK.sup.Pos and thereby ALK.sup.NegLTK.sup.Pos. We found that cells from the well-differentiated HOC cell line HepG were sensitive to the linear ALK inhibitor crizotinib in experiments where this drug acted in synergy with doxorubicin, a cytostatic drug (FIG. 17). Thus crizotinib sensitised the HepG HCC cells to apoptosis.