Glycosphingolipid biosynthesis inhibitors for the treatment of tumors dependent on the oncogene GOLPH3 and oncogenes of the GOLPH3 cluster

Abstract

An object of the present invention is the use of glycosphingolipid biosynthesis inhibitors for the treatment of tumors dependent on the oncogene GOLPH3 and two oncogenes similar to GOLPH3 (GOLPH3-L and LCS), which together with GOLPH3 form the GOLPH3 cluster. A further object of the present invention is the use of detection of genomic amplification of GOLPH3 as a predictive marker for identifying tumors that will respond to said inhibitors.

Description

[Technical Field]

[0001] An object of the present invention is the use of glycosphingolipid biosynthesis inhibitors for the treatment of tumors dependent on the oncogene GOLPH3 and two oncogenes similar to GOLPH3 (GOLPH3-L and LCS), which together with GOLPH3 form the GOLPH3 cluster. A further object of the present invention is the use of Golph3 as a marker for identifying tumors sensitive to said inhibitors. [Background technology]

[0002] Golgi phosphoprotein 3 (GOLPH3) is an oncoprotein encoded by a gene found in a region of the human genome that is frequently amplified in various solid tumors, including melanoma, lung cancer, breast cancer, glioma, and colorectal cancer (Rizzo R et al., Tissue Cell 2017;49 (2 Pt A):170-174). Reported frequencies vary depending on the source (database or publication). For example, according to an authoritative and conservative database, the amplification frequency of GOLPH3 is 11% in lung squamous cell carcinoma, 7% in lung adenocarcinoma, and 6% in uterine carcinoma (The Cancer Genome Atlas).

[0003] GOLPH3 is not only genomically amplified but also overexpressed (in a larger number of tumors). Its overexpression has been correlated with poor prognosis in several tumor types, including 52% of breast cancers and 41%-53% of glioblastomas (Sechi S et al. Int. J. Mol. Sci. 2020; 21: 933). The percentages also vary depending on the source (database or publication). Finally, the oncogene GOLPH3 confers chemotherapy resistance, a serious obstacle to pharmacological treatment of tumors.

[0004] The oncogenic mechanism of action of GOLPH3 (particularly with regard to its chemotherapy resistance-inducing properties) has been suggested to result from the oncogene's ability to selectively stimulate the transport of growth receptors to the plasma membrane, thereby activating them (Farber-Katz SE et al. Cell. 2014 156(3):413-27, Buschman MD et al. Cancer Res. 2015 75(4):624-7). Consistent with these findings, other authors have subsequently demonstrated that GOLPH3 acts by activating signaling pathways, such as those initiated by growth factor receptors (e.g., EGF receptors) or mediated by oncogenic kinases and GTPases, such as PI3K kinase, AKT, mTOR, and RAS (Zhi-Ping Zhou et al. Int J Oncol. 2018 53(3):1183-1192, Ming-Zhen Wang Molecular Medicine Reports 2018 17: 542-548).

[0005] More recently, the oncogenic effects of GOLPH3 were suggested to be linked to glycosphingolipid (GSL) metabolism by Rizzo R et al. in EMBO J. 2021: 15;40(8). The authors showed that GOLPH3 acts by increasing the levels and thus the activity of a group of GSL synthases, namely (in the order of their actions) lactosylceramide synthase (LCS), GB3 synthase (GB3S), GM3 synthase (GM3S), and GD3 synthase (GD3S), thereby increasing the amounts of their metabolites (LC, GB3, GM3, GD3) and, in return, decreasing their precursors, glucosylceramide (GC) and ceramide (Cer) (scheme in Figure 1).

[0006] These metabolites are bioactive: ceramide is a tumor suppressor, inhibiting proliferation and inducing apoptosis, while GB3 and GD3 activate SRC and growth factor receptors, which stimulate cell proliferation. The result of reducing ceramide and increasing GB3 and GD3 is therefore inhibition of apoptosis and enhanced proliferation stimulation. This action protects tumor cells from stress-induced ceramide production (see below) and stimulates cell proliferation, potentially contributing to tumor development. Other oncogenes belonging to the GOLPH3 cluster, GOLPH3-L and LCS, have very similar actions to those induced by GOLPH3 (Figure 1). The metabolic pathway of GSLs and the mechanism of action of GOLPH3 are shown in Figure 1.

[0007] A series of synthetic reactions (shown as horizontal processes) starting with serine (SER) and palmitic acid (PALM) produces sphingosine (SP), which is then used by ceramide synthase (CerS) to form Cer. Cer (black square) is a tumor suppressor whose formation induces apoptosis, whereas its consumption leads to cell proliferation, making it a functionally important molecule in the system. Cer is consumed in the Golgi complex and converted to glucosylceramide (GC) by glucosylceramide synthase (GCS, drug target, shown as concentric circles). GC is then translocated from the cytoplasmic leaflet to the luminal leaflet of the Golgi membrane by the action of the "flippase" P-glycoprotein (P-gp, also one of the molecular "pumps" known to extrude various molecules, including drugs, from cells).

[0008] In the lumen, GC is then converted to lactosylceramide (LacCer) by LCS. LacCer serves as a substrate for the synthesis of all GSL complexes, including GB3 and globoside, GM3 and ganglioside, and LC3 and lactoside. The mechanism of action of the GOLPH3 cluster is depicted in the central vertical rectangle. GOLPH3 (shown with a gray background) is a GSL metabolic regulator. It acts on the enzymes indicated by the letters (arrows), increasing their levels and, consequently, their activity. This increase has two complementary effects: a) attenuation of Cer consumption and therefore apoptosis, and b) increased formation of the enzyme's products, which have a stimulatory effect on SRC and growth factor receptors (RTKs), i.e., stimulation of cell proliferation. Regarding other oncogenes belonging to the GOLPH3 cluster, GOLPH3-L (top, gray background) is highly similar to GOLPH3 and has similar effects, while LCS (top, gray background) is the major mediator of the effects of GOLPH3 and has similar effects to GOLPH3.

[0009] The role of GSLs as tumor markers and cell signaling regulatory molecules has been described by Furukawa et al. in Cancer Sci., 2019; 110:1544-1551 and Ogretmen B in Nature Reviews, 2018; 18:33-50. The possibility of combating tumor growth by inhibiting GSL biosynthesis using glucosylceramide synthase (GCS) inhibitors has been investigated for some time; for example, Inokuchi et al., Cancer Letters, 38, 1987, 23-30, described the antitumor activity of the UDP-glucosylceramide glucosyltransferase inhibitor 1-phenyl-2-decanoylamino-3-morpholinepropanol (PDMD) in a mouse tumor model (Ehrlich ascites).

[0010] Stefanovic et al., in Oncotarget, 2016; 7(7):8254-8263, report the ability of several GSL biosynthesis inhibitors, in particular the GCS inhibitor (PDMD, imipramine), to improve the activity of sorafenib against hepatocellular carcinoma cells in vitro and in vivo. Tyler et al., Experimental Cell Research, 2015; 3 3 6: 23-32, describe enhanced cisplatin cytotoxicity by PDMD against melanoma cells and pleural mesothelioma. The potential antitumor activity of GCS inhibitors belonging to the chemical class of iminosugars, such as miglustat and its derivatives, in Lovo and HCT116 cells and mouse models of colorectal adenoma has been described in Int. J. Mol. Sci. 2021, 22, 10539, or those belonging to the class of ceramide analogs, such as PDMP and eliglustat, or those belonging to the class of P-gp inhibitors, such as zosuquidar and its analogs (Gottesman MM et al. FEBS Lett. 2006 13;580(4):998-1009), or the tamoxifen metabolite 4-hydroxytamoxifen (Morad SAF et al. Biochim Biophys Acta. 2015;1851(9):1134-1145.).

[0011] However, despite some positive data, the proven efficacy of GSL metabolism inhibitors used as antitumor agents in the cell and animal models tested to date is scarce and unpredictable. Most researchers have published data on experimental conditions and cell lines, and although the effects of inhibiting GSL metabolism have been documented, data on cases where the inhibitors are ineffective, as is often the case in research, are not documented.

[0012] By comparing the effects of GSL metabolic inhibitors on tumor cells characterized by amplification of GOLPH3 and genes belonging to the GOLPH3 cluster with the effects of said inhibitors on tumor cells that do not display said amplification, it has now been found that only a limited proportion of tumor lines, i.e., those characterized by GOLPH3 amplification (e.g., 10% in lung squamous cell carcinoma lines, as described above) or strong overexpression of genes belonging to the GOLPH3 cluster, are repeatedly and predictably sensitive to GSL metabolic inhibitors in both cell proliferation assays and stress resistance assays, e.g., those resulting from DNA damage induced by chemotherapy.

[0013] This is consistent with findings reported in the DEPMAP database (The Cancer Dependency Map Program at the Broad Institute), which demonstrates that GCS depletion significantly inhibits the growth of 2 out of 100 tumor lines grown in vitro under cell matrix adhesion. This success rate is far from clinically useful. The antitumor effects of these inhibitors have never been demonstrated in clinical practice. Only miglustat and its derivatives have been reported to be effective in reducing osteoclast activation and bone destruction associated with multiple myeloma (J Clin Invest. 2015;125(6):2279-2292). However, the described effects are limited to osteoclasts.

[0014] Overall, the data indicate that only a proportion of tumors characterized by amplification of GOLPH3 and the GOLPH3 cluster are sensitive to the effects of GSL biosynthesis inhibitors, and therefore only in these tumors do glycosphingolipid synthesis inhibitors hold any promise for pharmacological use in the clinical setting, a situation that is common in the development of many targeted drugs.

[0015] It should be noted that prior to the studies presented herein, it was not predictable that GSL metabolic inhibitors would be systematically and predictably active only in tumor cells with GOLPH3 amplification. Indeed, glycosphingolipid metabolism has the ability to activate mitogenic signaling in all mammalian cells. The reason that only tumors with GOLPH3 amplification respond is likely because Golph3 is a very potent activator of the metabolic pathway, simultaneously activating different enzymes of the metabolism (see Figure 1), and thus being employed as a defense mechanism by various types of tumors. Summary of the Invention

[0016] Here, we have discovered that GSL biosynthesis inhibitors (scheme in Figure 1) are selectively useful in treating tumors dependent on oncogenes in the GOLPH3 cluster. Specifically, we found that GSL biosynthesis inhibitors inhibit the growth of tumor lines only if the tumor lines treated with the inhibitors 1) express at least one oncogene belonging to the GOLPH3 cluster and 2) grow under stress conditions resulting from the presence of chemotherapeutic agents or loss of cell matrix anchorage. In contrast, the inhibitors have little effect on the growth of the same tumor lines under growth conditions of matrix adhesion in the absence of stress.

[0017] These observations indicate that 1) the presence of oncogenes belonging to the GOLPH3 cluster can be used as a predictive marker to identify tumors sensitive to inhibition of GSL metabolism, and 2) the inhibitor acts only in these tumor lines under stress conditions, reversing both their chemotherapy resistance and their resistance to loss of anchorage to the matrix that normally induces apoptosis, both of which are of considerable clinical importance. Chemoresistance is one of the major problems involved in cancer treatment, and resistance to anchorage loss is known to reflect the ability of tumors to grow and metastasize in vivo.

[0018] GCS inhibitors already used in clinical practice to treat other disorders (genetic disorders involving lysosomal storage) are used as GSL metabolism inhibitors, as are P-GP inhibitors, which are also GSL metabolism inhibitors. This strategy therefore involves repositioning GSL metabolism inhibitors for antitumor use, thereby making it possible to omit expensive development steps.

[0019] Tumors affected by oncogenes of the GOLPH3 cluster that can be advantageously treated with GSL metabolism inhibitors include liver, lung, prostate, ovarian, and colon cancers, as well as melanoma.

[0020] Examples of GSL metabolic inhibitors include iminosugars such as miglustat, eliglustat or its tartrate, benglustat, or ceramide analogs such as N-[(1R,2R)-2-(2,3-dihydro-1,4-benzodioxin-6-yl)-2-hydroxy-1-(1-pyrrolidinylmethyl)ethyl]-nonanamide (GENZ-123346), and inhibitors of P-GP, a flippase required for GSL metabolism, such as zosuquidar and its metabolite, tamoxifen derivatives. Miglustat and eliglustat are already in use, while benglustat is currently undergoing first-level clinical validation. These compounds are hereinafter collectively referred to as GSL metabolic inhibitors.

[0021] The following table lists other compounds with known inhibitory activity against the human enzyme glucosylceramide synthase (GCS) that have been used and are under evaluation in scientific experiments along with compounds with inhibitory activity against P-GP, and therefore GSL metabolism. The right-hand column lists one of the many references available in the scientific literature. [Table 1] TIFF2026506253000002.tif236165 [Brief explanation of the drawings]

[0022] [Figure 1] The GSL metabolic pathway and the mechanism of action of GOLPH3 are shown. SER: serine, PALM: palmitic acid, SP: sphingosine, CERS: ceramide synthase, CER: ceramide, GCS: glucosylceramide synthase, GC: glucosylceramide. [Figure 2A1] Figure 1 shows the inhibition of tumor line (prostate cancer, colon cancer, liver cancer tumor line) growth under anchorage-independent conditions by GSL metabolic inhibitors. [Figure 2A2] Figure 1 shows the growth inhibition of tumor lines (melanoma, lung cancer, and ovarian cancer tumor lines) under anchorage-independent conditions by GSL metabolic inhibitors. [Figure 2B] Growth of tumor lines adherent to the matrix in the presence of chemotherapeutic agents (chemo-resistance). [Figure 3] 1 shows the mechanism of action of GOLPH3 in chemotherapy resistance and tumor growth. Example 1

[0023] Figures 2 A1, 2 A2, and 2 B show data reported as examples that support the conclusions presented above.

[0024] The GSL metabolic inhibitors listed above were used in tumor lines expressing oncogenes belonging to the GOLPH3 cluster, and the results showed a suppression of apoptosis caused by the oncogene-induced loss of matrix anchorage and resistance to chemotherapeutic agents (in this case, etoposide).

[0025] The growth of tumor lines was evaluated under different experimental conditions. Twelve tumor lines were divided into six pairs derived from six different tumor types. Each pair consisted of a tumor line expressing an oncogene belonging to the GOLPH3 cluster and a line derived from the same tumor type but not expressing the oncogene. Growth of the lines was observed under three experimental conditions: 1) in the absence of a matrix anchorage, 2) under exposure to chemotherapeutic agents, and 3) under unhindered growth upon adhesion to the matrix.

[0026] Figure 2A1 shows that the growth of prostate, colon, and liver tumor lines expressing oncogenes belonging to the GOLPH3 cluster under anchorage-independent conditions is inhibited by miglustat (at concentrations ranging from 0 to 50 micromolar, similar to those known to inhibit GCS enzymes in cells) (top panel, light gray), whereas the growth of tumor lines derived from the same tumor type but not expressing GOLPH3 under the same conditions is not inhibited by miglustat (middle panel, dark gray). Growth of lines expressing oncogenes belonging to the GOLPH3 cluster under conditions of matrix adhesion and stress absence (bottom panel, black) is not inhibited by miglustat. Inhibition of GSL synthesis was approximately 70% (with 50 micromolar miglustat). Depletion of GOLPH3 cluster members by 25 nanomolar siRNA targeting oncogenes belonging to the GOLPH3 cluster expressed in the lines in question reached 90%, significantly more inhibiting growth than miglustat at the concentrations indicated (not shown).

[0027] FIG. 2A2 shows the same results as FIG. 2A1, but in melanoma, lung cancer, and ovarian cancer cell lines.

[0028] Figure 2B shows the growth of tumor lines adherent to a matrix in the presence of chemotherapeutic agents (chemoresistant). A line (LnCap) that does not express oncogenes belonging to the GOLPH3 cluster treated with the chemotherapeutic agent etoposide exhibits cell cycle arrest (strong increase in p21 synthesis), a situation that is not altered by cotreatment with eliglustat. In contrast, a line (DU145) that expresses oncogenes belonging to the GOLPH3 cluster is resistant to etoposide, but this resistance is abolished in the presence of eliglustat. Eliglustat alone has no effect. The same type of results are obtained by using the P-GP inhibitor zosuquidar.

[0029] Show me how reagent Cell lines and culture conditions: Ovarian cancer NIHOVCAR3 and TOV21G, liver cancer HepG2 and Hep3B, skin melanoma A375 and SKMEL2, colon cancer HCT116 and RKO, prostate cancer DU145 and LnCaP, and lung cancer A549 and PC14 cell lines were purchased from the American Tissue Type Collection (ATTC, USA). The cell lines DU145, A375, A549, HCT116, and HepG2 were cultured in DMEM supplemented with 10% FBS, 2 mM L-glutamine, and 100 U / ml penicillin and streptomycin; SKMEL5, Hep3B, and RKO were cultured in MEM supplemented with 10% FBS, 1 mM non-essential amino acids (NAA), 2 mM L-glutamine, and 100 U / ml penicillin and streptomycin; whereas the cell lines NIHOVCAR3, TOV21G, LnCaP, and PC14 were cultured in RPMI supplemented with 10% FBS, 2 mM L-glutamine, and 100 U / ml penicillin and streptomycin. All tumor cell lines used in this study were tested for and confirmed to be free of Mycoplasma and were cultured at controlled temperature (37°C) and atmosphere (5% CO2 and 95% air).

[0030] Medicines: Miglustat (from Cayman), Eliglustat (from Cayman), Genz-123346 (from Sigma), Etoposide (from Sigma), Zosuquidar (from SelleckChem).

[0031] Small RNA interference (RNAi): [Table 2]

[0032] Primary and secondary antibodies: [Table 3]

[0033] process siRNA-mediated gene silencing The siRNAs used in this study were obtained from Sigma-Aldrich (Table 1). Oligofectamine or RNAi Max (Thermofisher Scientific, USA) were used for siRNA transfection at concentrations ranging from 5 to 50 nM, as previously described (Rizzo et al., 2021). Silencing efficiency was assessed by Western blot or qPCR using specific primers.

[0034] Blockade of GSL synthesis by inhibitors GCS enzyme inhibitors, such as miglustat, eliglustat, and N-[(1R,2R)-2-(2,3-dihydro-1,4-benzodioxin-6-yl)-2-hydroxy-1-(1-pyrrolidinylmethyl)ethyl]-nonanamide (Genz-123346), or P-gp inhibitors, such as zosuquidar, were used at two or three different concentrations depending on the dose type to block glycosphingolipid biosynthesis in various tumor cells. For proliferation assays, fresh medium supplemented with the drug was added every three days.

[0035] Treatment with chemotherapy medications Cells were treated with etoposide at concentrations ranging from 10 to 40 μM for 12 to 16 hours.

[0036] Assay Cell proliferation assay 1. Growth in the absence of an anchorage (anchorage-independent) 2. Propagation of Isolated Cell Clones (Clonogenic Analysis) 3. Subadhesive growth (subadhesive).

[0037] Both anchorage-independent and clonogenic growth assays were performed as previously described in Rizzo et al., EMBOJ, 2021. Adherent proliferation and cell viability assays were performed by measuring XTT or ATP according to the manufacturer's protocols.

[0038] Apoptosis assay As an index of apoptosis activation, cleavage of PARP by caspase 3 was assessed by immunoblotting.

[0039] Cell cycle arrest assay p21 levels were determined by immunoblotting as a cycle arrest index.

[0040] Glycosphingolipid metabolism To analyze glycosphingolipid metabolites, we used a method based on radiolabeling of cellular sphingosine, followed by extraction and separation of metabolites by thin-layer chromatography (TLC) as previously described by Rizzo et al., EMOJ, 2021.

[0041] Immunoblotting Proteins were labeled with the primary antibodies described above (Table 2) and confirmed using secondary goat anti-rabbit or goat anti-mouse antibodies IgG-HRP (Santa Cruz, USA) at a dilution of 1:10,000.

[0042] statistics Error bars correspond to the standard deviation (SD) or standard error of the mean (SEM) as indicated in the figure legends. Statistical evaluation was performed using the Student's t test. *P<0.05, **P<0.01, and ***P<0.001 (ns, not significant).

[0043] The mechanisms underlying the stated conclusions Growth in the presence of chemotherapy (chemoresistance) Chemotherapeutic agents act by damaging the DNA of target cells. This triggers an apoptotic response, the function of which is to prevent cells with severely damaged DNA from continuing to proliferate in the body. This response is mediated by various mechanisms, primarily p53 activation (the most widely characterized) and increased ceramide, which triggers cell cycle arrest necessary to allow DNA repair and, if damage is excessive, apoptosis unless its effects are counteracted by oncogenes that directly promote survival and proliferation or by mechanisms that rapidly consume ceramide. GOLPH3 and oncogenes belonging to the GOLPH3 cluster activate both of these defense mechanisms.

[0044] In addition, an important regulatory mechanism specifically activates GOLPH3 in response to stress conditions; DNA damage and increased ceramide activate a PKD and DNAPK kinase-based mechanism, which phosphorylates GOLPH3 and potently enhances its activity, resulting in ceramide consumption and the production of growth-promoting GSLs. Under physiological conditions, this mechanism acts as a regulatory system to reduce DNA damage-induced increases in ceramide and the resulting apoptotic effects, which, unless alleviated, can become uncontrollable and destructive in non-tumor cells. However, in tumor cells overexpressing GOLPH3, this system induces excessive suppression of apoptotic signals and excessive increase in growth signals, which together cause chemotherapy resistance and increased tumor growth, rendering the tumor cells "invulnerable" to the apoptotic response induced by chemotherapeutic agents, as shown in Figure 3.

[0045] This regulatory mechanism explains, in particular, why the oncogene GOLPH3 is selectively effective in protecting tumor cells from apoptosis under stress conditions, such as DNA damage, which produces large amounts of ceramide, i.e., because it represents a potent chemotherapy resistance factor, while being less active in the absence of stress. This also explains why GCS inhibitors, such as eliglustat, which block ceramide consumption and the synthesis of growth-promoting GSLs, eliminate the chemotherapy resistance and oncogenic effects caused by the action of GOLPH3 on GSLs.

[0046] Growth in the absence of a scaffold GOLPH3 also plays a role in defense against apoptosis under other stress conditions, such as loss of anchorage to the matrix, which leads to strong production of ceramide. Although less well characterized, the mechanisms of apoptosis caused by anchorage loss and the apoptosis resistance mediated by GOLPH3 under these conditions are similar to those previously described for chemotherapy resistance. The same conclusions apply to other oncogenes belonging to the GOLPH3 cluster.

[0047] The scheme in Figure 3 illustrates the PKD- and DNAPK-based regulatory mechanisms that enhance the ability of GOLPH3 to induce resistance to DNA damage and loss of matrix anchorage, as well as the action of GCS inhibitors as suppressors of this property of GOLPH3. DNA damage (top rectangle, shown with a gray background on the left) induces a strong increase in the transcription of ceramide synthase and thus ceramide production. A portion of the ceramide is used to produce sphingomyelin (SM) and diacylglycerol (DAG), which then activates PKD kinase (top rectangle, shown with a gray background on the right). The kinase phosphorylates GOLPH3, enhancing its activity. A second effect of DNA damage is the direct activation of DNAPK kinase (top rectangle, shown with a gray background on the right), which has an effect similar to that of PKD and further enhances the effect of GOLPH3. GOLPH3 is therefore specifically enhanced during events involving DNA damage (or scaffold loss and consequent ceramide production), making the action of GOLPH3 particularly efficient under these severe stress conditions.

[0048] GCS inhibitors, such as eliglustat, or P-GP inhibitors, such as zosuquidar, block both GCS and the subsequent enzymatic consumption of ceramide and the synthesis of growth-promoting GSLs, thus blocking the anti-apoptotic and growth-promoting actions of GOLPH3.

[0049] For the therapeutic uses contemplated by the present invention, GCS or P-GP inhibitors can be formulated into compositions suitable for oral or parenteral administration using conventional techniques. The dosage depends on the type and stage of the disorder, as well as the patient's weight, sex, and age, and can be determined based on preclinical and clinical studies. In the case of miglustat, eliglustat, and other inhibitors already in clinical use, oral dosages broadly correspond to those already used for known approved therapeutic indications.

[0050] The activity of GCS and P-GP inhibitors in inhibiting the growth of tumor cells dependent on oncogenes of the GOLPH3 cluster was demonstrated using lung, ovarian, prostate, skin (melanoma), and colon cell lines.

Claims

1. Glycosphingolipid metabolism inhibitors for use in the treatment of tumors dependent on the expression of GOLPH3 and GOLPH3 cluster oncogenes and resistant to chemotherapy.

2. The glycosphingolipid metabolism inhibitor for use according to claim 1, which is an iminosugar.

3. 3. The glycosphingolipid metabolism inhibitor for use according to claim 2, wherein the iminosugar is miglustat.

4. 2. The glycosphingolipid metabolism inhibitor for use according to claim 1, which is selected from N-[(1R,2R)-2-(2,3-dihydro-1,4-benzodioxin-6-yl)-2-hydroxy-1-(1-pyrrolidinyl-methyl)ethyl]-nonanamide, eliglustat, eliglustat tartrate, benglustat, and zosuquidar.

5. The glycosphingolipid metabolism inhibitor for use according to any one of claims 1 to 4, wherein the tumor dependent on the expression of GOLPH3 and GOLPH3 cluster oncogenes and resistant to chemotherapy is liver cancer, lung cancer, prostate cancer, ovarian cancer, colon cancer, and melanoma.

6. A method for classifying tumors as sensitive or resistant to treatment with glycosphingolipid inhibitors, comprising measuring the expression levels of GOLPH3 and the GOLPH3 cluster.

7. 7. The method of claim 6, wherein the tumor is chemotherapy-resistant.

8. Use of detection of genomic amplification of GOLPH3 as a predictive marker for identification of tumors that will respond to glycosphingolipid metabolism inhibitors.

Images