Therapeutic targets for pancreatic neuroendocrine tumours

Abstract

The present invention relates to a use of a moiety selected from the group consisting of: (a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6; (b) a NR4A1 antagonist selected from the group consisting of 1,1- bis(3'-indolyl)-1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5- methoxy analog (DIM-C-pPhOH-3-Cl-5-OCH3) and a flavonoid selected from the group consisting of kaempferol and quercetin; (c) a vector comprising a nucleic acid molecule encoding menin; and (d) a NF-ĸB inhibitor, in the manufacture of a medicament for treating pancreatic neuroendocrine tumors (PNETs) in a subject, wherein the medicament is to be administered to the subject, thereby reducing cellular proliferation of PNETs. The present invention also relates to a method of treating pancreatic neuroendocrine tumors (PNETs) in a subject.

Description

DESCRIPTIONTITLE OF THE INVENTION: THERAPEUTIC TARGETS FOR PANCREATIC NEUROENDOCRINE TUMOURSFIELD OF THE INVENTION

[0001] The invention generally relates to the field of molecular biology. In particular, the invention relates to therapeutic methods and compositions for treating pancreatic neuroendocrine tumours.BACKGROUND OF THE INVENTION

[0002] Pancreatic neuroendocrine tumours (PNETs) are a group of rare, heterogenous tumours of the endocrine pancreas. Despite its low incidence of 1 per 100000 individuals, the incidence of PNETs has been steadily increasing over the past decades. Amongst neuroendocrine tumours, PNETs have the highest mortality rate, with a 5-year survival rate of 22.7% and a median overall survival of ~3.6 years. In contrast to healthy endocrine cells, which exhibit minimal proliferation (Ki-67 index <1 %), PNETs display markedly elevated proliferative rates (Ki-67 >1-20%). Higher tumour grade and Ki-67 index have been consistently associated with poorer patient outcomes across epidemiologic studies worldwide, underscoring the clinical relevance of cancer cell proliferation in PNET progression and patient survival outcomes. To date, however, the signalling pathways and molecular drivers that govern proliferation and tumorigenesis in PNETs remain incompletely understood. This is in part due to the scarcity of human PNET tissues and a lack of mechanistic studies to unveil the molecular basis of this intricate process.

[0003] Multiple whole-genome and exome studies have since unveiled the genetic landscape of human PNETs. Notably, common tumour suppressors and oncogenes that function as master regulators of proliferation such as p53, PTEN, KRAS, and APC are rarely mutated in PNETs, suggesting that the molecular program governing cancer cell proliferation and tumorigenesis in PNETs may be distinct from those of other cancer types. Instead, MEN1, ATRX, and DAXX have emerged as the most frequently mutated genes in PNETs. Among these, only the pancreas-specific conditional knockout of Men1 induces PNET formation in mouse models, while deletion of Atrx and / or Daxx does not lead to proliferative abnormalities in the endocrine pancreas. Consistent with this observation, the menin protein (encoded by the MEN1 gene) has been widely implicated with the repression of cellular proliferation invarious cell models, supporting its central role in safeguarding PNET proliferation and tumorigenesis. At the molecular level, menin functions as a scaffold protein, interacting with a plethora of protein interacting partners to modulate the transcription of genes in various cell types. Despite these studies, the protein interacting partner(s) of menin and the transcriptional targets regulated by menin to repress cellular proliferation and tumorigenesis in PNETs remain poorly understood.

[0004] Currently, surgical resection of the PNETs is the only curative solution in MEN1 patients. However, MEN1 patients typically do not qualify for surgery due to the presence of multiple tumours which increases the risk of complications during surgery. Other treatment options such as somatostatin analogues which were reported to be effective for treating non- MEN1 PNETs are also ineffective for treating MEN1-related neuroendocrine tumours.

[0005] There is thus a need for alternative therapeutic targets for slowing down the growth of MEN1 -related PNETs that overcome the drawbacks of the prior art. Furthermore, other desirable features and characteristics will become apparent from the subsequent detailed description and the appended claims, taken in conjunction with the accompanying drawings and this background of the disclosure.SUMMARY OF THE INVENTION

[0006] In one aspect, the present invention provides a use of a moiety selected from the group consisting of: (a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6; (b) a NR4A1 antagonist selected from the group consisting of 1,1-bis(3'-indolyl)-1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5-methoxy analog (DIM-C-pPhOH-3-CI-5-OCH3) and a flavonoid selected from the group consisting of kaempferol and quercetin, (c) a vector comprising a nucleic acid molecule encoding menin; and (d) a NF-KB inhibitor, in the manufacture of a medicament for treating pancreatic neuroendocrine tumors (PNETs) in a subject, wherein the medicament is to be administered to the subject, thereby reducing cellular proliferation of PNETs.

[0007] In one embodiment, the PNETs are Multiple Endocrine Neoplasia Type I (MEN1)- associated PNETs.

[0008] In one embodiment, the vector is an adeno-associated virus (AW) vector, or a lentiviral vector.

[0009] In one embodiment, the shRNA specific for NR4A1 can be sequenced by a forward primer sequence of SEQ ID NO: 1 or 3, and a reverse primer sequence of SEQ ID NO: 2 or 4.

[0010] In one embodiment, menin expressed from the vector forms a complex with NF-KB signalling molecule RelA / p65, thereby inhibiting transactivation of NR4A1 and reducing cellular proliferation of PNETs.

[0011] In one embodiment, each dose of the vector comprising a nucleic acid molecule encoding menin is to be administered at 50-1500 ng.

[0012] In one embodiment, the vector comprising a nucleic acid molecule encoding menin is to be administered at a frequency from once every day to once a month.

[0013] In one embodiment, the vector comprising a nucleic acid molecule encoding menin is to be administered for a period of 1 day to 1 year.

[0014] In one embodiment, each dose of the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered at 0.3 to 3 mg / kg.

[0015] In one embodiment, the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered at a frequency of about once a month.

[0016] In one embodiment, the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered for a period of 1 day to 1 year.

[0017] In one embodiment, each dose of the NR4A1 antagonist is to be administered at 2- 30mg / kg / day.

[0018] In one embodiment, the NR4A1 antagonist is to be administered at a frequency from once every day to once a month.

[0019] In one embodiment, the NR4A1 antagonist is to be administered for a period of 1 day to 1 year.

[0020] In one embodiment, the NF-KB inhibitor is selected from the group consisting of NF- KB Activation Inhibitor (NAI) and IKK inhibitor III.

[0021] In one embodiment, an additional therapeutic agent selected from the group consisting of somatostatin analogues (SSA) or mTOR inhibitors is to be further administered to the subject.

[0022] In one embodiment, the additional therapeutic agent is to be administered before, concomitantly, and after administration of the medicament.

[0023] In one embodiment, the medicament is to be administered via intravenous, intratumoral, intranodal, intradermal or subcutaneous administration.

[0024] In one embodiment, the subject has unresectable or metastatic PNETs.

[0025] In one embodiment, the subject is human.

[0026] In another aspect, the present invention provides a method of treating pancreatic neuroendocrine tumors (PNETs) in a subject, comprising administering a moiety selected from the group consisting of: (a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6; (b) a NR4A1 antagonist selected from the group consisting of 1,1-bis(3'-indolyl)-1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5-methoxy analog (DIM-C-pPhOH-3-CI-5-OCH3) and a flavonoid selected from the group consisting of kaempferol and quercetin; (c) a vector comprising a nucleic acid molecule encoding menin; and a NF-KB inhibitor, thereby reducing cellular proliferation of PNETs.

[0027] In one embodiment, the PNETs are Multiple Endocrine Neoplasia Type I (MEN1)- associated PNETs.

[0028] In one aspect, the present invention provides a moiety selected from the group consisting of: (a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6; (b) a NR4A1 antagonist selected from the group consisting of 1,1-bis(3'-indolyl)-1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5-methoxy analog (DIM-C-pPhOH-3-CI-5-OCH3) and a flavonoid selected from the group consisting of kaempferol and quercetin; (c) a vector comprising a nucleic acid molecule encoding menin; and (d) a NF-KB inhibitor, for treating pancreatic neuroendocrine tumors (PNETs) in a subject, wherein the moiety reduces cellular proliferation of PNETs in the subject, preferably the PNETs are Multiple Endocrine Neoplasia Type I (MENI)-associated PNETs.BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The invention will be better understood with reference to the detailed description when considered in conjunction with the non-limiting examples and the accompanying drawings, in which:

[0030] Fig. 1 shows that NF-KB signalling is hyperactive in pancreatic neuroendocrine tumours. (A) Schematic diagram illustrating the workflow of the single cell analysis of five healthy human pancreatic islets (EMTAB5061 , GSE73727, GSE81547, GSE81608, GSE83139) and five human PNET tissues (GSE162708, PHS003141). (B) GSEA reveals differentially regulated pathways between endocrine cells of PNETs and human islets. (C) GSEA enrichment plot of TNF-a signalling via NF-KB pathway (FDR < 0.05, Normalized Enrichment Score (NES): 2.32). (D) Western blot analysis of phosphorylated RelA (Ser536) and total RelA expression in pancreatic islets from eight healthy individuals (H2471, H2468, H2473, H2318, H2470, H1934, H2315, H2472) and ten human PNET tissues (T1 , T2, T3, T4,T5, T6, T7, T8, T9, T10). E) Immunofluorescence staining of total RelA and synaptophysin (SYP) in pancreatic islets from four healthy individuals (H2292, H2301 , H2312, H2344) and four human PNET tissues (T2, T7, T10, T11). Scale bar represents 20pm. See also Figure 8, Table 2.

[0031] Fig. 2 shows that RelA controls tumorigenic capacity of PNET cells in vitro and in vivo. (A) Western blot analysis of total and phosphorylated RelA in PNET cells BON-1 and QGP-1. (B) Immunofluorescence staining of RelA in BON-1 and QGP-1 cells. Scale bar represents 20pm. (C) Western blot analysis of total and phosphorylated RelA in control and RelA knockdown BON-1 cells. (D) Representative image of the colony forming assay to compare the colony forming capabilities in control and RelA knockdown BON-1 cells. (E) Western blot analysis of total and phosphorylated RelA in control, RelA knockdown and RelA- restored QGP-1 cells. (F) Representative image of the colony forming assay to compare the colony forming capabilities in control, RelA knockdown and RelA-restored QGP-1 cells. The number of colonies were measured using Imaged. Each dot represents the number of colonies formed in one biological replicate. Error bars represent SEM. One-way ANOVA was performed and asterisk (*) indicates p-value < 0.05. Growth curve of (G) control and RelA knockdown BON-1 cells and (H) control, RelA knockdown and RelA-restored QGP-1 cells. Each dot represents the average percentage phase confluency of two (BON-1) or three (QGP-1) biological replicates at the specified time-point. Two-way ANOVA was performed to determine statistical significance and asterisks (*) indicates p-value < 0.05. (I) Graph illustrating the percentage of Ki-67+ cells between control, RelA knockdown and RelA-restored QGP-1 cells. Each dot represents the average percentage phase confluency at the specific time-point and the error bars represent SEM. Two-way ANOVA was performed, asterisks (*) indicate p-value < 0.05. (J) Schematic illustrating the workflow of the xenograft studies. (K) Image of tumours harvested from the flanks of NOD-SCID mice injected with control and RelA knockdown QGP- 1 cells. (L) Tumour volume over a period of three weeks post tumour formation. Each dot represents the average reading of the five tumours and the error bars represent SEM. Two- way ANOVA was performed, asterisks (*) indicate p-value < 0.05. (M) Mass of the tumours derived from the control and RelA knockdown QGP-1 cells. Each dot represents the weight of one tumour and the error bar represents SEM. Student’s t-test was performed to determine the statistical significance and asterisks (*) indicates p-value < 0.05. See also Figure 9.

[0032] Fig. 3 shows that M2-like macrophages are major contributors of tumour microenvironment-derived signals that drive NF-KB hyperactivation in PNETs. (A) Schematic depicting the identification of 218 unique interactions between immune cells and PNET cells, with 13 interactions known to activate NF-KB. (B) Circles plot illustrating the strength of theinteractions between immune and PNET cells within the PNET tissue. (C) Dot plot illustrating the number of NF-KB activation interactions initiated by the respective immune cells. (D) Dot plot depicting the NF-KB ligand-receptor interactions between macrophages and PNET cells. (E) Chord diagram depicting the TNF signalling network between immune and PNET cells. (F) Schematic illustrating the analysis of macrophages in the scRNA-seq dataset (phs003141). Violin plot of the expression of NF-KB activators (TNF, TGFB1, PPIA, IGF1, HGF, FGF5) in macrophages. (G) GSEA enrichment plot of M2-like gene signature GSE13685 (FDR < 0.05, Normalized Enrichment Score (NES): 2.02 between macrophages expressing high levels of NF-KB activators (Top 25%) and macrophages expressing low levels of NF-KB activators (Bottom 25%). (H) Box and whiskers plot comparing the percentage of M2-like macrophages between normal (non-tumour) pancreas and PNET tissues. Each dot represents the percentage of M2 macrophages in one tissue sample and the error bar represents SEM. Student’s t-test was performed to determine the statistical significance and asterisks (*) indicates p-value < 0.05. See also Figure 10, Tables 2 and 3.

[0033] Fig. 4 shows that tumour suppressor menin antagonizes against NF-KB hyperactivity in human PNETs. (A) Schematic illustrating the top three prevalent mutations in PNETs accompanied with pie charts showing the respective percentages of MEN1, DAXX and ATRX mutations in 95 and 68 PNETs as reported by Scarpa et al. 2017 and Jiao et al. 2011. (B) Violin plot comparing the tumour size of non-MEN1 and MEN1 PNETs. Each dot represents the tumour size of one tumour sample. Student’s t-test was performed to determine the statistical significance and asterisks (*) indicate p-value <0.05. One-tailed t-test (p < 0.05), two-tailed t-test (p = 0.0916). (C) Schematic illustrating the transfection of PNET NF-KB luciferase reporter lines with the MEN1 plasmid followed by measuring of bioluminescence via the luciferase assay. (D) Western blotting of menin expression in BON-1 and QGP-1 cells upon transfection with increasing dosages (Ong, 50ng, 500ng and 1500ng) of menin. Luciferase activity measuring RELA-mediated transactivation in BON-1 and QGP-1 cell upon transfection with increasing dosages (Ong, 50ng, 500ng and 1500ng) of menin. Each dot represents the average reading of three biological replicates and the error bars represent SEM. One-way ANOVA was performed to determine the statistical significance and asterisks (*) indicate p-value < 0.05. (E) GSEA reveals differentially regulated pathways upon the overexpression of menin in QGP-1 cells. (F) GSEA enrichment plot of TNF-a signalling via NF-KB pathway (FDR < 0.05, Normalized Enrichment Score (NES): -2.41). (G) Heatmap depicting the leading-edge genes in the TNF-a signalling via NF-KB pathway that is downregulated upon the overexpression of menin. See also Figure 11 and Table 4.

[0034] Fig. 5 shows that tumour suppressor menin interacts with RelA to form the menin- RelA complex to suppress NF-KB-driven proliferation by repressing selected NF-KB target genes. (A) Immunofluorescence staining of menin and RelA in BON-1 and QGP-1 cells. Scale bar represents 20pm. (B) Expression of RelA and menin in the nuclear and cytoplasmic fractions of BON-1 and QGP-1 cells overexpressing menin. (C) Menin and RelA expression following the IP of menin in BON-1 cells overexpressing menin. (D) Menin and RelA expression following the IP of menin and RelA in QGP-1 cells overexpressing menin. (E) Schematic illustrating the different conditions resulting in the absence of the menin-RelA complex (shCtrl + Empty, shRelA + Empty and shRelA + MEN1) or the presence of the menin- RelA complex (shCtrl + MEN1). Representative image of the colony forming assay to compare the colony forming capabilities in (F) shCtrl + Empty (complex absent) and shCtrl + MEN1 (complex present), (G) shRelA + Empty (complex absent) and shRelA + MEN1 (complex absent). The number of colonies were measured using ImageJ. Each dot represents the number of colonies formed in one biological replicate. Error bars represent SEM. Student’s t- test was performed and asterisk (*) indicates p-value < 0.05. Growth curve of (H) shCtrl + Empty (complex absent) and shCtrl + MEN1 (complex present) and (I) shRelA + Empty (complex absent) and shRelA + MEN1 (complex absent). (J) Heatmap illustrating the selected menin-RelA targets whose expression is repressed in the presence of the menin-RelA complex but no longer repressed when the menin-RelA complex is absent. See also Figure 12.

[0035] Fig. 6 shows that menin-RelA complex localizes to the KB sites on the promoter of proliferation-associated NF-KB target genes to silence their expression. (A) Schematic illustrating the analysis of RelA ChlP-seq from 0.5h treated MEFs to identify conserved KB sites on or near the promoters of the functional menin-RelA targets and the subsequent validation via menin ChlP-qPCR on QGP-1 cells. IGV tracks showing the RelA ChlP-seq peaks at or near the promoters of menin-RelA target (B) Btg2, (C) Egr3, (D) Fosb and (E) Nr4a1. The scale used to visualize the peaks in IGV is indicated on the left side of the tracks. Menin ChlP-qPCR using primers flanking the promoters of (F) BTG2, (G) EGR3, (H) FOSB and (I) NR4A1 containing the conserved KB sites. Each dot represents the enrichment fold change of the genomic region relative to the IgG control. Error bars represent SEM. Student’s t-test was performed and asterisk (*) indicates p-value < 0.05.

[0036] Fig. 7 shows that menin mutants P320R, R415P and W423R have impaired ability to bind to RelA. (A) Schematic diagram illustrating three clinically-relevant menin mutants with missense mutations at the RelA binding domain of menin. (B) Table of the HADDOCK binding scores of WT menin and menin mutants P320R, R415P and W423R. (C) In silico dockinganalysis of the menin-RelA complex formed between RelA and WT or mutant menin. (D) Immunoblotting of menin and RelA expression following IP of menin and RelA in 293FT cells overexpressing WT menin and menin mutants P320R, R415P and W423R. (E) Polyphen analysis depicting the HumDiv and Humvar scores of the menin mutants P320R, R415P and W423R. (F) Table of the missing menin interaction residues between the mutant menin and RelA proteins. (G) Menin amino acid residue Ser381 and RelA amino acid residue Asp53 and (H) menin amino acid residue Glu384 and RelA amino acid residue Lys56. These amino acid interactions are present in WT menin but absent in P320R, R415P and W423R mutants. (I) Representative image of the colony forming assay to compare the colony forming capabilities in QGP-1 cells overexpressing empty vector, WT menin and menin mutants P320R, R415P and W423R. The number of colonies were measured using Imaged. Each dot represents the number of colonies formed in one biological replicate. Error bars represent SEM. One-way ANOVA was performed to determine statistical significance and asterisk (*) indicates p-value< 0.05. See also Figure 13.

[0037] Fig. 8 shows that scRNA-seq analysis reveals global transcriptomic differences in human pancreatic islets and PNETs. (A) UMAP analysis reveals cells from different scRNA- seq dataset following Seurat clustering and batch effect correction by project code. GSE162708 and PHS003141 represents the PNET dataset, EMTAB5061 , GSE73727, GSE81547, GSE81607 and GSE83139 represent the five human islet datasets used. (B) UMAP plots depicting the expression pattern of endocrine markers GCG, INS, SST and PPY. (C) Violin plot comparing the transcript expression of CHGA, SYP and SSTR2 in endocrine cells of human islets and PNETs. Kruskal-Wallis test was used to determine statistical significance and asterisk (*) indicates p-value < 0.05.

[0038] Fig. 9 shows that chemical inhibition of the NF-KB pathway and genetic knockdown of RelA reduce proliferation in QGP-1 cells. (A) Representative image of the colony forming assay to compare the colony forming capabilities of QGP-1 cells treated with either DMSO vehicle control, 20nM of NAI or 10nM of IKK inhibitor III. The number of colonies were measured using Imaged. Each dot represents the number of colonies formed in one biological replicate. Error bars represent SEM. Student’s t-test was performed and asterisk (*) indicates p-value < 0.05. (B) Growth curve of QGP-1 cells treated with either DMSO vehicle control, 20nM of NAI or 10nM of IKK inhibitor III. (C) Western blot analysis of total RelA expression in QGP-1 control and RelA knockdown cells. (D) Representative image of the colony forming assay to compare the colony forming capabilities of QGP-1 control and RelA knockdown cells. The number of colonies were measured using Imaged. Each dot represents the number of colonies formed in one biological replicate. Error bars represent SEM. Student’s t-test wasperformed and asterisk (*) indicates p-value < 0.05. (E) Growth curve of QGP-1 control and RelA knockdown cells. Each dot represents the average percentage phase confluency of three (NAI treated QGP-1) or two (IKK inhibitor Illi treated QGP-1) biological replicates at the specified time-point. Two-way ANOVA was performed to determine statistical significance and asterisks (*) indicates p-value < 0.05. (F) Representative flow cytometry analysis and graph of Ki-67+ cells in QGP-1 control and RelA knockdown cells. Each dot represents the percentage of Ki-67+ cells in each biological replicate. Error bars represent SEM. Student’s t- test was performed and asterisk (*) indicates p-value < 0.05. (G) Representative flow cytometry analysis of Ki-67+ cells in control, RelA knockdown and RelA reconstituted QGP-1 cells.

[0039] Fig. 10 shows that macrophages are the major immune cells secreting NF-KB activators to PNET cells. (A) Chord diagrams depicting the various signalling networks (FGF, HGF, IGF, PPIA, TGFfJ) between immune and PNET cells. (B) Schematic illustration of the CIBERSORTx analysis of RNA-seq datasets from four non-tumour pancreas (TCGA-H6- A45N-11 A-12R-A26U-07, TCGA-YB-A89D-11 A-11 R-A36G-07, TCGA-HV-A5A3-11 A-11 R- A26U-07 and TCGA-H6-8124-11A-01R-2404-07) and eight PNET tissues (TCGA-3A-A9IL- 01 A-11 R-A38C-07, TCGA-3A-A9IR-01 A-11 R-A38C-07, TCGA-2L-AAQM-01 A-11 R-A39D-07, TCGA-3A-A9IJ-01A-11 R-A39D-07, TCGA-3A-A9IS-01A-21 R-A39D-07, TCGA-3A-A9IN-01A- 11 R-A39D-07, TCGA-3A-A9IV-01 A-11 R-A41 B-07, TCGA-3A-A9IQ-01 A-11 R-A38C-07) retrieved from TCGA. (C) Box and whiskers plot comparing the total percentage of macrophages, percentage of M0 macrophages and M1 macrophages over total macrophages between normal (non-tumour) pancreas and PNET tissues. Error bars represent SEM. Student’s t-test was performed to determine statistical significance and asterisk (*) indicates p-value < 0.05.

[0040] Fig. 11 shows that the overexpression of menin, a commonly mutated tumour suppressor in PNETs, does not reduce RelA or p-RelA expression. (A) Schematic depicting the percentage of MEN1 , ATRX and DAXX mutations in 95 and 68 PNETs as reported by Scarpa et al., 2017, and Jiao et al., 2011 respectively. (B) PCA plot of QGP-1 control cells expressing the empty vector and QGP-1 cells overexpressing WT menin. (C) Western blot analysis of p-RelA and total RelA expression in QGP-1 control cells and QGP-1 cells overexpressing WT menin.

[0041] Fig. 12 shows that BTG2, EGR3, FOSB and NR4A 1 are functional targets of the menin-RelA complex and promote PNET tumorigenesis. (A) RT-qPCR validation of the knockdown of BTG2 in QGP-1 cells. (B) Representative image of the colony forming assay to compare the colony forming capabilities of QGP-1 control and BTG2 knockdown cells. (C) RT-qPCR validation of the knockdown of EGR3 in QGP-1 cells. (D) Representative image of the colony forming assay to compare the colony forming capabilities of QGP-1 control and EGR3 knockdown cells. (E) RT-qPCR validation of the knockdown of FOSB in QGP-1 cells. (F) Representative image of the colony forming assay to compare the colony forming capabilities of QGP-1 control and FOSB knockdown cells. (G) RT-qPCR validation of the knockdown of NR4A 1 in QGP-1 cells. (H) Representative image of the colony forming assay to compare the colony forming capabilities of QGP-1 control and NR4A1 knockdown cells. (I) Immunohistochemistry staining of NR4A1 on PNETs derived from patients with Multiple Endocrine Neoplasia Type I syndrome. Scale bar represents 200pm.

[0042] Fig. 13 shows the interactions between menin and RelA proteins. Interaction plots providing the detailed amino acid residues involved in the interaction between (A) WT menin, (B) menin mutant P320R, (C) menin mutant R415P, (D) menin mutant W423R, and RelA proteins.

[0043] Fig. 14 shows that NR4A1 is upregulated in human pancreatic neuroendocrine tissues. (A) Comparison of NR4A 1 transcript expression between healthy islets and MEN1 PNET tissue harbouring the R527X mutation. (B) Immunohistochemistry staining of NR4A1 in human PNET tissues and adjacent normal healthy tissues.

[0044] Fig. 15 shows that shRNA mediated depletion of NR4A1 reduces PNET tumorigenesis. (A) Western blotting of shCtrl and shNR4A1 QGP-1 cells. (B) Clonogenic assay of shCtrl and shNR4A1 QGP-1 cells. (C) Images of PNET tumours formed following administration of shNR4A1. (D) Weight of tumours harvested.

[0045] Fig. 16 shows that treatment of PNET cells with NR4A1 inhibitor DIM-C-pPhOH reduces PNET tumorigenesis. (A) Clonogenic assay following treatment of QGP-1 cells with 20pM and 40pM of DIM-C-pPhOH. (B) Growth rate of QGP-1 cells following treatment of cells with 20pM and 40pM of DIM-C-pPhOH.

[0046] Fig. 17 shows that shRNA mediated depletion of NR4A1 reduces PNET tumorigenesis. Clonogenic assay of shCtrl and shNR4A1 BON-1 cells demonstrates that the knockdown of NR4A1 reduces growth rate of BON-1 cells.

[0047] Fig. 18 shows that DIM-C-phOH impairs tumour growth in vivo. (A) Tumour growth curves of mice treated with vehicle control or DIM-C-phOH (40 mg / kg / day). (B) Representative images of tumours harvested from each treatment group. (C) Body weights of mice throughout the treatment period. (D) Quantification of tumour weights at the study endpoint.DETAILED DESCRIPTION OF THE INVENTION

[0048] The present study has demonstrated that the overexpression of wild-type menin represses cellular proliferation in QGP1 cells whereas the overexpression of menin mutants such as R527X failed to do so. The present study also showed that wild-type menin represses several NFkB target genes such as NR4A1 whereas the repression of NR4A1 was not observed in the R527X mutant. The present ChlP-qPCR analyses and a publicly available ChlP-on-chip data set revealed that wild-type menin binds onto the promoter of NR4A1. Furthermore, analysis of publicly available sc-RNA seq data set revealed that PNETs harbour higher expression of NR4A1 and this was further validated by immunostaining of NR4A1 in MEN1 -associated PNETs. The present inventors have also demonstrated that NR4A1 could be a potential therapeutic target as the reduction of NR4A1 expression either through genetic manipulation such as shRNA-mediated knockdown of NR4A1 or through chemical inhibition via the NR4A1 antagonist DIM-C-pPhOH resulted in decreased cellular proliferation in QGP1 cells. Altogether, the present findings reveal a novel link between menin and NFkB target NR4A1, and suggest that inhibiting NR4A1 and / or other NFkB targets could potentially reduce proliferation rate and PNET growth.

[0049] Apart from cancer cell-intrinsic factors such as genetic mutations, extrinsic factors such as the tumour microenvironment (TME) has emerged as another crucial driver of tumorigenesis. In the context of PNETs, tumour associated macrophages (TAMs) have been reported to be positively correlated with Ki67-index, while being negatively correlated with disease-free and disease specific survival. Thus, suggesting that extrinsic factors from the TME can be an additional factor that modulates PNET proliferation and tumorigenesis. In this regard, multiple single cell RNA-sequencing (scRNA-seq) of human PNET tissues have been performed, allowing for high-resolution delineation of the tumour and immune landscape in PNETs. However, whether cell-to-cell communication between the TME and PNET cells can activate signalling pathways that contribute to PNET proliferation and tumorigenesis was previously unknown. In addition, the molecular interplay between PNET cell-intrinsic factors and TME-derived extrinsic factors was virtually unexplored.

[0050] In the present study, to gain insights into the intrinsic and extrinsic molecular determinants of PNET proliferation and tumorigenesis, the present inventors integrated publicly-available single cell transcriptomic datasets of healthy human pancreatic islets and PNET tissues. A combination of comparative transcriptomic analyses, western blotting and immunofluorescence staining experiments revealed NF-KB signalling to be hyperactive in the endocrine cells of PNETs relative to healthy human islets. Importantly, pharmacologicalinhibition of NF-KB signalling and genetic depletion of RelA / p65 (a central player of NF-KB signalling) significantly reduced the proliferation of human PNET cell lines - BON-1 and QGP- 1 in vitro and in vivo, verifying the functional role of NF-KB signalling in driving PNET tumorigenesis. CellChat analysis on human PNET tissues further unveiled M2-like macrophages as the major immune cell type that secretes NF-KB activators, thereby highlighting the potential role of immune cell-derived extrinsic factors in driving NF-KB hyperactivation. Critically, the present study demonstrates that PNET-associated menin protein forms a hitherto complex with RelA in the nucleus to repress the transcription of proliferation-associated NF-KB target genes in PNET cells. Abrogation of this menin-RelA complex resulted in a defective repression of NF-KB target genes involved in proliferation (BTG2, EGR3, FOSB and NR4A1), and significantly diminished menin’s ability to repress PNET proliferation. Mechanistically, the present study demonstrates that the menin-RelA complex binds onto KB sites in the promoter of these genes to repress their transcription. Finally, in silica modelling of the menin-RelA complex using the High Ambiguity Driven proteinprotein (HADDOCK) method uncovered three clinically-relevant MEN1 mutations (P320R, R415P and W423R) that possess unstable conformations when docked onto RelA, suggesting possibly weaker binding affinity to RelA. Through biochemical and clonogenic assays, the present inventors further demonstrate that menin mutants P320R, R415P and W423R have diminished ability to bind RelA and consequently, are unable to repress cellular proliferation. Collectively, the present results reveal a previously unappreciated role of the menin-RelA complex as a cancer cell-intrinsic regulator that mitigates NF-KB-driven proliferation and tumorigenesis in PNETs. This hitherto menin-RelA complex represses the transcription of proliferation-associated NF-KB target genes, thereby counteracting NF-KB hyperactivation, which can be driven not only intrinsically but also by extrinsic signals originating from the tumour microenvironment.

[0051] In one aspect, the present invention provides a use of a moiety selected from the group consisting of: (a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6; (b) a NR4A1 antagonist selected from the group consisting of 1,1-bis(3'-indolyl)-1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5-methoxy analog (DIM-C-pPhOH-3-CI-5-OCHs) and a flavonoid selected from the group consisting of kaempferol and quercetin; (c) a vector comprising a nucleic acid molecule encoding menin; and (d) a NF-KB inhibitor, in the manufacture of a medicament for treating pancreatic neuroendocrine tumors (PNETs) in a subject, wherein the medicament is to be administered to the subject, thereby reducing cellular proliferation of PNETs.

[0052] In one embodiment, the PNETs are Multiple Endocrine Neoplasia Type I (MEN1)- associated PNETs.

[0053] It would be generally understood by the skilled person that the use of the vector comprising a nucleic acid molecule encoding menin would be able to treat MEN1 -associated PNETs. It would also be generally understood by the skilled person that the use of the nucleic acid encoding shRNA or siRNA specific for NR4A1, the NR4A1 antagonist and the NF-KB inhibitor would be able to treat PNETs with elevated NR4A1 expression, including non-MEN1 PNETs and MEN1-PNETs.

[0054] As used herein, the term “MEN1” refers to multiple endocrine neoplasia type 1 syndrome. Accordingly, the term “MEN1 patient” is meant to refer to a patient with MEN1 syndrome. The term “PNET” refers to pancreatic neuroendocrine tumours, or a type of cancer characterized by pancreatic neuroendocrine tumours. As used herein, the terms “MEN1- associated PNET” and “MEN1 -related PNET” are meant to refer to PNETs developed by patients with MEN1 syndrome.

[0055] In various examples, the moiety may repress the expression and / or activity of NR4A1 directly or indirectly. Examples of the moiety include a transcriptional repressor of NR4A1, a translational repressor of NR4A1 , an inhibitor of NR4A1 protein, as well as moieties that negatively regulate the expression or activity of NR4A1 through other means such as mRNA degradation and protein degradation of NR4A1. The moiety as described herein may also be an indirect repressor of NR4A1 , such as an inhibitor of a transcriptional activator of NR4A1. An example of a transcriptional activator of NR4A1 is NF-KB. In various examples, the moiety may be a vector comprising a nucleic acid molecule encoding menin, a nucleic acid encoding shRNA or siRNA specific for NR4A1, an NR4A1 antagonist, or an inhibitor of NF- KB.

[0056] As used herein, the term “treat” or “treating” in the context of treating a disease such as PNETs including MEN1 -associated PNETs is meant to include improving clinical condition of patients having the disease. This includes reducing the severity and preventing or slowing the progression of the disease. In the methods of the invention, therapy is used to provide a positive therapeutic response with respect to a disease or condition. The term “positive therapeutic response” is intended to include an improvement in the disease or condition, and / or an improvement in the symptoms associated with the disease or condition, and / or prevent the worsening of symptoms associated with the disease or condition. Positive therapeutic responses in any given disease or condition can be determined by standardized response criteria specific to that disease or condition. In addition to these positive therapeuticresponses, the subject undergoing therapy may experience the beneficial effect of an improvement in the symptoms associated with the disease.

[0057] As used herein, the terms “administered”, “administering” and “administration” refer to any method of providing a pharmaceutical preparation to a subject. Such methods are well known to those skilled in the art and include, but are not limited to, oral administration, transdermal administration, administration by inhalation, nasal administration, topical administration, intravaginal administration, ophthalmic administration, intraaural administration, intracerebral administration, rectal administration, and parenteral administration, including injectables such as intravenous administration, intra-arterial administration, intramuscular administration, and subcutaneous administration. Administration can be continuous or intermittent. In various aspects, a preparation can be administered therapeutically; that is, administered to treat an existing disease or condition. In further various aspects, a preparation can be administered prophylactically; that is, administered for prevention of a disease or condition. In one aspect, administration of a tablet refers to oral administration.

[0058] In one embodiment, the vector is an adeno-associated virus (AVV) vector, or a lentiviral vector.

[0059] As used herein, the term “vector” refers to discrete elements that are used to introduce a heterologous nucleic acid into cells for either expression or replication thereof. The vectors typically remain episo al, but can be designed to effect integration of a gene or portion thereof into a chromosome of the genome. Also contemplated are vectors that are artificial chromosomes, such as yeast artificial chromosomes and mammalian artificial chromosomes. Selection and use of such vehicles are well known to those of skill in the art.

[0060] As used herein, the term "nucleic acid" refers to a deoxyribonucleotide or ribonucleotide polymer in either single- or double-stranded form, and unless otherwise limited, encompasses known analogues of natural nucleotides that hybridize to nucleic acids in a manner similar to naturally occurring nucleotides.

[0061] A pharmaceutically acceptable carrier refers, generally, to materials that are suitable for administration to a subject wherein the carrier is not biologically harmful, or otherwise, causes undesirable effects. Such carriers are typically inert ingredients of a medicament. Typically a carrier is administered to a subject along with an active ingredient without causing any undesirable biological effects or interacting in a deleterious manner with any of the other components of a pharmaceutical composition in which it is contained. As used herein, the term “pharmaceutically acceptable carrier” refers to sterile aqueous or nonaqueous solutions, dispersions, suspensions or emulsions, as well as sterile powders for reconstitutioninto sterile injectable solutions or dispersions just prior to use. Examples of suitable aqueous and nonaqueous carriers, diluents, solvents or vehicles include water, ethanol, polyols (such as glycerol, propylene glycol, polyethylene glycol and the like), carboxymethylcellulose and suitable mixtures thereof, vegetable oils (such as olive oil) and injectable organic esters such as ethyl oleate. Proper fluidity can be maintained, for example, by the use of coating materials such as lecithin, by the maintenance of the required particle size in the case of dispersions and by the use of surfactants. These compositions can also contain adjuvants such as preservatives, wetting agents, emulsifying agents and dispersing agents. Prevention of the action of microorganisms can be ensured by the inclusion of various antibacterial and antifungal agents such as paraben, chlorobutanol, phenol, sorbic acid and the like. It can also be desirable to include isotonic agents such as sugars, sodium chloride and the like. Prolonged absorption of the injectable pharmaceutical form can be brought about by the inclusion of agents, such as aluminum monostearate and gelatin, which delay absorption. Injectable depot forms are made by forming microencapsule matrices of the drug in biodegradable polymers such as polylactide-polyglycolide, poly(orthoesters) and poly(anhydrides). Depending upon the ratio of drug to polymer and the nature of the particular polymer employed, the rate of drug release can be controlled. Depot injectable formulations are also prepared by entrapping the drug in liposomes or microemulsions which are compatible with body tissues. The injectable formulations can be sterilized, for example, by filtration through a bacterial-retaining filter or by incorporating sterilizing agents in the form of sterile solid compositions which can be dissolved or dispersed in sterile water or other sterile injectable media just prior to use. Suitable inert carriers can include sugars such as lactose.

[0062] In one embodiment, the shRNA specific for NR4A1 can be sequenced by a forward primer sequence of SEQ ID NO: 1 or 3, and a reverse primer sequence of SEQ ID NO: 2 or 4.

[0063] In one embodiment, menin expressed from the vector forms a complex with NF-KB signalling molecule RelA / p65, thereby inhibiting transactivation of NR4A1 and reducing cellular proliferation of PNETs.

[0064] As used herein, the term “transactivation” is meant to include the process by which a transactivator increases the expression of a target gene. The transactivator may recruit or enhance the activity of the cellular machinery responsible for gene transcription. An example of a transctivator is a transcription factor.

[0065] In one embodiment, each dose of the vector comprising a nucleic acid molecule encoding menin is to be administered at 50-1500 ng.

[0066] In one embodiment, the vector comprising a nucleic acid molecule encoding menin is to be administered at a frequency from once every day to once a month.

[0067] In one embodiment, the vector comprising a nucleic acid molecule encoding menin is to be administered for a period of 1 day to 1 year.

[0068] In one embodiment, each dose of the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered at 0.3 to 3 mg / kg.

[0069] In one embodiment, the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered at a frequency of about once a month.

[0070] In one embodiment, the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered for a period of 1 day to 1 year.

[0071] In one embodiment, each dose of the NR4A1 antagonist is to be administered at 2- 30mg / kg / day.

[0072] In one embodiment, the NR4A1 antagonist is to be administered at a frequency from once every day to once a month.

[0073] In one embodiment, the NR4A1 antagonist is to be administered for a period of 1 day to 1 year.

[0074] In one embodiment, the NF-KB inhibitor is selected from the group consisting of NF- KB Activation Inhibitor (NAI) and IKK inhibitor III.

[0075] In one embodiment, an additional therapeutic agent selected from the group consisting of somatostatin analogues (SSA) or mTOR inhibitors is to be further administered to the subject.

[0076] In one embodiment, the additional therapeutic agent is to be administered before, concomitantly, and after administration of the medicament.

[0077] In one embodiment, the medicament is to be administered via intravenous, intratumoral, intranodal, intradermal or subcutaneous administration.

[0078] In one embodiment, the subject has unresectable or metastatic PNETs.

[0079] In one embodiment, the subject is human.

[0080] In another aspect, the present invention provides a method of treating pancreatic neuroendocrine tumors (PNETs) in a subject, comprising administering a moiety selected from the group consisting of: (a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6; (b) a NR4A1 antagonist selected from the group consisting of 1,1-bis(3'-indolyl)-1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5-methoxy analog (DIM-C-pPhOH-S-CI-S-OCHs) and a flavonoid selected from the group consisting of kaempferol and quercetin; (c) a vector comprising anucleic acid molecule encoding menin; and (d) a NF-KB inhibitor, thereby reducing cellular proliferation of PNETs.

[0081] In one embodiment, the PNETs are Multiple Endocrine Neoplasia Type I (MEN1)- associated PNETs.

[0082] In one aspect, there is provided a moiety selected from the group consisting of: (a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6; (b) a NR4A1 antagonist selected from the group consisting of 1 ,1- bis(3'-indolyl)-1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5- methoxy analog (DIM-C-pPhOH-3-CI-5-OCH3) and a flavonoid selected from the group consisting of kaempferol and quercetin; (c) a vector comprising a nucleic acid molecule encoding menin; and (d) a NF-KB inhibitor, for treating pancreatic neuroendocrine tumors (PNETs) in a subject, wherein the moiety reduces cellular proliferation of PNETs in the subject, preferably the PNETs are Multiple Endocrine Neoplasia Type I (MENI)-associated PNETs.

[0083] In one example, there is provided a method of repressing activity of NR4A1 in a cell, comprising administering menin, or a vector comprising a nucleic acid molecule encoding menin to the cell.

[0084] In one example, the cell is a PNET cell line selected from the group consisting of QGP-1 , BON-1 and NT-3.

[0085] In one example, wherein the cell is a PNET cell inside the body of a subject.

[0086] In one example, menin forms a complex with NF-KB signalling molecule RelA / p65, thereby inhibiting transactivation of NR4A1.

[0087] In one example, the subject is human.

[0088] While NR4A1 has been studied as a drug target for pancreatic ductal adenocarcinomas (PDACs), PDACs are fundamentally different from pancreatic neuroendocrine tumours (PNETs) whereby both tumours are characteristically different and exhibit different behaviours. For example, PDACs typically have KRAS mutations whereas PNETs do not have KRAS mutations. In addition, PNETs are typically treated with somatostatin analogues, whereas this is not a treatment option for PDACs. Therefore, PDACs and PNETs are fundamentally different.

[0089] Additionally, while the inhibition of NR4A1 has shown to inhibit tumour growth in several different cancer cell types, depleting the levels of NR4A1 has also enhanced tumorigenesis in other tumours. For example, NR4A1 has been reported to be a tumour suppressor of acute myeloid leukemia and the depletion of NR4A1 together with NR4A3 in mice led to mixed myelodysplastic / myeloproliferative tumours. Similarly, NR4A1 has also been reported to be a tumour suppressor for aggressive lymphomas. A study reported thatlow NR4A1 expression in significantly associated with poorer overall survival for patients with lymphomas. Furthermore, the study also demonstrated that inducing NR4A1 expression abrogated lymphoma tumour growth in vivo.

[0090] Given that NR4A1 has both pro-tumorigenic and anti-tumorigenic function depending on cancer type, this suggests that the inhibition of NR4A1 to treat PNETs is not expected to succeed. The present inventors are the first to explore and demonstrate that the inhibition of NR4A1 is able to reduce proliferation rate in PNETs, which behaves differently from PDACs and other typical solid cancers (i.e PNETs are slow growing tumours while PDACs and other solid cancers are fast growing tumours).

[0091] The invention illustratively described herein may suitably be practiced in the absence of any element or elements, limitation or limitations, not specifically disclosed herein. Thus, for example, the terms "comprising", "including", "containing", etc. shall be read expansively and without limitation. Additionally, the terms and expressions employed herein have been used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by preferred embodiments and optional features, modification and variation of the inventions embodied therein herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention.

[0092] The invention has been described broadly and generically herein. Each of the narrower species and subgeneric groupings falling within the generic disclosure also form part of the invention. This includes the generic description of the invention with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recited herein.

[0093] Other embodiments are within the following claims and non- limiting examples. In addition, where features or aspects of the invention are described in terms of Markush groups, those skilled in the art will recognize that the invention is also thereby described in terms of any individual member or subgroup of members of the Markush group.

[0094] MATERIAL AND METHODS

[0095] Human pancreatic islet and PNET tissue specimens

[0096] Pancreatic islets from 12 healthy donors from the University of Alberta, Edmonton and 11 tissues from patients with grade 1 to 2 PNETs from the National Cancer CentreSingapore were obtained. Five PNET tissues were also obtained from Kitasato University School of Medicine, Japan, that were either from MEN1 patients or patients suspected to harbour the MEN1 syndrome. All tissue specimens were obtained following informed written consent and are covered under CIRB Ref: 2018 / 2456, A*STAR IRB 2020-096 or approved by the Ethics Committee of Kitasato University Hospital / School of Medicine (approval number: B22-118). Patients’ data were obtained through an opt-out methodology. The list of human tissues examined and the relevant clinical information such as age, gender, tumour grade and Ki-67 index are listed in Table 1 .

[0097] Cell culture

[0098] BON-1 and QGP-1 cell lines were obtained from Prof Jorg Schrader, University Medical Centre Hamburg. BON-1 cells were grown in DMEM / HAM F-12 media (Gibco) supplemented with 10% FBS (Hyclone) and 1% glutamax (Gibco). QGP-1 cells were grown in RPMI1640 (Gibco) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Hyclone) and 1% glutamax (Gibco). 293FT cells were grown in DMEM / High glucose media (Hyclone) with 10% heat- inactivated FBS (Life Technologies) and 1 % NEAA (Life Technologies).

[0099] Animal studies

[0100] Mouse studies were conducted under the guidelines and approval of the Institutional Animal Care and Use Committee (IACUC) #211660. Female NOD-SCID mice were purchased from Jackson Laboratories. 1 million control and knockdown (shRelA) QGP- 1 cells were suspended in Matrigel and injected subcutaneously into the flank of each NOD- SCID mice (five per group). Tumours were measured using the calipers every week and tumour volume was computed as length multiplied by the square of the width. Tumour weight was measured using an electronic weighing scale.

[0101] Single cell RNA-seg analyses of human islets and PNETs

[0102] Raw reads of the publicly-available datasets were downloaded from NCBI Gene Expression Omnibus (GEO) and European Nucleotide Archive (ENA), and processed as follows: lOxChromium datasets were aligned to the reference genome GRCh38 using cellranger v6.1.2 and the filtered gene count matrices were generated using the ‘count’ program. FluidigmCI and smart-seq datasets were converted from SRA to fastq using the fasterq-dump program of sra-tools v2.11.0. Adapter sequences were trimmed using TrimGalore vO.6.6 and processed fastq files were aligned to the GRCh38 reference genome with STAR v2.7.2a aligner using the following settings (-outSAMmultNmax -1 -- readNameSeparator space -outSAM unmapped Within KeepPairs -outSAMtype SAM - outSAMorder PairedKeepinputOrder) to match the alignment settings used by cellrangerv6.1 .2. Strandness of the mapped reads was assessed using the infer_experiment.py program by RSeQC v2.6.4 for proper counting of assigned genomic features. Mapped reads of the SAM files generated by the STAR alignment were assigned to genomic features using featureCounts of the subread v2.0.1 package.

[0103] Count matrices for each dataset were processed individually. Datasets were filtered for low quality cells by removing cells with less than 200 genes using Seurat v4.1.1 in R v4.1.1. For each dataset the percentage mitochondria and per cell ‘transcriptome complexity’ ( log10(nFeature_RNA) / log10(nCount_RNA) ) were determined. For doublet detection, each dataset was log-normalised, variable features were detected and scaled. HLA, immunoglobin, RNA, mitochondrial and ribosomal protein genes were removed from the detected variable features which was subsequently used for PCA analysis. The PCs selected for further processing was determined quantitatively by calculating the PC which exhibits a variation less than 0.1% from the subsequent PC. The calculated PC was used in DoubletFinder v2.0.3 when running the paramsweep_v3 function with ground truth being set to FALSE.

[0104] To integrate single cell RNA-sequencing datasets from various studies and conditions, the Seurat (v4.3.1) package in R was utilized, incorporating the Harmony algorithm to correct for batch effects. Initially, datasets from studies including EMTAB5061 , GSE73727, GSE81547, GSE81608, GSE83139, GSE162708, and phs003141 , were merged, focusing on normal and tumour pancreatic islet cells. Cells identified as doublets or belonging to non- endocrine cell types were excluded to ensure data quality.

[0105] Post-filtering, the data was normalized using Seurat's 'NormalizeData' function and identified 2,000 highly variable genes with FindVariableFeatures'. The data was then scaled, regressing out variables such as total RNA count, number of detected features, and mitochondrial gene percentage, to mitigate potential confounding effects. Principal Component Analysis (PCA) was performed on the scaled data to reduce dimensionality.

[0106] To address batch effects arising from different studies and conditions, the Harmony integration method was applied via Seurat's 'RunHarmony' function, specifying the original identity ('orig.ident') as the batch variable. Harmony aligns subpopulations across datasets by iteratively adjusting principal components, facilitating the integration of diverse single cell datasets. Subsequently, Uniform Manifold Approximation and Projection (UMAP) was employed on the Harmony-corrected embeddings to visualize the integrated data, enabling the identification of shared and distinct cellular populations across the combined datasets.

[0107] Non-epithelial cells and cells from patients with T1 D orT2D based on the meta data published by the authors were filtered out. For datasets GSE162708 and phs003141 , cells highly positive (Normalized expression > 1) for CHGA, SYP and SSTR2 were labelled asPNETs, as these are well established PNET markers. Differential gene expression between PNET and endocrine cells was performed using the ‘FinalMarkers’ function of Seurat with ‘MAST’ as test, use parameter.

[0108] RNA-seguencinq

[0109] RNA-seq was performed on QGP-1 cells by the Next Generation Sequencing Platform (Genome Institute of Singapore, A*STAR). Poly-A mRNA was enriched from 100ng of total RNA with oligo-dT beads (Invitrogen). Up to 100ng of poly-A mRNA recovered was used to construct multiplexed strand specific RNA-seq libraries as per manufacturer’s instructions (NEBNext® Ultra™ Directional RNA Library Prep Kit for Illumina® (reverse stranded). The 150-bp paired-end sequenced reads were processed using the nf-core / rnaseq v1.4.2 pipeline81 with nextflow v19.10.082. Briefly, raw reads were trimmed using Trim Galore vO.6.483 to remove low-quality bases and adapters. Trimmed reads were aligned to the human reference genome, hg38 with STAR v2.6.1d aligner84. Finally, FeatureCounts vl.6.485 is utilized for assigning reads. Principle component analysis was conducted as quality control. All downstream analyses were performed using R statistical software. Differential gene expression analysis was performed using the DESeq2 package86 where genes showing at least two-fold change relative to GFP samples in expression and FDR <0.05 (the cutoff value) were considered as having significantly altered expression. The results were visualized using the EnhancedVolcano function. All genes were ranked by their signed log p- value for gene set enrichment analysis (GSEA) using fgsea v. 3.1387using various molecular signature database datasets (Hallmark and Gene Ontology) from the Molecular Signatures Database (MSigDB) website88. RNA-seq data under series GSE242549 can be accessed via secure token: iranyioahpartmp.

[0110] Gene Set Enrichment Analysis (GSEA)

[0111] Differentially expressed genes from either scRNA-seq or bulk RNA-seq analysis was ranked by their changes in expression levels (log2FC) and used as input for subsequent analysis. GSEA was performed using either the fgsea package on R or the GSEA_4.3.2 software from Broad Institute.

[0112] CellChat analysis

[0113] To investigate cell-cell communication within PNETs, the CellChat (v2.1.2) R packages were employed. scRNA-seq data from the phs003141 dataset was analysed. Cells with unique molecular identifier (UM I) counts exceeding 10,000, detected features over 3,000, or mitochondrial gene expression above 15% were excluded to ensure data quality. The data underwent normalization using Seurat's 'NormalizeData' function, followed by identification of 2,000 highly variable genes via 'FindVariableFeatures'. The data was scaled, regressing outvariables such as UM I count, number of detected features, and mitochondrial gene percentage to mitigate potential confounding effects. Principal Component Analysis (PCA) was performed on the scaled data to reduce dimensionality, and the first 20 principal components were utilized to construct a Uniform Manifold Approximation and Projection (UMAP) for visualization. Normalized expression data and associated metadata were input into CellChat to create an object representing the cellular communication landscape. The built-in CellChatDB.human database was employed to identify potential ligand-receptor interactions. The analysis encompassed identifying overexpressed genes and interactions, computing communication probabilities, and aggregating the inferred networks. This integrated approach facilitated a comprehensive understanding of the intercellular communication networks within PNETs, highlighting key signalling pathways.

[0114] Single cell analysis of macrophages in PNET tissues

[0115] Single cell RNA-seq data from the dataset phs003141 were processed and analyzed using Seurat (v4), Harmony, and tidyverse-based packages in R. The initial dataset comprised 33,538 genes and 24,048 cells. Cells were annotated using provided metadata and filtered to retain only those annotated broadly as monocytes or macrophages and derived from specific biosamples (pnet1-4), resulting in 4,235 cells. Data normalization, identification of highly variable genes (n=2000), scaling (regressing out RNA counts, gene features, and mitochondrial percentage), and PCA dimensionality reduction were performed. Harmony integration corrected for batch effects associated with biosample origin. A UMAP embedding was generated using the first 20 Harmony-corrected dimensions. Differential gene expression analysis focused specifically on macrophages stratified by TGFB1 expression. Macrophages expressing TGFB1 were divided into top (above 75th percentile) and bottom (below 25th percentile) groups, and DEGs were identified between these groups.

[0116] Cl BERSORTx analyses

[0117] To estimate the relative abundance of macrophages in healthy human pancreas and PNETs, CIBERSORTx, a deconvolution algorithm that infers cell-type composition from bulk RNA-seq data, was performed. RNA-seq expression data from The Cancer Genome Atlas (TCGA) Pancreatic Adenocarcinoma (TCGA-PAAD) dataset was obtained using the 'TCGAbiolinks' package in R. TCGA RNA-seq data for four healthy pancreas samples (solid normal tissue) and eight PNET tissue samples were retrieved using the 'GDCquery' function, specifying the "Illumina HiSeq" platform and "normalized_results" file type. The raw data were downloaded via 'GDCdownload' and preprocessed with 'GDCprepare', extracting gene expression quantification values. The resulting expression matrix was formatted with gene identifiers and sample barcodes for downstream analysis.

[0118] For deconvolution, “LM22 signature matrix”, which profiles 22 immune cell types, including macrophage subtypes, was used. CIBERSORTx was run in 'absolute mode' with '100 permutations' to ensure robust cell fraction estimations. The estimated macrophage proportions were statistically compared between healthy pancreas and PNET samples to identify significant differences in immune infiltration.

[0119] Generation of overexpression, knockdown and luciferase reporter cells

[0120] Full length human MEN1 and RELA genes were cloned into either pCDH-3xFLAG vector containing the puromycin resistance gene or pCDNA3.1 vector containing the neomycin resistance gene to generate expression plasmids. The MEN1 mutants were generated by site- directed mutagenesis, using the WT MEN1 plasmid as the backbone. All positive clones were validated by Sanger sequencing. For transfection, 10 million BON-1 or QGP-1 cells were seeded into tissue culture treated plates and 1 Opg of the respective expression plasmids were transfected using lipofectamine 2000 one day later. 48 h after transfection, cells were selected with either 2pg / ml of puromycin or 550pg / ml of neomycin for a week. Thereafter, the cells were maintained in DMEM / HAM-F12 (BON-1) or RPMI 1640 (QGP-1) media containing 2pg / ml of puromycin or 550pg / ml of neomycin respectively.

[0121] For shRNA constructs, shRNA primers were annealed and ligated to the pLKO.1 plasmid containing puromycin resistance gene (Addgene #8453). All positive clones were validated by Sanger sequencing. To generate stable knockdown cells, the shRNA constructs were packaged into lentivirus and used to transduce BON-1 and QGP-1 cells. 48 h after transfection, cells were selected with 2pg / ml of puromycin and the knockdown cells were maintained in DMEM / HAM-F12 (BON-1) and RPMI 1640 (QGP-1) media containing 2pg / ml of puromycin.

[0122] To generate the NF-KB luciferase reporter plasmid, the NF-KB response element sequence “GGGAATTTCCGGGACTTTCGGGAATTTCCGGGACTTTCGGGAATTTCC” (SEQ ID NO: 7) was cloned into the pGL4.10[luc2] luciferase plasmid (Promega #E6651), modified with a puromycin resistance gene downstream of the SV40 late poly(A) signal. BON- 1 and QGP-1 cells were transfected with the NF-KB luciferase reporter plasmid using Lipofectamine 2000. Sequences of cloning primers are listed in Table 5.

[0123] Drug treatment

[0124] QGP-1 cells were treated with either 20nM of NAI or 10nM of IKK inhibitor III while control cells were treated with DMSO one day before phenotypic assays were carried out.

[0125] Clonogenic assay and measurement of in vitro growth rate

[0126] 1x105cells were seeded onto 6-well plates and cultured for ten days. Colonies were fixed with 100% methanol and stained with 0.5% crystal violet solution (Sigma-Aldrich). Thenumber of colonies was quantified using the Imaged software. To measure the growth rate of cells, 1x10scells were seeded onto 6-well plates and the growth rate of the cells over ten days was monitored using the Incucyte® ZOOM Live-Cell Analysis System. The growth curves were plotted using GraphPad Prism 8.

[0127] Flow cytometry analyses

[0128] Cells were dissociated with 0.25% trypsin-EDTA (Gibco, 25200-056) at 37°C and passed through a 40pm cell strainer to obtain single cells. Single cells were fixed with 4% paraformaldehyde at 4°C for 20min then blocked with FACS buffer containing 0.1 % Triton X- 100 (5% FBS in DBPS) at 4°C for 30min, followed by incubation with primary antibody at a dilution of 1:100, at4°C for 1 h. Cells were washed thrice with FACS buffer containing 1% triton X-100 and incubated with secondary antibody at a dilution of 1 :500, at 4°C for 1h. The cells were washed in FACS buffer containing Triton X-100 thrice and resuspended with DPBS. The cells were analyzed with the BD LSR II Flow Cytometer and data analysis was performed using the FlowJo v10 software.

[0129] Biochemical fractionation

[0130] 2x10sBON-1 and QGP-1 cells were seeded onto 6cm plates one day prior to experiment. Cytoplasmic and nuclear fractions were extracted from cells using the NE-PER Nuclear and Cytoplasmic Reagent kit (ThermoFisher, 78833) following manufacturer instructions with protease and phosphatase inhibitors (Sigma-Aldrich) added to the commercial buffers. Protein lysates were quantified using the BCA assay (Thermo Scientific) and the proteins were separated via sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE).

[0131] Western blotting analyses

[0132] Human pancreatic islets, PNET tissues and PNET cells were lysed on ice in M-PER (Thermo Scientific) containing protease and phosphatase inhibitors (Sigma-Aldrich). Protein lysates were quantified using the BCA assay (Thermo Scientific) and the proteins were separated via SDS-PAGE. The proteins were transferred using the Mini-PROTEAN Tetra Cell system (Bio-rad) onto PVDF membranes (Bio-rad). The membrane was blocked with blocking buffer containing 0.1% Tween-20 (5% non-fat milk in TBS) and incubated with primary antibody overnight at 4°C and with the appropriate secondary antibody for 1h. Primary and secondary antibodies used are listed in Key Resources Table (Table 7).

[0133] Immunofluorescence and immunohistochemistry staining

[0134] Primary human islets were cryo-embedded in tissue freezing medium (Leica Biosystems), mounted onto glass slides and sectioned. Sectioning was performed by the Advanced Molecular Pathology Laboratory (AMPL), A*STAR and sections were stored at -80°C. For the PNET tissues retrieved from the National Cancer Centre Singapore, the frozen PNET tissues were sectioned and mounted onto glass slides by Singhealth Tissue Repository. For QGP-1 cells, the cells were seeded onto coverslips prior to immunofluorescence staining. Cryosections and cells were fixed with 4% paraformaldehyde for 20min and then incubated with blocking buffer containing 0.1 % Triton X-100 either 5% BSA (Proliant) or donkey serum.

[0135] For immunohistochemistry staining, tissues were deparaffinized and microwaved at 95°C for 20min in citrate buffer (pH 6.0). The slides were blocked using the Endogenous Advidin / Biotin Blocking Kit (Nichirei Bioscience, Tokyo, Japan) and 1% horse serum and incubated primary antibody. Following primary antibody incubation, the tissue sections were incubated with biotinylated anti-rabbit IgG (Vector Laboratories Inc., Burlingame, CA, USA). The antibody binding was visualised using the avidin-biotin-complex peroxidase method with Vectastain ABC kit (Vector Laboratories Inc.) and DAB peroxidase substrate kit (Vector Laboratories Inc). The tissues were stained with Meyer’s hematoxylin (MutoPureChemicals, Tokyo, Japan). Primary and secondary antibodies used are listed in Key Resources Table (Table 7).

[0136] Quantitative real-time PCR assay

[0137] Total RNA was extracted using the Nucleospin RNA extraction kit (Macherey- Nagel) and cDNA was prepared using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Quantitative real time PCR was performed using the CFX384 Touch™ Real-Time PCR Detection System with iTaq™ Universal SYBR® Green Supermix (Bio-Rad). Reported fold changes were based on relative expression values calculated using the 2'AAC(T)and normalized to ACTIN expression for each sample. qPCR primers were designed using the Primer-BLAST (NCBI) software and their sequences are listed in Table 5.

[0138] Luciferase assay

[0139] The NF-KB response element was cloned into the pGL4.10 luciferase vector (Promega) and transfected in BON-1 and QGP-1 cells to obtain NF-KB reporter lines from the respective PNET cells. The NF-KB reporter cells were plated in triplicate one day prior to transfection and the cells were transfected with increasing dosages of pCDH-3xFLAG-MEN1 plasmid (Ong, 50ng, 500ng, 1500ng). The total amount of DNA for transfection was kept constant at 1500ng in all transfection reactions by adding the appropriate amounts of pCDH- 3xFLAG-empty vector. 48h after transfection, the cells were harvested and luciferase was measured using the Dual-Glo® Luciferase Reporter Assay Kit (Promega, E2920) following the manufacturer’s instructions. The readings from each triplicate were normalized to the mean of the cells transfected with Ong of pCDH-3xFLAG-MEN1 plasmid.

[0140] Co-immunoprecipitation (CoIP) assay

[0141] Cells overexpressing WT or mutant menin were seeded into two 10cm plates at a density of 10 million cells per plate. Cells were harvested once they reached -80% confluence and lysed with a modified RIPA buffer (50mM Tris-HCL pH7.4, 1% NP-40, 0.25% sodium deoxycholate, 150mM sodium chloride and 1 mM EDTA) containing protease and phosphatase inhibitors for 30min at 4°C. The cell lysates were pre-cleared with protein A / G beads (sc-2003) and rabbit IgG for 1 h at 4°C. Protein concentration in the cell lysates were quantified by Pierce™ BCA protein assay kit. A minimum of 1mg of protein was incubated with the 1 pg of the respective antibodies for pull down overnight at 4°C. Antibodies used are listed in the Key Resources Table (Table 7). Subsequently, protein A / G agarose beads (sc-2003) were used to capture the proteins bound to the antibodies. The protein bound agarose beads were washed with 1xTBS before being heated at 99°C for 5min in 60pL of SDS-PAGE loading buffer (RIPA buffer + protein and protease inhibitors, 4x SDS-loading buffer). The samples were subsequently loaded for western blot analysis.

[0142] Chromatin immunoprecipitation (ChlP)-qPCR assay

[0143] ChIP experiments were performed as previously described100. Briefly, 10 million cells were dissociated into single cells via incubation with 0.25% trypsin for 5min. Cells were cross-linked with 3.3mg / ml of dimethyl 3,3'-dithiobispropionimidate (DTBP, Sigma) and 1mg / ml of3,3'-dithiodi- propionic acid (DSP, Sigma) for 15min at room temperature. The cells were cross-linked again with 1% formaldehyde for 15min at room temperature before 0.125M glycine was added to the cells to quench the reaction. Cell lysis buffer (10mM Tris-HCL pH 8, 10mM NaCI and 0.2% NP-40) was added to lyse the cells. Nuclear lysis buffer (50mM Tris- HCI pH 8, 10mM EDTA and 1% SDS) was then added to allow for nuclear lysis. Nuclear lysates were diluted in IP dilution buffer (20mM Tris-HCI pH 8, 2mM EDTA, 150mM NaCI, 0.01% SDS and 1% Triton X-100). Protease inhibitors (0.1 mM leupeptinhemisulfate, 1 pM pepstatin and 1 mM phenylmethylsulfonyl fluoride) were added fresh to the cell lysis buffers, nuclear lysis buffers and IP dilution buffers. Sonication was performed on the Misonix Q500 sonicator (QSonica) for 12 cycles using the following settings: 30s on / 45s off, 30% power. 10pg of rabbit IgG and 50pl of protein A / G beads were incubated with the sonicated lysate for 3h at 4 °C to pre-clear the samples. After pre-clearing, 10% of the lysate volume was retrieved as input for downstream analysis. The remaining lysate was split equally and incubated with either 10pg of anti-menin antibody (A300-1500) or 10pg of rabbit IgG overnight at 4 °C. Antibodies used are listed in the Key Resources Table (Table 7). The following day, the samples were washed twice with IP wash buffer and Tris-EDTA buffer before the immunoprecipitated genomic DNA (gDNA) was eluted. Elution of gDNA was completed using the IP elution buffer (100mM NaHCO3, 1% SDS, 100mM DTT). Following elution, the sampleswere treated with RNaseA, 5M NaCI and Proteinase K to initiate reverse crosslinking and to obtain the immunoprecipitated gDNA. Menin- bound gDNA was extracted using the phenol / chloroform method and subjected to ChlP-qPCR. Sequences of the ChlP-qPCR primers are listed in Table 5.

[0144] Molecular docking and HADDOCK analysis

[0145] The apo structure of WT menin was modelled using the homology modelling server “SWISS-MODEL” using the Protein Data Bank (PDB) structure, 7O9T, as the template structure. This was because 7O9T is missing some loop regions. Structures of the menin variants P320R, R415P and W423R were also modelled using 7O9T as the template structure. The structure for RelA was retrieved from the PDB (PDB ID: 1 NFI). To observe the binding of menin and its variants to that of RelA, a protein-protein docking approach was carried out using HADDOCK (High Ambiguity Driven protein-protein DOCKing) using the default settings. Amino acid residues 305-381 of menin and residues 20-59 from the N-terminal amino acid residues of RelA were defined as the interaction sites for molecular docking. The most stable conformation of the WT menin was selected as the best pose and the stable conformations of the mutants and RelA complex that are similar in confirmation to WT menin were selected out for analysis. Interaction plots were generated using LIGPLOT.

[0146] Data and code availability

[0147] RNA-seq data under series GSE242549 can be accessed via secure token: iranyioahpartmp. scRNA-seq data of healthy pancreatic islets and PNETs were downloaded from EGA and GEO under the following accession numbers: EMTAB5061, GSE73727, GSE81547, GSE81608, GSE83139, GSE162708, PHS003141.

[0148] Mice study to assess DIM-C-phOH efficacy in inhibiting PNET tumour growth

[0149] Mouse studies were conducted in accordance with the guidelines and approval of the Institutional Animal Care and Use Committee (IACUC #211660). Male NOD-SCID mice were purchased from Jackson Laboratories. 1 million BON-1 cells were suspended in Matrigel and injected subcutaneously into the flank of each mouse (five mice per group). One week after tumour implantation, the mice were treated with either vehicle control (corn oil containing 10% DMSO) or the test compound, DIM-C-phOH, at a dose of 40 mg / kg / day via oral gavage. The compound was administered three times per week. Tumour size was measured three times weekly using digital calipers, and tumour volume was calculated using the formula: V = V = !4(ab)2, whereby a = length of tumour and b = width of tumour. Body weight of the mice were also recorded three times per week using an electronic balance, and the average weekly weight was computed for each group. After four weeks of treatment, the mice were euthanized and tumours were harvested. Five tumours were collected from the vehicle-treated group, andfour from the DIM-C-phOH-treated group (one mouse in the treatment group died prematurely and was excluded from analysis). Tumour weights were determined using an electronic balance.

[0150] EXAMPLES

[0151] Non-limiting examples of the invention and comparative examples will be further described in greater detail by reference to specific Examples, which should not be construed as in any way limiting the scope of the invention.

[0152] Example 1 : Human clinical PNETs exhibit hyperactive NF-KB signalling

[0153] To identify molecular pathways that are dysregulated in the endocrine cells of PNETs, a comparison was carried out for the transcriptome of endocrine cells derived from PNET tissues and healthy human islets at the single cell level by generating an integrated single cell RNA-sequencing (scRNA-seq) atlas from five publicly-available datasets of human pancreatic islets and five PNET tissues (Figure 1A). The endocrine cells of the PNET tissues were manually annotated as cells with at least one transcript expression of well-established PNET markers - CHGA, SYP, or SSTR2. After batch correction using Harmony, UMAP analysis demonstrated that the single cells clustered by cell type and not by datasets (Figure 8A), comprising a total of 8300 endocrine cells, with 4083 cells from the PNETs and 4217 cells from the five healthy islets. Reassuringly, distinct clusters of cells were identified that express well-established endocrine markers such as GCG (alpha cell), / MS (beta cell), SST (delta cell) and PPY (pancreatic polypeptide cell) (Figure 8B). Consistent with previous literature, a significantly higher expression of PNET markers, CHGA, SYP and SSTR2 (Figure 8C) were detected in the endocrine cells of PNET origin compared to those from healthy islets. Differentially expressed genes (DEG) analysis revealed a total of 462 upregulated and 802 downregulated genes in PNETs based on a threshold of FC ±1.5 and FDR <0.05. Gene Set Enrichment Analysis (GSEA) on this DEG gene set uncovered TNF-a signalling via NF-KB pathway as the top upregulated signalling pathway that is significantly enriched in PNETs (Figures 1 B and 1C).

[0154] To determine if the hyperactivation of NF-KB signalling can be independently generalized to other clinical PNETs at large, human islets from a total of twelve healthy individuals and eleven human clinical PNET samples (T1 - T11) (Table 1) were leveraged on to compare the NF-KB activity between the two groups. Remarkably, western blot analysis showed phosphorylated RelA (Ser536), a phosphorylation event that represents active NF-KB signalling, to be highly expressed in the ten PNETs as compared to human islets from healthy individuals (Figure 1 D). Furthermore, nuclear localization of RelA, another hallmark of activeNF-KB signalling, was observed in SYP-expressing endocrine cells of PNETs but not that of healthy islets (Figure 1 E). Together, these results indicate that NF-KB signalling is hyperactive in the endocrine cells of PNETs but not in healthy human islets.

[0155] Table 1 : Clinical information of healthy islets and PNETTable t .

[0156] Example 2: Hyperactive NF-KB signalling drives PNET tumorigenesis in vitro and in vivo

[0157] Since the NF-KB signalling pathway is widely implicated in proliferation and commonly overexpressed in a variety of cancers, it was next asked if the inhibition of this pathway can inhibit PNET tumorigenesis in vitro and in vivo. To this end, NF-KB activity was profiled in two commercially-available PNET cell lines, BON-1 and QGP-1. Both PNET celllines expressed phosphorylated RelA (Ser536) (Figure 2A) and exhibited nuclear localization of RelA (Figure 2B), indicating that NF-KB signalling is active in these cells.

[0158] To diminish NF-KB activity in PNET cells, QGP-1 cells were treated with two NF-KB inhibitors: NF-KB activation inhibitor (NAI), and IKK Inhibitor III. Treatment of QGP-1 cells with NAI or IKK inhibitor III significantly reduced growth in vitro, as assessed by the clonogenic assay (Figure 9A) and via monitoring of growth rate using the IncucyteZOOM software (Figure 9B). In addition, shRNA-mediated knockdown of RelA (Figures 2C and 9C), an essential component of the NF-KB signalling pathway, also significantly reduced the growth rate of both BON-1 and QGP-1 cells in vitro as observed via the clonogenic assay (Figures 2D and 9D). Importantly, the reconstitution of RelA in shRelA QGP-1 cells (Figure 2E) was able to rescue the impaired proliferative phenotype observed via the clonogenic assay (Figure 2F). Next, the knockdown of RelA in both BON-1 and QGP-1 cells also resulted in a significant reduction in the growth rates assessed by IncucyteZOOM (Figures 2G, 2H and 9E) and Ki-67+cells (Figures 2I, 9F and 2G), while the reconstitution of RelA in shRelA QGP-1 cells rescued these impaired proliferative phenotypes observed (Figures 2H, 2I and 9G). Subcutaneous injection of shCtrl and shRelA QGP-1 cells into immunocompromised NOD-SCID mice (Figure 2J) further demonstrated that the knockdown of RelA impaired tumour forming capacity of the cells in vivo, as observed from the smaller size, volume and growth rate of the tumours formed (Figures 2K - 2M). Collectively, these results implied that the inhibition of the hyperactive NF- KB signalling pathway can reduce proliferation and tumorigenicity of PNET cells.

[0159] Example 3: M2-like macrophage is a major contributor of tumour microenvironment- derived signals that possibly drive NF-KB hyperactivation in PNETs

[0160] Cells in the TME play an essential role in activating oncogenic signalling pathways in cancer cells through the secretion of various growth factors and cytokines. To identify the cell types in the TME that could communicate with PNET cells, and their associated ligands that may activate NF-KB signalling in PNETs, CellChat analysis was applied on four existing publicly-available PNET single cell datasets (phs003141) with pre-annotated cell identities. Based upon a p-value cut-off of <0.05, CellChat analysis revealed 218 unigue interactions between the various immune cell types and PNET cells (Table 2). Using a curated list of known NF-KB activators, 13 interactions were pinpointed that were previously reported to activate NF-KB signalling (Figure 3A and Table 3). Amongst these interactions, macrophages were identified to be the main cell type that secretes these NF-KB activators (Figures 3B and 3C; 12 out of 13 interactions detected). These interactions were predominantly cytokine and growth factor- receptor interactions (Figure 3D), and many of these cytokines or growth factorswere found to be secreted by at least two different cell types within the TME (Figures 3E and 10A).

[0161] Table 2: Cell-cell communication ligands and receptors between PNET and immune cellsTable 2.

[0162] Table 3: List of NF-KB activators secreted by immune cells to PNET cellsTable 3.

[0163] Given that macrophages are the primary source of NF-KB activators in the PNET TME (Figures 3B and 3C), it was next examined which macrophage subtype could have contributed to this activity. Using pre-annotated cell identities, macrophages were subclassified into 'low,' 'medium,' and 'high' groups based on their expression of key NF-KB activators (TNF, TGFB1 , PPIA, IGF1 , HGF, FGF5) (Figure 3F). GSEA revealed that macrophages with high NF-KB activator expression were significantly enriched for M2-like gene signatures compared to those with low expression (Figure 3G), indicating that M2-like macrophages are the predominant source of NF-KB activators in PNETs. To validate this observation, CIBERSORTx analysis was performed on bulk RNA-seq datasets from a separate cohort of non-tumour pancreas and PNET tissues available publicly at TCGA to estimate the proportion of M2-like macrophages in the PNET TME (Figure 10B). While there were no significant differences between the total percentage of macrophages between healthy pancreas and PNET tissues (Figure 10C), a significantly higher proportion of M2-like macrophages was found in PNETs compared to healthy islets (Figure 3H) and a significant reduction in MO-like macrophages (Figure 10C). Collectively, these results highlight that M2- like macrophages are enriched in the PNET TME, and that they are possibly a major contributor of NF-KB activators within the TME.

[0164] Example 4: Tumour suppressor menin antagonizes NF-KB hyperactivity in human PNETs

[0165] Having identified a potential role for extrinsic TME in driving NF-KB hyperactivity, it was next sought to investigate if any of the cancer cell intrinsic mutations prevalent in PNETs can modulate NF-KB hyperactivity. Previous whole genome and whole exome sequencing studies on PNETs revealed MEN1 as the most prevalent gene mutation in PNETs (Figures 4A and 11A). Additionally, meta-analysis of human clinical PNETs from two separate cohorts revealed that MEN1 mutations correlated with a larger tumour size (Figure 4B and Table 4). Thus, it was hypothesized that wild-type (WT) menin could potentially repress PNET growth by antagonizing NF-KB hyperactivity in human PNETs cells.

[0166] Table 4: Clinical information of non-MEN1 and MEN1 PNETsTable 4.

[0167] To test this hypothesis, an NF-KB luciferase reporter line was generated by transfecting BON-1 and QGP-1 cells with a luciferase vector containing four tandem repeats of the NF-KB response element sequence upstream of a Iuc2 gene (Figure 4C). Transfection of increasing dosages of menin to the reporter lines resulted in a significant repression of NF- KB transactivation in a dosage-dependent manner for both PNET lines (Figure 4D). RNA-seq analyses on QGP-1 cells overexpressing either the empty vector (QGP-1 Empty) or WT menin (QGP-1 MEN1) demonstrated global transcriptomic differences (Figure 11 B). GSEA on this dataset revealed TNF-a signalling via NF-KB to be the top downregulated signalling pathway upon menin overexpression (Figures 4E and 4F), with 50 genes in the TNF-a signalling via NF-KB pathway being repressed by menin (Figure 4G). Notably, the overexpression of menin did not alter the protein expression levels of RelA or p-RelA (Figure 11 C), indicating that menin did not target RelA for degradation or diminish the phosphorylation of RelA. Collectively, these data indicate that menin antagonizes NF-KB transactivation activity without decreasing the expression of total RelA or p-RelA.

[0168] Example 5: Tumour suppressor menin interacts with RelA to form the menin-RelA complex to suppress NF-KB-driven proliferation by repressing selected NF-KB target genes

[0169] As menin is well-known to be a scaffold protein, it was next asked if menin interacts with RelA at the protein level to modulate its ability to transactivate target genes associated with cellular proliferation. Immunofluorescence staining of menin and RelA revealed the colocalization of both proteins in the nucleus of PNET cells (Figure 5A). Fractionation of PNET cell lysates from BON-1 and QGP-1 cells overexpressing menin also confirmed the localization of RelA and menin exclusively in the nucleus (Figure 5B). Co-immunoprecipitation (CoIP) studies done on both PNET cell lines further confirmed that menin interacts with RelA at the protein level, forming a menin-RelA complex (Figures 5C and 5D).

[0170] To determine if the menin-RelA complex plays a functional role in PNETs, menin was overexpressed in both RelA-sufficient (shCtrl) and RelA-deficient (shRelA) QGP-1 cells wherein the menin-RelA complex is either intact or diminished, respectively (Figure 5E). It was found that the overexpression of menin and formation of menin-RelA complex in shCtrl QGP-1 cells significantly reduced cellular proliferation as observed from the fewer colonies in colony forming assay (Figure 5F). In contrast, when menin was overexpressed in shRelA QGP-1 cells, no repression of cellular proliferation was observed (Figure 5G). This finding is also consistent when measuring the growth rate of QGP-1 cells, where cells with intact menin-RelA complex had a slower growth rate (Figure 5H), while cells lacking the menin-RelA complex displayed no significant difference in terms of cell growth (Figure 5I). Together, these findings implicate the menin-RelA complex as a crucial regulator of NF-KB-driven proliferation in PNETs.

[0171] To then determine if the menin-RelA complex suppresses proliferation by transcriptionally repressing genes in the NF-KB pathway, the expression levels of NF-KB target genes upon the overexpression of menin in both RelA-sufficient and RelA-deficient cells was evaluated next. The analysis returned seven NF-KB genes (NR4A1, EGR3, SIK1, FOSB, DUSP5, EGR2 and BTG2) whose expression is significantly repressed in the presence of the menin-RelA complex, but not when the complex is abrogated (Figure 5J). Knockdown of four of these target genes (BTG2, EGR3, FOSB and NR4A1) significantly reduced cellular proliferation as assayed by clonogenic assays (Figures 12A - 12H), suggesting that these are functional targets of the menin-RelA complex in PNET cells. Immunohistochemistry staining of one of these targets, NR4A1 , on five MEN1 mutated PNETs and adjacent non-tumour tissue revealed higher NR4A1 expression in the PNET tissues as compared to the adjacent nontumour tissue (Figure 121). Altogether, these findings demonstrate that the menin-RelA complex plays a functional role in PNET tumour suppression by repressing the transcriptional activity of selected NF-KB targets implicated in PNET growth.

[0172] Example 6: Menin-RelA complex localizes to the KB binding site on the promoters of proliferation-associated NF-KB target genes to silence their expression

[0173] Given that the menin-RelA complex represses the transcriptional activity of several NF-KB target genes (Figure 5J), it was next hypothesized that the menin-RelA complex may localize to KB binding sites within the promoters of these genes to regulate their expression. To investigate this, publicly-available RelA ChlP-seq data performed on TNF-a-treated mouse embryonic fibroblasts (MEFs) was first analysed to identify RelA bound regions. Specifically, the focus was on RelA-bound KB sites within promoter regions that are evolutionarily conserved between the mouse and human genomes (Figure 6A). Visualization of the RelA ChlP-seq dataset revealed binding of RelA to the proximal promoter of Btg2, Egr3, Fosb and Nr4a1, with at least one KB binding site within each of the RelA ChlP-seq peaks detected (Figures 6B - 6E). To evaluate if menin co-localizes to the same region, menin ChlP-qPCR was performed using primers flanking or in proximity to the conserved KB binding sites in QGP-1 cells with intact menin-RelA complex. Indeed, menin ChlP-qPCR confirmed the binding of menin protein onto these RelA-bound KB binding sites (Figures 6F - 6I). Altogether, these findings demonstrate that menin and RelA co-localize at the same KB binding site within the promoter regions of proliferation-associated NF-KB target genes, supporting the hypothesis that the menin-RelA complex directly binds these promoters to repress their gene expression.

[0174] Example 7: Menin mutants P320R, R415P and W423R have impaired ability to bind to RelA and fail to repress PNET proliferation

[0175] Having established the functional relevance of the menin-RelA complex, we next sought to understand the structural basis of the menin-RelA interaction. To do so, we first modelled the interaction between WT menin and RelA using the High Ambiguity Driven protein-protein docking (HADDOCK) method to characterize the native binding interface and underlying amino acid-level interactions between the two proteins (Figure S6A). Next, we profiled MEN1 missense mutations reported in ClinVar and selected those containing single nucleotide polymorphisms (SNPs) within amino acid residues 276-479, a region previously reported to be important for menin binding onto RelA, for HADDOCK analysis (Figure 7A). This approach identified three menin mutants (P320R, R415P, and W423R) that exhibited higher HADDOCK scores as compared to WT menin (Figure 7B), suggesting reduced structural stability (Figure 7C) and potentially reduced binding affinity to RelA. CoIP experiments confirmed that the menin mutants P320R, R415P and W423R exhibit diminished ability to interact with RelA (Figure 7D), and PolyPhen analysis predicted all three menin variants as highly pathogenic (Figure 7E). Interaction mapping analyses revealed the loss of several key contacts in the respective menin mutants (Figures S6B - S6D). Notably, the menin-RelA interactions at Ser381-Asp53 and Glu384-Lys56 were consistently absent across all three menin mutants (Figures 7F - 7H), suggesting that these interactions are critical for maintaining the menin-RelA complex. Clonogenic assays performed on QGP-1 cells further demonstrated that all three menin mutants, which have reduced binding affinity to RelA, were significantly less effective at suppressing cell growth compared to WT menin (Figure 7I). Together, these findings unveil potential amino acid-level interactions between menin and RelA that are crucial for sustaining the menin-RelA interaction and further exemplify the functional role of the menin-RelA complex in suppressing PNET proliferation and tumorigenesis.

[0176] Example 8: NR4A1 is uprequlated in PNETs

[0177] NR4A1 is upregulated in human MEN1 PNET tissues as compared to healthy nontumour tissues, suggesting that NR4A1 can be a target (Figure 14). Quantitative PCR (qPCR) revealed that NR4A1 transcript levels were elevated in MEN1 PNET tissue carrying the R527Xmutation as compared to healthy islets (Figure 14A). Consistently, immunohistochemistry staining showed increased NR4A1 protein expression in human MEN1 PNET tissue relative to adjacent non-tumour tissue.

[0178] Example 9: Use of shRNAs targeting NR4A1 can reduce PNET tumoriqenesis

[0179] The knockdown of NR4A1 using shRNAs targeting NR4A1 in PNET cell lines results in reduced growth rate in vitro and reduced PNET tumorigenesis in vivo (Figure 15). Western blotting revealed reduced NR4A1 protein expression in the PNET cells transfected with shNR4A1 plasmids (Figure 15A). Clonogenic assay confirmed a reduction in the number of colonies formed in the shNR4A1 cells, indicating a reduction in growth rate of the cells in vitro (Figure 15B & Figure 17). Moreover, subcutaneous injection of shCtrl and shNR4A1 QGP-1 cells into immunocompromised NOD-SCID mice revealed that the depletion of NR4A1 markedly impaired tumorigenic potential in vivo as evidenced by reduced tumour size and volume (Figures 15C and 15D).

[0180] Example 10: Use of NR4A1 inhibitors such as DIM-C-pPhOH can reduce PNET tumoriqenesis

[0181] Treatment of PNET cell lines with the NR4A1 inhibitor DIM-C-pPhOH resulted in a dosage dependent reduction in PNET cells (Figure 16). Treatment of the PNET cells with the NR4A1 inhibitor DIM-C-pPhOH resulted in reduced growth rate of the cells as observed by the reduced number of colonies formed from the clonogenic assay (Figure 16A) and slower growth rate of the cells (Figure 16B).

[0182] Example 11 : DIM-C-phOH impairs tumour growth in vivo

[0183] Next, the efficacy of the NR4A1 inhibitor, DIM-C-phOH, in suppressing PNET tumorigenesis in vivo was assessed. 1 million BON-1 PNET cells were subcutaneously injected into the flanks of NOD-SCID mice. After one week, when tumours became palpable, mice were administered either vehicle control (corn oil) or DIM-C-phOH (40 mg / kg / day) by oral gavage three times per week for three weeks. Tumour dimensions were measured using vernier calipers, and volumes of the tumours were calculated (Figure 18A). Treatment with DIM-C-phOH resulted in reduced tumour growth and smaller tumour size compared to vehicle- treated mice (Figure 18B and 18D). Importantly, no significant change in body weight was observed in DIM-C-phOH treated mice (Figure 18C), suggesting that the compound is well tolerated.

[0184] Example 12: Discussion

[0185] The present inventors integrated and analysed publicly-available scRNA-seq data performed on healthy human islets and PNET tissues to advance the mechanistic understanding of the molecular players underlying PNET development and tumorigenesis.Comparative transcriptomic analyses performed on endocrine cells derived from the two tissues identified oncogenic NF-KB signalling to be specifically hyperactivated in PNET endocrine cells compared to healthy human islets. These findings were validated in an independent cohort of PNET tissue and healthy islets by western blotting and immunofluorescence staining. To assess the functional relevance of NF-KB signalling in PNETs, the present inventors inhibited the pathway using both pharmacological inhibitors (NAI and IKK inhibitor) and shRNA-mediated depletion of RelA, a central mediator of NF-KB activity. These interventions significantly reduced the proliferative rates and tumour formation capacity of human PNET cell lines (QGP-1 and BON-1) in vitro and in vivo, highlighting hyperactive NF-KB signalling as a significant contributor to PNET tumorigenesis.

[0186] Over the past decades, the TME which represents a tumour cell extrinsic factor, has garnered traction as a crucial factor that drives tumorigenesis in a variety of cancers. With the advent of single cell technologies such as scRNA-seq, two separate groups have independently characterized the diverse heterogenous cell populations within the PNET TME. However, whether and how these cell populations may influence signalling pathways to modulate PNET progression remained unclear. To explore whether oncogenic NF-KB hyperactivation in PNETs could be driven by extrinsic cues from the TME, the present inventors performed CellChat analysis on the scRNA-seq dataset from PNET tissues. Our analysis revealed M2-like tumour-associated macrophages (TAMs) as a key contributor of NF- KB activators via secretion of cytokines such as TNF. Through CIBERSORTx analysis on TOGA datasets, the present inventors further found the proportion of M2-like TAMs to be significantly enriched in PNET tissues relative to non-tumour pancreas tissues. Notably, prior studies have linked higher TAM abundance to increased tumour grade, Ki-67 index, and poorer survival outcomes of PNET patients. Moreover, depletion of the macrophage growth factor CSF-1 in RIP1-Tag2 mice (a transgenic mouse model of PNET), significantly reduced tumour growth at the endocrine pancreas. While these studies established a phenotypic role for macrophages in PNET progression, they did not propose a possible mechanism underlying these observations. Here, the present inventors’ findings bridge this gap by establishing a link between macrophage-derived NF-KB activators and oncogenic NF-KB signalling in PNET cells.

[0187] Much of our current understanding of PNETs has traditionally been shaped by whole genome and exome sequencing studies, which provided insights into the mutational landscape of PNETs, including the importance of loss of MEN1 being sufficient to promote malignant transformation. Therefore, in addition to the role of the extrinsic TME in driving NF- KB hyperactivity, the present inventors also studied and identified a cancer cell-intrinsicmechanism that modulates NF-KB activity in PNETs. The present inventors found that menin, a tumour suppressor mutated in approximately 50% of PNETs, represses NF-KB transactivation by forming a complex with RelA. In the absence of RelA, this complex does not form, and menin fails to suppress PNET proliferation under such conditions. The present study is the first to demonstrate the functional relevance of menin in suppressing PNET proliferation. Mechanistically, the menin-RelA complex binds to KB sites on the promoters of NF-KB target genes involved in cell proliferation, such as BTG2, EGR3, FOSB and NR4A1, to repress their transcription. While the exact mechanism through which the menin-RelA complex represses these target genes remains unclear, menin is known to recruit epigenetic histone modifiers, suggesting that the menin-RelA complex may recruit epigenetic modulators to transcriptionally silence these genes. In line with this hypothesis, previous immunohistochemistry experiments performed on clinical samples have revealed epigenetic modulators such as histone deacetylases to be overexpressed in PNETs, supporting the involvement of epigenetic regulation in controlling PNET tumorigenesis. The exact identity of these epigenetic modifiers that are recruited to the menin-RelA complex to transcriptionally silence these target genes warrant further future studies.

[0188] Finally, to understand the structural basis of menin-RelA interactions, the present inventors performed protein-protein docking analysis using HADDOCK and uncovered the underlying amino acid-level interactions between these two proteins. The inventors identified three menin mutants (P320R, R415P and W423R) which exhibited less favourable docking scores, indicating that they are likely to be in an unstable conformation when bound to RelA. CoIP experiments confirmed that these mutants have diminished ability to form the menin- RelA complex. The inventors then demonstrated that these menin mutants which harbour diminished ability to form the menin-RelA complex, also have reduced ability to suppress PNET proliferation. These findings suggest that the menin-RelA complex is perturbed in at least a subset of / WE / -mutated PNETs, thereby demonstrating the clinical relevance of this complex in suppressing PNET development. Importantly, the present structural modelling studies also unveiled two critical amino acid-level interactions between menin-RelA at Ser381-Asp53 and Glu384-Lys56 that are universally lost in all three menin mutants (P320R, R415P and W423R) with reduced binding affinity to RelA. Further studies could evaluate the functional and structural role of these interactions in enabling menin-RelA complex formation.

[0189] In summary, the present study uncovered NF-KB hyperactivation as an oncogenic driver in PNETs, mediated by both tumour-extrinsic and tumour-intrinsic mechanisms. The present inventors demonstrated that M2-like TAMs within the PNET TME is the dominant source of NF-KB-activators, providing the first documented link for possible interplay betweenTME and oncogenic signalling in PNETs. Concurrently, the present inventors identified the tumour suppressor menin as a critical cancer cell intrinsic regulator of NF-KB transactivation for a subset of NF-KB target genes involved in PNET proliferation. Menin achieves this function through its interaction with RelA to repress the transcription of proliferation-associated NF-KB target genes. Disruption of the menin-RelA complex, either by RelA loss or through MEN1 mutations, impairs this regulatory complex that safeguards against PNET proliferation. Altogether, the present findings highlight a previously uncharacterized molecular crosstalk between oncogenic NF-KB signalling and menin-RelA complex in PNETs. Future studies could explore therapeutic strategies that restore or mimic menin-RelA function, and investigate the precise epigenetic modifiers recruited by this complex, which may offer new avenues for targeted intervention in PNETs.

[0190] Example 13: Sequences

[0191] Table 5 provides embodiments of sequences used in the present study.

[0192] Table 5. List of primersTable 5. Sequences

[0193] Table 6 provides embodiments of sequences used in the present study.Table 6. Sequences

[0194] The shRNAs used to target NR4A1 are commercially available ones bought from https: / / sg.idtdna.com / paqe Integrated DNA Technologies (IDT). SEQ ID NOs 1-4 are the primers to anneal to get the shRNA oligos.

[0195] Table 7: Key Resources TableTable 7. Key Resources Table

[0196] While embodiments of the invention have been particularly shown and described with reference to specific embodiments, it should be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims. The scope of the invention is thus indicated by the appended claims and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced.

Claims

CLAIMS1 . Use of a moiety selected from the group consisting of:(a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6;(b) a NR4A1 antagonist selected from the group consisting of 1 , 1 -bis(3'-indolyl)- 1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5- methoxy analog (DIM-C-pPhOH-3-CI-5-OCH3) and a flavonoid selected from the group consisting of kaempferol and quercetin;(c) a vector comprising a nucleic acid molecule encoding menin; and(d) a NF-KB inhibitor, in the manufacture of a medicament for treating pancreatic neuroendocrine tumors (PNETs) in a subject, wherein the medicament is to be administered to the subject, thereby reducing cellular proliferation of PNETs.

2. The use of claim 1 , wherein the PNETs are Multiple Endocrine Neoplasia Type I (MENI)-associated PNETs.

3. The use of claim 1 or 2, wherein the vector is an adeno-associated virus (AVV) vector, or a lentiviral vector.

4. The use of claim 1 or 2, wherein the shRNA specific for NR4A1 can be sequenced by a forward primer sequence of SEQ ID NO: 1 or 3, and a reverse primer sequence of SEQ ID NO: 2 or 4.

5. The use of any one of claims 1 to 3, wherein menin expressed from the vector forms a complex with NF-KB signalling molecule RelA / p65, thereby inhibiting transactivation of NR4A1 and reducing cellular proliferation of PNETs.

6. The use of any one of claims 1 to 3, wherein each dose of the vector comprising a nucleic acid molecule encoding menin is to be administered at 50-1500 ng.

7. The use of claim 6, wherein the vector comprising a nucleic acid molecule encoding menin is to be administered at a frequency from once every day to once a month.

8. The use of claim 7, wherein the vector comprising a nucleic acid molecule encoding menin is to be administered for a period of 1 day to 1 year.

9. The use of any one of claims 1 , 2 and 4, wherein each dose of the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered at 0.3 to 3 mg / kg.

10. The use of claim 9, wherein the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered at a frequency of about once a month.

11. The use of claim 10, wherein the nucleic acid encoding shRNA or siRNA specific for NR4A1 is to be administered for a period of 1 day to 1 year.

12. The use of claim 1 or 2, wherein each dose of the NR4A1 antagonist is to be administered at 2- 30mg / kg / day.

13. The use of claim 12, wherein the NR4A1 antagonist is to be administered at a frequency from once every day to once a month.

14. The use of claim 13, wherein the NR4A1 antagonist is to be administered for a period of 1 day to 1 year.

15. The use of claim 1 or 2, wherein the NF-KB inhibitor is selected from the group consisting of NF-KB Activation Inhibitor (NAI) and IKK inhibitor III.

16. The use of any one of claims 1 to 15, wherein an additional therapeutic agent selected from the group consisting of somatostatin analogues (SSA) or mTOR inhibitors is to be further administered to the subject.

17. The use of claim 16, wherein the additional therapeutic agent is to be administered before, concomitantly, and after administration of the medicament.

18. The use of any one of claims 1 to 17, wherein the medicament is to be administered via intravenous, intratumoral, intranodal, intradermal or subcutaneous administration.

19. The use of any one of claims 1 to 18, wherein the subject has unresectable or metastatic PNETs.

20. The use of any one of claims 1 to 19, wherein the subject is human.

21. A method of treating pancreatic neuroendocrine tumors (PNETs) in a subject, comprising administering a moiety selected from the group consisting of:(a) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6;(b) a NR4A1 antagonist selected from the group consisting of 1 , 1 -bis(3'-indolyl)-1 - (p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5-methoxyanalog (DIM-C-pPhOH-3-CI-5-OCH3) and a flavonoid selected from the group consisting of kaempferol and quercetin;(c) a vector comprising a nucleic acid molecule encoding menin; and(d) a NF-KB inhibitor, thereby reducing cellular proliferation of PNETs.

22. The method of claim 21 , wherein the PNETs are Multiple Endocrine Neoplasia Type I (MENI)-associated PNETs.

23. A moiety selected from the group consisting of:(e) a nucleic acid encoding shRNA or siRNA specific for NR4A1 comprising the sequence of SEQ ID NO: 5 or SEQ ID NO: 6;(f) a NR4A1 antagonist selected from the group consisting of 1 , 1 -bis(3'-indolyl)- 1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), resveratrol, a 3-chloro-5- methoxy analog (DIM-C-pPhOH-3-CI-5-OCH3) and a flavonoid selected from the group consisting of kaempferol and quercetin;(g) a vector comprising a nucleic acid molecule encoding menin; and(h) a NF-KB inhibitor, for treating pancreatic neuroendocrine tumors (PNETs) in a subject, wherein the moiety reduces cellular proliferation of PNETs in the subject, preferably the PNETs are Multiple Endocrine Neoplasia Type I (MENI)-associated PNETs.

Images