Commposition comprising a TGF-beta inhibitor for use in the treatment of tumours

Abstract

The invention relates to methods and compositions for the prevention, management and / or treatment of tumours, and in particular tumours associated with Neurofibromatosis Type 1 (NF1). The methods and compositions inhibit TGF-β (and / or disrupting the TGF-β signalling) in a subject.

Description

[0001] COMMPOSITION COMPRISING A TGF-BETA INHIBITOR FOR USE IN THE TREATMENT OF TUMOURS

[0002] Technical Field of the Invention

[0003] This invention relates to compositions and methods for the prevention, management and / or treatment of tumours by inhibiting TGF-p. The compositions and methods are particularly suited to the prevention, management and / or treatment of tumours which are initiated and / or mediated by NF1 -neurofibroma formation.

[0004] Background to the Invention

[0005] Neurofibromatosis Type 1 (NF1) is a common tumour predisposition syndrome characterised by loss of function of the NF1 gene. NF1 encodes neurofibromin, a tumour suppressor that negatively regulates Ras by activating its intrinsic GTPase activity. Dysfunction of this tumour suppressor directly leads to hyperactivation of Ras which drives NF1 -associated neurofibroma development (Le & Parada, 2007). The Schwann cell (SC) has been identified as the cell of origin of these tumours (Parrinello & Lloyd, 2009; Zhu et al., 2002). SCs are the most abundant glial cell type of the peripheral nervous system (PNS). Existing in peripheral nerves as myelinating or non-myelinating SCs, they provide a crucial supportive role to the PNS during homeostasis by ensheathing, myelinating, and providing structural and metabolic support to PNS axons. SCs also play a crucial role during peripheral nerve regeneration; in response to injury, their intrinsic plasticity underpins their remarkable ability to de-differentiate into a transient progenitor-like state, which orchestrates the regeneration process (K. R. Jessen et al., 2015; K. R. Jessen & Mirsky, 2019). This feature of SCs endows the PNS with regenerative capabilities in contrast to the central nervous system. However, this plasticity also renders SCs prone to tumourigenic transformation. Indeed, it has been reported that the tumour-forming potential of SCs can be induced with two coinciding events: 1) genetic loss of the Nf 1 tumour suppressor, and 2) signaling from the injury microenvironment (Ribeiro et al., 2013).

[0006] Genetic and epigenetic abnormalities underpin all cancers; these alterations allow cells to transgress the normal limits of cellular behaviour, driving their transformation into tumour cells (Hanahan & Weinberg, 2011). However, tumour formation and growth are also reliant on the complex network of interactions between tumour cells and non-neoplastic tissue, collectively constituting the tumour microenvironment (TME), which differs substantially between tumour type and tumour stage. The TME comprises a heterogenous mixture of cellular and non-cellular components including immune cells, inflammatory cells, fibroblastic cells, endothelial cells, pericytes, nerve cells, and extracellular matrix (ECM) components (Hass et al., 2020; Whiteside, 2008). Bi-directional communication between tumour cells and the TME play a pivotal role in determining tumour formation, progression, and spread. Indeed, microenvironmental signals are crucial to the formation of neurofibroma. Studies have documented that loss of Nf1 during SC development using Krox20-Cre:Nf1fl / - mice only leads to neurofibroma formation in an Nf1 + / - background, implying that contributions of stromal Nf1 -heterozygous cells are imperative for tumourigenesis, without which tumourigenesis cannot be achieved (Zhu et al., 2002). Consistent with a role for stromal cells, work from our lab has demonstrated that loss of Nf 1 in SCs in adulthood is insufficient to induce the formation of neurofibromas. However, when combined with nerve injury, tumour formation occurred frequently, indicating that signals derived from the injured microenvironment are required (Ribeiro et al., 2013). Consistent with this, a close link between the structure of neurofibroma and regenerating peripheral nerve, suggesting that mechanisms of nerve repair are hijacked by tumourigenesis (Fletcher et al., 2020; Margaret R. Wallace, 2000; Ribeiro et al., 2013).

[0007] Currently, it is unknown what specific signal within the injury microenvironment synergises with Nf1 -deficiency to cause SC tumourigenic transformation. Identification of this signal would provide valuable insight into the mechanism by which microenvironmental and genetic elements combine to generate a tumour-forming phenotype. If druggable, this target could provide a novel therapeutic approach in the treatment of peripheral nerve tumours.

[0008] An object of the present invention is to provide an improved treatment of tumours. It would be desirable if the treatments could be used to treat tumours which are initiated and / or mediated by NF1 neurofibromatosis formation. It would also be advantageous if such treatments were compatible and / or complementary with existing tumour treatments.

[0009] Summary of the Invention

[0010] According to an aspect of the invention, there is provided a composition comprising an inhibitor of Transforming Growth Factor Beta (TGF-P) for use in the prevention, management and / or treatment of tumours.

[0011] According to a related aspect of the invention, there is provided a composition for use in a method of prevention, management and / or treatment of tumours, comprising administering an inhibitor of TGF-p to an individual.

[0012] In another related aspect of the invention, there is provided a composition comprising an inhibitor of TGF-p for use as a tumour medicament. In yet another related aspect of the invention, there is provided the use of a composition comprising an inhibitor of TGF-p, for use in the manufacture of a medicament for the prevention, management and / or treatment of tumours.

[0013] The tumour(s) may be malignant.

[0014] The term “inhibitor” is intended to mean any molecule which limits, prevents or blocks the action or function of a target protein. The term is also intended to mean any molecule which limits, prevents or blocks the action of any downstream pathway (or pathways) of a target protein. The term is further intended to mean any molecule which limits, prevents or blocks the action of any cells and / or cellular organelle which produces a target protein. The target protein is TGF-p. The inhibitor may comprise multiple inhibitors or a single inhibitor. The inhibitor may comprise a first inhibitor directed towards TGF-p and a second inhibitor directed towards TGF-p. Preferably, the inhibitor functions as an antagonist, which may be selective.

[0015] The inhibitor of TGF-p may modulate TGF-p signaling.

[0016] The inhibitor of TGF-p may modulate and / or inhibit macrophages which are (or are capable of) producing TGF-p. It will be apparent to the skilled address that the modulation and / or inhibition of macrophages may be achieved in a number of ways, such as inhibiting the Colony Stimulating Factor 1_ (CSF1) receptor.

[0017] The tumour may be associated with Neurofibromatosis Type 1 (NF1). The tumour may originate in a Schwann cell (SC). The tumour may be in the peripheral nervous system (PNS).

[0018] The composition may be for administration prior to, during or after injury to the PNS. Such injury could be a wound or other type of physical or physiological damage.

[0019] The composition may be for administration prior to, during or after excision of a pre-existing tumour located within, or adjacent to at least part of the PNS. The composition may be for administration to the area surrounding a pre-existing tumour prior to excision of the tumor and / or to the area previously surrounding the pre-existing tumor immediately after excision.

[0020] The composition may be administered in combination with, or may further comprise, one or more additional anti-cancer agents.

[0021] The atopic dermatitis may be chronic atopic dermatitis. In another aspect of the present invention, there is provided a composition comprising an inhibitor of TGF-p for use in the prevention, management and / or treatment of a neurofibroma.

[0022] According to a related aspect of the invention, there is provided a composition for use in a method of prevention, management and / or treatment of a neurofibroma, comprising administering an inhibitor of TGF-p to an individual.

[0023] In another related aspect of the invention, there is provided a composition comprising an inhibitor of TGF-p for use as a neurofibroma medicament.

[0024] In yet another related aspect of the invention, there is provided the use of a composition comprising an inhibitor of TGF-p, for use in the manufacture of a medicament for the prevention, management and / or treatment of a neurofibroma.

[0025] The neurofibroma may be associated with Neurofibromatosis Type 1 (NF1). The neurofibroma may originate in a Schwann cell (SC). The neurofibroma may be in the peripheral nervous system (PNS).

[0026] In each (and all) aspects, the TGF-p inhibitor may comprise a number of compounds which are known to inhibit or modulate or disrupt TGF-p and its associated pathway. The TGF-p inhibitor may comprise compounds described and encompassed within WO02 / 094833, WO04 / 048382 and W02007 / 018818 (the contents of which are all herein incorporated by reference).

[0027] The TGF-p inhibitor may comprise a compound having the structure: wherein is a four, five, or six membered saturated ring and X is C, O or S;

[0028] R1 is unsubstituted or substituted phenyl; unsubstituted or substituted pyridine; unsubstituted or substituted pyridine N-oxide; unsubstituted or substituted quinoline; unsubstituted or substituted quinoline N-oxide; unsubstituted or substituted naphthyridine; unsubstituted or substituted pyrazine; furyl; unsubstituted or substituted thiazolyl; unsubstituted or substituted imidazolyl; unsbustituted or substituted pyrazolyl; or unsbustituted or substituted thiophenyl; wherein the substitution may be one or more of the following: (C1-C6)alkyl, (C2-C6)alkenyl, (C2-C6)alkynyl, (C1-C6)alkoxy, (C2- C6)alkenyloxy, (C2-C6)alkynyloxy, (C1-C6)alkylthio, (C1-C6)alkylsulphinyl, (C1- C6)alkylsulphonyl, (C1-C6)alkylamino, di-[(C1-C6)alkyl]amino, (C1-C6)alkoxycarbonyl, N- (C1-C6)alkylcarbamoyl, N,N-di-[(C1-C6)alkyl]carbamoyl, (C2-C6)alkanoyl, (C2- C6)alkanoyloxy, (C2-C6)alkanoylamino, N-(C1-C6)alkyl-(C2-C6)alkanoylamino, (C3- C6)alkenoylamino, N-(C1-C6)alkyl-(C3-C6)alkenoylamino, (C3-C6)alkynoylamino, N-(C1- C6)alkyl-(C3-C6)alkynoylamino, N-(C1-C6)alkylsulphamoyl, N,N-di-[(C1- C6)alkyl]sulphamoyl, (C1-C6)alkanesulphonylamino, N-(C1-C6)alkyl-(C1- C6)alkanesulphonylamino, carboxamide, ethylene, thiophenyl, aminophenyl, trifluoromethyl, halo, trifluoromethoxy, hydroxymethyl, N-pyrrolidino, N- morpholino, phenylthio, (C1-C4)dialkylaminomethyl, methoxyphenyl, amino, hydroxy, carboxyl, phenyl, arylalky;

[0029] R2 is unsubstituted or substituted quinoline; unsbustituted or substituted quinoline N-oxide; unsbustituted or substituted phenyl; unsubstituted or substituted naphthalene; unsubstituted or substituted pyridine; unsubstituted or substituted pyridine N-oxide; unsbustituted or substituted quinazoline; unsubstituted or substituted cinnoline; unsubstituted or substituted benzodioxole; unsbustituted or substituted benzodioxane; unsubstituted or substituted pyrimidine; unsubstituted or substituted benzothiophene; or unsubstituted or substituted phenanthrolene; wherein the substitution may independently be one or more of the following: hydrogen, (C1-C6)alkyl, (C2-C6)alkenyl, (C2-C6)alkynyl, (C1-C6) alkylhalide, (C1-C6)alkoxy, (C2-C6)alkenyloxy, (C2-C6)alkynyloxy, (C1- C6)alkylthio, (C1-C6)alkylsulphinyl, (C1-C6)alkylsulphonyl, (C1-C6)alkylamino, di-[(C1- C6)alkyl]amino, (C1-C6)alkoxycarbonyl, N-(C1- C6)alkylcarbamoyl, N,N-di-[(C1- C6)alkyl]carbamoyl, aminooxy, N-(C1-C6)alkyl aminooxy , N,N-di-[(C1-C6)alkyl]aminooxy, (C2-C6)alkanoyl, (C2-C6)alkanoyloxy, (C2-C6)alkanoylamino, N-(C 1 -C6)alkyl-(C2- C6)alkanoylamino, (C3-C6)alkenoylamino, N-(C1-C6)alkyl-(C3-C6)alkenoylamino, (C3- C6)alkynoylamino, N-(C1-C6)alkyl-(C3-C6)alkynoylamino, sulphamoyl, N-(C1- C6)alkylsulphamoyl, N,N-di-[(C1-C6)alkyl]sulphamoyl, (C1-C6)alkanesulphonylamino, N- (C1-C6)alkyl-(C1-C6)alkanesulphonylamino, carboxamide, ethylene, phenyl, thiophenyl, aminophenyl, phenylthio, halo, cyano, pyridinyl, arylalkyl, hydroxy, N-pyrrolidino, N- morpholino, carboxyl, [5-phenyl-1 ,2,4-oxadiazole-3-yl]methoxy, 6-methyl-pyridazin-3-yloxy, (5-oxo-2-pyrrolidinyl)methoxy, 2-(4,5-dihydro-1 H-imidazolyl), N, N-dialkylcarbamoyloxy, 1- hydroxy-1 -methylethyl, 4-fluorophenyl, 3,4-methylenedioxyphenyl, trifluoromethyl, trifluoromethoxy, or a group of the formula wherein: Xi is O, N, S, SO2, NR13, C(O), or bond; Qi is hydrogen, phenyl, 5-(2,2-difluoro- 1 ,3-benzodioxolyl), C(O)Qs, or pyridyl when m and n are independently 0-2, except when one is 0 the other cannot be 0; Qi is ORn, NR11R12, halo, N-morpholino, N-piperazino-N'R , N-imidazolyl, N-pyrazolyl, N-triazolyl, N-(4-piperidinylpiperidine), SO2R14, SOR 14, NHSO2R15, acetamido, N-phthalimido, N-oxazolidino, N-imidazolino, N-benzoxazolidino, N- pyrolidinonyl, N(N'-methylbenzimidazolino), N,N-di(C1-C4)alkylamino(C1-C4)alkoxy, N- benzimidazolino; when m and n are independently 0-2, but one or the other of m or n is not 0; Q5 is hydroxy, methoxy, amino, diethylamino, dimethylamino; R10 is hydrogen, halo, (C1- C6)alkyl; Rn and R12 are independently hydrogen, (C1-C6)alkyl, (C1-C6)alkoxy, arylalkyl, (C3-C8)cycloalkyl, (C3-C8)cycloalkylmethyl, 4-(N-methylpiperidinyl), pyridyl, or Rn and R10 can be taken together to form a 4, 5, 6, or 7 membered ring, or Rn and R12 can be taken together to form a 3, 4, 5, 6, or 7 membered ring; R13 is hydrogen, (C1-C6)alkyl, 2- methoxyphenyl, 2-pyridim idinyl; R14 is 2-pyrimidinyl, N-methyl-2-imidazolyl, 4-chlorophenyl, 2-pyridylmethyl; R15 is (C1-C6)alkyl, N-methyl-4-imidazolyl; R is hydrogen, halo, arylalkyl, aryl, or a group of the formula wherein: Q2 is hydrogen, 4-imidazolyl, or C(O)NR24R25 when o and p are independently 0- 2; Q2 is OR23, NR24R25, or N-morpholino, when o and p are independently 0-2, but one or the other of o or p is not 0; R20 is hydrogen, or (C1-C6)alkyl; R21 is hydrogen, (C1-C6)alkyl, or R21 and R20 can be taken together to form a 4, 5, 6, or 7 membered ring; R22 is hydrogen, (C1-C6)alkyl, arylalkyl, aryl, or R21 and R22 can be taken together to be a 3, 4, 5, 6, 7 membered ring; R23 is hydrogen or (C1-C6)alkyl; R24 is hydrogen, (C1-C6)alkyl, or R24 and R25 can be taken together to form a 3, 4, 5, 6, or 7 membered ring, or R24 and R20 can be taken together to form a 6 or 7 membered ring; R25 is hydrogen, (C1-C6)alkyl, or acetyl, or a group of the formula wherein: R30 is hydrogen, or (C1-C6)alkyl; R31 is hydrogen, (C1-C6)alkyl, 2-pyridyl, pyridylmethyl, amino, or hydroxy, or a group of the formula

[0030] — NR„R J„o wherein: R32 and R33 are each independently hydrogen, (C1-C6)alkyl, acetyl, (C1- C4)alkylsulphonyl, or R32 and R33 can be taken together to form a 4, 5, 6, or 7 membered ring, or a group of the formula wherein: X2 is CH2, O, or N; q is 2-3 except when Q3 is a bond, q is 0-3; Q3 is NR36R37, or OR38, and R35 is hydrogen, or R35 and Q3 can be taken together to form a 5 membered ring; R36, R37, and R38 are each independently hydrogen, or (C1-C6)alkyl, or a group of the formula wherein: X3 is cyano, carboxamide, N,N-dimethylcarboxamide, N,N- dimethylthiocarboxamide, N,N-dimethylaminomethyl, 4-methylpiperazin-1yl-methyl or carboxylate, or a group of the formula wherein: Qe is NR41R42; r is 2-3; R40 is hydrogen, or (C1-C6)alkyl; R41 and R42 are hydrogen, (C1-C6)alkyl, or R41 and R40 can be taken together to form a 6 or 7 membered ring, or a group of the formula wherein: Q? is hydroxy, methoxy, dimethylamino, or N-piperidinyl; with the proviso that when one of Rl or R2 is unsubtituted or substituted phenyl, then the other cannot be unsubstituted or substituted phenyl or thiophen-2-yl; and with the proviso that when R2 is quinolin-4-yl, substitution at the quinoline 7-position cannot include an aryl, heteroaryl, fused aryl, or fused heteroaryl; k is 1-8; R3 is one or more of the following: hydrogen; (C1-C4)alkyl; (C1-C4)alkylhydroxy; hydroxy; N,N-di(C1-C4)alkylamino(C1-C4)alkoxy; benzyl oxymethyl; phenyloxymethyl; oxo; carboxyl; (C1-C4) alkylaryl; benzyloxy; acetoxy; amino(C1-C4)alkyl; (C2-C4)alkenyl; halo; - O-(C1-C4)alkyl; chlorophenethyl; acetonitrile; unsubsituted or substituted phenyl; wherein the substitution may be one or more of the following: (C1-C6)alkoxy, halo, carboxy, or (C1- C6)alkoxycarbonyl; and the pharmaceutically acceptable salts, esters and prodrugs thereof.

[0031] The TGF-p inhibitor may comprise 2-(6-methyl-pyridin-2-yl)-3-[6-amido-quinolin-4-yl)-5,6- dihydro-4H~pyrrolo[l,2-b]pyrazole monohydrate (IIIPAC name: 4-[2-(6-methylpyridin-2-yl)- 5,6-dihydro-4H-pyrrolo[1 ,2-b]pyrazol-3-yl]quinoline-6-carboxamide) which is also known as Galunisertib and / or LY2157299 having a CAS number: 700874-72-2 and a general formula of:

[0032] Galunisertib is an orally available, small molecule antagonist of the tyrosine kinase transforming growth factor-beta (TGF-P) receptor type 1 (TGFBR1), with potential antineoplastic activity. Upon administration, galunisertib specifically targets and binds to the kinase domain of TGFBR1 , thereby preventing the activation of TGF-p-mediated signalling pathways. Dysregulation of the TGF-p signalling pathway is seen in a number of cancers and is associated with increased cancer cell proliferation, migration, invasion and tumour progression.

[0033] Galunisertib (LY-2157299) is a pyrrolopyrazole that is 5,6-dihydro-4H-pyrrolo[1 ,2- b]pyrazole which is substituted at positions 2 and 3 by 6-methylpyridin-2-yl and 6- (aminocarbonyl)quinolin-4-yl groups, respectively. A Transforming growth factor-pRI (TGF- PRI) kinase inhibitor, it blocks TGF-p-mediated tumour growth in glioblastoma. It has a role as a TGF-p receptor antagonist and an antineoplastic agent. It is a member of quinolines, a pyrrolopyrazole, a member of methylpyridines, an aromatic amide and a monocarboxylic acid amide.

[0034] Galunisertib has been used in trials studying the basic science and treatment of Glioma, Neoplasms, Solid Tumour, Glioblastoma, and Prostate Cancer.

[0035] According to another related aspect of the invention there is provided a pharmaceutical composition, comprising a TGF-p inhibitor and a pharmaceutically acceptable carrier, excipient, or diluent.

[0036] Preferably, the inhibition of TGF-p is effected by the administration of Galunisertib (LY- 2157299) or derivatives thereof.

[0037] The administration of Galunisertib (LY-2157299) may be in combination with one or more of the following: Lomustine; Temozolomide; Gemcitabine; Durvalumab; Sorafenib

[0038] Alternative TGF-p inhibitors will be apparent to the skilled addressee and include compounds such as PF-03446962, which is a fully humanised novel monoclonal antibody. As used herein, the terms "treatment", "treating", “treat” and the like, refer to obtaining a desired pharmacologic and / or physiologic effect. The effect can be prophylactic in terms of completely or partially preventing a disease or symptom thereof and / or can be therapeutic in terms of a partial or complete cure for a disease and / or adverse effect attributable to the disease. The preventative treatment may for those individuals who have a pre-disposition to (or have already) develop (developed) many tumours. "Treatment" as used herein, covers any treatment of a disease in a mammal, particularly in a human, and includes: (a) preventing the disease from occurring in a subject which can be predisposed to the disease but has not yet been diagnosed as having it; (b) inhibiting the disease, i.e., arresting or slowing its development; and (c) relieving the disease, i.e., causing regression of the disease.

[0039] The term “subject” or “individual” used herein includes any human or nonhuman animal. The term “nonhuman animal” includes all mammals.

[0040] A “therapeutically effective amount” refers to the amount of an inhibitor of TGF-p that, when administered to a subject for treating a disease, is sufficient to affect such treatment for the disease. The “therapeutically effective amount” will vary depending on the inhibitor(s) used, the disease and its severity and the age, weight, etc., of the subject to be treated.

[0041] In general, routes of administration contemplated by the invention include, but are not necessarily limited to, enteral, parenteral, or inhalational routes.

[0042] Parenteral routes of administration other than inhalation administration include, but are not necessarily limited to, topical, transdermal, subcutaneous, intramuscular, intraorbital, intracapsular, intraspinal, intrasternal, intrathecal, and intravenous routes, i.e., any route of administration other than through the alimentary canal. Parenteral administration can be carried to effect systemic or local delivery. Where systemic delivery is desired, administration typically involves invasive or systemically absorbed topical or mucosal administration of pharmaceutical preparations. Enteral routes of administration include, but are not necessarily limited to, oral and rectal (e.g., using a suppository) delivery.

[0043] Conventional and pharmaceutically acceptable routes of administration include intranasal, intramuscular, intra-tracheal, intrathecal, intracranial, subcutaneous, intradermal, topical, intravenous, intraperitoneal, intra-arterial (for example, via the carotid artery), spinal or brain delivery, rectal, nasal, oral, and other enteral and parenteral routes of administration. The inhibitor may be administered in a single dose or in multiple doses. A suitable frequency of administration may be at least once per day, every other day, once per week, once every two, three, or four weeks, once every month, two months, or once every three to six months. The inhibitor may be administered over a period of at least a week, at least a month, at least three to six months, at least one, two, three, four, or five years, or over the course of the disease, or the lifetime of the subject.

[0044] In some embodiments, an inhibitor of TGF-p is administered by injection.

[0045] The inhibitor may be artificially generated. That is to say that it is not naturally occurring. The inhibitor may however be a naturally occurring molecule whose concentration and formulation in a medicament or pharmaceutical preparation enables it to be used for the prevention, amelioration or treatment of a brain cancer, whereas otherwise it would have no or limited efficacy.

[0046] In embodiments, the inhibitor comprises a small molecule. The small molecule may be any appropriate organic molecule that inhibits TGF-p.

[0047] The inhibitor may comprise an antibody or antibody mixture. Such antibody or antibodies may be polyclonal or may be monoclonal.

[0048] The monoclonal antibodies may be obtained by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the monoclonal antibodies may be obtained using hybridoma technology, such as fusing antibody producing cells of antigen- immunised mammals with mammalian myeloma cells to produce a hybridoma cell line. Preferably, the hybridoma cell line is produced by first immunising a mammal with an immunising antigen to produce an immunised mammal. Suitable immunising antigens are as defined above. The mammal may be immunised by any suitable method such as, for example, by intraperitoneal, subcutaneous, intravascular, intramuscular or intrasplenic injection or by oral administration. Preferably, the immunising antigen may be administered as a suspension or solution in a buffer, such as phosphate buffered saline (PBS), optionally with an adjuvant, such as Freund’s complete adjuvant. Any suitable mammal may be used such as, for example, mice, rats, rabbits, sheep or goats. It will be appreciated by a person skilled in the art that the mammal should typically be chosen so as to be compatible with the myeloma cells used in the subsequent cell fusion step. Preferably, the immunising antigen is administered to the mammal several times, such as 2, 3, 4 or more times, at 4 to 21 day intervals. Typically, the antibody producing cells of the immunised mammal may be splenic cells. Preferably, the splenic cells of the immunised mammal may be collected and fused with mammalian myeloma cells. The mammalian myeloma cells may be from any suitable source such as, for example, mice, rats or rabbits. Preferably, the mammalian myeloma cells may be from the same mammalian source as the mammal immunised with the immunising antigen. Preferably, the myeloma cells may be from mice. Preferably, the mammalian myeloma cells are selected so as to have a hypoxyanthine-guanine- phosphoribosyltransferase deficiency (HGPRT) and / or a thymidine kinase deficiency (TK ). For example, the mammalian myeloma cells may be mouse P3 / NS1 / 1-Aq4-1 cells. The splenic cells may be fused to the mammalian myeloma cells by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the splenic cells may be fused to the mammalian myeloma cells using electrofusion, optionally in the presence of a fusion promoter such as, for example, polyethylene glycol (PEG) or hemagglutinating virus of Japan (HVJ). Preferably, the splenic cells and the mammalian myeloma cells may be mixed at a ratio of 1 :1 to 10: 1.

[0049] Typically, the fused cells may be cultured and screened to selectively obtain hybridomas. The fused cells may be cultured in any suitable medium. The fused cells may be screened for hybridomas by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the fused cells may be screened for hybridomas using enzyme immunoassay, such as enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA) or surface plasmon resonance (SPR). Preferably, the fused cells may be screened for hybridomas using enzyme-linked immunosorbent assay (ELISA). It will be appreciated by a person skilled in the art that it is the supernatant of the culture of the fused cells that is typically screened. Preferably, the fused cells may be screened for hybridomas by screening for binding to TGF-p. More preferably, the fused cells may be screened for hybridomas by screening for binding to TGF-p, using enzyme-linked immunosorbent assay (ELISA).

[0050] The monoclonal antibodies produced by the cultured hybridomas may be obtained by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the monoclonal antibodies produced by the cultured hybridomas may be obtained by centrifugation of the supernatant of the culture.

[0051] In certain alternative embodiments, when the monoclonal antibodies are produced by culturing the obtained hybridomas in the abdominal cavities of a suitable mammal such as, for example, a mouse, the obtained hybridomas may be intraperitoneally administered to the mammal such as, for example, mouse. The monoclonal antibodies produced by the culture in the abdominal cavities of a suitable mammal such as, for example, a mouse may then be obtained by collecting the fluid in the peritoneal cavity.

[0052] The monoclonal antibodies produced by culturing the obtained hybridomas in a suitable medium or in the abdominal cavities of a suitable mammal such as, for example, a mouse may be used directly or may be purified. Preferably, the monoclonal antibodies may be purified. The monoclonal antibodies may be purified by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the monoclonal antibodies may be purified by ammonium sulphate precipitation, ion exchange chromatography or an anti-IgG antibody column.

[0053] If monoclonal antibodies are employed which have binding with TGF-p (or its ligand) then such antibodies may be used in combination (or combined) with small molecule inhibitors of TGF-p.

[0054] In certain embodiments, the antibody of the present invention may be polyclonal. The polyclonal antibodies may be produced by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the polyclonal antibodies may be produced by immunising a mammal with an immunising antigen to induce the production of antibodies specific for the said immunising antigen. Suitable immunising agents are as defined above. Any suitable mammal may be used such as, for example, mice, rats or rabbits. The polyclonal antibodies produced by the immunised mammal may be obtained by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the polyclonal antibodies may be obtained by collecting the serum of the immunised mammal.

[0055] The polyclonal antibodies produced by immunising a mammal with an immunising antigen and collecting the serum of the immunised mammal may be used directly or may be purified. Preferably, the monoclonal antibodies may be purified. The polyclonal antibodies may be purified by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the polyclonal antibodies may be purified by ammonium sulphate precipitation, ion exchange chromatography or an anti-IgG antibody column.

[0056] The polyclonal antibodies may be screened for binding to TGF-p using any suitable method. Suitable methods will be known to a person skilled in the art. For example, the polyclonal antibodies may be screened for binding to TGF-p using enzyme immunoassay, such as enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA) or surface plasmon resonance (SPR). Preferably, the polyclonal antibodies may be screened for binding to the TGF-p using enzyme-linked immunosorbent assay (ELISA).

[0057] The antibody of the present invention may be any suitable isotype. Preferably, the antibody is of isotype IgG. For the avoidance of doubt, antibodies of the isotype IgG typically comprise four peptide chains, of which two are heavy chains and two are light chains, and have two fragment antigen-binding (Fab) regions. The Fab regions comprise complementary determining regions (CDRs) which are the part of the antibody which bind to the antigen.

[0058] Whilst the inhibitor or inhibitors may comprise a whole antibody or whole antibodies, it may comprise an antibody fragment or a modified form thereof. Suitable examples of whole antibodies include, but are not limited to, monovalent or divalent antibodies. Suitable examples of antibody fragments include, but are not limited to, Fab, F(ab’)2, Fv, Fab / c having one Fab and complete Fc, and single chain Fv (scFv) having heavy (H) or light (L) chain Fvs connected by a suitable linker. Other antibody scaffold proteins may be employed, such as Nanobodies RTM (these constructs, marketed by Ablynx (Belgium), comprise synthetic single immunoglobulin variable heavy domain derived from a camelid (e.g. camel or llama) antibody), Domain Antibodies (marketed by Domantis (Belgium), comprising an affinity matured single immunoglobulin variable heavy domain or immunoglobulin variable light domain), UniBodies (marketed by Genmab, UniBodies are modified fully human lgG4 antibodies where the hinge region of the antibody has been eliminated), Trifunctional Antibodies (monoclonal antibodies with binding sites for two different antigens), Affibodies (marketed by Affibody, Affibodies are based on a 58-amino acid residue protein domain, a three helix bundle domain, derived from one of the IgG- binding domains of staphylococcal protein A), Anticalins (antibody mimetics synthesised from human lipocalins, which can also be formatted as dual targeting proteins, so-called Duocalins) or DARPins (Designed Ankyrin Repeat Proteins) (which are another example of antibody mimetic based on repeat proteins, such as ankyrin or leucine-rich repeat proteins, which are ubiquitous binding molecules).

[0059] In certain embodiments, the antibody may be optimised or may be humanised. ‘Optimised’, and like terms as used herein, means that the amino acid sequence of the antibody is adapted, such as by mutation or modification including, for example, glycosylation, so as to be suitable for use in the patient to which it is to be administered. ‘Humanised’, and like terms as used herein, means that the amino acid sequence of the antibody is adapted, such as by mutation or modification including, for example, glycosylation, to reduce the composition of non-human amino acid sequences in the antibody.

[0060] The antibody, when humanised, may be partially humanised or may be substantially fully humanised. By ‘partially humanised’ is mean that part of the amino acid sequence of the antibody has been adapted, such as by mutation or modification including, for example, glycosylation, to be the same as the amino acid sequence of the human antibody. The antibody, when partially humanised, may be partially humanised in any region of the antibody. Preferably, the antibody, when partially humanised, may be partially humanised in one or more of the variable fragment antigen-binding (Fab) regions of the antibody. By ‘substantially fully humanised’ is meant that substantially all of the amino acid sequence of the antibody has been adapted, such as by mutation or modification including, for example, glycosylation, to be the same as the amino acid sequence of the human antibody. Preferably, the antibody is substantially fully humanised.

[0061] The optimised and / or humanised antibody and derivatives may be produced by any suitable method. Suitable methods will be known to a person skilled in the art. For example, the optimised and / or humanised antibody and derivatives thereof may be produced using genetic engineering technology, chimeric technologies, CDR grafting or veneering.

[0062] In certain embodiments, when the optimised and / or humanised antibody and derivatives thereof are produced using genetic engineering technology, a polynucleotide encoding the antibody may be isolated and cloned into an expression vector to obtain a recombinant plasmid, transforming a host organism with the obtained recombinant plasmids, culturing the transformants and causing expression of the polynucleotide encoding the antibody. When more than one polynucleotide is used, each polynucleotide may be cloned into the same or different expression vectors. Any suitable host organism may be transformed with the obtained recombinant plasmids. For example, the host organism may be prokaryotic, such as E.coli, bacilli including Bacillus subtilis and enterobacteriaceae including Salmonella typhimurium, or eukaryotic, such as yeast including Saccharamyces cerevisiae. It will be appreciated by a person skilled in the art that the step of cloning into an expression vector to produce a recombinant plasmid may be performed using standard methods in the same or different organism to the host organism. Preferably, the step of cloning into an expression vector to produce a recombinant plasmid may be performed using standard methods in E.coli.

[0063] The optimised and / or humanised antibody may bind to any suitable antigen. Preferably, the optimised and / or humanised antibody binds to TGF-p. The antibody or antibody mixtures may be present in the composition at any suitable concentration. Preferably, the antibody may be present at a concentration of about 0.1 nanograms (ng) to 100 micrograms (mg), more preferably about 1 ng to 50 mg, most preferably about 10 mg to 50 mg.

[0064] In other embodiments, the inhibitor or inhibitors comprise a peptide or peptide mimetic thereof, or C-terminal amidated peptide thereof.

[0065] The terms “peptide” and “peptides” include compounds that have amino acid residues (H- Ca-[side chain]) but which may be joined by peptide (-CO-NH-) or non-peptide linkages.

[0066] Peptides may be synthesised by the Fmoc-polyamide mode of solid-phase peptide synthesis. Reagents for peptide synthesis are readily commercially available.

[0067] Purification of the peptides may be effected by any one, or a combination of, techniques such as size exclusion chromatography, ion-exchange chromatography and (principally) reverse-phase high performance liquid chromatography. Analysis of peptides may be carried out using thin layer chromatography, reverse-phase high performance liquid chromatography, amino-acid analysis after acid hydrolysis and by fast atom bombardment (FAB) mass spectrometric analysis.

[0068] The peptide may contain at least one D-amino acid residue, such as 1 , 2 or 3 or 4 or 5 or 6 or 7 or 8 D-amino acids. Typically, the composition of the invention may contain 0, 1 , 2 or 3 D-amino acids. The presence of D-amino acids in the composition of the invention may be useful in preventing degradation of the compound by proteases. Other methods for making peptides resistant to proteolytic degradation include blocking the N- and / or C- terminal amino acid residues. Thus, in some embodiments the N- and / or C-terminal amino acid residues are blocked. Suitable blocking methods include acetylation of the N-terminus or incorporating a pyroglutamate residue at the N-terminus.

[0069] The peptide may be a peptide aptamer. Peptide aptamers typically consist of short, 5-20 amino acid residues long sequences that can bind to a specific target molecule.

[0070] There are a number of different approaches to the design and synthesis of peptide composition that do not contain amide bonds. In one approach, one or more amide bonds are replaced in an essentially isoteric manner by a variety of chemical functional groups.

[0071] Retro-inverso peptidomimetics, in which the peptide bonds are reversed, can be synthesised by methods known in the art. This approach involves making pseudopeptides containing changes involving the backbone, and not the orientation of side chains. Retro- inverse peptides, which contain NH-CO bonds instead of CO-NH peptide bonds, are more resistant to proteolysis.

[0072] The peptide may be linear. Although, it may be advantageous to introduce a cyclic moiety into a peptide-based framework. The cyclic moiety restricts the conformational space of the peptide structure and this may lead to an increased efficacy. An added advantage of this strategy is that the introduction of a cyclic moiety into a peptide may also result in the peptide having a diminished sensitivity to cellular peptidases.

[0073] In some embodiments of the invention the peptide may be joined to another moiety. Convenient moieties to which the peptide may be joined include polyethylene glycol (PEG) and peptide sequences, such as TAT and antennapedia which enhance delivery to cells.

[0074] PEGylation is a method well known to those skilled in the art wherein a (peptide or other compound) is modified such that one or more polyethylene glycol (PEG) molecules are covalently attached to the side chain of one or more amino acids. It is one of the most important molecule altering structural chemistry techniques (MASC). Other MASC techniques may be used; such techniques may improve the pharmacodynamic properties of a compound, for example extending its serum half-life in vivo. A PEG-peptide conjugate is formed by first activating the PEG moiety so that it will react with, and couple to, the compound of the invention. PEG moieties vary considerably in molecular weight and conformation, with the early moieties (monofunctional PEGs; mPEGs) being linear with molecular weights of 12kDa or less, and later moieties being of increased molecular weights. PEG2, a recent innovation in PEG technology, involves the coupling of a 30kDa (or less) mPEG to a lysine amino acid (although PEGylation can be extended to the addition of PEG to other amino acids) that is further reacted to form a branched structure that behaves like a linear mPEG of much greater molecular weight. Methods that may be used to covalently attach the PEG molecules to the peptides. The potential advantages of PEGylation of the compound of the invention include reduced renal clearance which, for some products, results in a more sustained adsorption after subcutaneous administration as well as restricted distribution, possibly leading to a more constant and sustained plasma concentrations and hence an increase in clinical effectiveness. Further potential advantages include reduced immunogenicity of the therapeutic compound, and lower toxicity.

[0075] In some embodiments, the inhibitor or inhibitors is / are pro-drugs of the peptide. A pro-drug is a compound which is metabolised in vivo to produce the molecule, such as a protein. One of skill in the art will be familiar with the preparation of pro-drugs.

[0076] The peptide may be a peptide mimetic. A peptide mimetic is an organic compound having similar geometry and polarity to the molecules defined herein, and which has a substantially similar function. A mimetic may be a molecule in which the NH groups of one or more peptide links are replaced by CH2 groups. A mimetic may be a molecule in which one or more amino acid residues is replaced by an aryl group, such as a napthyl group.

[0077] In other embodiments, an inhibitor or inhibitors comprise nucleic acid, such as single stranded DNA or RNA (such as an antisense molecule, a ribozyme, an siRNA, or a triple helix molecule that hybridizes to a nucleic acid encoding a target protein, or a transcription regulatory region, and blocks or reduces mRNA expression of the gene encoding TGFB1), which is capable of binding to and inhibiting TGF-p. It is envisaged that the same targets on TGF-p are also suitable for targeting with peptides and peptide aptamers will also be suitable for targeting with RNA or modified RNA aptamers. Nucleic acids such as single stranded DNAs and RNAs may be provided that bind to and inhibit TGF-p. Typically, the nucleic acids are single stranded and have from 100 to 5000 bases.

[0078] An "antisense" nucleic acid can include a nucleotide sequence which is complementary to a "sense" nucleic acid encoding TGFB1, e.g., complementary to the coding strand of a double- stranded cDNA molecule or complementary to an mRNA sequence. The antisense nucleic acid can be complementary to an entire TGFB1 gene coding strand, or to only a portion thereof.

[0079] Exemplary antisense compounds include DNA or RNA sequences that specifically hybridize to the target nucleic acid, e.g., the mRNA encoding the TGFB1 gene. The complementary region can extend for between about 8 to about 80 nucleobases. The compounds can include one or more modifier nucleobases. Modified nucleobases are known in the art. siRNAs are small double stranded RNAs (dsRNAs) that optionally include overhangs. For example, the duplex region of an siRNA is about 18 to 25 nucleotides in length, e.g., about 19, 20, 21 , 22, 23, or 24 nucleotides in length. Typically, the siRNA sequences are exactly complementary to the target mRNA. dsRNAs and siRNAs in particular can be used to silence gene expression in mammalian cells (e.g., human cells). siRNAs also include short hairpin RNAs (shRNAs) with 29-base-pair stems and 2-nucleotide 3' overhangs.

[0080] Features, integers, characteristics, compounds, molecules, chemical moieties or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith. All of the features disclosed in this specification (including any accompanying claims, abstract and figures), and / or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. The invention is not restricted to the details of any foregoing embodiments. The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

[0081] Detailed Description of the Invention

[0082] Embodiments of the invention are described below, by way of example only, with reference to the accompanying figures in which:

[0083] Figure 1 shows a number of graphs and images which illustrate the identification of early time point critical for tumour formation, in which: A is a schematic of mouse model; B is a graph depicting percentage of proliferating recombined SCs in the bridge at Day 7, 10 and 14 post injury in WT and Nf1-KO mice; C is a graph depicting percentage of proliferating recombined SCs in the distal stump at Day 7, 10 and 14 post injury in WT and Nf1-KO mice; D are representative images from the bridge and distal stump of control and Nf 1 KO mice at Day 14 post injury stained antibodies for p75 (white) recombined SCs (GFP) and Edll (magenta) (images are representative of n=mice / group, scale bar=10pm); E is a graph showing the quantification of the percentage of Edll+ P75+ SCs at Day 14 in the bridge; F is a graph showing the distal stump; G and H are volcano plots depicting differential gene expression between experimental groups; and I shows a heat map of candidates identified from RNAseq data. Data is presented as mean ± SEM, *p<0.05, **p<0.01 , ***p<0.001.

[0084] Figure 2 shows a graph and images which illustrate that TGF-p signaling is critical for SC proliferation at early time point, in which: A shows representative confocal images of cells labelled with pSmad3 (red) and nuclei (blue) in WT and Nf1-KO mice at Day 14 post injury (images are representative of n= / group); B shows confocal images of the distal stump and bridge region of Nf1-KO mice at Day 14 post injury stained with pSmad3 (green) and nuclei (blue) (images are representative of n= / group); C shows representative images of recombined SCs in WT, Nf1-KO, Nf1-KO Tgfbr2-KO and Tgfbr2-KO to assess proliferation at Day 14. SCs are labelled with tdTomato (red), Edll (white) and nuclei (blue) (images are representative of n= / group, scale bar=25pm); and D is a graph showing the quantification of SC proliferation in C. Data is presented as mean ± SEM, *p<0.05, **p<0.01 , ***p<0.001. Figure 3 shows a number of graphs and images which illustrate the loss of TGF-p signaling inhibits tumour formation, in which: A shows a table depicting frequency of tumour occurrence in WT control, Nf1-KO, Nf1-KO Tgfbr2-KO and Tgfbr2-KO mice at 6 months post injury; B shows representative macroscopic images of nerves 6 months following injury; C shows the nerve width at injury site (in) for the Control, Nf1-KO, Nf1-KO Tgfbr2- KO and Tgfbr2-K0 mice; D shows the quantification of maximal nerve width; E is a graph showing injured area at 6 months post injury n=6 / group; F are representative confocal tile scan images of sciatic nerves at 6 months following injury in which recombined SCs are labelled with tdTomato (red) and nuclei (blue) (n= / group, scale bar=400pm); G are representative confocal images of SC morphology in the regenerated area at 6 months labelled with tdTomato (red) and Hoescht (blue) (images are representative of n=6 / group); and I is a graph depicting percentage of proliferating SCs at 6 months. Data is represented at mean ± SEM *p<0.05, **p<0.01 , ***p<0.001 .

[0085] Figure 4 shows a number of graphs and images, in which: A is a representative tile scan images of Nf1-KO nerves at 6 months post injury depicting the three regions of uncut, regenerated nerve and tumour labelled with tdTomato (Red) and Hoechst (blue) (dashed lines indicate regions. Scale bar=400um); B is a graph showing the quantification of cell density in WT control, Nf1 -KO regenerated, Nf1-KO tumour, Nf1-KO Tgfbr2-KO and Tgfbr2- KO nerves at 6 months post injury; C are representative confocal images of SCs (tdTomato, red) perinurial cells (Glutl , cyan) and nuclei (blue) in WT control, Nf1-KO regenerated, Nf1- KO tumour, Nf1-KO Tgfbr2-KO and Tgfbr2-KO nerves at 6 months post injury (images are representative of n=mice / group); D is a representative tile scan images in WT control, Nf 1 - KO, Nf 1 -KO T gfbr2-K0 and T gfbr2-K0 nerves at 6 months post injury, labelled with markers for SCs (tdTomato, red) perinurial cells (Glutl , cyan) and nuclei (blue) (scale bar=400pm); E is a graph showing the analysis of Glutl + area depicted in D (n=6 mice / group); F is a graph showing the quantification of the number of cells positive for Glutl (green) and tdTomato (magenta) at 6 months post injury in WT, Nf1 KO, Nf1-KO Tgfbr2- KO, Tgfbr2 KO nerves; G. show representative images of the cells from F (arrows denote double positive cells); H show a table of components of WT, Nf1 KO, Nf1 -KO Tgfbr2-KO, Tgfbr2 KO nerves at 6 months post injury. Data is represented at mean ± SEM, *p<0.05, **p<0.01 , ***p<0.001 ;

[0086] Figure 5 shows a number of graphs and images illustrating that TGF-p signaling mediates the escaped population of SCs at Day 14, in which: A is a confocal image of Nf 1 KO nerve showing the uncut, regenerating region and escaped region at Day 14 after injury (dashed lines indicate border of the three regions, scale bar=200pm); B is a graph showing the quantification of the area of the escaped region in WT, Nf1 KO, Nf1-KO Tgfbr2-KO, Tgfbr2 KO nerves at Day 14 (n=4 / group); C is a graph showing the analysis of SC proliferation in the escaped region at Day 14 post injury in Nf1-KO compared to Nf1-KO Tgfbr2-KO; D shows representative confocal images of axons (white) SCs (red) and nuclei (blue) in Nf 1 KO nerves (tile scan shows distribution of axons in the regenerated and escaped regions (left panel) with higher magnification images of the SCs and axons in escaped region (right). Images are representative of n= / mice); E shows representative confocal images of blood vessels (white) SCs (red) and nuclei (blue) in Nf 1 KO nerves. Tile scan shows distribution of blood vessels in the regenerated and escaped regions (left panel) with higher magnification images of the SCs the vasculature in escaped region (right) (images are representative of n= / mice); G is a graph showing the analysis of macrophage density in Nf1-KO nerves at Day 14 post injury (n=2 mice / group); H shows representative images from HCR for Tgfbl (white) and SCs (red) co-stained for Glutl (cyan) in the regenerating and escaped region of Nf1-KO nerves at Day 14; I is a graph showing the quantification of the cells in H; J shows representative images from HCR for Tgfbl (white) SCs (red) and costained for macrophages (green) in the escaped region of Nf1- KO nerves at Day 14; and K is a graph showing the quantification of cells from J. Data is represented at mean ± SEM, *p<0.05, **p<0.01 , ***p<0.001.

[0087] Figure 6 shows a number of graphs and images illustrating that TGF-p negatively regulates SC differentiation, in which: A shows a schematic diagram of Adult DRG experimental design; B shows representative tilescan reconstruction images of ex vivo DRG - SC cocultures 21 days after myelination initiation, in the presence or absence of TGF-p, myelination was induced using 50 pg / mL L-ascorbic acid, myelin basic protein (MBP) was used to stain myelin sheaths (green), neurofilament to stain axons (white), and Hoechst was used to stain nuclei (blue) (white arrowheads indicate examples of myelin sheaths (scale bar: 100 pm); C shows a graph of the quantification of myelin area as determined by MBP staining (each dot represents the average of 10 fields from a separate mouse, Graph shows mean ± SEM, statistical signficance was determined by one-way ANOVA, p value: ****<0.0001); F shows representative immunofluorescence images of ex vivo DRG - SC cocultures 12h after SC seeding (Phalloidin-546 was used to stain F-actin (red), neurofilament to stain axons (white), and Hoechst was used to stain nuclei (blue) Quantification of SC- axon interactions) and the associated graph shows mean ± SEM (statistical signficance was determined by ANOVA); G shows schematic diagrams of experimental set up; H shows graphs of the analysis of RNA extracted from cells as in G (myelin protein zero (MPZ), periaxin (PRX), myelin basic protein (MBP), and Krox20 / Egr2 were used as markers of SC differentiation state, data was normalised to B2M n=3, each dot represents one biological replicate, and each biological replicate is shown in a separate colour, graph shows mean ± SEM). Statistical signficance was determined by one-way ANOVA, p value: ** <0.01.

[0088] Figure 7 shows a number of graphs and images illustrating the Pharmacological inhibition of TGF-p impairs tumour formation, in which: A is a representative tile scan confocal images of Control and Galunisertib treated Nf1-KO mice at Day 14 post injury labelled with Glutl (Cyan), tdTomato (red), Edll (white) and Hoescht (blue) (images are representative of n=6 animals / group); B is a graph showing the quantification of SC proliferation in Nf1-KO mice at Day 14 in control treated mice compared to Galunisertib (n=4 mice / group); C is a graph showing the analysis of SC proliferation in the escaped region of Nf1-KO nerves at Day 14 post injury between Control and Glaunisertib treated mice (n=3 / group); D is a table of frequency of tumour formation in Nf1-KO mice at 6 months post injury in Control vs Galunisertib treated mice; E are representative macroscopic images (and graph) of Nf1-KO nerves at 6 months post injury following treatment with control or Glaunisertib with nerve width quantified (images are representative of n=6 mice / group, yellow line denotes average nerve width of Nf1-KO mice, blue line represents average nerve width of Nf-KO Tgfbr2-KO and green line denotes width of WT control at 6 months post injury; F are representative tile scan images of Nf1-KO nerves treated with control or Galunisertib at 6 months and stained for Glutl (cyan), SCs are labelled with tdTomato (red) and nuclei (blue). Data is represented at mean ± SEM, *p<0.05, **p<0.01 , ***p<0.001 .

[0089] Figure 8 shows a number of images and graphs illustrating that axonal interactions are impaired in Schwann cells in the escaped region at Day 14, in which A are representative images of Nf1-KO sciatic nerve longitudinal sections harvested at day 14 after nerve partial transection, stained for axons (neurofilament, white) and nuclei (Hoechst, blue) in the upper panel, and axons traced using NeuronJ in the lower panel. Regions of interests (uncut, regenerating, escaped) are shown above; B-D are graphs showing quantifications of the axon deviation from the proximal-to-distal nerve axis and specifically B shows the the axon direction relative to the proximal-to-distal axis, C shows the average density of axons and D shows each region of interest. For B and C, each point represents one axon from n=3 animals. For D, each point represents one mouse (n=3), graph representative of mean ± SEM. ** P<0.01 , **** P<0.000.1 ; E shows the quantification of the proportion of Schwann cells that have not established axonal contact in regions of interest. Each point represents one mouse (n=3), graph is representative of mean ± SEM. * P<0.05 and F shows representative images of the regenerating and escaped region of Nf1-KO sciatic nerve at day 14 post-injury, stained for recombined Schwann cells (tdTomato, red), axons (Neurofilament, white), and nuclei (Hoechst, blue). Yellow arrowheads indicate Schwann cells that have not established axonal contact in the escaped region. Figure 9 are images and graphs illustrating that Tgfp signalling is upregulated in the escaped region at Day 14 and specifically, A are representative confocal images of the regenerating, escaped, and distal regions of the Nf1-KO nerve 14 days post-injury, stained for phosphorylated (Ser423 / 425)-Smad3 (pSmad3, green) and nuclei (Hoechst, blue); B and C are graphs show the average intensity of pSmad3 staining in all cells (B) and the percentage of cells that are pSmad3 positive (C) in each region of interest. Each point represents one animal (n=3), graphs are representative of mean ± SEM. *P<0.05, ** P<0.01 , *** P<0.001 ; D are representative images of the regenerating and escaped region of Nf1-KO mice at day 14 post-injury, stained for recombined Schwann cells (tdTomato, red), perineurial-like cells (Glutl , cyan), and nuclei (Hoechst, blue) in the upper panel, and Tgfbl mRNA (white) in the lower panel; E is a graphs showing the quantification of the average coverage of Tgfbl mRNA signal in the regenerating and escaped region of Nf1- KO mice at day 14 post-injury. Each point represents one mouse (n=3), graph is representative of mean ± SEM. *P<0.05.

[0090] Figure 10 are images and graphs illustrating that Tgfp is produced by macrophages in the escaped region and more specifically where A are representative images of the Nf1-KO escaped region at day 14 post-injury, stained for macrophages (F4 / 80), mast cells (CD117), T cells (CD3), and neutrophils (Ly6G); B is a graph showing the density of immune cells in the escaped region at day 14 post-injury in Nf1-KO mice. Each point represents one mouse (n=3-4 mice), graph shows mean ± SEM. , **** P<0.0001 ; C are representative images of the escaped region of Nf1-KO mice at day 14 post-injury, stained for recombined Schwann cells (tdTomato, red), macrophages (F4 / 80, green), nuclei (Hoechst, blue), and Tgfbl mRNA (white); D is a graph showing the average number of Tgfbl punctae per cell of each cell type analysed. Each point represents the average of at least 30 cells from a single mouse from n=3 mice, graph shows mean ± SEM. *** P<0.001 , **** P<0.0001 ; E are representative confocal images of the indicated regions of Nf1-KO nerve at day 14 postinjury, stained for recombined Schwann cells (tdTomato, red), macrophages (F4 / 80, white), and nuclei (Hoechst, blue); F-G are graphs showing the density (I) and proportion (J) of macrophages in each region of interest in Nf1-KO mice at day 14 post-injury. Each point represents one mouse (n=4 mice), graph shows mean ± SEM. ** P<0.01 , *** P<0.001 , **** P0.0001.

[0091] Figure 11 are images and a graph showing that Tgfp regulates SC-axonal interactions and more specifically where A are representative images of SCs cultured for 12h with DRG axons, in the presence or absence of TGF , and PD184352 (MEK inhibitor) or SIS3 (Smad3 inhibitor), and subsequently stained for nuclei (Hoechst, blue), actin (phalloidin , red) , and axons (neurofilament, white); and B is a graph showing the quantification of the percentage of SCs exhibiting both axonal contact and correct cytoplasmic / nuclear alignment with an axon. Data is presented as mean ±SEM and is representative of n=3 independent experiments.

[0092] Example 1

[0093] Experiments were conducted to assess the microenvironmental TGF-p signalling during early tumorigenesis to assess whether the signalling determines NF1- associated tumour formation. The experiments were as detailed below.

[0094] In vivo tumourigenesis studies

[0095] To assess tumour formation, 4 - 5 week-old P0-CreERT2:tdTomato:Nf1fl / fl:Tgfbr2fl / fl mice were administered with Tamoxifen to induce Cre-mediated recombination. Animals then underwent a sciatic nerve partial transection. The injured and uninjured nerve of the animals were then harvested at indicated timepoints - 14 days post-injury for analysis of early tumour initiation; 6 months post-injury for tumour formation assays.

[0096] Tamoxifen administration

[0097] To induce Cre-mediated recombination, 2mg tamoxifen (Sigma) was orally administered to 4 - 5 week-old mice daily for 5 consecutive days. Tamoxifen was dissolved in Ethanol and diluted 10 times in sunflower seed oil (Sigma) to a final concentration of 20 mg / mL. The solution was vortexed for at least 4h at RT to ensure complete dissolution, and stored at - 80°C.

[0098] Sciatic nerve transection

[0099] 14 days after tamoxifen, mice were anaesthetised with isoflurane under aseptic conditions and the right sciatic nerve was exposed at the sciatic notch. The nerve was half-transected just below the notch, and the wound closed with surgical clips. Mice received a subcutaneous injection of buprenorphine prior to the surgery.

[0100] Galunisertib treatment

[0101] To pharmacologically inhibit TGF-p signaling, Galunisertib (LY2157299) was obtained from MedChemExpress, and was administered orally at a dosage of 50 mg / kg to mice twice a day for 3 days (at days 7 - 9 after sciatic nerve transection). EdU administration

[0102] For cell proliferation experiment in vivo, mice received a double EdU intra-peritoneal injection (2mg EdU in saline) 6h and 3h before dissection. Cell proliferation was determined by measuring EdU incorporation using the Click-iT EdU imaging kit (Invitrogen) according to the manufacturer’s instructions.

[0103] Nerve processing for frozen IHC-IF

[0104] For fresh frozen tissue, nerves were dissected and rapidly placed into optimal cutting temperature (OCT) solution in a cryosection mould, and snap-frozen in liquid nitrogen. For fixed tissue, nerves were dissected and incubated in 4% PFA at 4°C overnight for fixation, and cryo- protected in 30% w / v sucrose-PBS at 4°C overnight. Nerves were next incubated for 2h at RT in a 1 :1 mix of 30% sucrose and OCT solution. Nerves were subsequently embedded in OCT in a cryosection mould and snap-frozen in liquid nitrogen. Cryo-blocks were stored at -80°C.

[0105] Immunofluorescence staining

[0106] Cryo-sections were obtained using a Leica cryostat. For fixed tissue, sections were kept in PBS prior to permeabilisation. For fresh-frozen unfixed tissue, sections were immediately fixed in 4% PFA for 10 mins, and rinsed with PBS prior to permeabilisation. Cryo-sections were permeabilised in 0.3% triton X-100 in PBS for 30 mins at RT, blocked in 10% goat or donkey serum (as appropriate) in PBS for 1 h at RT, and incubated in primary antibody diluted in blocking solution overnight at 4°C. For nuclear stains. For nuclear stains, 0.01% triton X-100 was added to the blocking solution. Sections were then washed 3 times in PBS for 15 mins each, and incubated for 1 h at RT in secondary antibody and Hoechst (1 :1000) diluted in blocking solution. Sections were washed 3 times in PBS for 15 mins each, and mounted using Fluoromount.

[0107] Hybridisation chain reaction (HCR) - IF

[0108] For in situ RNA analysis, hybridisation chain reaction (HCR) reagents from Molecular Instruments were utilised. Cryo-sections of PFA-fixed RNase-clean tissue were cut using a Leica cryostat. Sections were mounted on microscope slides and sealed using SecureSeal hybridisation chambers (Merck). Sections were washed twice in DEPC-PBS for 5 mins. Sections were then permeabilised in 70% ethanol in DEPC-PBS for 3h at 4°C, and subsequently washed twice with DEPC-PBS. For hybridisation, sections were incubated in Probe hybridisation buffer for 30 mins at 37°C. Samples were then incubated in probe solution diluted in probe hybridisation buffer for 16h at 37°C. Samples were then washed 5 times in Probe wash buffer for 15 mins each. For amplification, sections were incubated in Amplification buffer for 30 mins at RT. Hairpin solution was prepared by snap-heating hairpins at 95°C for 90 sec, allowed to cool to RT for 30 mins in the dark, and diluted 1 :200 in amplification buffer. Sections were incubated in hairpin solution for 16h at RT in the dark. Sections were then washed 4 times in 5X SSCT solution and fixed in 4% PFA for 10 mins at RT. Samples were washed twice with DEPC-PBS, blocked in 5% goat or donkey serum (as appropriate) for 1 h at RT, and incubated in primary antibody diluted in blocking solution overnight at 4°C. Sections were washed 3 times in DEPC-PBS, incubated in secondary antibody diluted in blocking solution for 1h at RT. Subsequently, samples were washed 3 times in DEPC-PBS, incubated for 30 mins at RT in DAPI (1 :500 in PBS), washed twice in DEPC-PBS, and mounted using Fluoromount.

[0109] Primary rat Schwann cell culture

[0110] Primary rat Schwann cells were harvested from the sciatic and brachial nerves of P7 Sprague- Dawley rats. Unless stated otherwise, cells were cultured in Schwann cell medium: DMEM (low glucose) containing 3% foetal bovine serum, 4 mM L-glutamine, 1 pM forskolin, glial growth factor (GGF), kanamycin / gentamicin. For stimulation and / or differentiation experiments, cells were maintained in a serum-free medium, consisting of DMEM supplemented with 4 mM L- glutamine and SATO supplement. SATO supplement was made from DMEM containing progesterone, glucose, thyroxine, triiodothyronine, putrescine, selenium. Cells were cultured in 37°C 1 10% CO2 conditions.

[0111] Dibutyryl-cAMP (dbcAMP) differentiation

[0112] To activate the Schwann cell myelination program in vitro in the absence of axons, 5 x 105 cells were seeded per 60 mm culture dish, pre-coated with PLL overnight. Cells were grown for 2 days until completely confluent, washed 3 times in SATO medium, and incubated in SATO-d medium, containing SATO medium supplemented with transferrin, insulin, and dbcAMP. For de- differentiation experiments, cells were incubated for 72h in SATO-d medium to ensure robust differentiation prior to stimulation, with the SATO-d medium replenished every 48h. For differentiation experiments, stimulating factors were adding concomitantly with SATO-d medium.

[0113] RNA extraction and RT-qPCR

[0114] For RNA analysis, cells were harvested using Trizol (Invitrogen). RNA was extracted following the manufacturer’s instructions, and RNA concentration determined by nanodrop measurement. cDNA was synthesised from 1 pg RNA using the SuperScript II RT kit (Invitrogen). RNA was incubated with 25 pM random hexamers and 0.5 pM dNTPs for 5 mins at 65°C. First strand synthesis was performed by adding 40 units RNaseOUT, 0.02 M DTT, and 1X FS buffer to each reaction, and incubated for 2 mins at 25°C. SuperScript II Reverse Transcriptase was added at 20 units / pL, and the samples incubated at 25°C for 10 mins, 42°C for 50 mins, and 70°C for 15 mins.

[0115] The MesaBlue qPCR Master Mix Plus kit (Eurogentec) was used for the RT-qPCR. A final volume of 25 pL per reaction was achieved using 12.5 pL MesaBlue Mix, 0.25 pL 10 pM primer mix, 7.25 pL nuclease-free water, and 5 pL template cDNA per reaction. Wells were loaded in duplicates or triplicates. Steps of the qPCR program used are outlined in Table 1 below. Nuclease-free water was used as a template for negative control, and B2M used as a normalisation control.

[0116] Table 1

[0117] Adult mouse DRG explant culture

[0118] DRGs were extracted from 8 - 12 week old mice. DRGs were decapsulated and explanted onto PLL- and laminin-coated coverslips, and cultured overnight at 37 °C 1 10% CO2 in DRG explant medium, consisting of neurobasal medium supplemented with penicillin / streptomycin, B27 supplement, 2 mM L-glutamine, and 50 ng / mL NGF. The following day, fresh medium was carefully added, additionally supplemented with fresh 10 ng / mL GDNF, 20 uM uridine, and 20 uM FLIDR. Medium was replenished every 2 days. After 4 days in culture, uridine and FLIDR were no longer added to the culture medium. Explants were stable for use for several weeks in these conditions.

[0119] DRG - SC association assay To assay SC axonal interaction, DRGs were explant and cultured as above. 5 days after explanting, primary rat SCs were seeded onto the DRG axons. Co-cultures were incubated for 12h at 37°C 1 10% CO2, and then fixed in 4% PFA for 15 mins. Co-cultures were then immunostained, imaged, and SC axonal association scored as described by Parrinello et al., 2008.

[0120] DRG - SC myelination assay

[0121] To assay myelination of adult DRG axons by SCs, primary rat SCs were seeded onto DRG axons as described above. After 72h, co-cultured were then incubated in a myelination- permissive medium, consisting of DMEM / F-12 supplemented with NGF, insulin, SATO supplement, forskolin, and L-ascorbic acid. Co-cultures were maintained for 3 weeks, with fresh medium added every 2 days. Co-cultures were then fixed and immunostained for myelin basic protein to detect myelin sheaths.

[0122] Identification of early time point critical for tumour formation

[0123] Previous work had identified adult SCs as the cell of origin for neurofibromas following peripheral nerve injury (Ribeiro et al., 2013). In these experiments, P0- CreERT2YFPNf1fl / fl mice were used in which a tamoxifen-inducible Cre was able to induce Nf1 knockout specifically in myelinating SCs in adult mice (Nf1-KO), with recombined cells labelled with YFP. Following recombination in adult mice, a sciatic nerve injury was then performed, and the animals were analysed for neurofibroma formation. Remarkably, tumours only formed in injured mice demonstrating an important role for the microenvironment in the ability of Nf1-KO SCs to behave normally or form a tumour. Moreover, these experiments also showed that the tumour-promoting environment was specific to the wound site.

[0124] Following sciatic nerve injury, all SCs at the injury site and distal to the injury site undergo de- differentiation and proliferation, but neurofibromas only formed at the injury site, implying that the microenivronment specific to the injury region was required for tumourigenesis (Ribeiro et al., 2013). Consistent with this we found that Nf1-KO SCs at the injury site exhibited increased levels of proliferation by day 14 after injury compared to either WT SCs in the injury region or Nf1-KO SCs at the distal stump, indicating that tumourigenic signaling from the microenvironment affects Nf1-KO SC behaviour early during the tumourigenic process (Figurel). Using RNA-seq analysis, TGF-p signaling was found to be upregulated in the Nf1- KO bridge at day 14 post-injury compared to the Nf1-KO distal region and the WT bridge region (Figure 1). TGF-p was therefore a promising candidate as the microenvironmental tumourigenic signal. TGF-fi signaling is critical for SC proliferation at early time point

[0125] The RNA-seq data revealed the differential upregulation of Tgfbl , Tgfb2, and Tgfb3 mRNA transcripts in the Nf1-KO nerve bridge at day 14, suggesting an increase in TGF-p in this region. To determine whether this resulted in elevated TGF-p signaling in SCs, nerves from Nf1-KO mice and control mice were collected at 14 days after sciatic nerve half-transection, and stained for phosphorylated Smad3 (pSmad3), as an indicator of active TGF-p signaling (Attisano & Lee- Hoeflich, 2001 ; Massague, 2012). In Nf1-KO nerves, a significant increase in pSmad3 staining was observed in the nerve bridge when compared to WT controls (Figure 2A). Moreover, the pSmad3 positive cells appeared to exhibit more intense staining in the Nf1-KO nerves. Consistent with a role in promoting tumorigenesis, pSmad3 staining differed between the Nf1-KO bridge and Nf1-KO distal region, with less pSmad3 positive cells observed in the distal stump (Figure 2B). At day 14 post injury, Nf1-KO SCs in the nerve bridge exhibit a significantly higher rate of proliferation compared to WT cells, indicating that this is a critical time point when Nf1-KO SC behaviour begins to diverge from normal SCs (Figure 1). Since TGF-p signaling was elevated in the Nf1-KO bridge at this timepoint (Figure 1), it was hypothesized that that TGF-p signaling was responsible for the increased rate of proliferation of Nf1-KO SCs at this early time point. In order to test this, mice lacking both Nf1 and Tgfbr2 in SCs (P0-CreERT2:tdTomato:Nf1fl / fl:Tgfbr2fl / fl) were generated. Consistent with the earlier results, Nf1-KO SCs in the bridge showed increased proliferation rates compared to WT SCs in the bridge in this model (Figure 2C-D). Strikingly, Nf1-KO Tgfbr2-KO SC proliferation in the bridge was significantly lower compared to Nf1- KO SCs, with an approximate 2-fold reduction in the percentage of Edll positive SCs (Figure 2C- D). These findings were consistent with previous results, which showed that Tgfbr2-K0 SCs in the wound site did not exhibit any proliferative defect, with loss of TGF- p signaling did not significantly decrease proliferation of WT SCs. Together, these results indicate that TGF-p signaling is important for the enhanced proliferation of Nf1-KO SCs at this critical timepoint.

[0126] Loss TGF- / 3 signaling in Nf1-KO SCs inhibits tumour formation

[0127] To determine whether Nf1-KO SCs form tumours if they are unresponsive to TGF-p signaling, nerves were collected at 6 months following injury and assessed for tumour formation (Figure 3A). In line with previous work, 100% of Nf1-KO mice exhibited visible tumours upon harvest. More strikingly, no tumours were evident in Nf1-KO Tgfbr2-KO mice, and closely resembled the regenerated nerves from WT control and Tgfbr2-KO mice (Figure 3A-F), suggesting that the tumour-forming potential of Nf1-KO SCs was lost in the absence of TGF-p signaling.

[0128] To quantify tumour formation at the macroscopic level, the maximum width and area at the injury site from the regenerated nerve was measured (Figure 3D-E). Nf1-KO nerves were wider at the wound site, compared to those harvested from the other three genotypes (Figure 3D). In contrast, nerves harvested from Nf1-KO Tgfbr2-K0 mice were indistinguishable from control nerves, suggesting a complete inhibition of tumour formation (Figure 3D-E). A small decrease in the nerve width of Tgfbr2-KO nerves compared to Nf 1 - KO Tgfbr2-KO and control nerves, was observed (Figure 3D-F). This decrease is consistent with our previous studies on the effects of TGF-p on nerve regeneration, in that loss of this signaling pathway leads to delayed and misdirected nerve regeneration (Clements et al., 2017).

[0129] Adult peripheral nerve is a highly quiescent tissue with all cell types within the endoneurium exhibiting very low turnover rates. In contrast, after injury, all SCs, at the site of, and distal to the injury become proliferative as part of the reprogramming process associated with nerve repair. However, these cells subsequently return to the quiescent state as they redifferentiate as the nerves repair. To determine whether SCs continued to proliferate in the tumour environment, the proliferation rate of SCs in the tumour / regenerated tissue at 6 months post-injury was measured. Consistent with previous studies, no tdTomato+ mSCs in uncut nerve incorporated Edll, highlighting the quiescent nature of uninjured nerve tissue (Figure 3H-I). SCs in WT control regenerated tissue exhibited low levels of proliferation, with less than 0.5% of tdTomato+ cells incorporating Edll. In contrast, Nf1-KO SCs were far more proliferative, with approximately 3% of tdTomato+ cells having incorporated EdU (Figure 3I). Moreover, Nf1 - KO Tgfbr2-KO SCs were significantly less proliferative than Nf 1 - KO SCs, indicating that TGF-p signaling loss in Nf1-KO SCs reduces their proliferation. However, the Nf1-KO Tgfbr2-KO SCs were still significantly more proliferative than the control. This suggests that other factors also stimulate Nf1-KO SC proliferation but are not sufficient to drive tumourigenesis.

[0130] Collectively, this data demonstrates that Nf1-KO Tgfbr2-KO mice failed to form tumours following peripheral nerve injury, indicating that TGF-p is an essential component of the microenvironment that is necessary for NF1 -neurofibroma formation.

[0131] Composition of the tumour at 6 months

[0132] Nf1 -deficient SCs found in neurofibromas have been reported to have a strikingly different phenotype to normal SCs, are found less in contact with axons and remain in a de- differentiated proliferative state (Zhu et al., 2002). Moreover, neurofibromas have a more disorganised structure than regenerated nerve, which while hypercellular than undamaged nerve, becomes polarised and organized as the axons return to their targets and SCs reassociate with the axons (Ribeiro ref and others). To determine the effects of loss of TGF-p signaling in SCs on the morphology and organisation of regenerated nerve tissue longitudinal cryosections of regenerated nerves were assessed in order to visualise the nerve bridge and tumour regions together with the proximal and distal stumps. In Nf1-KO tissue, three distinct regions were observed, the uncut fascicle, a regenerated region, and a tumour region adjacent to the main nerve bundle (Figure 4A). The uncut fascicle displays no remarkable differences between genotypes, consistent with previous studies of Nf1-KO and Tgfbr2-KO mice (Clements et al., 2017; Ribeiro et al., 2013). However, between regenerated regions, Nf1-KO SCs exhibited a similar morphology to WT SCs. However, cellularity in this region was increased compared to control (Figure 4B). Moreover, in the tumour region, cellularity was even higher than in the regenerated Nf1-KO region (Figure 4B) indicative of higher rates of proliferation. In addition to changes in cellularity, Nf1-KO SCs exhibited a distinct phenotype, with a less polarised morphology and a more rounded shape, with the tumour appearing much more disorganised (Figure 4C). In contrast, Nf1- KO Tgfbr2-KO SCs in the regenerated region were morphologically similar to SCs in the regenerated regions of WT and Nf1-KO regenerated regions, as they exhibited elongated polarised morphology.

[0133] Collectively, this data illustrated that Nf1-KO tissue has 2 distinct regions: the regenerated region, in which SCs are morphologically similar to the controls but the region shows increased cellularity, and the tumour region, unique to Nf1-KO tissue, in which SCs are morphologically distinct from all other genotypes and the region shows even higher levels of cellularity. As such, our findings show that loss of TGF-p signaling in Nf1-KO tissue prevents the emergence of this tumour region, but does not alter cellularity nor SC morphology in non-tumour regions. Previous work has identified the formation of Glutl positive mini fascicles at the injury site, where they are purported to play a role in bundling regrowing axons during nerve regeneration. Consistent with this, Glutl + cells were mostly longitudinally polarised, seeming to form cords / tubes from the proximal to distal region (Figure 4C-D). In the Nf1-KO, Glutl + cells were distinct between the regenerative tissue and the tumour. In the regenerated region, Glut1+ cells formed minifascicles that were longitudinally polarized, resembling those seen in WT control regenerated tissue. However, in the Nf1-KO tumour region, Glutl + cells frequently lacked longitudinal polarity, and instead appeared to enwrap Nf1-KO tumour cells, reflecting the disorganization of the SCs within the tumour (Figure 4C-D). Furthermore, the Glutl area appeared to cover the entire tumour region, acting as a proxy for the tumour, and allowing the ability to observe more easily the spread of the tumour within the regenerated region (Figure 4D). Quantification of the Glut1+ cell area further confirmed the presence of additional regenerated tissue in these mice with a combination of regenerated and tumour tissue (Figure 4D-E), the presence of the minifasicles within the tumour region provides additional evidence that neurofibromas resemble an unrepaired wound.

[0134] In previous studies utilising the same experimental model, recombined Nf1-KO SCs in the neurofibroma that expressed the Glutl PnC marker was identified which developed a perineurial-like morphology (Ribeiro et al., 2013). Since perineurial cells are not neural crest- derived cells (Kucenas, 2015), expression of Glutl by Nf1-KO SCs indicated that these cells were highly abnormal. To assess trans-differentiated SCs, Glutl / tdTomato colocalisation was quantified. Consistent with previous results, numerous trans-differentiated SCs were detected in Nf1-KO tissue, and were absent in WT control tissue, indicating that the trans-differentiated SC phenotype is dependent on Nf1 loss (Figure 4F-G). Interestingly, although trans-differentiated cells were detected in both the tumour region and regenerated region of Nf1-KO tissue, our quantification revealed that trans-differentiated Nf1-KO SCs were significantly more numerous in the tumour region (Figure 4F-G). Strikingly, in Nf1-KO:Tgfbr2-KO regenerated tissue, very few trans-differentiated cells were detected, significantly fewer than in Nf1-KO regenerated tissue. This finding suggests that Nf1-KO SC trans-differentiation relies on TGF-p signaling for this process.

[0135] TGF-ft signaling mediates the escaped population of SCs

[0136] To understand how early microenvironmental TGF-p signals contribute to tumour formation, the composition of the regenerating nerve at Day 14 was assessed. Interestingly, we identified an ‘escaped’ population of proliferating SCs outside of the re-forming perineurium, which corresponded with the location of neurofibroma formation in Nf1-KO mice at 6 months (Figure 5A). In Nf1-KO Tgfbr2-KO mice, the escaped region was significantly smaller than that observed in the Nf1-KO mice (Figure 5B), indicating that Nf 1 KO SCs have a propensity to escape and proliferate outside of the reforming fascicles, which may be partially dependent on TGF-p signaling. Further comparison between Nf1-KO and Nf1-KO Tgfbr2- KO mice revealed a higher number of recombined SCs, with a higher proliferative index within the NF1-KO escaped region (Figure 5C). This high dependence on TGF-p signaling for the number and proliferation of these escaped cells, together with the location on subsequent tumour formation suggests that these escaped cells go on to form the neurofibromas observed at 6 months. As such, characterising the escaped region at this time point may provide insight into the role of TGF-p signaling in subsequent tumour formation.

[0137] Nf1-KO SCs are more proliferative in the escaped region compared to the regenerating region, suggesting the microenvironment of the escaped region is more conducive for Nf 1 - KO SC proliferation and ultimately tumour formation. During normal nerve regeneration, SC proliferation is halted by contact with re-growing axons, which induces SC redifferentiation and return to quiescence (Jessen & Mirsky, 2019). To assess whether axons were present in the escaped region, nerves were harvested from Nf1-KO mice at day 14 post-injury, and immunostained for the axonal marker neurofilament (NF). Remarkably, these results showed a high density of axons in the escaped region (Figure 5D). However, axons in this region were disordered and projected in multiple directions, as opposed to regrowing axons within the regenerating region, that regrow from the proximal to distal stump. Moreover, within the escaped region, axons appeared to be associated with Nf1-KO SCs, suggesting that the increased proliferation of Nf1-KO SCs is not due to lack of axonal contact.

[0138] To further characterise the features of the pre-tumour microenvironment, Nf1-KO nerves at Day 14 were stained for macrophages (F4 / 80) and blood vessels (CD31). Strikingly a substantial number of F4 / 80+ macrophages were detected in the escaped region (Figure 5G). Moreover, the escaped region exhibited extensive vascularisation in this region. Numerous CD31+ blood vessels were detected, which displayed considerable disorder (Figure 5E). Similarly, to axons, the neo-vasculature that is generated after injury is longitudinally polarised, whereas blood vessels in the escaped region are not, indicating the escaped region is distinct from the regenerating nerve and represents the early stages of tumour development.

[0139] Nf1-KO SCs in the escaped region are more numerous and proliferative, with the cell types observed in the escaped region corresponding to stromal cell types seen in the neurofibroma microenvironment (Figure 5A-G), indicating that the tumour originates from the escaped Nf 1 - KO SCs. Importantly, loss of TGF-p signaling in these cells resulted in a strong reduction in their frequency, density, and proliferation, which appears to be sufficient in completely preventing tumour formation at later stages. In order to identify the cell type(s) that express TGF-p at this timepoint, in situ hybridisation chain reaction (HCR) for Tgfbl was performed. In the uncut region of the nerve, low Tgfbl levels were observed, consistent with similar findings from earlier studies (Clements et al., 2017). In the regenerating region of the Nf1-KO nerve, an increase in the Tgfbl signal was detected, with markedly higher expression of Tgfbl observed in the escaped region versus the regenerating region (Figure 5H-I), consistent with a role for TGF-p signaling in promoting tumourigenesis in this region. Although TGF-p could be detected in many different cell types, some cells exhibited much higher levels, virtually all macrophages expressed Tgfbl . Macrophages appeared to be particularly enriched in Tgfbl mRNA, as Iba1+ cells contained more punctae than neighbouring I ba1 - cells (Figure 5J quantified in Figure 5K). This result, combined with the large proportion of macrophages that are present in the Nf1-KO escaped region, suggested that macrophages are a major source of TGF-p ligands, highlighting their tumourigenic importance at this tumour initiation timepoint.

[0140] TGF- / 3 negatively regulates SC differentiation

[0141] The data has shown that TGF-p signaling is required for neurofibroma formation, without which tumourigenesis does not occur. As mentioned previously, loss of this signaling in Nf1 - KO SCs resulted a reduction in the number and proliferation of escaped Nf1-KO SCs during the very early timepoints of tumourigenesis, implying that TGF-p is required for the initiation of tumour formation from this abnormal population of cells. TGF-p was therefore examined to assess how it affects SC behaviour to promote a tumour-forming phenotype. Given the earlier results showing that Nf1-KO SCs in the tumour region exhibit a more progenitor-like morphology and that loss of TGF-p drastically reduced the occurrence of transdifferentiated Nf1-KO SCs, it was hypothesised that TGF-p functions by altering SC plasticity and modulating SC state. TGF-p was examined to assess whether it similarly prevents SC myelination of adult DRG axons.

[0142] An adult mouse DRG - rat SC co-culture system was developed (Figure 6A). In the presence of ascorbic acid, MBP+ myelin sheaths were prominently visible. However, when cultured in the presence of ascorbic acid and TGF-p, no myelin sheaths could be detected, indicating that TGF-p prevented SC myelination of adult PNS axons (Figure 6B-C). Importantly, the ability of SCs to form tight axonal association was also assessed (Figure 6F). In these experiments, SCs were seeded onto adult DRG explants, and after 12h were scored for their association with axons, as performed in our earlier study (Parrinello et al., 2008). SCs were categorised into 3 groups based on their behaviour relative to axons: 1) Associated and aligned: cells in this category display tight parallel alignment of their cytoplasm and nucleus to axonal processes; 2) Associated, not aligned: cells in this category have successfully made multiple cytoplasmic contacts with axons but have not aligned their cytoplasm and / or nucleus parallel to the axons; 3) Non-associated: cells have been unsuccessful in making axonal contacts despite close proximity to axons. As expected, in the absence of TGF-p, the vast majority of SCs successfully associated and aligned to an axon. However, in the presence of TGF-p, we found that a substantially reduced proportion of SCs were able to associate and align themselves to an axon. It was shown that, using Nf1- targeted siRNA, the loss of Nf1 also negatively affects SC axonal interactions, consistent with the findings from the Parrinello et al. study. Furthermore, Nf1- knockdown SCs cultured in the presence of TGF-p exhibited further impairment in their axonal association (Figure 6F). Interestingly, TGF-p did not cause detachment of SCs that had been previously allowed to associated to axons. This data shows that TGF-p disrupts normal SC - axon interaction, and is exacerbated by Nf1 -deficiency.

[0143] We then examined whether TGF-p was able to similarly prevent SC differentiation in an in vitro dibutyryl-cAMP (dbcAMP) model of SC differentiation (Figure 6G). Confluent cultures of SCs that were cultured in the presence of dbcAMP exhibited considerable increases in the mRNA levels of myelin-associated genes MPZ, MBP, PRX, and Krox20 / Egr2. Presence of TGF-p in the culture medium strongly impaired the upregulation of these genes, implying that TGF-p prevents SC differentiation (Figure 6H). Using the MEK inhibitor PD184352, it was also found that inhibition of ERK signaling had no significant impact on the antidifferentiation effects of TGF-p (Supplementary Figure S6B), ruling out the possibility that the effects of TGF-p were mediated through non-canonical activation of the Ras / Raf / MEK / ERK pathway.

[0144] Pharmacological inhibition of TGF- / 3 impairs tumour formation

[0145] The results indicate that TGF-p signaling is necessary for neurofibroma formation by altering Nf1-KO SC behaviour at the earliest stage of tumourigenesis, where it is important for sustaining the proliferation of Nf1-KO SCs that escape the normalising environment of a regenerating nerve. This suggests a critical time-window, in which the escaped population is primed for tumourigenesis, and that therapeutically targeting TGF-p signaling for a limited period at this timepoint may allow Nf1-KO SCs to undergo a fate decision that impairs tumourigenesis.

[0146] To test this, TGF-p signaling was inhibited from day 7 to 9 using Galunisertib (LY2157299), a well-characterised specific TGF-p signaling inhibitor. Following treatment with Galunisertib, a significant decrease in total Nf1-KO SC proliferation was observed compared to the control at Day 14 (Figure 7A-B). Similarly, in the escaped region that goes on to form a tumour, proliferation Nf1-KO SCs was reduced following inhibition of TGF-p (Figure 7C). Consistent with this, loss of TGF-p signaling in Nf1-KO SCs via Tgfbr2-KO led to reduced number, density, and proliferation of cells in the escaped region at day 14. This result suggested that TGF- pR1 inhibition successfully impaired Nf1-KO SC proliferation during the initial stage of tumour formation, with similar effectiveness as SC-specific genetic ablation of the TGF-PR2.

[0147] In the genetic model, loss of TGF-p signaling in Nf1-KO SCs resulted in a reduction in Nf 1 - KO SC proliferation and in the escaped region size and content, which correlated with the absence of a tumour formation. Given that pharmacological inhibition of TGF-p signaling achieved similar effects at this early timepoint regarding Nf1-KO SC proliferation and the escaped region, we next aimed to test whether Galunisertib treatment during this early therapeutic window is sufficient to inhibit tumour formation.

[0148] Upon harvest at 6 months post-injury, tumours were visible at the site of injury in 100% of untreated mice. Remarkably, in mice treated with Galunisertib shortly after nerve injury, only 50% of mice had a visible tumour at the site of injury, with the other half showing little to no visible enlargement to the injury site (Figure 7D-E). A significant reduction in the width and area of nerves harvested from Galunisertib- treated mice, indicated impaired tumour formation following TGF-p signaling inhibition (Figure E-G). Importantly, while Galunisertib treatment reduced the frequency of tumour formation, this was not as much as the reduction seen in the Nf1-KO Tgfbr2-KO genetic model, in which tumour formation was completely abolished from (Figure 3.X). Additionally, nerves from Galunisertib-treated mice were slightly larger than that of Nf1-KO Tgfbr2-KO mice. This finding indicated that pharmacological inhibition of TGF-p in the early microenvironment was sufficient to prevent tumour formation. This finding has major importance regarding the therapeutic potential of TGF-p inhibition in neurofibroma.

[0149] Schwann cell-axonal interactions are impaired in the escaped region at Day 14

[0150] Since SC contact with re-growing axons is a major stimulus for SC re-differentiation during regeneration (Cattin & Lloyd, 2016), we hypothesised that a lack of SC-axonal interactions in the escaped region may confer a proliferative advantage of escaped Nf1-KO SCs, maintaining them in a ‘repair’ state while regenerating Nf1-KO SCs begin to exit the cell cycle. Interestingly, axonal staining of longitudinal sections from Nf1-KO nerves revealed that axons were in fact ubiquitously present in the escaped region in similar numbers to the regenerating region (Figures 8A-B). However, axons in the escaped region were far more disordered and frequently deviated from the proximal-to-distal trajectory (Figures 8C-D). Importantly, while Nf1-KO SCs in the regenerating region were seldom devoid of axonal contact (<5% of Nf1-KO SCs), a far greater proportion of Nf1-KO SCs in the escaped region exhibited this abnormal behaviour, with a 3-fold increase in the proportion of Nf1-KO SCs that were not in contact with axons (Figure 1 F, quantified in Figure 1 E). In comparison, virtually all control SCs were in contact with axons, highlighting that loss of Nf 1 impairs SC / axonal interactions (Parrinello et al., 2008).

[0151] Collectively, these findings indicate that the microenvironment of the escaped region promotes a pro-tumourigenic state of Nf1-KO SCs, which is characterised by abnormally high proliferation and impaired axonal association. This, combined with its location corresponding to the later tumour, implies that the escaped Nf1-KO SCs are the tumourinitiating population that will give rise to the neurofibroma.

[0152] Tgffl signalling is upregulated in the escaped region at Day 14

[0153] The findings have demonstrated that the escaped region corresponds to the early stages of tumourigenesis. However, although loss of TGFp signalling in Nf1-KO SCs reduced the size of this early tumour region and prevented subsequent tumour formation, it is unclear how TGFp is involved in this process. Therefore, TGFp signalling was analysed to see whether it was elevated in the escaped region compared to the regenerating region of Nf1- KO nerves at day 14 post-injury. Smad3 is a critical intracellular mediator of TGFp signalling, which is regulated through phosphorylation by the TGFp receptor complex. Therefore, using phosphorylated Smad3 (pSmad3) as a readout, it was found that TGFp signalling was significantly elevated in the escaped region compared to the regenerating region, with a greater percentage of pSmad3+ cells and more intense pSmad3 signal per cell observed (Figures 9A-C). In addition, we performed in situ hybridisation experiments to compare the expression of TGFp ligands in the escaped region and in the regenerating region. Consistent with pSmad3 staining, significantly more Tgfbl mRNA was detected in the escaped region than in the regenerating region at this timepoint (Figures 9D-E). These results demonstrate that TGFp signalling is enriched in the escaped region in early tumourigenesis.

[0154] The cell type which was the source of the TGFp in the escaped region was then assessed. In many tissues, TGFp is commonly produced by immune cell types (Massague & Sheppard, 2023). The Day 14 Nf1-KO escaped region was stained to examine the abundance of macrophages, mast cells, T cells, and neutrophils, all of which are known to be recruited to the site of injury. It was found that macrophages were highly abundant in the escaped region at this timepoint, while all other immune cell types tested were virtually absent (Figures 10A-B). Since the major immune cell type present in the escaped region at Day 14 are macrophages, it was speculated that they may be producing TGFp. Indeed, previous scRNA-seq data showed that TGFp is primarily produced by macrophage / monocyte populations at days 1 , 3, and 7 after nerve injury (Zhao et al., 2022). In order to test this, co-staining for macrophages and Tgfbl mRNA was performed, and macrophages were found to express significantly more TGFp than non-macrophage cells (Figures 10C-D), including tdTomato+ SCs. In keeping with this, macrophages were also found to be more numerous in the escaped region than in the regenerating region, and account for almost 30% of all cells in the regenerating region (Figures 3E-G). Since macrophages are the main source of TGFp at this timepoint, it is likely that the increased TGFp signalling in the escaped region is a result of the increased density of macrophages in this area. This further supports a crucial role for macrophages during early tumorigenesis, and highlights a TGFp-dependent macrophage-SC axis for tumour initiation.

[0155] Tgfp regulates SC-axonal interactions

[0156] Previous results revealed that numerous Nf1-KO SCs in the tumour exhibited a repair SC morphology, implying a failure to re-differentiate and instead remain in a progenitor-like state (Kershner et al., 2022; Parrinello & Lloyd, 2009; Zheng et al., 2008). Importantly, the TGFp-enriched escaped region it was found that Nf1-KO SCs were more frequently found without axonal contact. TGFb was then assessed to see whether it directly exerts a tumourigenic effect on SCs by disrupting SC / axonal interactions, thus counteracting the pro-differentiation stimulus provided by the axon. SC / axonal interactions were assayed by co-culturing SCs with DRG axons (Parrinello et al., 2008) using adult DRG axons to better mimic the in vivo system.

[0157] Since TGFp is a highly pleiotropic cytokine acting through multiple pathways, the pathway downstream of TGFp ligand binding responsible for its tumourigenic effects was investigated. Given that the TGFp is able to activate Ras / Raf / MEK / ERK signalling, it was assessed whether the disruption to SC / axonal interactions induced by TGFp were prevented by the MEK inhibitor PD184352 or Smad3 inhibitior SIS3. The same DRG SC / axon interaction assays as earlier was assessed but in the presence of either inhibitor and found that SMAD inhibition restored SCs’ ability to correctly associate and align with axons, whereas MEK inhibition had no effect (Figure 11A-B), further demonstrating the reliance on downstream SMAD signalling.

[0158] The forgoing embodiments are not intended to limit the scope of the protection afforded by the claims, but rather to describe examples of how the invention may be put into practice.

Claims

Claims1. A composition comprising an inhibitor of TGF-p for use in the prevention, management and / or treatment of tumours.

2. The use of the composition according to claim 1 , wherein the tumour is associated with Neurofibromatosis Type 1 (NF1).

3. The use of the composition according to claim 1 or 2, wherein the tumour originates in a Schwann cell (SC).

4. The use of the composition according to any preceding claim, wherein the tumour is in the peripheral nervous system (PNS).

5. The use of the composition according to any preceding claim, wherein the composition is for administration prior to, during or after injury to the PNS.

6. The use of the composition according to claim 5, wherein the composition is for administration prior to, during or after excision of a pre-existing tumour located within or adjacent to at least part of the PNS.

7. The use of the composition according to any preceding claim, wherein the composition is for administration to the area surrounding a pre-existing tumour prior to excision of the tumor and / or to the area previously surrounding the pre-existing tumor immediately after excision.

8. The use of the composition according to any preceding claim, wherein the composition is administered in combination with, or further comprises, one or more additional anti-cancer agents.

9. A composition comprising an inhibitor of TGF-p for use in the prevention, management and / or treatment of a neurofibroma.

10. The use of the composition according to claim 9, wherein the neurofibroma is associated with Neurofibromatosis Type 1 (NF1).

11. The use of the composition according to claim 9 or 10, wherein the neurofibroma originates in a Schwann cell (SC).

12. The use of the composition according to claims 9 to 11 , wherein the neurofibroma is in the peripheral nervous system (PNS).

13. The use of the composition according to any preceding claim, wherein the inhibitor of TGF-p comprises a dihydropyrrolopyrazole.

14. The use of the composition according to any preceding claim, wherein the inhibitor of TGF-p comprises 2-(6-methyl-pyridin-2-yl)-3-[6-amido-quinolin-4-yl)-5,6-dihydro- 4H~pyrrolo[l,2-b]pyrazole monohydrate, or derivatives thereof.

15. A pharmaceutical composition, comprising a TGF-p inhibitor and a pharmaceutically acceptable carrier, excipient, or diluent, wherein the TGF-p inhibitor comprises dihydropyrrolopyrazole.

16. The pharmaceutical composition of claim 15, wherein the dihydropyrrolopyrazole comprises 2-(6-methyl-pyridin-2-yl)-3-[6-amido-quinolin-4-yl)-5,6-dihydro-4H~pyrrolo[l,2-b]pyrazole monohydrate, or derivatives thereof.

Images