WO2019083457A1 - Varlitinib for use in treating cancer in a patient identified as having a beta-catenin pathway mutation - Google Patents

Varlitinib for use in treating cancer in a patient identified as having a beta-catenin pathway mutation

Info

Publication number
WO2019083457A1
WO2019083457A1 PCT/SG2018/050537 SG2018050537W WO2019083457A1 WO 2019083457 A1 WO2019083457 A1 WO 2019083457A1 SG 2018050537 W SG2018050537 W SG 2018050537W WO 2019083457 A1 WO2019083457 A1 WO 2019083457A1
Authority
WO
WIPO (PCT)
Prior art keywords
cancer
varlitinib
catenin
compound
formula
Prior art date
Application number
PCT/SG2018/050537
Other languages
French (fr)
Inventor
Ann Gee Lisa OOI
Qihui SEET
Original Assignee
Aslan Pharmaceuticals Pte Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Aslan Pharmaceuticals Pte Ltd filed Critical Aslan Pharmaceuticals Pte Ltd
Publication of WO2019083457A1 publication Critical patent/WO2019083457A1/en

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61K—PREPARATIONS FOR MEDICAL, DENTAL OR TOILETRY PURPOSES
    • A61K31/00—Medicinal preparations containing organic active ingredients
    • A61K31/33—Heterocyclic compounds
    • A61K31/395—Heterocyclic compounds having nitrogen as a ring hetero atom, e.g. guanethidine or rifamycins
    • A61K31/495—Heterocyclic compounds having nitrogen as a ring hetero atom, e.g. guanethidine or rifamycins having six-membered rings with two or more nitrogen atoms as the only ring heteroatoms, e.g. piperazine or tetrazines
    • A61K31/505—Pyrimidines; Hydrogenated pyrimidines, e.g. trimethoprim
    • A61K31/517—Pyrimidines; Hydrogenated pyrimidines, e.g. trimethoprim ortho- or peri-condensed with carbocyclic ring systems, e.g. quinazoline, perimidine
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61K—PREPARATIONS FOR MEDICAL, DENTAL OR TOILETRY PURPOSES
    • A61K45/00—Medicinal preparations containing active ingredients not provided for in groups A61K31/00 - A61K41/00
    • A61K45/06—Mixtures of active ingredients without chemical characterisation, e.g. antiphlogistics and cardiaca
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61P—SPECIFIC THERAPEUTIC ACTIVITY OF CHEMICAL COMPOUNDS OR MEDICINAL PREPARATIONS
    • A61P35/00—Antineoplastic agents
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
    • G01N33/574—Immunoassay; Biospecific binding assay; Materials therefor for cancer
    • G01N33/57484—Immunoassay; Biospecific binding assay; Materials therefor for cancer involving compounds serving as markers for tumor, cancer, neoplasia, e.g. cellular determinants, receptors, heat shock/stress proteins, A-protein, oligosaccharides, metabolites

Definitions

  • VARLITINIB FOR USE IN TREATING CANCER IN A PATIENT IDENTIFIED AS
  • the present disclosure relates to a therapy, for example a monotherapy or combination therapy comprising a type I tyrosine kinase inhibitor for the treatment of cancer, for example liver cancer, such as hepatocellular carcinoma, in patients who are identified as having a mutation in the beta- catenin pathway.
  • a therapy for example a monotherapy or combination therapy comprising a type I tyrosine kinase inhibitor for the treatment of cancer, for example liver cancer, such as hepatocellular carcinoma, in patients who are identified as having a mutation in the beta- catenin pathway.
  • Tumor heterogeneity may also contribute to resistance, where small subpopulations of cells may acquire or stochastically already possess some of the features enabling them to emerge under selective drug pressure. This is a problem that many patients with cancer encounter, and it obviously limits the therapeutic alternatives that are effective and worsens the prognosis.
  • Cancer therapy guidelines describe the sequence of therapies, which are recommended and in which sequence, so that if a patient shows disease progression on the first therapy ("first line”), then a next therapy (“second line”) is recommended, and so on. These therapy recommendations are based on available scientific data and experience, and illustrate that resistance to one therapy does not exclude that another therapy may be effective and prolong life or shrink a tumor. At late stages cancers do not respond, no more avenues of therapy exist, and are completely therapy refractory. Thus, unless new therapies can be found, which are effective, these cancers cannot be treated.
  • Varlitinib is in clinical trials for the treatment of cancer.
  • the present inventors have data that suggest the molecule is particularly effective in the treatment of a subcategory of cancers, namely those where cancer cells have a mutation in the beta-catenin pathway. These cancers seem to be sensitive to treatment with Varlitinib.
  • Varlitinib is an inhibitor of a mutated beta- catenin pathway.
  • a method of treating a patient for cancer by administering a compound of formula (I):
  • liver cancer such as hepatocellular carcinoma
  • biliary tract cancer such as gall bladder cancer
  • breast cancer such as none ER+ breast cancer
  • prostate cancer colorectal cancer
  • ovarian cancer cervical cancer
  • lung cancer gastric cancer
  • pancreatic bone cancer
  • bladder cancer head and neck cancer
  • thyroid cancer skin cancer
  • renal cancer and oesophagus cancer and combinations of two or more of the same.
  • each dose of the compound of formula (I) is in the range 100 to 900mg, for 100, 200, 300, 400, 500, 600, 700, 800 or 900mg.
  • a method according to paragraph 13, wherein each dose of the compound of formula (I) is in the range 200 to 500mg or 100 to 500mg, for example 250 mg or 490 mg.
  • each dose is lOOmmg, 200mg, 300g or 400mg (such as 300mg or 400mg).
  • the chemotherapeutic agent is independently selected from the group comprising a platin (such as cisplatin or oxaliplatin), gemcitabine, capecitabine, 5-FU, FOLFOX, FOLFIRI and FOLFIRINOX.
  • liver cancer such as hepatocellular carcinoma
  • biliary tract cancer such as gall bladder cancer
  • breast cancer such as none ER+ breast cancer
  • prostate cancer colorectal cancer
  • ovarian cancer cervical cancer
  • lung cancer gastric cancer
  • pancreatic bone cancer
  • bladder cancer head and neck cancer
  • thyroid cancer skin cancer
  • renal cancer and oesophagus cancer and combinations of two or more of the same.
  • each dose of the compound of formula (I) is in the range 200 to 500mg or 100 to 500mg, for example 250 mg or 490 mg.
  • each dose is lOOmg, 200mg, 300mg or 400mg (such as 300mg or 400mg).
  • a chemotherapeutic agent is independently selected from the group comprising a platin [such as cisplatin or oxaliplatin), gemcitabine, capecitabine, 5-FU, FOLFOX, FOLFIRI and FOLFIRINOX.
  • liver cancer such as hepatocellular carcinoma
  • biliary tract cancer such as biliary tract cancer
  • gall bladder cancer such as none ER+ breast cancer
  • prostate cancer colorectal cancer
  • ovarian cancer cervical cancer
  • lung cancer gastric cancer
  • pancreatic bone cancer
  • bladder cancer head and neck cancer
  • thyroid cancer skin cancer
  • renal cancer and oesophagus cancer and combinations of two or more of the same.
  • each dose of the compound of formula [I) is in the range 100 to 900mg, for 100, 200, 300, 400, 500, 600, 700, 800 or 900mg.
  • each dose of the compound of formula [I) is in the range 200 to 500mg or 100 to 500mg, for example 250 mg or 490 mg.
  • each dose is lOOmg, 200mg, 300mg or 400mg [such as 300mg or 400mg).
  • combination therapy comprises a chemotherapeutic agent.
  • chemotherapeutic agent is independently selected from the group comprising a platin (such as cisplatin or oxaliplatin), gemcitabine, capecitabine, 5-FU, FOLFOX, FOLFIRI and FOLFIRINOX.
  • the mutated beta-catenin localises in a cell's cytoplasm and/or nucleus.
  • the cancer is both HER2 activated and HER3 activated.
  • the cancer has at least one of the following:
  • the biliary duct cancer is in a location selected from intrahepatic bile ducts, left hepatic duct, right hepatic duct, common hepatic duct, cystic duct, common bile duct, Ampulla of Vater and combinations thereof.
  • the biliary duct cancer is in an intrahepatic bile duct. In one embodiment the biliary duct cancer is in a left hepatic duct. In one embodiment the biliary duct cancer is in a right hepatic duct. In one embodiment the biliary duct cancer is in a common hepatic duct. In one embodiment the biliary duct cancer is in a cystic duct. In one embodiment the biliary duct cancer is in a common bile duct. In one embodiment the biliary duct cancer is in an Ampulla of Vater. In one embodiment the biliary duct cancer is a cancer of the Papilla of Vater.
  • the cancer is a metastatic form of a cancer, in the patient population disclosed herein disclosed herein.
  • the cancer according the present disclosure has not metastasized in said patient population.
  • the compound of formula (I) is (i?)-N4-[3-Chloro-4-(thiazol-2- ylmethoxy)-phenyl]-N6-(4-methyl-4, 5,-dihydro-oxazol-2-yl)-quinazoline-4,6-diamine:
  • the patient population is not activated for all of HER1, HER2, HER3 and HER4, for example the population is at least negative for HER1 and/or HER4.
  • the patient population is HER2 and/or HER3 and HER1 activated but not HER4 activated.
  • the patient population is HER2 and/or HER3 activated and HER4 activated but not HER1 activated.
  • the patient population is only HER2 and/or HER3 activated.
  • the patient population is only HER2 activated.
  • the patient population is only HER3 activated.
  • the patient population is both HER3 and HER4 activated.
  • the compound of formula (I) an enantiomer thereof or a pharmaceutically acceptable salt thereof is employed as a monotherapy, for example first line therapy or second line therapy, such as a first line monotherapy, in said patient population.
  • the compound of formula (I) an enantiomer thereof or a pharmaceutically acceptable salt thereof is employed in a combination therapy, for example in combination with a chemotherapy and/or a biological therapeutic, in particular as a first line therapy or a second line therapy in said patient population.
  • the compound of formula (I) an enantiomer thereof or a pharmaceutically acceptable salt thereof is employed as a second line monotherapy, in said patient population.
  • CA19-9 is a marker employed in the management of cholangiocarcinoma.
  • a HER3 and HER4 positive or amplified patient has a 10, 20, 30, 40, 50, 60, 70, 80 or 90% decrease in CA19-9 level whilst on the therapy according to the present disclosure, wherein the level is decreased relative to the level of CA19-9 before initiation of said therapy.
  • This decrease in CA19-9 may be observed 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or weeks after initiating therapy according to the present disclosure.
  • the compound of formula (I), according to the present disclosure is employed in a second line therapy together with a chemotherapy agent or chemotherapy regimen, for example gemcitabine, capecitabine, 5-FU, FOLFOX, a platin, such as cisplatin or oxaliplatin, and a combination thereof, in said patient population.
  • a chemotherapy agent or chemotherapy regimen for example gemcitabine, capecitabine, 5-FU, FOLFOX, a platin, such as cisplatin or oxaliplatin, and a combination thereof, in said patient population.
  • a chemotherapy agent or chemotherapy regimen for example gemcitabine, capecitabine, 5-FU, FOLFOX, a platin, such as cisplatin or oxaliplatin, and a combination thereof, in said patient population.
  • the compound of formula [I), according to the present disclosure is administered orally.
  • the compound of formula (I), according to the present disclosure such as Varlitinib, is administered at a dose in the range lOOmg to 900mg on each occasion, in particular 200 mg, 300mg, 400mg or 500mg each dose, such as 400mg, 250mg or 490 mg, for example administered once or twice daily, such as twice daily.
  • the compound of formula (I), according to the present disclosure such as Varlitinib is administered twice daily.
  • the compound of formula (I), according to the present disclosure such as Varlitinib, is administered for a 28 days, referred to herein as a 28 day treatment cycle.
  • the compound of formula (I), according to the present disclosure such as Varlitinib, is administered as pharmaceutical formulation comprising one or more pharmaceutically acceptable excipients.
  • the compound of formula (I), according to the present disclosure such as Varlitinib, or a formulation comprising the same is administered orally, for example as tablet or capsule.
  • the treatment is adjuvant therapy, for example after surgery or after chemotherapy, in said patient population.
  • the treatment is neoadjuvant therapy, for example before surgery, in particular to shrink the tumour or tumours, in said patient population.
  • the treatment according to the present disclosure is suitable for the treatment of secondary tumours, in said patient population.
  • the cancer is metastatic cancer, in said patient population.
  • the treatment according to the present disclosure is suitable for the treatment of primary cancer and metastases, in said patient population.
  • the treatment according to the present disclosure is suitable for the treatment of secondary cancer and metastases.
  • the treatment according to the present disclosure is suitable for the treatment of primary cancer, secondary cancer and metastases in said patient population.
  • the treatment according to the present disclosure is suitable for the treatment of cancerous cells in a lymph node, for a cancer of the present disclosure in said patient population.
  • the patient is a human.
  • a method comprising profiling an ex vivo sample from a patient to identify dysregulation in at least two genes independently selected from FM03, RORC, NAG, ADIRF, CES3, FAT4, S100A14, SLC16A14, AZGP1, PPL, LOC200772, TNNI1, SLC1A7, BCHE, CYP4A11, NOY6R, ENPEP, HA01, CYP3A5, BHMT, TMEM132A, EPO, FAM19A5, ASNS, HK2, TTBK1, NSMF, HAL, OBSL1, LOCI 00507194, GPC4, CERCAM, ADRA2C, SERPINE1 , SPARC, SLC6A8, SCARF2, HS6ST2, ADM and IRX3.
  • the dysregulation is upregulation (also referred to herein as activation) of at least one gene (for example 1, 2, 3, , 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes) independently selected from FM03, RORC, NAG, ADIRF, CES3, FAT4, SI 00A14, SLC16A14, AZGP1, PPL, LOC200772, TNN11, SLC1A7, BCHE, CYP4A11, NOY6R, ENPEP, HAOl, CYP3A5, and BHMT.
  • upregulation also referred to herein as activation
  • at least one gene for example 1, 2, 3, , 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes
  • the dysregulation is downregulation [also referred to herein as repression ⁇ of at least one gene (for example 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes] independently selected from TMEM132A, EPO, FAM19A5, ASNS, HK2, TTBK1, NSMF, HAL, OBSL1, LOC100507194, GPC4, CERCAM, ADRA2C, SERPINE1, SPARC, SLC6A8, SCARF2, HS6ST2, ADM and IRX3.
  • TMEM132A EPO
  • FAM19A5 for example 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes
  • the dysregulation is upregulation of at least one gene (for example 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes) independently selected from FM03, RORC, NAG, ADIRF, CES3, FAT4, S100A14, SLC16A14, AZGP1, PPL, LOC200772, TNNI1, SLC1A7, BCHE, CYP4A11, NOY6R, ENPEP, HAOl, CYP3A5, and BHMT, and downregulation of at least one gene (for example 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes) independently selected from TMEM132A, EPO, FAM19A5, ASNS, HK2, TTBK1, NSMF, HAL, OBSL1, LOC100507194, GPC4, CERCAM, ADRA2C, SERPINE1, SPARC, SLC6A8, SCARF2, HS6ST2, ADM and IRX3.
  • at least one gene for example 1,
  • the profiling is for diagnostic or prognostic purposes, for example to identify the patient as having cancer, or a subtype thereof.
  • the profiling is for identifying a patient as suitable for treatment with regime according to the present disclosure.
  • treatment of a new patient population comprising a cancer patient with dysregulation according to any one of paragraphs i), ii), iii) and iv).
  • treatment of a patient wherein said patient has been identified within the target population.
  • the CTNNB1 gene provides instructions for making a protein called beta-catenin.
  • This protein is present in many types of cells and tissues, where it is primarily found at junctions that connect neighboring cells (adherens junctions). Beta-catenin plays an important role in sticking cells together (cell adhesion) and in communication between cells.
  • the beta-catenin protein is also involved in cell signaling as an essential part of the Wnt signaling pathway. Certain proteins in this pathway attach (bind) to beta-catenin, which triggers a multi-step process that allows the protein to move into the nucleus. Once in the nucleus, beta-catenin interacts with other proteins to control the activity (expression) of particular genes.
  • the Wnt signaling pathway promotes the growth and division (proliferation) of cells and helps determine the specialized functions a cell will have (differentiation). Wnt signaling is known to be involved in many aspects of development before birth. In adult tissues, this pathway plays a role in the maintenance and renewal of stem cells, which are cells that help repair tissue damage and can give rise to other types of cells.
  • a mutations in ⁇ -catenin pathway refers to any mutation that impacts on the function of the said pathway, in particular a mutation in ⁇ -catenin and/or the gene encoding the same CTNNB1, including genetic alteration, including missense mutations and amplification.
  • the ⁇ -catenin mutation is T41A.
  • the T41A ⁇ -catenin mutation is thought to be constitutively active.
  • a pan-HER inhibitor such as Varlitinib in high p-ErbB2/3 expressing cancers, such as hepatocellular carcinoma (HCC).
  • HCC hepatocellular carcinoma
  • the present inventors have shown that Varlitinib can selectively target cancers that have the T41A ⁇ - catenin mutation.
  • cancer patients such as HCC patients with the T41A CTNNB1 mutation may represent a specific patient population subset that is particularly responsive to treatment with a pan-HER inhibitor, such as Varlitinib.
  • HER2 and/or HER3 activated as employed herein refers to where the cancer expresses an increased level of said receptor, such as total levels of the receptor and/or levels of the activated form of the receptor, such as the phosphorylated form of the receptor, in comparison a corresponding non-cancerous cell and includes any of the following:
  • Phospho-ErbB2/p-ErbB2 or phospho-ErbB3/p-ErbB3 as used herein refers to the phosphorylated form of ErbB2 or ErbB3 respectively. It includes phosphorylation of the receptor at any suitable phosphorylation site.
  • Known phosphorylation sites for ErbB2 include but are not limited to: Tyr 877, Tyrl023, Tyrll39, Tyrll96, Tyrl221, Tyrl222 and Tyrl248.
  • Known phosphorylation sites for ErbB3 include but are not limited to Tyrl054, Tyrll97, Tyrll99, Tyrl222, Tyrl260, Tyrl262, Tyrl289 and Tyrl328.
  • phospho-ErbB2 refers to ErbB2 which has been phosphorylated at
  • phospho-ErbB3 refers to ErbB3 which has been phosphorylated at Tyrl289.
  • increased levels as employed herein refers to a level that is at least 1, 2 or 3, such as 2 standard deviations higher than the mean level in a non-cancerous cell.
  • increased levels as employed herein refers to a level that is above the level of the 95 th percentile of a population of non-cancerous cells. "In a patient population” refers to administration of the therapy to a patient characterised
  • identification in the context of the present disclosure refer to the patient is known to be in the requisite patient population before the administration of the therapy according to the present disclosure.
  • Liver cancer refers to cancer which starts in the liver, including starting from structures located within the liver, such as blood vessels, including hepatocellular carcinoma.
  • the liver cancer is, for example selected from the group hepatocellular carcinoma, cholangiocarcinoma, angiosarcoma, and hepatoblastoma, in particular hepatocellular carcinoma.
  • the primary liver cancer is stage 1, 2, 3 or 4.
  • the liver cancer is secondary or metastasized liver cancer.
  • liver cancer does not include biliary tract cancer, such as cholangiocarcinoma.
  • the cancer is liver cancer, for example a liver metastasis from a primary cancer, for example colon cancer, which has spread to the liver.
  • the liver cancer is HCC hepatocellular carcinoma.
  • Biliary duct cancer also referred to as biliary cancer as employed herein refers to cancer which starts in the bile ducts and includes cholangiocarcinoma and gallbladder cancer.
  • Cholangiocarcinoma as referred to herein is a form of cancer that is composed of mutated epithelial cells (or cells showing characteristics of epithelial differentiation) that originate in the bile ducts which drain bile from the liver into the small intestine, but not including gallbladder cancer.
  • biliary duct cancer General guidelines for operability of biliary duct cancer include: Absence of lymph node or liver metastases; Absence of involvement of the portal vein; Absence of direct invasion of adjacent organs; and Absence of widespread metastatic disease.
  • the cancer is gall bladder cancer.
  • Gallbladder cancer as employed herein cancer which starts in the gallbladder. The following stages are used for gallbladder cancer:
  • Stage 0 (carcinoma in situ): Abnormal cells are found in the inner (mucosal) layer of the gallbladder; these abnormal cells may become cancer and spread into nearby normal tissue; Stage I Cancer has formed and has spread beyond the inner (mucosal) layer to a layer of tissue with blood vessels or to the muscle layer; Stage II Cancer has spread beyond the muscle layer to the connective tissue around the muscle; Stage IIIA Cancer has spread through the thin layers of tissue that cover the gallbladder and/or to the liver and/or to one nearby organ (e.g., stomach, small intestine, colon, pancreas, or bile ducts outside the liver); Stage IIIB Cancer has spread to nearby lymph nodes and beyond the inner layer of the gallbladder to a layer of tissue with blood vessels or to the muscle layer; or beyond the muscle layer to the connective tissue around the muscle; or through the thin layers of tissue that cover the gallbladder and/or to the liver and/or to one nearby organ; Stage IVA Cancer has spread to a main blood vessel
  • the gastric cancer may have spread to nearby lymph nodes; and Stage IVB Cancer has spread to either lymph nodes along large arteries in the abdomen and /or near the lower part of the backbone or to organs or areas far away from the gallbladder.
  • the gastric cancer is selected from the group comprising adenocarcinoma of the stomach, squamous cell carcinomas, lymphoma of the stomach, gastric stromal tumor, and neuroendocrine tumors.
  • Prostate cancer refers to cancer of the prostate, for example ductal adenocarcinoma, transitional cell (urothelial cancer], squamous cell cancer, carcinoid of the prostate, small cell cancer or sarcoma and sarcomatoid cancer.
  • the prostate cancer is any one of the same.
  • Pancreatic cancer as employed herein includes exocrine cancers (including rare forms thereof such as cystitic tumours, and cancer of the acinar cells), endocrine pancreatic tumours (including gastrinomas, insulinomas, somatostatinomas, VIPomas, glucagonomas), pancreatoblastoma, sarcomas of the pancreas and lymphoma.
  • Colorectal cancer refers to cancer or the colon and/or rectum and includes squamous cell cancers, carcinoid tumours, sarcomas and lymphomas.
  • Breast cancer as employed herein refers to cancer of the breast and includes ductal cardinoma in situ, lobular carcinoma in situ, invasive ductal breast cancer, invasive lobular breast cancer, invasive breast cancer, Paget's disease, angiosarcoma of the breast and rare types of breast cancer such as medullary breast cancer, mucinous breast cancer, tubular breast cancer, adenoid cystic carcinoma of the breast metaplastic breast cancer, basal type breast cancer and papillary breast cancer.
  • the breast cancer is one selected from any one of the same.
  • the breast cancer is phyllodes or cystosarcoma phyllodes.
  • Lung cancers are classified according to histological type and are categorized by the size and appearance of the malignant cells seen by a histopathologist under a microscope.
  • two broad classes are distinguished: non-small cell lung carcinoma and small cell lung carcinoma.
  • the epithelial cancer is lung cancer, for example selected from small-cell lung cancer (SCLC) and non-small-cell lung cancer (NSCLC), such as NSCLC.
  • SCLC small-cell lung cancer
  • NSCLC non-small-cell lung cancer
  • Non-small-cell lung carcinoma The three main subtypes of NSCLC are adenocarcinoma, squamous-cell carcinoma and large-cell carcinoma.
  • adenocarcinoma Nearly 40% of lung cancers are adenocarcinoma, which usually originates in peripheral lung tissue.
  • a subtype of adenocarcinoma, the bronchioloalveolar carcinoma, is more common in female never-smokers, and may have a better long term survival.
  • Squamous-cell carcinoma accounts for about 30% of lung cancers. They typically occur close to large airways. A hollow cavity and associated cell death are commonly found at the center of the tumor. About 9% of lung cancers are large-cell carcinoma. These are so named because the cancer cells are large, with excess cytoplasm, large nuclei and conspicuous nucleoli.
  • SCLC Small-cell lung carcinonia-In small-cell lung carcinoma
  • the cells contain dense neurosecretory granules (vesicles containing neuroendocrine hormones), which give this tumor an endocrine/paraneoplastic syndrome association.
  • These cancers grow quickly and spread early in the course of the disease. Sixty to seventy percent have metastatic disease at presentation.
  • the cancer is non-small lung carcinoma.
  • treatment of renal cancer for example renal cell carcinoma and/or urothelial cell carcinoma.
  • renal cancer include squamous cell carcinoma, juxtaglomerular cell tumor [reninoma), angiomyolipoma, renal oncocytoma, Bellini duct carcinoma, clear-cell sarcoma of the kidney, mesoblastic nephroma, Wilms' tumor, mixed epithelial stromal tumor, clear cell adenocarcinoma, transitional cell carcinoma, inverted papilloma, renal lymphoma, teratoma, carcinosarcoma, and carcinoid tumor of the renal pelvis.
  • Renal cancer as employed herein refers to cancer of the kidney
  • the cancer is bladder cancer, for example is any of several types of malignancy arising from the epithelial lining (i.e., the urothelium) of the urinary bladder.
  • the epithelial lining i.e., the urothelium
  • the bladder cancers are transitional cell carcinoma.
  • the other 10% are squamous cell carcinoma, adenocarcinoma, sarcoma, small cell carcinoma, and secondary deposits from cancers elsewhere in the body.
  • the staging of is given below.
  • NX Regional lymph nodes cannot be assessed; NO No regional lymph node metastasis; Nl Metastasis in a single lymph node 2 cm or less in greatest dimension; N2 Metastasis in a single lymph node more than 2 cm but not more than 5 cm in greatest dimension, or multiple lymph nodes, none more than 5 cm in greatest dimension; N3 Metastasis in a lymph node more than 5 cm in greatest dimension
  • the current disclosure extends to any stage of bladder cancer.
  • Cancerous ovarian tumors can start from three common cell types: Surface Epithelium - cells covering the lining of the ovaries; Germ Cells - cells that are destined to form eggs; and Stromal Cells - Cells that release hormones and connect the different structures of the ovaries.
  • the present disclosure relates to treatment of ovarian cancer from any source, for example as described herein, in particular epithelium cells.
  • Epithelial ovarian carcinomas account for 85 to 90 percent of all cancers of the ovaries.
  • Epithelial ovarian tumors develop from the cells that cover the outer surface of the ovary. Most epithelial ovarian tumors are benign (noncancerous). There are several types of benign epithelial tumors, including serous adenomas, mucinous adenomas, and Brenner tumors. Cancerous epithelial tumors are carcinomas - meaning they begin in the tissue that lines the ovaries. These are the most common and most dangerous of all types of ovarian cancers. Unfortunately, almost 70 percent of women with the common epithelial ovarian cancer are not diagnosed until the disease is advanced in stage.
  • LMP tumors ovarian epithelial tumors whose appearance under the microscope does not clearly identify them as cancerous. These are called borderline tumors or tumors of low malignant potential (LMP tumors). The present disclosure includes treatment of the latter.
  • Germ Cell Tumors Ovarian germ cell tumors develop from the cells that produce the ova or eggs. Most germ cell tumors are benign (non-cancerous), although some are cancerous and may be life threatening. The most common germ cell malignancies are maturing teratomas, dysgerminomas, and endodermal sinus tumors. Germ cell malignancies occur most often in teenagers and women in their twenties. Today, 90 percent of patients with ovarian germ cell malignancies can be cured and their fertility preserved.
  • Stromal Tumors - Ovarian stromal tumors are a rare class of tumors that develop from connective tissue cells that hold the ovary together and those that produce the female hormones, estrogen and progesterone. The most common types are granulosa-theca tumors and Sertoli- Ley dig cell tumors. These tumors are quite rare and are usually considered low-grade cancers, with approximately 70 percent presenting as Stage I disease (cancer is limited to one or both ovaries).
  • Primary Peritoneal Carcinoma The removal of one's ovaries eliminates the risk for ovarian cancer, but not the risk for a less common cancer called Primary Peritoneal Carcinoma.
  • Primary Peritoneal Carcinoma is closely rated to epithelial ovarian cancer (most common type). It develops in cells from the peritoneum (abdominal lining) and looks the same under a microscope. It is similar in symptoms, spread and treatment.
  • stage of a tumor can be determined during surgery, when the doctor can tell if the cancer has spread outside the ovaries.
  • the treatment plan and prognosis (the probable course and outcome of your disease) will be determined by the stage of cancer you have.
  • Stage I - Growth of the cancer is limited to the ovary or ovaries.
  • Stage IA - Growth is limited to one ovary and the tumor is confined to the inside of the ovary. There is no cancer on the outer surface of the ovary. There are no ascites present containing malignant cells. The capsule is intact.
  • Stage IB - Growth is limited to both ovaries without any tumor on their outer surfaces. There are no ascites present containing malignant cells. The capsule is intact.
  • Stage IC The tumor is classified as either Stage IA or IB and one or more of the following are present: (1) tumor is present on the outer surface of one or both ovaries; (2) the capsule has ruptured; and (3) there are ascites containing malignant cells or with positive peritoneal washings.
  • Stage II - Growth of the cancer involves one or both ovaries with pelvic extension.
  • Stage HA The cancer has extended to and/or involves the uterus or the fallopian tubes, or both.
  • Stage IIB The cancer has extended to other pelvic organs.
  • Stage IIC The tumor is classified as either Stage HA or IIB and one or more of the following are present: (1] tumor is present on the outer surface of one or both ovaries; (2 ⁇ the capsule has ruptured; and (3) there are ascites containing malignant cells or with positive peritoneal washings.
  • Stage III - Growth of the cancer involves one or both ovaries, and one or both of the following are present: (1) the cancer has spread beyond the pelvis to the lining of the abdomen; and [2) the cancer has spread to lymph nodes.
  • the tumor is limited to the true pelvis but with histologically proven malignant extension to the small bowel or omentum.
  • Stage IIIA - During the staging operation, the practitioner can see cancer involving one or both of the ovaries, but no cancer is grossly visible in the abdomen and it has not spread to lymph nodes. However, when biopsies are checked under a microscope, very small deposits of cancer are found in the abdominal peritoneal surfaces.
  • Stage IIIB The tumor is in one or both ovaries, and deposits of cancer are present in the abdomen that are large enough for the surgeon to see but not exceeding 2 cm in diameter. The cancer has not spread to the lymph nodes.
  • Stage IIIC The tumor is in one or both ovaries, and one or both of the following is present: (1) the cancer has spread to lymph nodes; and/or (2) the deposits of cancer exceed 2 cm in diameter and are found in the abdomen.
  • Stage IV This is the most advanced stage of ovarian cancer. Growth of the cancer involves one or both ovaries and distant metastases [spread of the cancer to organs located outside of the peritoneal cavity) have occurred. Finding ovarian cancer cells in pleural fluid [from the cavity which surrounds the lungs) is also evidence of stage IV disease.
  • the ovarian cancer is: type I, for example IA, IB or IC; type II, for example IIA, IIB or IIC; type III, for example IIIA, IIIB or IIIC; or type IV.
  • Thyroid cancer refers to cancer of the thyroid originating from follicular or parafollicular thyroid cells and includes papillary thyroid cancer [75% to 85% of cases); follicular thyroid cancer [10% to 20% of cases); medullary thyroid cancer [5% to 8% of cases)- cancer of the parafollicular cells, often part of multiple endocrine neoplasia type 2; poorly differentiated thyroid cancer; anaplastic thyroid cancer [less than 5% of cases) is not responsive to treatment and can cause pressure symptoms, thyroid lymphoma, squamous cell thyroid carcinoma, sarcoma of thyroid.
  • the present disclosure extends to treatment of thyroid cancer.
  • Bladder cancer as employed herein refers to cancer of the bladder including transitional cell bladder cancer, carcinoma in situ, papillary cancer and rarer types of bladder cancer such as squamous cell cancer and adenocarcinoma.
  • the present disclosure extends to treatment of bladder cancer.
  • Esophageal cancer refers to cancer of the oesphagus including esophageal squamous-cell carcinomas, esophageal adenocarcinomas, and variants of squamous-cell carcinoma, and non-epithelial tumors, such as leiomyosarcoma, malignant melanoma, rhabdomyosarcoma, lymphoma, among others.
  • the present disclosure relates to treatment of the esphageal cancer.
  • Head and neck cancer as employed herein refers to cancer of the neck and/or head, including mouth cancer, nasopharyngeal cancer, oropharyngeal cancer, paranasal sinus cancer and salivary gland cancer.
  • the present disclosure includes treatment of head and neck cancer.
  • the treatment of the present disclosure is neo-adjuvant therapy, for example to shrink the tumour/carcinoma before surgery to remove the cancerous tissue or before chemotherapy to improve the chances of success of the latter or to reduce the severity of the treatment required.
  • the treatment of the present disclosure is adjuvant therapy, for example following surgery to remove the cancerous tissue.
  • the treatment of the present disclosure is adjuvant therapy, for example following chemotherapy
  • combination adjuvant therapy comprising a compound of formula (I) and chemotherapy or radiotherapy.
  • First line therapy as employed herein is the first therapy employed for the treatment of the cancer and in some instances the first line therapy may be neo-adjuvant therapy, in this context surgery will generally be considered a treatment.
  • Second line therapy as employed herein is treatment following first line therapy and may be adjuvant therapy.
  • second line therapy is simply therapy other than first line therapy and includes, third line therapy, fourth line therapy etc.
  • Monotherapy as employed herein is wherein the compound of formula (I) an enantiomer thereof and/or a pharmaceutically acceptable salt thereof, is the only active agent being administered to the patient for the treatment of cancer.
  • treatment according to the present disclosure with a combination therapy comprising Varlitinib, for example Varlitinib and an anti-cancer therapy, such as chemotherapy.
  • a combination therapy comprising Varlitinib, for example Varlitinib and an anti-cancer therapy, such as chemotherapy.
  • Combination therapy refers to wherein the compound of formula [I) an enantiomer thereof or a pharmaceutically acceptable salt thereof is employed for the treatment of the cancer in conjunction with one or more further anticancer treatments, for example where the treatment regimens for the two or more active anticancer agents overlap or where the two or more anticancer agents are administered concomitantly.
  • the combination therapy according to the present disclosure comprises a RON inhibitor, for example as disclosed WO2008/058229, incorporated herein by reference.
  • the combination therapy comprises a checkpoint inhibitor, such as a CTLA4 inhibitor, a PD-1 inhibitor or a PD-Ll inhibitor, in particular an antibody or binding fragment thereof.
  • a checkpoint inhibitor such as a CTLA4 inhibitor, a PD-1 inhibitor or a PD-Ll inhibitor, in particular an antibody or binding fragment thereof.
  • Examples of pharmaceutically acceptable salts include but are not limited to acid addition salts of strong mineral acids such as HC1 and HBr salts and addition salts of strong organic acids, such as a methansulfonic acid salt, tosylates, furoates and the like, including di, tri salts thereof, such as ditosylates.
  • Chemotherapeutic agent and chemotherapy or cytotoxic agent are employed interchangeably herein unless the context indicates otherwise.
  • Chemotherapy as employed herein is intended to refer to specific antineoplastic chemical agents or drugs that are "selectively" destructive to malignant cells and tissues, for example alkylating agents, antimetabolites including thymidylate synthase inhibitors, anthracyclines, anti- microtubule agents including plant alkaloids, taxanes, topoisomerase inhibitors, parp inhibitors and other antitumour agents. Selectively in this context is used loosely because of course many of these agents have serious side effects.
  • the preferred dose may be chosen by the practitioner, based on the nature of the cancer being treated.
  • alkylating agents which may be employed in the method of the present disclosure include a platinum alkylating agent, nitrogen mustards, nitrosoureas, tetrazines, aziridines, platins and derivatives, and non-classical alkylating agents (such as procarbazine, altretamine and dacarbazine).
  • platinum containing chemotherapeutic agents include cisplatin, carboplatin, oxaliplatin, satraplatin, picoplatin, nedaplatin, triplatin and lipoplatin (a liposomal version of cisplatin), in particular cisplatin, carboplatin and oxaliplatin.
  • Others include eptaplatin, lobaplatin, miriplatin and dicycloplatin.
  • the dose for cisplatin ranges from about 20 to about 270 mg/m 2 depending on the exact cancer. Often the dose is in the range about 70 to about 100mg/m 2 .
  • Nitrogen mustards include mechlorethamine, cyclophosphamide, melphalan, chlorambucil, ifosfamide and busulfan.
  • Nitrosoureas include V-Nitroso-N-methylurea (MNU), carmustine (BCNU), lomustine (CCNU] and semustine (MeCCNU), fotemustine and streptozotocin.
  • Tetrazines include dacarbazine, mitozolomide and temozolomide.
  • Aziridines include thiotepa, mytomycin and diaziquone (AZQ).
  • antimetabolites examples include anti-folates (for example methotrexate and pemetrexed), purine analogues (for example thiopurines, such as azathiopurine, mercaptopurine, thiopurine, fludarabine (including the phosphate form), pentostatin and cladribine), pyrimidine analogues (for example fluoropyrimidines, such as 5-fluorouracil (5-FU) and prodrugs thereof such as capecitabine [Xeloda®]), floxuridine, gemcitabine, cytarabine, decitabine, raltitrexed (tomudex) hydrochloride, cladribine and 6-azauracil.
  • anti-folates for example methotrexate and pemetrexed
  • purine analogues for example thiopurines, such as azathiopurine, mercaptopurine, thiopurine, fludarabine (including the
  • anthracyclines examples include daunorubicin (Daunomycin), daunorubicin (liposomal), doxorubicin (Adriamycin), doxorubicin [liposomal), epirubicin, idarubicin, valrubicin currently are used only to treat bladder cancer and mitoxantrone an anthracycline analog, in particular doxorubicin.
  • anti-microtubule agents examples include vinca alkaloids and taxanes.
  • Vinca alkaloids include completely natural chemicals for example vincristine and vinblastine and also semi-synthetic vinca alkaloids, for example vinorelbine, vindesine, and vinflunine
  • Taxanes include paclitaxel, docetaxel, abraxane, carbazitaxel and derivatives of thereof.
  • Derivatives of taxanes as employed herein includes reformulations of taxanes like taxol, for example in a micelluar formulations, derivatives also include chemical derivatives wherein synthetic chemistry is employed to modify a starting material which is a taxane.
  • Topoisomerase inhibitors which may be employed in a method of the present disclosure include type I topoisomerase inhibitors, type II topoisomerase inhibitors and type II topoisomerase poisons.
  • Type I inhibitors include topotecan, irinotecan, indotecan and indimitecan.
  • Type II inhibitors include genistein and ICR ng structure:
  • Type II poisons include amsacrine, etoposide, etoposide phosphate, teniposide and doxorubicin and fluoroquinolones.
  • the chemotherapeutic is a PARP inhibitor.
  • chemotherapeutic agents employed is, for example a platin and 5-FU or a prodrug thereof, for example cisplatin or oxaplatin and capecitabine or gemcitabine, such as FOLFOX.
  • the chemotherapy comprises a combination of chemotherapy agents, in particular cytotoxic chemotherapeutic agents.
  • the chemotherapy combination comprises a platin, such as cisplatin and fluorouracil or capecitabine.
  • the chemotherapy combination is capecitabine and oxaliplatin [XELOX].
  • the chemotherapy is a combination of folinic acid and 5-FU, optionally in combination with oxaliplatin (FOLFOX).
  • the chemotherapy is a combination of folinic acid, 5-FU and irinotecan (FOLFIRI), optionally in combination with oxaliplatin (FOLFIRINOX).
  • the regimen for example includes: irinotecan (180 mg/m 2 IV over 90 minutes) concurrently with folinic acid (400 mg/m 2 [or 2 x 250 mg/m 2 ] IV over 120 minutes); followed by fluorouracil (400-500 mg/m 2 IV bolus) then fluorouracil (2400-3000 mg/m 2 intravenous infusion over 46 hours). This cycle is typically repeated every two weeks.
  • the dosages shown above may vary from cycle to cycle.
  • the combination therapy employs a microtubule inhibitor, for example vincristine sulphate, epothilone A, N-[2-[(4-Hydroxyphenyl)amino]-3-pyridinyl]-4- methoxybenzenesulfonamide (ABT-751), a taxol derived chemotherapeutic agent, for example paclitaxel, abraxane, or docetaxel or a combination thereof.
  • the combination therapy employs an mTor inhibitor.
  • mTor inhibitors include: everolimus (RAD001), WYE-354, KU-0063794, papamycin (Sirolimus), Temsirolimus, Deforolimus(MK-8669), AZD8055 and BEZ235(NVP-BEZ235).
  • the combination therapy employs a MEK inhibitor.
  • MEK inhibitors include: AS703026, CI-1040 (PD184352), AZD6244 (Selumetinib), PD318088, PD0325901, AZD8330, PD98059, U0126-EtOH, BIX 02189 or BIX 02188.
  • the combination therapy employs an AKT inhibitor.
  • AKT inhibitors include: MK-2206 and AT7867.
  • the combination therapy employs an aurora kinase inhibitor.
  • aurora kinase inhibitors include: Aurora A Inhibitor I, VX-680, AZD1152-HQPA (Barasertib], SNS- 314 Mesylate, PHA-680632, ZM-447439, CCT129202 and Hesperadin.
  • the chemotherapy combination employs a p38 inhibitor, for example as disclosed in W02010/038086, such as N-[4-( ⁇ 4-[3-(3-teri-Butyl-l-p-tolyl-ltf-pyrazol-5- yl)ureido]naphthalen-l-yloxy ⁇ methyl)pyridin-2-yl]-2-methoxyacetamide.
  • a p38 inhibitor for example as disclosed in W02010/038086, such as N-[4-( ⁇ 4-[3-(3-teri-Butyl-l-p-tolyl-ltf-pyrazol-5- yl)ureido]naphthalen-l-yloxy ⁇ methyl)pyridin-2-yl]-2-methoxyacetamide.
  • the combination therapy employs a Bcl-2 inhibitor.
  • Bcl-2 inhibitors include: obatoclax mesylate, ABT-737, ABT-263(navitoclax) and TW-37.
  • the combination therapy comprises an antimetabolite such as capecitabine (xeloda), fludarabine phosphate, fludarabine (fludara), decitabine, raltitrexed (tomudex], gemcitabine hydrochloride and/or cladribine.
  • an antimetabolite such as capecitabine (xeloda), fludarabine phosphate, fludarabine (fludara), decitabine, raltitrexed (tomudex], gemcitabine hydrochloride and/or cladribine.
  • the combination therapy comprises ganciclovir, which may assist in controlling immune responses and/or tumour vasculation.
  • a monotherapy comprising a compound of formula (I), such as Varlitinib (as defined herein including doses described above) is employed, for example a maintenance therapy.
  • a monotherapy comprising a compound of formula (I), such as Varlitinib (as defined herein including doses described above) is employed, for example a maintenance therapy.
  • the combination therapy comprises a checkpoint inhibitor, for example selected from the group comprising:
  • AZD-7762 a CHK1 and CHK2 inhibitor (with competitive CHK1 inhibitor with a Ki of 0.49 nM an IC50 of about 5 nM in a cell-free assay, with in a cell free assay.
  • the molecule also VEGFR2, less potent activity against CAM, Yes, Fyn, Lyn, Aurora-A, FGFR3, Flt3, Fms (CSF-1R], Ret and Hck an Lck) Yes. It shows approximately 100 fold selectivity for CHK1 over CHK2)
  • V155422, V155991 and V158411 are CHKl and CHK2 inhibitors
  • LY2603618 is a selective potent CHKl inhibitor MK -8776 is a selective CHKl inhibitor
  • CHIR124 is CHKl inhibitor (is a novel and potent The molecule shows 500-fold selectivity inhibitor with IC50 of 0.3 nM in a cell-free assay. against CHK2) It shows 2,000-fold selectivity against CHK2 and
  • CP-466722 is a potent ATM inhibitor, which does not inhibit ATR.
  • the molecule and PIKK are a potent ATM inhibitor, which does not inhibit ATR.
  • KU55933 is a potent and selective ATM inhibitor (vis-a-vie DNA-PK, Pi3K/Pi4K, ATR and mTOR) with an IC50 of 13 nM and KI of 2.2 mM
  • KU60019 is an ATM inhibitor
  • N U6027 is a potent ATR inhibitor
  • VE-821 is a potent and selective ATP competitive inhibitor of ATR with a Ki of 13 nM and an IC50 of 26 nM
  • CCT241533 is a CHK2 inhibitor with about 63 fold selectivity over CHK1
  • BMN-673 (Talazoparib) a PARP-1 and
  • PAR-2 inhibitor (which is currently in Phase III trials for ovarian, breast and other solid cancers)
  • BGP-15 a PARP inhibitor (which has been E7016 (previously known as GPI-21016) a shown to protect against ischemia-reperfusion PARP inhibitor (undergoing Phase I trials in injury) combination with temozolomide for
  • INO-1001 (3-aminobenzamide) a potent inhibitor of PARP [with IC50 of ⁇ 50 nM in CHO cells and a mediator of oxidant-induced myocyte dysfunction during
  • checkpoint inhibitors Drugs Fut 2003, 28(9): 881 Cell cycle inhibitors for the treatment of cancer Kong, N., Fotouhi, N., Wovkulich, P.M., Roberts, J; Novel pyrrole derivatives as selective CHK1 inhibitors: design, regioselective synthesis and molecular modelling Med. Chem. Commun., 2015,6, 852-859; PLoS One 2010; 5(8): el2214. Binding of protein kinase inhibitors to synapsin I inferred from pair-wise binding site similarity measurements.
  • the checkpoint inhibitor is a PARP inhibitor.
  • the checkpoint inhibitor is an antibody or binding fragment specific to a checkpoint protein, in particular one disclosed herein.
  • the checkpoint kinase inhibitor is independently selected from: 3- [(Aminocarbonyl)amino]-5-(3-fluorophenyl)-N-(3S)-3-piperidinyl-2-thiophenecarboxamide hydrochloride; (3R,4S)-4-[[2-(5-Fluoro-2-hydroxyphenyl)-6,7-dimethoxy-4-quinazolinyl]amino]- a,a-dimethyl-3-pyrrolidinemethanol dihydrochloride; 4,4'-diacetyldiphenylurea bis(guanylhydrazone) ditosylate; 9-Hydroxy-4-phenyl-pyrrolo[3,4-c]carbazole-l,3(2H,6H)-dione; (R)- -Amino-N-[5,6-dihydro-2-(l-methyl-lH-pyrazol-4-yl)-6-o
  • check point inhibitor is prexasertib.
  • Checkpoint kinase inhibitor as employed herein refers to an inhibitor that reduces or eliminates the biological activity of a cell regulatory checkpoint kinase 1 and/or 2.
  • Cells that suffer DNA damage activate the checkpoint kinases CHKl and CHK2, which signal to initiate the DNA repair processes, limit cell-cycle progression and prevent cell replication, until the damaged DNA is repaired.
  • Embodiments are described herein as comprising certain features/elements. The disclosure also extends to separate embodiments consisting or consisting essentially of said features/elements.
  • FIG. 1 ErbB family expression and its phosphorylation in representative HCC PDXs and the anti-tumour effects of Varlitinib on 3 PDXs with activated ErbB2/3.
  • a and B Western blot shows the expression of EGFR, ErbB2, ErbB3, and its phosphorylation in representative HCC PDXs.
  • HCC29-0909A PDX model Mice bearing HCC29-0909A PDX were treated with vehicle control, 25, 50, or lOOmg/kg BID. Tumours were collected at day 2 (A) and day 14 (B) post-Varlitinib treatment. Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis. Blots were incubated with indicated antibodies. Representative blots were shown.
  • FIG. 3 Effects of Varlitinib on tumour cell proliferation, tumour cell death, and vessel normalisation in HCC29-0909A PDX model.
  • Mice bearing HCC29-0909A PDX were treated with vehicle control and lOOmg/kg BID.
  • Tumours collected at day 14 were processed for paraffin [for cleaved PARP and p-histone 3 SerlO staining] or Tissue-Tek embedding (for CD31 staining).
  • A Representative images of tumour sections from vehicle-treated and Varlitinib-treated mice stained for p-Histone H3 (SerlO), cleaved PARP, and CD31.
  • HCC PDX models The area-proportional Venn diagram analysis showing the common activated genes (left) and the common repressed genes (right in lOOmg/kg BID and 50mg/kg BID Varlitinib-treated HCC29-0909A model for 14 days.
  • B The heatmap showing the common Varlitinib-dysregulated 2331 genes in HCC29-0909A model.
  • C The area-proportional Venn diagram analysis showing the common activated genes (left) and the common repressed genes (right) in lOOmg/kg BID and 50mg/kg BID Varlitinib-treated HCCOl-0708 model.
  • FIG. 5 Dose-dependent ⁇ -catenin pathway inhibition and membrane translocation of ⁇ -catenin by Varlitinib in the tested HCC29-0909A and HCCOl-0708 PDX models.
  • Treatment was started when the tumours reached the size of approximately 120-150 mm 3 .
  • Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis for the ⁇ -catenin, its related signalling molecules, and downstream targets of ⁇ -catenin.
  • C Representative images of tumour sections from vehicle-treated (top) and Varlitinib-treated (bottom) HCC29- 0909A, HCC07-0409, and HCCOl-0708 PDXs stained for ⁇ -catenin.
  • FIG. 7 ErbB family expression in additional 28 hepatocellular carcinoma (HCC) patient-derived xenografts (PDXs) and the anti-tumour effects of Varlitinib on the high ErbB2 /3-expressing PDXs.
  • HCC hepatocellular carcinoma
  • PDXs patient-derived xenografts
  • a and B Western blot shows the expression of EGFR, EbB2, ErbB3, and its phosphorylation in 28 HCC PDXs.
  • HCCOl-0708 PDX model Mice bearing HCCOl-0708 PDX were treated with vehicle control, 25, 50, or lOOmg/kg BID. Tumours were collected at day 2 (A) and day 11 (B) post-Varlitinib treatment. Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis. Blots were incubated with indicated antibodies. Representative blots were shown.
  • HCC29-0909A PDX Dose-dependent inhibition of tumour cell proliferation, induction of apoptosis and formation of capillary-like blood vessels by Varlitinib in HCC29-0909A PDX model.
  • Mice bearing HCC29-0909A PDX were treated with vehicle control, 25, 50, or lOOmg/kg BID.
  • Tumours collected at day 14 were processed for paraffin [for cleaved PARP and p-histone 3 SerlO staining] or Tissue-Tek embedding (for CD31 staining).
  • Tumour regression by Varlitinib treatment in three other PDX models HCC01- 0708, HCC07-0409, and HCC21-0208.
  • Mice bearing indicated PDX were treated with vehicle control or lOOmg/kg BID.
  • Tumours collected at indicated day were processed for paraffin [for cleaved PARP and p-histone 3 SerlO staining] or Tissue- Tek embedding (for CD31 staining).
  • PLC/PRF/5, Hep3B, HepG2/C3A, and Huh7 are ERBB3H ⁇ gh cell lines and SNU182, SNU423, SNU449, and SNU475 are ERBB3Low cell lines.
  • Different doses of Varlitinib in normal culture medium with 1% DMSO were added into the 6-well plates seeded with different cell lines. The cell lines were cultured for 2-4 weeks, followed by fixation and Giemsa staining.
  • C Inhibition of ErbB receptor family and their downstream signalling molecules by Varlitinib.
  • the £7?Bi?J?High PLC/PRF/5 cell line was treated with 3 ⁇ of Varlitinib for indicated time. Blots were incubated with indicated antibodies. a-Tubulin was used as equal loading marker. See also Figure 25.
  • Varlitinib responder and non-responder PDXs Identification of Varlitinib responder and non-responder PDXs and the importance of ErbB2/3 dependence in HCC.
  • A In vivo effect of Varlitinib in treated PDXs.
  • FIG. 23 Transcriptomic analysis identifies dose-dependant responses and the potential inhibition mechanisms in responder PDXs.
  • A Three-way area proportional Venn diagram analysis of differential expressed gene (DEG) analysis from RNA-Seq analysed Varlitinib-treated PDXs.
  • B The normalised read counts were analysed based on gene sets of glycolysis and gluconeogenesis (left) and hypoxia (right). Data from control (indicated in blue box), 50mg/kg (indicated in red box), and lOOmg/kg BID (indicated in green box) ofVarlitinib-treated HCC29-0909A are shown.
  • PDXs Dose-dependent ⁇ -catenin pathway inhibition by Varlitinib. Western blot analysis of the protein lysates collected on day 14 post-treatment of Varlitinib in HCC29-0909A (A) and on day 11 post-treatment of Varlitinib in HCCOl-0708 (B). Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis. Blots were incubated with indicated antibodies. a-Tubulin was used as equal loading marker. Representative blots were shown. See also Figure 30. Analysis of ERBB3 m and ERBB3 Low liver cancer cell lines.
  • Varlitinib treatment in HCC PDXs (A) ErbB family expression in 56 HCC (PDXs. Western blot shows the expression of EGFR, ErbB2, ErbB3, and its phosphorylation in 56 HCC PDXs. Blots were incubated with indicated antibodies. ⁇ -Tubulin was used as equal loading marker. (B) In vivo effect of Varlitinib in treated non-responder PDXs.
  • C Principal component analysis ofquantile normalised gene expression dataset Smaller circle indicates Varlitinib responders, larger circle circle indicates non-responders. Triplicate samples of 3 responder PDXs and 9 non- responder PDXs were analysed by Affymetrix Human Genome U133 Plus 2.0 microarray.
  • D Heatmap of the DEG.
  • K Heatmap of the inhibition of embryonic stem cell core gene set in Varlitinib- treated HCCOl-0708. Control of HCC29-0909A, Control of HCC16-1014, and Varlitinib-treated HCC16-1014 are included for the comparison.
  • PDX models Western blot analysis of the protein lysates collected on day 14 post- treatment of lOOmg/kg QD Varlitinib in HCC07-0409 PDX model. Two tumours from each condition were lysed and equal amount of the protein lysates were used. Blots were incubated with indicated antibodies. a-Tubulin was used as equal loading marker. Representative blots were shown.
  • PDXs. (A-B) ⁇ -catenin pathway inhibition by Varlitinib.
  • Figure 31 shows hypothetical model of the Varlitinib-mediated tumour growth inhibition and vessel normalisation in HCC.
  • the antibodies against ERK1/2 and a-tubulin were from Santa Cruz Biotechnology Inc, Santa Cruz, CA, USA.
  • Anti-mouse CD31 antibody was from BioLegend, San Diego, CA, USA.
  • Varlitinib was obtained from ASLAN Pharmaceuticals Ltd, Singapore.
  • HCC tumours have previously been used to create patient-derived xenograft models(Huynh et al., 2006) of which the 56 models were used to screen for the expression of EGFR, ErbB2, ErbB3, p-EGFR, p-ErbB2 and p-ErbB3 by Western blot analysis.
  • mice bearing the HCC29-0909A, HCC07-0409, and HCC01- 0708 xenografts (8-10 mice per group) were orally given vehicle [3 parts Polyethylene glycol) average Mn 300 (PEG, Aldrich Cat#202371) and 7 parts 30% w/v Research Grade Captisol (Ligand Pharmaceuticals, San Diego, CA) or 3 doses of Varlitinib (25 or 50, 75 and 100 mg/kg BID) for indicated days. Treatment started when the tumours reached the size of approximately 120-150 mm 3 .
  • tumour xenografts received intravenously with 100 mg of Biotinylated Lycopersicon Esculentum (Tomato) Lectin (VectorLabs #B-1175) prepared in 100 ⁇ of 0.9% NaCl.
  • Tomato Biotinylated Lycopersicon Esculentum
  • the tumours were harvested 10 minutes after lectin perfusion, fixed in 10 % formalin buffer solution, embedded in paraffin.
  • Five ⁇ sections were prepared. After blocking endogenous peroxidase activity and nonspecific staining, the sections were incubated 1 hour at room temperature with Streptavidin Peroxidase (Lab Vision Corporation, Fremont, CA).
  • mice bearing indicated tumours were treated with vehicle, or 100 mg/kg Varlitinib BID for indicated days.
  • Mice were i.p. injected with pimonidazole hydrochloride (60 mg/kg, 2.5 ⁇ /g of mouse body weight) 1 hour before tumours harvested. Hypoxic regions of tumour were identified by staining the sections with Hypoxyprobe plus Kit HP2 (Chemicon) as described by the manufacturer.
  • RNA samples would be contaminated with mouse cells due to the growth of PDXs in immunocompromised SCID mouse, an extra filtering step to remove mouse component was introduced.
  • the raw sequencing reads were aligned to hgl9_mml0 mixed reference by Burrows-Wheeler Aligner (BWA) and the read pairs were removed as long as any of the paired reads mapped to mmlO chromosome and/or rRNA sequence to filter mouse contamination and rRNA reads, respectively.
  • BWA Burrows-Wheeler Aligner
  • the detected mouse read rates at genome and gene levels ranged from 0.68% to 11.2% and 0.21% to 5.12%, respectively. In average, 74 million to 90 million clean reads per sample were obtained.
  • the BAM files were then uploaded to Partek Flow for further analysis. Aligned reads were quantify to Partek E/M annotation model in Partek Flow (Partek Inc. St. Louis, MO, USA), followed by total count and add 0.0001 normalisation of gene counts and GSA differential expression detection from the comparisons of 50mg/kg varlitinib treatment vs control and lOOmg/kg varlitinib treatment vs control in both models using the cut-off threshold of >2 and -2 fold-change and FDR adjusted p-value ⁇ 0.05, followed by heatmap generation.
  • the global gene expression analysis was done by quantile normalized data from Affymetrix GeneChip Human Genome U133 Plus 2.0 Array with high stringent cut-off threshold of >2 and -2 fold-change and FDR adjusted p-value ⁇ 0.0001.
  • the identified gene lists were then compared using BioVenn online tool for the area-proportional Venn diagram analysis (Hulsen et al., 2008).
  • the KEGG Pathway Enrichment analysis was carried out in Partek Genomics Suite (Partek Inc. St. Louis, MO, USA) with the cut-off threshold as FDR adjusted p-value ⁇ 0.05.
  • Varlitinib suppresses tumour growth through dose-dependent inhibition of ErbB family pathways in HCC PDXs
  • Varlitinib is a best-in-class reversible pan-HER inhibitor with a good clinical safety profile.
  • HCC21-0208 was selected to serve as negative control because it has the very low level of ErbB2 and undetectable level of ErbB3 (Figure IB).
  • Mice bearing HCCOl-0708, HCC07-0409, and HCC29- 0909A tumours were treated with three different doses of Varlitinib (25, 50, and lOOmg/kg BID).
  • Figure 1C-1E showed that Varlitinib inhibited tumour growth in a dose-dependent manner. This correlated well with tumour weight at harvest.
  • HCCOl-0708 is a fast-growing PDX that tolerated Varlitinib at low dosing, but the tumour growth and weight of that were significantly suppressed at the highest tested dosing, suggesting that higher dosing of Varlitinib is able to overcome the drug tolerance (Figure 1C).
  • HCC07-0409 and HCC29-0909A were more sensitive to Varlitinib and showed better dose-dependent inhibition effect ( Figure ID and IE).
  • Statistically significant growth inhibition is also observed among all different dosing in these two models, meaning that these models are sensitive and respond to Varlitinib very well.
  • Varlitinib promotes vessel normalisation and tumour perfusion in HCC PDX
  • RNA-Seq total mRNA sequencing
  • KEGG pathways are ribosome pathway, RNA transport, steroid biosynthesis, central carbon metabolism in cancer, and HIFl signalling pathway (Table 1C).
  • the Ingenuity Pathway Analysis (IPA) demonstrated the suppression of EIF2, eIF4, p70S6K, and mTOR signalling with predicted HIF1A as the inhibited upstream regulator as well as activation of ILl-mediated inhibition of RXR function, FXR/RXR and LXR/RXR pathway with predicted HNF1A and PPARA activated upstream regulators (Table ID), showing that the repressed ErbB downstream pathways correlates with hepatic lipid differentiation.
  • IPA Ingenuity Pathway Analysis
  • the numbers of differentially expressed genes identified from different dosing in both models indicate that higher dosing of Varlitinib used, more genes will be affected.
  • Varlitinib could inhibit similar EPO-related pathway and distinct VEGF-dependent [PDGFA, VEGFB, PGF, and NRP2) and VEGF- independent [PDGFC and BMP2/FST) anti-angiogenic pathways in two different PDX models to facilitate the vessel normalisation (Andrae et al., 2008; Kertesz et al., 2004; Krneta et al., 2006; Li et al., 2010; Lin et al., 2014; Zuo et al, 2016).
  • Varlitinib enhances immune infiltration in the treated tumours
  • RNA-Seq reads that align to mouse genome reference, mmlO were used to analyse the stromal components in the PDX models. Myeloid cell-related markers were specifically analysed.
  • Varlitinib effectively inhibits mutated ⁇ -catenin pathways and mediates membrane translocation of mutated ⁇ -catenin
  • Varlitinib-sensitive PDXs Figure 13A and Table 13A.
  • the Varlitinib-resistant PDX, HCC21-0208 contains wild-type CTNNB1 but with much higher TGFBR1 /2, IGF2/IGFR/IRS1, NOTCH1 /JAG1, TEAD1 /2, and CYR61 expression and lower MST1 expression comparatively from the gene expression microarray data ( Figure 13B and Table 13B].
  • Varlitinib-resistant PDX exhibited highly activated TGF- ⁇ , NOTCH, and Hippo pathways.
  • HCC proposed molecular classes of HCC
  • the Varlitinib-resistant PDX, HCC21-0208 is related to Wnt/TGF- ⁇ class
  • the Varlitinib-sensitive PDXs, HCCOl-0708, HCC07-0409, and HCC29- 0909 are related to CTNNB1 class or more specifically CTNNB1 mutation class.
  • the gene signature of Varlitinib-potency of the two RNA-sequenced PDXs was determined and shown in Figure 14.
  • HCCOl-0708, HCC07-0409 and HCC29-0909A harbour T41A ⁇ -catenin mutation and the Wnt/ ⁇ -catenin-related genes/pathways are inhibited according to WES and RNA-Seq analysis separately we sought to further investigate if the mutated ⁇ -catenin and its related pathway members are suppressed at protein level and whether the localisation of mutated ⁇ -catenin is affected by the Varlitinib treatment.
  • Figure 5 and Figure 30B displayed marked inhibition of ⁇ - catenin and its related pathways in the lOOmg/kg BID Varlitinib-sensitive PDXs, HCC29-0909A, HCCOl-0708, and HCC07-0409 models, ⁇ -catenin upstream regulators, p-LRP6 and DVL3, as well as downstream targets, Axin2, survivin, c-Jun, c-Met, and N-cadherin were suppressed and E-cadherin expression was elevated by Varlitinib in the treated HCC29-0909A ( Figure 5A).
  • c-Jun was reported to physically interact with ⁇ -catenin and TCF4 and to stabilise interaction for the transcription of ⁇ -catenin-target genes and cancer development (Gan et al., 2008; Nateri et al., 2005].
  • the inhibition of p-c-Jun by Varlitinib suggested the transcription repression of ⁇ -catenin.
  • Varlitinib targets mutated ⁇ -catenin and its related pathways in ErbB-dependent tumours by inhibition of p-LRP6, p- ⁇ -catenin at Tyrl42, and p-RanBP3 Ser58, resulting in ⁇ -catenin membrane translocation and inhibiting downstream targets of ⁇ -catenin.
  • the varlitinib-resistant PDX has statistically significantly higher pathway activities of angiogenesis, epithelial adherent junction signalling, epithelial-mesenchymal transition signalling, embryonic stem cell pluripotency, RhoA signalling, and Hippo signalling with higher CD24 signalling predicted.
  • HNF4A is a key liver differentiation driver and CD24 is a tumour-initiating cell marker in HCC (Enane et al., 2017; Lee et al., 2011).
  • varlitinib sensitivity correlates with ErbB3 expression and differentiation status in in HCC PDXs.
  • HEY1 and SOX9 were showed significantly higher expression in HCCOl-0708 (48.88 and 28.90-fold, respectively) than those in HCC29-0909A (Table 5A). Furthermore, the pathway enrichment analysis also demonstrated (Table 5C) that HCCOl-0708 has much higher Hippo, Wnt, IGFR, and FGFR pathway activities, whereas HCC29-0909A retains stronger metabolic activities, revealing the impaired differentiation in the former model (Table 5D).
  • RNA-Seq comparison further showed that ⁇ -catenin, Notch, and Hippo targets, MYC, CTGF, CYR61, and YAPl were highly expressed in HCCOl-0708, but were significantly inhibited by high dosing of Varlitinib (Table 5E and Table 5F).
  • Activated Wnt, Hippo, Notch, IGF, and FGFR are related the tumour-initiating cells and chemoresistance in HCC (Lau et al., 2016; Liu et al., 2016; Martinez- Quetglas et al., 2016; Villanueva et al., 2012). Accordingly, our study demonstrated that Varlitinib potency correlates with a spectrum of sternness/differentiation level in ErbB3-expressing and ⁇ -catenin mutated HCC PDXs.
  • HCC29-0909A HCC07-0409
  • HCCCOl-0708 HCC PDX models which display high levels of p- ErbB2 and p-ErbB3.
  • HCC29-0909A possesses the highest p-ErbB3.
  • HCC21-0208 model has undetectable p-ErbB2 and p-ErbB3 demonstrated poor efficacy in the Varlitinib treatment.
  • Varlitinib sensitivity is determined by the levels of p- ErbB3 and/or pErbB2 and raises the possibility that p-ErbB3 and p-ErbB2 are more reliable biomarkers for patient selection than total ErbB2 or ErbB3.
  • p-ErbB3 and p-ErbB2 are more reliable biomarkers for patient selection than total ErbB2 or ErbB3.
  • p-ErbB3 and p-ErbB2 are more reliable biomarkers for patient selection than total ErbB2 or ErbB3.
  • higher Wnt, Hippo, Notch, IGF, and FGFR pathway signalling were observed in the HCCOl- 0708 PDX, which required high dosing of Varlitinib to achieve better tumour growth inhibition.
  • HIF-1 activity stimulates neovascularization by enabling tumour and host cells to produce a variety of proangiogenic factors like VEGF-A, PDGF-B, FGF-2, and angiopoietins that stimulate new blood vessel formation within hypoxic areas (Calvani et al., 2006; Okuyama et al., 2006).
  • VEGF-A vascular endothelial growth factor
  • PDGF-B vascular endothelial growth factor
  • FGF-2 fibroblast growth factor-2
  • angiopoietins that stimulate new blood vessel formation within hypoxic areas
  • Varlitinib induces blood vessel normalization by previous study showing that anti-HER2 therapies (trastuzumab) promote normalisation in HER2-positive breast cancer (Goel et al., 2011).
  • T41A mutations have been identified from 2% to 18.8% in analysed HCC and the ⁇ -catenin mutations are found to be associated with low-stage of HCC (Austinat et al., 2008; Boyault et al., 2007; Cleary et al., 2013; Enane et al., 2017; Guichard et al., 2012; Hoshida et al., 2009; Kan et al., 2013; Legoix et al., 1999; Nault et al., 2013; Nhieu et al., 1999; Waisberg and Saba, 2015; Wong et al., 2001).
  • ⁇ -catenin is ubiquitinated by ⁇ -TrCP ubiquitin ligases, followed by proteasomal degradation.
  • mutations at T41 is found to prevent GSK3-mediated p-phosphorylation at serines 37 and 33, further avoiding ⁇ -TrCP recognition (Aberle et al., 1997; Liu et al., 2002; MacDonald et al., 2009; Orford et al., 1997).
  • T41A mutation is further shown to be constitutively active mutant, which enhances nuclear localisation of TCF4, resulting in elevated expression of ⁇ -catenin-targets (Hsu et al., 2006). Since phosphorylation of ⁇ -catenin at tyrosine 654 and at tyrosine 142 is essential for binding to E-cadherin (Roura et al., 1999) and a-catenin (Aberle et al., 1996; Piedra et al., 2003; Pokutta and Weis, 2000) respectively, inhibition of tyrosine phosphorylation at these sites by Varlitinib would facilitate the interaction of E-cadherin and a-catenin with ⁇ -catenin. This could lead to the assembly of the E-cadherin- a-catenin ⁇ -catenin complex at the plasma membrane and to decreased ⁇ -catenin-dependent transcription.
  • Wnt/ ⁇ -catenin is one of the main classes.
  • the first FDA-approved targeted therapy in HCC, Sorafenib is found to modulate Wnt ⁇ -catenin signalling in in vitro and in vivo CTNNBl-class liver cancer models, demonstrated by reduced TCF/LEF luciferase-reporter activity and ⁇ -catenin expression (Boyault et al., 2007; Hoshida et al., 2009; Lachenmayer et al., 2012; Sia et al., 2017).
  • Wnt ⁇ -catenin has been comprehensively discussed (Pez et al, 2013; Vilchez et al., 2016).
  • Wnt/ ⁇ -catenin pathway inhibitor ICGOOl, FH535, and other small inhibitors have been tested in HCC in different in vitro and in vivo models (Delgado et al., 2014; Gedaly et al., 2014; Handeli and Simon, 2008). Cheng's group and Gedaly's group further proved that the combination of sorafenib and ICGOOl as well as the combination of sorafenib and FH535 had better treatment outcome in their experimental models (Galuppo et al, 2014; Lin et al, 2016).
  • Varlitinib would show high efficacy and potency in the specific subset of HCC which depends on both ErbB pathway and mutated ⁇ -catenin.
  • Varlitinib effectively targeted ⁇ -catenin with T41A mutation in ErbB-dependent HCC by inhibiting ⁇ - ⁇ -catenin Tyrl42 and p-RanBP3 Ser58 as well as downregulated p-LRP6 and DVL3.
  • the mutated ⁇ -catenin is transported from nucleus and cytoplasm to membrane.
  • Varlitinib is a clinically available pan-HER and mutated ⁇ - catenin inhibitor, that can be used to target p-ErbB2 and p-ErbB3-highly expressing and CTNNB1 subset of HCC.
  • the robust and tolerate anti-tumour activity of Varlitinib in HCC PDX models and its safety profile warrants the development of clinical trial with trial enrichment for Varlitinib in HCC.
  • Example 2 - Varlitinib induces apoptosis in HCC cell lines
  • HCC cells were grown in cell culture medium with 10% FBS and with varying concentrations of varlitinib. Apoptosis profiles were analysed at 24 hour and 48 hour timepoints, using Muse Annexin V & Dead Cell Assay kit and Muse Cell Analyser. Early apoptosis cells (identified as Annexin V-PE positive and Dead Cell Marker negative ⁇ were measured and plotted.
  • Figure 18 shows the results of the experiment.
  • varlitinib was able to induce early apoptosis in all of the HCC cell lines tested after 48 hours of incubation.
  • Varlitinib was particularly effective in sorafenib resistant cells (Huh7-SorR) where almost 70% early apoptosis was observed when high dose of varlitinib was used, suggesting that varlitinib could be effective in patients who progress on sorafenib.
  • Figure 19 shows the apoptosis profile for the PLC/PRF/5 (PLC) cells after 48 hour culture in the presence of varlitinib. Note the increase in percentage of apoptotic cells correlates with increasing varlitinib concentration.
  • Example 3 Varlitinib-mediated tumour growth inhibition and vascular normalisation in activated ErB2/3-dependent and mutated ⁇ -catenin hepatocellular carcinoma
  • Hep3B Sorafenib-resistant cell line (Hep3B- SorR), which we established by step-wise exposure to increasing amounts of Sorafenib until the maximal dose that the cell lines could tolerate, showed higher phosphorylation status of EGFR, ErbB2, and ErbB3. Upregulated ErbB pathway activity could therefore be an acquired Sorafenib resistance mechanism.
  • Varlitinib effectively suppresses in vivo tumour growth in high p-ErbB2/3 and NRG1 expressing HCC patient-derived xenografts (PDXs)
  • mice bearing PDX tumours were treated with Varlitinib (lOOmg/kg BID). Additional dosing of 25 and 50mg/kg BID were adopted to treat three PDXs, HCCOl-0708, HCC07-0409, and HCC29-0909A ( Figures 21A and 26B).
  • Varlitinib showed effective tumour growth inhibition in only three PDXs (HCCOl-0708, HCC07-0409, and HCC29-0909A), at the dose of lOOmg/kg BID.
  • HCC07- 0409 and HCC29-0909A were more sensitive to Varlitinib and showed superior dose-dependent inhibition (Figure 21A).
  • Statistically significant growth inhibition was also observed among all three doses in the two models (p-122 value ⁇ 0.0001).
  • the latter model exhibited dramatic tumour suppression even when treated with Varlitinib at 50mg/kg BID, signifying that Varlitinib is highly potent in the HCC29-0909A model.
  • HCCOl-0708 was a fast-growing PDX mouse model that tolerated Varlitinib at low dosing. Tumour growth was significantly suppressed at the highest dose, suggesting that a higher dosing of Varlitinib may overcome intrinsic drug resistance. Conversely Varlitinib at the dose of 100 mg/kg BID had no significant anti-tumour activity in the rest of the tested PDX models ( Figures 21A and 26B).
  • the PDXs were categorised into responders and non-responders according to their treatment responses. Three responders and two non-responders were reassessed by Western blot (Figure 21B). As shown in Figure 21B, the three responders showed much higher p-ErbB2/3 and ErbB2/3 compared to the two non-responders. The expression of p-EGFR, total EGFR, p-Erkl/2, and p-Akt were similar in all tested PDXs, suggesting that responders are highly dependent on activated ErbB2/3 signalling.
  • Varlitinib responders We further sought to understand the molecular differences between Varlitinib responders and non-responders.
  • Three responders HCCOl-0708, HCC07-0409, and HCC29-0909A
  • 9 non- responders HCCOl-0909, HCC06-0606, HCC13-0109, HCC21-0208, HCC25-0705A, HCC26-0808A, HCC26-0808B, HCC29-1104, and HCC30-0805B
  • all 3 responders grouped together based on the principal component analysis (Figure 26C).
  • GSEA Gene set enrichment analysis
  • Transcriptome analysis reveals the molecular basis of Varlitinib-mediated tumour suppression Although HCCOl-0708 and HCC29-0909A PDX models highly expressed activated ErbB3, they had distinct, differentiated responses to Variitinib. Maximal dosing (lOOmg/kg BID) of Variitinib was needed to suppress HCCOl-0708 tumour growth, whereas a lower dose [50mg/kg BID) was sufficient to suppress HCC29-0909A tumour growth ( Figure 21A).
  • RNA-Seq total mRNA sequencing
  • the top 20 commonly activated and repressed genes in HCC29-0909A are shown in Tables 1A and IB, respectively.
  • the top 6 significantly suppressed KEGG pathways were the ribosome pathway, RNA transport, steroid biosynthesis, central carbon metabolism in cancer, and HIF1 signalling pathway (Table 1C).
  • IPA Ingenuity Pathway Analysis
  • Varlitinib inhibits glycolysis, HIF1A pathway, and angiogenesis in the treated PDXs
  • FIGS 28C-E further revealed that Varlitinib could inhibit the EPO pathway, VEGF-dependent ⁇ PDGFA, VEGFB, PGF, and NRPZ) and VEGF-independent ⁇ ANG, PDGFC and BMP2) pro-angiogenic pathways to facilitate vessel normalisation (Krneta et al, Kertesz et al and Andrae et al).
  • LGR5 ⁇ -catenin upstream regulator
  • YAP1, IRX3, and MYC were downregulated in both treated HCC29-0909A and HCCOl-0708 models.
  • expression of CDH1 was enhanced ( Figures 23E and 28F).
  • LEF1 was reduced in the treated HCC29- 0909A model ( Figure 28G).
  • Figures 24A, 24B, 30A and 30B displayed marked inhibition of ⁇ -catenin and its related pathways in the Varlitinib responder PDXs, HCC29-0909A, HCCOl-0708, and HCC07-0409 models.
  • ⁇ -catenin upstream regulators, p-LRP6 and DVL3, as well as downstream targets, Axin2, survivin, c- Jun, c-Met, and N-cadherin, were suppressed.
  • E-cadherin expression was elevated by Varlitinib in the treated PDXs ( Figure 24A).
  • c-Jun was reported to physically interact with ⁇ -catenin and TCF4 to stabilise ⁇ - ⁇ 3 ⁇ / ⁇ 4 interaction for the transcription of ⁇ -catenin target genes and cancer development (Nateri et al).
  • the inhibition of p-c-Jun by Varlitinib suggested the transcriptional repression of ⁇ -catenin.
  • Varlitinib targets mutated ⁇ -catenin and its related pathways in ErbB-dependent tumours - by inhibiting ⁇ -catenin upstream regulator LGR5 and p-LRP6, ⁇ - ⁇ -catenin at Tyrl42 and Tyr654, nuclear export regulator p-RanBP3 at Ser58, as well as upregulating E-Cadherin expression. These, in turn, result in ⁇ -catenin membrane translocation, thereby inhibiting downstream targets of ⁇ - catenin. T41A ⁇ -catenin mutation is thought to be constituitively active. However, our study demonstrates that its nuclear localisation and ⁇ -catenin-driven genes could be inhibited by Varlitinib in high p-ErbB2/3 expressing HCC.
  • ERBB3 is upregulated in a subset of HCC.
  • Varlitinib a small molecule pan-ErbB inhibitor, effectively inhibited high £7?£?£fJ?-expressing HCC in vitro.
  • in vivo efficacy was demonstrated in the responder PDXs, HCC29-0909A, HCC07-0409, and HCCOl-0708 in immune-deficient NOD/SCID mice, which displayed high levels of p-ErbB2/3, total ErbB2/3, and NRG1 gene expression.
  • Varlitinib sensitivity is determined by the quantitative expression of p-ErbB3 and/or pErbB2 and raises the possibility that the presence of p-ErbB2/3 could potentially be specific for patient treatment stratification.
  • expression of ERBB2 and ERBB3 are known to have prognostic value, whereas that of NRG1 has the same trend in our Singapore HCC dataset ( Figure 21D).
  • ERBB2/3 expression belong to both the G6 subgroup and CTNNB1 subclass in the non-proliferation class of HCC, whereas the 14 tested non-responder PDXs are related to the HCC subgroups with TGF and IGF/IGF1R pathways and S1/S2 tumours in the proliferation class of HCC (Llovet et al 2017,Boyault et al 2007,Lachenmayer et al ). It indicates the strong correlation of the ErbB family and ⁇ -catenin pathways in the specific subclass of HCC PDXs, that could be effectively inhibited by Varlitinib.
  • Varlitinib was especially potent against HCC29-0909A. It may be due to the low activity of mutated ⁇ -catenin in the model, as indicated by the IPA pathway enrichment analysis ( Figure 281).
  • ⁇ -catenin-driven cancers require YAPl for tumour progression (Rosenbluh et al).
  • YAPl and its related sternness properties are also associated with drug resistance in multiple cancers (Zanconato et al). Lin and colleagues previously identified through genetic screen that YAPl is the key resistance driver of RAF- and MEK- targeted therapies (Lin L et al).
  • HIF-1 activity stimulates neovascularisation by enabling tumour and host cells to produce a variety of proangiogenic factors like VEGF-A, PDGF-B, FGF-2, and angiopoietins that stimulate new blood vessel formation within hypoxic areas (Okuyama et al, Calvani et al).
  • ⁇ -catenin mutations are found to be associated with the earlier stages of HCC and the T41A mutation has been identified in up to 18.8% of analysed HCC (Guichard et al).
  • wild-type ⁇ -catenin is ubiquitinated by ⁇ - TrCP ubiquitin ligases, destined for proteasomal degradation.
  • mutations at T41 prevent GSK3-mediated phosphorylation at serines 33/37, avoiding ⁇ -TrCP recognition (Liu et al).
  • T41A mutation is also found to be a constitutively activated mutant, which enhances nuclear localisation of TCF4, resulting in elevated expression of ⁇ -catenin-targets (Hsu et al H-T et al). Since phosphorylation of ⁇ -catenin at tyrosines 142 and 654 is essential for binding to E-cadherin (Roura et al) and oc-catenin (Pokutta et al), inhibition of tyrosine phosphorylation at these sites by Varlitinib would facilitate the interaction of E-cadherin and a-catenin with ⁇ -catenin.
  • Sorafenib - a BRAF, C-RAF and VEGF-R inhibitor - can modulate Wnt/ -catenin signalling in in vitro and in vivo CTNNBl-class liver cancer models. This was demonstrated by showing reduced TCF/LEF luciferase-reporter activity and ⁇ -catenin expression following Sorafenib treatment (Boyault et al, Lachenmayer et al). There has been considerable exploration of targeting Wnt/ -catenin against HCC. Wnt/ -catenin pathway inhibitors ICGOOl, FH535, and other small inhibitors have been tested in HCC in different in vitro and in vivo models.
  • Varlitinib shows good efficacy in the specifically defined subset of HCC that is dependent on both the ErbB pathway and ⁇ -catenin mutation.
  • Varlitinib promotes apoptosis and vessal normalisation with reduction of tumour progression and hypoxia in this subset of HCC.
  • Varlitinib effectively inhibits ErbB receptors and their downstream oncogenic signalling such as MEK/Erk and AKT/mTOR pathways.
  • ⁇ -catenin targets ⁇ -catenin with T41A mutation in ErbB-dependent HCC by inhibiting ⁇ - ⁇ -catenin Tyrl42 /654 and p-RanBP3 Ser58, as well as inducing LGR5, p-LRP6, and DVL3 downregulation.
  • mutated ⁇ -catenin is transported from the nucleus and cytoplasm to the cell membrane, reducing the expression of ⁇ -catenin-driven genes.
  • Varlitinib has a comparably favorable safety profile compared to irreversible pan-ErbB inhibitors such as Neratinib and Dacomitinib.
  • the recommended dose of Varlitinib for clinical trials is between 200mg BID to 500mg BID, which is equivalent to 41mg/kg BID to 102.5mg/kg BID in mouse (calculation is based on Nair and Jacob (Hulsen et al).
  • Our study reveals that 50mg/kg BID in mice ( ⁇ 250mg BID in human) showed efficient growth supression in high p-ErbB2/3 expressing PDXs.
  • Intrinsic resistance demonstrated by high ⁇ - catenin/YAPl-related gene signature in high p-ErbB2/3 expressing PDX could be overcome by increasing the dose to lOOmg/kg BID in mice ( ⁇ 490mg BID in human). In contrast, even with very high dose of Varlitinib given to the mice (lOOmg/kg BID), there is no anti-tumour benefit in the non- responders. Expression of p-ErbB2/3 and/or ErbB2/3 could be a predictive biomarker in the clinical treatment of HCC with Varlitinib.
  • Varlitinib is a pan-ErbB and, as we report here, an inhibitor of mutated ⁇ - catenin - that can selectively target the p-ErbB2/3 highly-expressing and CTNNB1 subset of HCC.
  • Our data suggests that the selection of HCC patients with high p-ErbB2/3 expression could be a useful target and predictive biomarker for Varlitinib clinical efficacy.
  • the role of tailoring systemic therapies, whether molecular or immune modulating, can potentially optimise treatment efficacy based on HCC subtypes.
  • Three tumours from each condition (control, 50mg/kg, and lOOmg/kg Varlitinib treatment) of 95 HCC29-0909A and HCCOl-0708 and two tumours from each condition (control and lOOmg/kg treatment) of HCC16-1014 were used for transcriptome analysis.
  • a total of 200 ng of total RNA was used for illumina TruSeq mRNA library prep, followed by 150bp paired-end sequencing on an illumina HiSeq4000 platform by BGI HK, Ltd.
  • the raw sequencing reads were aligned to mouse mmlO genome reference and human hg38 genome reference by Spliced Transcripts Alignment to a Reference (STAR) aligner version 2.5.3a, separately using Partek Flow (Partek Inc. St. Louis, MO). In average, 74 million to 90 million clean reads per sample were obtained. The total human alignments ranged from 70 million to 87 million per samples. Aligned reads were then quantify to hg38 - RefSeq Transcripts 85 -2018-05-02 annotation model using Partek E/M algorithm, followed by low expressed gene filtering and total count and add 0.0001 gene counts normalisation.
  • GSA differential expression detection model was used to compare 50mg/kg Varlitinib treatment vs control and lOOmg/kg Varlitinib treatment vs control in PDX models using the cut-off threshold of >2 and -2 fold-change and FDR adjusted p-value ⁇ 0.05, followed by heatmap generation.
  • the gene expression analysis was done by quantile normalised data from Affymetrix GeneChip Human Genome U133 Plus 2.0 Array with cut-off threshold of >2 and -2 fold-change and high stringent FDR adjusted p-value ⁇ 0.0001.
  • the identified gene lists were compared using BioVenn online tool for the area-proportional Venn diagram analysis (Hulsen et al).
  • the KEGG Pathway Enrichment analysis was carried out in Partek Genomics Suite (Partek Inc. St Louis, MO) with the cut-off threshold of FDR adjusted p-value ⁇ 0.05.
  • IPA Ingenuity Pathway Analysis
  • GSEA Gene Set Enrichment Analysis
  • MSigDB Molecular Signature Database
  • Table 1 Transcriptome analysis of Varlitinib treatment in HCC29-0909A.
  • A Top 20 common activated genes and
  • B top 20 common repressed genes in lOOmg/kg and 50mg/kg BID Varlitinib- treated HCC29-0909A. FDR adjusted p-values are shown.
  • C KEGG pathway analysis showing the enriched pathways generated from the gene lists of significant differentially expressed genes with the threshold of >2 and ⁇ -2 fold-change with FDR adjusted p-values ⁇ 0.05.
  • D Ingenuity Pathway Analysis showing the predicted repressed and activated pathways with enrichment FDR adjusted p- values ⁇ 0.05.
  • Table 2 Transcriptome analysis of Varlitinib treatment in HCCOl-0708.
  • A Top 20 common activated genes and
  • B top 20 common repressed genes in lOOmg/kg and 50mg/kg BID Varlitinib- treated HCCOl-0708.
  • FDR adjusted p-values are shown.
  • C Ingenuity Pathway Analysis showing the predicted repressed and activated pathways with enrichment FDR adjusted p-values based on the gene lists of significant differentially expressed genes with the threshold of >2 and ⁇ -2 fold- change with FDR adjusted p-values ⁇ 0.05.
  • Table 4 Identification of Varlitinib-resistant gene signature by global gene expression analysis.
  • A The top 20 highly expressed genes and
  • B the top 20 lowly expressed genes identified in Varlitinib-sensitive PDXs, HCC29-0909A, HCCOl-0708, and HCC07-0409 models.
  • C Ingenuity Pathway Analysis showing the predicted repressed and activated pathways with enrichment FDR adjusted p-values ⁇ 0.05. ** indicates p-value ⁇ 0.01, *** indicates p-value ⁇ 0.001, **** indicates p- value ⁇ 0.0001.
  • HCM Hypertrophic cardiomyopathy
  • B the top 20 lowly expressed genes identified in higher Varlitinib-potent PDX HCC29-0909A when compared to lower Varlitinib-potent HCCOl- 0708.
  • C Ingenuity Pathway Analysis showing the predicted activated pathways with enrichment FDR adjusted p-values ⁇ 0.05. Pathways activated in HCC29-0909A Innchment Score Enrichment p-value
  • Table 6 Gene enrichment analysis in Varlitinib responder and non-responder PDXs.
  • the enriched gene sets with q-value ⁇ 0.05 are shown.
  • ES refers to enrichment scores
  • NES refers to normalised enrichment scores.
  • ERBBf £f?8B3 L0W Table 7 - Analysis of ERBB3TM and
  • MCC29-0909A Responds r R5J 1 PliTOA, I6S5 ⁇ Nit NI! T41A Hepatitis B positive
  • Table 10 Transcriptome analysis of Varlitinib treatment in HCC29-0909A, HCCOl-0708, and HCC16-1014.
  • H2-AU HC class il 40.00 7,18 x 10 " '
  • H2-DMO MHC class 11 20 62 5.46 10
  • H2-Ea-ps MHC class il 22.40 2,28 x 10 ;
  • A Differentially expressed myeloid-related genes. Gene symbol Fold-change FDR adjusted rvalue
  • Table 13 Identification of ⁇ -catenin mutation and the constitutively active ⁇ -catenin pathways in the Varlitinib-sensitive PDXs.
  • A Expression of ⁇ -catenin and its downstream targets and
  • B expression of Wnt/TGF -related targets, in the four analysed PDXs indicated by normalised probe intensity.

Abstract

A type I tyrosine kinase inhibitor, which is Varlitinib, or a pharmaceutical formulation comprising the same as monotherapy or combination therapy, for use in a method of treating a particular cancer patient population wherein the patients have a mutation in the beta-catenin pathway.

Description

VARLITINIB FOR USE IN TREATING CANCER IN A PATIENT IDENTIFIED AS
HAVING A BETA-CATENIN PATHWAY MUTATION
The present disclosure relates to a therapy, for example a monotherapy or combination therapy comprising a type I tyrosine kinase inhibitor for the treatment of cancer, for example liver cancer, such as hepatocellular carcinoma, in patients who are identified as having a mutation in the beta- catenin pathway.
BACKGROUND
There are many cancers that are difficult to treat and although therapy is available, there appears to exist or to come into existence, a degree of resistance to the therapy. Primary resistance may occur in that cancer does not respond to treatment from the outset. Secondary or acquired resistance also occurs quite frequently, which means that a therapy to which the patient seems to respond, at a certain time, loses its efficacy.
There are numerous reasons for resistance, for example some cancers are discovered at a late stage and/or a simply not responsive to treatment.
Mechanisms by which cancers avoid the therapeutic effect of therapy include but are not limited to:
i) mutations which render the cancer less vulnerable to the treatment (eg mutation of the site of action of the therapy),
ii) active transportation of the drug out of the tumor, for example by p-glycosylation, iii) building up physical defences, for example stroma which inhibit certain immune responses, and
iv) certain cancers develop paths to repair damage caused by some anti-cancer therapies.
Tumor heterogeneity may also contribute to resistance, where small subpopulations of cells may acquire or stochastically already possess some of the features enabling them to emerge under selective drug pressure. This is a problem that many patients with cancer encounter, and it obviously limits the therapeutic alternatives that are effective and worsens the prognosis.
Cancer therapy guidelines describe the sequence of therapies, which are recommended and in which sequence, so that if a patient shows disease progression on the first therapy ("first line"), then a next therapy ("second line") is recommended, and so on. These therapy recommendations are based on available scientific data and experience, and illustrate that resistance to one therapy does not exclude that another therapy may be effective and prolong life or shrink a tumor. At late stages cancers do not respond, no more avenues of therapy exist, and are completely therapy refractory. Thus, unless new therapies can be found, which are effective, these cancers cannot be treated.
Varlitinib is in clinical trials for the treatment of cancer. However, the present inventors have data that suggest the molecule is particularly effective in the treatment of a subcategory of cancers, namely those where cancer cells have a mutation in the beta-catenin pathway. These cancers seem to be sensitive to treatment with Varlitinib.
These mutations are important subclass of cancers, for example in heptacellular carcinoma TCGA data (Ally et al., 2017), 46% of HCC patients contain CTNNBl genetic alteration, including missense mutations and amplification. Targeting Wnt/^-catenin has been comprehensively discussed (Pez et al., 2013; Vilchez et al., 2016). Wnt/ -catenin pathway inhibitor ICGOOl, FH535, and other small inhibitors have been tested in HCC in different in vitro and in vivo models [Delgado etal., 2014; Gedaly et al., 2014; Handeli and Simon, 2008). Cheng's group and Gedaly's group further proved that the combination of sorafenib and ICGOOl as well as the combination of sorafenib and FH535 had better treatment outcome in their experimental models [Galuppo et al., 2014; Lin et al., 2016). However, studies thus far displayed that the promoted downregulation or degradation of wild-type β-catenin is the mean of β-catenin pathway inhibition in HCC and there is no small inhibitor available and proven to be a mutant β-catenin inhibitor.
The present inventors have data to suggest that Varlitinib is an inhibitor of a mutated beta- catenin pathway.
SUMMARY OF THE DISCLOSURE
The following numbered paragraphs summarise the present invention:
1. A method of treating a patient for cancer (for example a solid tumour) by administering a compound of formula (I):
Figure imgf000003_0001
an enantiomer thereof or a pharmaceutically acceptable salt of any one of the same, for example wherein the compound of formula (I) is Varlitinib or a pharmaceutically acceptable salt thereof, wherein the patient has been identified as having a beta-catenin pathway mutation (in particular a mutation in beta-catenin).
2. The method according to paragraph 1, wherein the mutation is at a serine residue.
3. The method according to paragraph 1 or 2, wherein the mutation is at a threonine residue.
4. The method according to any one of paragraphs 1 to 3, wherein the mutation is in the GSK 3β region of the beta-catenin.
5. The method according to any one of paragraphs 1 to 4, wherein the mutation is T41A.
6. The method according to any one of paragraphs 1 to 5, wherein the patient has a cancer selected from: liver cancer (such as hepatocellular carcinoma), biliary tract cancer, gall bladder cancer, breast cancer (such as none ER+ breast cancer), prostate cancer, colorectal cancer, ovarian cancer, cervical cancer, lung cancer, gastric cancer, pancreatic, bone cancer, bladder cancer, head and neck cancer, thyroid cancer, skin cancer, renal cancer, and oesophagus cancer and combinations of two or more of the same.
7. The method according to paragraph 6, wherein the cancer is hepatocellular carcinoma. The method according to any one of paragraphs 1 to 7, wherein the compound of formula (I) Varlitinib:
Figure imgf000004_0001
or a pharmaceutically acceptable salt thereof.
A method according to any one of paragraphs 1 to 8, wherein the compound of formula [I) is provided as the free base.
A method according to any one of paragraphs 1 to 9, wherein the compound of formula (I) is administered as a pharmaceutical formulation, for example comprising at least one excipient, diluent or carrier.
A method according to any one of paragraphs 1 to 10, wherein the compound of formula (I) or a pharmaceutical formulation comprising same is administered orally.
A method according to any one of paragraphs 1 to 11, wherein the compound of formula (I) or a pharmaceutical formulation comprising the same is administered bi-daily.
A method according to any one of paragraphs 1 to 12, wherein each dose of the compound of formula (I) is in the range 100 to 900mg, for 100, 200, 300, 400, 500, 600, 700, 800 or 900mg. A method according to paragraph 13, wherein each dose of the compound of formula (I) is in the range 200 to 500mg or 100 to 500mg, for example 250 mg or 490 mg.
A method according to paragraph 14, wherein each dose is lOOmmg, 200mg, 300g or 400mg (such as 300mg or 400mg).
A method according to any one of paragraphs 1 to 15, wherein the compound of formula (I) or formulation comprising the same is employed as a monotherapy.
A method according to any one of paragraphs 1 to 16, wherein the compound of formula (I) is employed in a combination with another anti-cancer agent.
A method according to paragraph 17, wherein the combination therapy comprises a chemotherapeutic agent.
A method according to paragraph 18, wherein the chemotherapeutic agent is independently selected from the group comprising a platin (such as cisplatin or oxaliplatin), gemcitabine, capecitabine, 5-FU, FOLFOX, FOLFIRI and FOLFIRINOX.
A compound of formula (1} as defined abovean enantiomer thereof or a pharmaceutically acceptable salt of any one of the same, for example wherein the compound of formula (I] is Varlitinib or a pharmaceutically acceptable salt thereof, for use in treating a cancer patient (for example a solid tumour) wherein the patient has been identified as having a beta-catenin pathway mutation (in particular a mutation in beta-catenin.
The compound for use according to paragraph 20, wherein the mutation is at a serine residue. The compounds for use according to paragraphs 20 or 21, wherein the mutation is at a threonine residue.
The compound for use according to any one of paragraphs 20 to 22, wherein the mutation is in the GSK 3β region of the beta-catenin.
The compound for use according to any one of paragraphs 20 to 23, wherein the mutation is T41A.
The compound for use according to any one of paragraphs 20 to 24, wherein the patient has a cancer selected from: liver cancer (such as hepatocellular carcinoma), biliary tract cancer, gall bladder cancer, breast cancer (such as none ER+ breast cancer), prostate cancer, colorectal cancer, ovarian cancer, cervical cancer, lung cancer, gastric cancer, pancreatic, bone cancer, bladder cancer, head and neck cancer, thyroid cancer, skin cancer, renal cancer, and oesophagus cancer and combinations of two or more of the same.
The compounds for use according to paragraph 25, wherein the cancer is hepatocellular carcinoma.
The compound for use according to any one of paragraphs 20 to 26, wherein the compound of formula (I) is Varlitinib as defined above or a pharmaceutically acceptable salt thereof.
A compound for use according to any one of paragraphs 20 to 27, wherein the compound of formula (I) is provided as the free base.
A compound for use according to any one of paragraphs 20 to 28, wherein the compound of formula (I) is administered as a pharmaceutical formulation, for example comprising at least one excipient, diluent or carrier.
A compound for use according to any one of paragraphs 20 to 29, wherein the compound of formula (I) or a pharmaceutical formulation comprising same is administered orally.
A compound for use according to any one of paragraphs 20 to 30, wherein the compound of formula (I) or a pharmaceutical formulation comprising the same is administered bi-daily. A compound for use according to any one of paragraphs 20 to 31, wherein each dose of the compound of formula (I) is in the range 100 to 900mg, for example 100, 200, 300, 400, 500, 600, 700, 800 or 900mg.
A compound for use according to paragraph 32, wherein each dose of the compound of formula (I) is in the range 200 to 500mg or 100 to 500mg, for example 250 mg or 490 mg.
A compound for use according to paragraph 33, wherein each dose is lOOmg, 200mg, 300mg or 400mg (such as 300mg or 400mg).
A compound for use according to any one of paragraphs 20 to 34, wherein the compound of formula (I) or formulation comprising the same is employed as a monotherapy.
A compound for use according to any one of paragraphs 20 to 34, wherein the compound of formula (I) is employed in a combination with another anti-cancer agent
A compound for use according to paragraph 36, wherein the combination therapy comprises a chemotherapeutic agent. A compound for use according to paragraph 37, wherein the chemotherapeutic agent is independently selected from the group comprising a platin [such as cisplatin or oxaliplatin), gemcitabine, capecitabine, 5-FU, FOLFOX, FOLFIRI and FOLFIRINOX.
Use of a compound of formula (I) as defined abovean enantiomer thereof or a pharmaceutically acceptable salt of any one of the same, for example wherein the compound of formula (I) is Varlitinib or a pharmaceutically acceptable salt thereof, for the manufacture of a medicament for treating a cancer patient [for example a solid tumour) wherein the patient has been identified as having a beta-catenin pathway mutation [in particular a mutation in beta-catenin. Use according to paragraph 39, wherein the mutation is at a serine residue.
Use according to paragraphs 39 or 40, wherein the mutation is at a threonine residue.
Use according to any one of paragraphs 39 to 41, wherein the mutation is in the GSK 3β region of the beta-catenin.
Use according to any one of paragraphs 39 to 42, wherein the mutation is T41A.
Use according to any one of paragraphs 39 to 43, wherein the patient has a cancer selected from: liver cancer [such as hepatocellular carcinoma), biliary tract cancer, gall bladder cancer, breast cancer [such as none ER+ breast cancer), prostate cancer, colorectal cancer, ovarian cancer, cervical cancer, lung cancer, gastric cancer, pancreatic, bone cancer, bladder cancer, head and neck cancer, thyroid cancer, skin cancer, renal cancer, and oesophagus cancer and combinations of two or more of the same.
Use according to paragraph 44, wherein the cancer is hepatocellular carcinoma.
Use according to any one of paragraphs 39 to 45, wherein the compound of formula [I) is Varlitinib as defined above or a pharmaceutically acceptable salt thereof.
Use according to any one of paragraphs 39 to 46, wherein the compound of formula [I) is provided as the free base.
Use according to any one of paragraphs 39 to 47, wherein the compound of formula [I) is administered as a pharmaceutical formulation, for example comprising at least one excipient, diluent or carrier.
Use according to any one of paragraphs 39 to 48, wherein the compound of formula (I) or a pharmaceutical formulation comprising same is administered orally.
Use according to any one of paragraphs 39 to 49, wherein the compound of formula (I) or a pharmaceutical formulation comprising the same is administered bi-daily.
Use according to any one of paragraphs 39 to 50, wherein each dose of the compound of formula [I) is in the range 100 to 900mg, for 100, 200, 300, 400, 500, 600, 700, 800 or 900mg.
Use according to paragraph 51, wherein each dose of the compound of formula [I) is in the range 200 to 500mg or 100 to 500mg, for example 250 mg or 490 mg.
Use according to paragraph 52, wherein each dose is lOOmg, 200mg, 300mg or 400mg [such as 300mg or 400mg).
Use according to any one of paragraphs 39 to 53, wherein the compound of formula [I) or formulation comprising the same is employed as a monotherapy. 55. Use according to any one of paragraphs 39 to 53, wherein the compound of formula [I] is employed in a combination with another anti-cancer agent.
56. Use according to claim 55, wherein the combination therapy comprises a chemotherapeutic agent.
57. Use according to claim 56, wherein the chemotherapeutic agent is independently selected from the group comprising a platin (such as cisplatin or oxaliplatin), gemcitabine, capecitabine, 5-FU, FOLFOX, FOLFIRI and FOLFIRINOX.
In one embodiment the mutated beta-catenin localises in a cell's cytoplasm and/or nucleus. In one embodiment, the cancer is both HER2 activated and HER3 activated. For example, the cancer has at least one of the following:
• higher expression levels of total ErbB2 and total ErbB3 compared to a non-cancerous cell; · higher expression levels of phospho-ErbB2 and total ErbB3 compared to a non-cancerous cell;
• higher expression levels of total ErbB2 and phospho-ErbB3 compared to a non-cancerous cell;
• higher expression levels of phospho-ErbB2 phospho-ErbB3 compared to a non-cancerous cell;
• higher expression levels of total ErbB2, phospho-ErbB2 and total ErbB3 compared to a noncancerous cell;
· higher expression levels of total ErbB2, phospho-ErbB2 and phospho-ErbB3 compared to a noncancerous cell;
• higher expression levels of phospho-ErbB2, total ErbB3 and phospho-ErbB3 compared to a noncancerous cell;
• higher expression levels of total-ErbB2, total ErbB3 and phospho-ErbB3 compared to a non- cancerous cell; or
• higher expression levels of total ErbB2, phospho-ErbB2, total ErbB3 and phospho-ErbB3 compared to a non-cancerous cell.
In one embodiment the biliary duct cancer is in a location selected from intrahepatic bile ducts, left hepatic duct, right hepatic duct, common hepatic duct, cystic duct, common bile duct, Ampulla of Vater and combinations thereof.
In one embodiment the biliary duct cancer is in an intrahepatic bile duct. In one embodiment the biliary duct cancer is in a left hepatic duct. In one embodiment the biliary duct cancer is in a right hepatic duct. In one embodiment the biliary duct cancer is in a common hepatic duct. In one embodiment the biliary duct cancer is in a cystic duct. In one embodiment the biliary duct cancer is in a common bile duct. In one embodiment the biliary duct cancer is in an Ampulla of Vater. In one embodiment the biliary duct cancer is a cancer of the Papilla of Vater.
In one embodiment the cancer is a metastatic form of a cancer, in the patient population disclosed herein disclosed herein.
In one embodiment the cancer according the present disclosure has not metastasized in said patient population.
In one embodiment the compound of formula (I) is (i?)-N4-[3-Chloro-4-(thiazol-2- ylmethoxy)-phenyl]-N6-(4-methyl-4, 5,-dihydro-oxazol-2-yl)-quinazoline-4,6-diamine:
Figure imgf000008_0001
(Varlitinib) or a pharmaceutically acceptable salt thereof or a pro-drug thereof.
In one embodiment (fi)-N4-[3-Chloro-4-(thiazol-2-ylmethoxy)-phenyl]-N6-(4-methyl-4, 5,- dihydro-oxazol-2-yl)-quinazoline-4,6-diamine is employed/administered as the free base (also referred to herein as Varlitinib).
In one embodiment the patient population is not activated for all of HER1, HER2, HER3 and HER4, for example the population is at least negative for HER1 and/or HER4. Thus, in one embodiment the patient population is HER2 and/or HER3 and HER1 activated but not HER4 activated. In one embodiment the patient population is HER2 and/or HER3 activated and HER4 activated but not HER1 activated. In one embodiment the patient population is only HER2 and/or HER3 activated. In one embodiment the patient population is only HER2 activated. In one embodiment the patient population is only HER3 activated. In one embodiment the patient population is both HER3 and HER4 activated.
In one embodiment the compound of formula (I) an enantiomer thereof or a pharmaceutically acceptable salt thereof is employed as a monotherapy, for example first line therapy or second line therapy, such as a first line monotherapy, in said patient population.
In one embodiment the compound of formula (I) an enantiomer thereof or a pharmaceutically acceptable salt thereof is employed in a combination therapy, for example in combination with a chemotherapy and/or a biological therapeutic, in particular as a first line therapy or a second line therapy in said patient population.
In one embodiment the compound of formula (I) an enantiomer thereof or a pharmaceutically acceptable salt thereof is employed as a second line monotherapy, in said patient population.
In, for example a second line monotherapy employing Varlitinib in the treatment of cholangiocarcinoma in said patient patients may show a significant reduction in CA19-9 levels. CA19-9 is a marker employed in the management of cholangiocarcinoma. Thus in one embodiment a HER3 and HER4 positive or amplified patient has a 10, 20, 30, 40, 50, 60, 70, 80 or 90% decrease in CA19-9 level whilst on the therapy according to the present disclosure, wherein the level is decreased relative to the level of CA19-9 before initiation of said therapy. This decrease in CA19-9 may be observed 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or weeks after initiating therapy according to the present disclosure.
In one embodiment the compound of formula (I), according to the present disclosure such as Varlitinib, is employed in a second line therapy together with a chemotherapy agent or chemotherapy regimen, for example gemcitabine, capecitabine, 5-FU, FOLFOX, a platin, such as cisplatin or oxaliplatin, and a combination thereof, in said patient population. In one embodiment the compound of formula [I), according to the present disclosure such as Varlitinib, is administered orally.
In one embodiment the compound of formula (I), according to the present disclosure such as Varlitinib, is administered at a dose in the range lOOmg to 900mg on each occasion, in particular 200 mg, 300mg, 400mg or 500mg each dose, such as 400mg, 250mg or 490 mg, for example administered once or twice daily, such as twice daily.
In one embodiment, the compound of formula (I), according to the present disclosure such as Varlitinib is administered twice daily.
In one embodiment the compound of formula (I), according to the present disclosure such as Varlitinib, is administered for a 28 days, referred to herein as a 28 day treatment cycle.
In one embodiment the compound of formula (I), according to the present disclosure such as Varlitinib, is administered as pharmaceutical formulation comprising one or more pharmaceutically acceptable excipients.
In one embodiment the compound of formula (I), according to the present disclosure such as Varlitinib, or a formulation comprising the same is administered orally, for example as tablet or capsule.
In one embodiment the treatment is adjuvant therapy, for example after surgery or after chemotherapy, in said patient population.
In one embodiment the treatment is neoadjuvant therapy, for example before surgery, in particular to shrink the tumour or tumours, in said patient population.
In one embodiment the treatment according to the present disclosure is suitable for the treatment of secondary tumours, in said patient population. In one embodiment the cancer is metastatic cancer, in said patient population. In one embodiment the treatment according to the present disclosure is suitable for the treatment of primary cancer and metastases, in said patient population. In one embodiment the treatment according to the present disclosure is suitable for the treatment of secondary cancer and metastases. In one embodiment the treatment according to the present disclosure is suitable for the treatment of primary cancer, secondary cancer and metastases in said patient population.
In one embodiment the treatment according to the present disclosure is suitable for the treatment of cancerous cells in a lymph node, for a cancer of the present disclosure in said patient population.
In one embodiment the patient is a human.
i) Also provided is a method comprising profiling an ex vivo sample from a patient to identify dysregulation in at least two genes independently selected from FM03, RORC, NAG, ADIRF, CES3, FAT4, S100A14, SLC16A14, AZGP1, PPL, LOC200772, TNNI1, SLC1A7, BCHE, CYP4A11, NOY6R, ENPEP, HA01, CYP3A5, BHMT, TMEM132A, EPO, FAM19A5, ASNS, HK2, TTBK1, NSMF, HAL, OBSL1, LOCI 00507194, GPC4, CERCAM, ADRA2C, SERPINE1 , SPARC, SLC6A8, SCARF2, HS6ST2, ADM and IRX3.
ii) In one embodiment the dysregulation is upregulation (also referred to herein as activation) of at least one gene (for example 1, 2, 3, , 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes) independently selected from FM03, RORC, NAG, ADIRF, CES3, FAT4, SI 00A14, SLC16A14, AZGP1, PPL, LOC200772, TNN11, SLC1A7, BCHE, CYP4A11, NOY6R, ENPEP, HAOl, CYP3A5, and BHMT. iii) In one embodiment the dysregulation is downregulation [also referred to herein as repression} of at least one gene (for example 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes] independently selected from TMEM132A, EPO, FAM19A5, ASNS, HK2, TTBK1, NSMF, HAL, OBSL1, LOC100507194, GPC4, CERCAM, ADRA2C, SERPINE1, SPARC, SLC6A8, SCARF2, HS6ST2, ADM and IRX3.
iv) In one embodiment the dysregulation is upregulation of at least one gene (for example 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes) independently selected from FM03, RORC, NAG, ADIRF, CES3, FAT4, S100A14, SLC16A14, AZGP1, PPL, LOC200772, TNNI1, SLC1A7, BCHE, CYP4A11, NOY6R, ENPEP, HAOl, CYP3A5, and BHMT, and downregulation of at least one gene (for example 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 genes) independently selected from TMEM132A, EPO, FAM19A5, ASNS, HK2, TTBK1, NSMF, HAL, OBSL1, LOC100507194, GPC4, CERCAM, ADRA2C, SERPINE1, SPARC, SLC6A8, SCARF2, HS6ST2, ADM and IRX3.
In one embodiment the profiling is for diagnostic or prognostic purposes, for example to identify the patient as having cancer, or a subtype thereof.
In one embodiment the profiling is for identifying a patient as suitable for treatment with regime according to the present disclosure.
In one embodiment there is provided treatment of a new patient population comprising a cancer patient with dysregulation according to any one of paragraphs i), ii), iii) and iv).
Thus, in one embodiment there is provided treatment of a patient, according to the present disclosure, wherein said patient has been identified within the target population.
DETAILED DISCLOSURE
The following disclosure is given in the context of treatment of the requisite patient population and intended to complement the disclosure in the summary of the invention section above.
The CTNNB1 gene provides instructions for making a protein called beta-catenin. This protein is present in many types of cells and tissues, where it is primarily found at junctions that connect neighboring cells (adherens junctions). Beta-catenin plays an important role in sticking cells together (cell adhesion) and in communication between cells.
The beta-catenin protein is also involved in cell signaling as an essential part of the Wnt signaling pathway. Certain proteins in this pathway attach (bind) to beta-catenin, which triggers a multi-step process that allows the protein to move into the nucleus. Once in the nucleus, beta-catenin interacts with other proteins to control the activity (expression) of particular genes. The Wnt signaling pathway promotes the growth and division (proliferation) of cells and helps determine the specialized functions a cell will have (differentiation). Wnt signaling is known to be involved in many aspects of development before birth. In adult tissues, this pathway plays a role in the maintenance and renewal of stem cells, which are cells that help repair tissue damage and can give rise to other types of cells.
A mutations in β-catenin pathway refers to any mutation that impacts on the function of the said pathway, in particular a mutation in β-catenin and/or the gene encoding the same CTNNB1, including genetic alteration, including missense mutations and amplification.
In one embodiment the β-catenin mutation is T41A. The T41A β-catenin mutation is thought to be constitutively active. However, the present inventors have demonstrated that its nuclear localisation and β-catenin-driven genes can be inhibited by a pan-HER inhibitor such as Varlitinib in high p-ErbB2/3 expressing cancers, such as hepatocellular carcinoma (HCC). Surprisingly, the present inventors have shown that Varlitinib can selectively target cancers that have the T41A β- catenin mutation. Accordingly, cancer patients, such as HCC patients with the T41A CTNNB1 mutation may represent a specific patient population subset that is particularly responsive to treatment with a pan-HER inhibitor, such as Varlitinib.
Targeting Wnt/^-catenin has been comprehensively discussed (Pez et al., 2013; Vilchez et al., 2016).
HER2 and/or HER3 activated as employed herein refers to where the cancer expresses an increased level of said receptor, such as total levels of the receptor and/or levels of the activated form of the receptor, such as the phosphorylated form of the receptor, in comparison a corresponding non-cancerous cell and includes any of the following:
• increased levels of total HER2 only;
• increased levels of phospho-HER2 only;
· increased levels of total HER2 and phospho-HER2 only;
• increased levels of total HER3 only;
• increased levels of phospho-HER3 only;
• increased levels of total HER3 and phospho-HER3 only;
• increased levels of total ErbB2 and total ErbB3 only;
· increased levels of phospho-ErbB2 and total ErbB3 only;
• increased levels of total ErbB2 and phospho-ErbB3 only;
• increased levels of phospho-ErbB2 and phospho-ErbB3 only;
• increased levels of total ErbB2, phospho-ErbB2 and total ErbB3 only;
• increased levels of total ErbB2, phospho-ErbB2 and phospho-ErbB3 only;
· increased levels of phospho-ErbB2, total ErbB3 and phospho-ErbB3 only;
• increased levels of total-ErbB2, total ErbB3 and phospho-ErbB3 only; and
• increased levels of total ErbB2, phospho-ErbB2, total ErbB3 and phospho-ErbB3.
Methods of measuring said expression and amplification are known in the art.
Phospho-ErbB2/p-ErbB2 or phospho-ErbB3/p-ErbB3 as used herein refers to the phosphorylated form of ErbB2 or ErbB3 respectively. It includes phosphorylation of the receptor at any suitable phosphorylation site. Known phosphorylation sites for ErbB2 include but are not limited to: Tyr 877, Tyrl023, Tyrll39, Tyrll96, Tyrl221, Tyrl222 and Tyrl248. Known phosphorylation sites for ErbB3 include but are not limited to Tyrl054, Tyrll97, Tyrll99, Tyrl222, Tyrl260, Tyrl262, Tyrl289 and Tyrl328.
In one embodiment, phospho-ErbB2 refers to ErbB2 which has been phosphorylated at
Tyrl221 or Tryl222.
In one embodiment, phospho-ErbB3 refers to ErbB3 which has been phosphorylated at Tyrl289.
In one embodiment, increased levels as employed herein refers to a level that is at least 1, 2 or 3, such as 2 standard deviations higher than the mean level in a non-cancerous cell.
In another embodiment, increased levels as employed herein refers to a level that is above the level of the 95th percentile of a population of non-cancerous cells. "In a patient population" refers to administration of the therapy to a patient characterised
Unless the context indicates otherwise, "identified" in the context of the present disclosure refer to the patient is known to be in the requisite patient population before the administration of the therapy according to the present disclosure.
Liver cancer as employed herein refers to cancer which starts in the liver, including starting from structures located within the liver, such as blood vessels, including hepatocellular carcinoma.
In one embodiment the liver cancer is, for example selected from the group hepatocellular carcinoma, cholangiocarcinoma, angiosarcoma, and hepatoblastoma, in particular hepatocellular carcinoma. In one embodiment the primary liver cancer is stage 1, 2, 3 or 4. In one embodiment the liver cancer is secondary or metastasized liver cancer.
In one embodiment liver cancer does not include biliary tract cancer, such as cholangiocarcinoma.
In one embodiment the cancer is liver cancer, for example a liver metastasis from a primary cancer, for example colon cancer, which has spread to the liver. In one embodiment the liver cancer is HCC hepatocellular carcinoma.
Biliary duct cancer (also referred to as biliary cancer) as employed herein refers to cancer which starts in the bile ducts and includes cholangiocarcinoma and gallbladder cancer.
In one embodiment Biliary tract cancer as employed herein refers to cholangiocarcinoma
(intrahepatic, extrahepatic), gall bladder cancer and ampullary carcinoma.
Cholangiocarcinoma as referred to herein is a form of cancer that is composed of mutated epithelial cells (or cells showing characteristics of epithelial differentiation) that originate in the bile ducts which drain bile from the liver into the small intestine, but not including gallbladder cancer.
General guidelines for operability of biliary duct cancer include: Absence of lymph node or liver metastases; Absence of involvement of the portal vein; Absence of direct invasion of adjacent organs; and Absence of widespread metastatic disease.
In one embodiment, the cancer is gall bladder cancer. Gallbladder cancer as employed herein cancer which starts in the gallbladder. The following stages are used for gallbladder cancer:
Stage 0 (carcinoma in situ): Abnormal cells are found in the inner (mucosal) layer of the gallbladder; these abnormal cells may become cancer and spread into nearby normal tissue; Stage I Cancer has formed and has spread beyond the inner (mucosal) layer to a layer of tissue with blood vessels or to the muscle layer; Stage II Cancer has spread beyond the muscle layer to the connective tissue around the muscle; Stage IIIA Cancer has spread through the thin layers of tissue that cover the gallbladder and/or to the liver and/or to one nearby organ (e.g., stomach, small intestine, colon, pancreas, or bile ducts outside the liver); Stage IIIB Cancer has spread to nearby lymph nodes and beyond the inner layer of the gallbladder to a layer of tissue with blood vessels or to the muscle layer; or beyond the muscle layer to the connective tissue around the muscle; or through the thin layers of tissue that cover the gallbladder and/or to the liver and/or to one nearby organ; Stage IVA Cancer has spread to a main blood vessel of the liver or to 2 or more nearby organs or areas other than the liver. Cancer may have spread to nearby lymph nodes; and Stage IVB Cancer has spread to either lymph nodes along large arteries in the abdomen and /or near the lower part of the backbone or to organs or areas far away from the gallbladder. In one embodiment the gastric cancer is selected from the group comprising adenocarcinoma of the stomach, squamous cell carcinomas, lymphoma of the stomach, gastric stromal tumor, and neuroendocrine tumors.
Prostate cancer as employed herein refers to cancer of the prostate, for example ductal adenocarcinoma, transitional cell (urothelial cancer], squamous cell cancer, carcinoid of the prostate, small cell cancer or sarcoma and sarcomatoid cancer. In one embodiment the prostate cancer is any one of the same.
Pancreatic cancer as employed herein includes exocrine cancers (including rare forms thereof such as cystitic tumours, and cancer of the acinar cells), endocrine pancreatic tumours (including gastrinomas, insulinomas, somatostatinomas, VIPomas, glucagonomas), pancreatoblastoma, sarcomas of the pancreas and lymphoma.
Colorectal cancer as employed herein refers to cancer or the colon and/or rectum and includes squamous cell cancers, carcinoid tumours, sarcomas and lymphomas.
Breast cancer as employed herein refers to cancer of the breast and includes ductal cardinoma in situ, lobular carcinoma in situ, invasive ductal breast cancer, invasive lobular breast cancer, invasive breast cancer, Paget's disease, angiosarcoma of the breast and rare types of breast cancer such as medullary breast cancer, mucinous breast cancer, tubular breast cancer, adenoid cystic carcinoma of the breast metaplastic breast cancer, basal type breast cancer and papillary breast cancer. In one embodiment the breast cancer is one selected from any one of the same. In one embodiment the breast cancer is phyllodes or cystosarcoma phyllodes.
Lung cancers are classified according to histological type and are categorized by the size and appearance of the malignant cells seen by a histopathologist under a microscope. For therapeutic purpose, two broad classes are distinguished: non-small cell lung carcinoma and small cell lung carcinoma.
In one embodiment the epithelial cancer is lung cancer, for example selected from small-cell lung cancer (SCLC) and non-small-cell lung cancer (NSCLC), such as NSCLC.
Non-small-cell lung carcinoma-The three main subtypes of NSCLC are adenocarcinoma, squamous-cell carcinoma and large-cell carcinoma.
Nearly 40% of lung cancers are adenocarcinoma, which usually originates in peripheral lung tissue. A subtype of adenocarcinoma, the bronchioloalveolar carcinoma, is more common in female never-smokers, and may have a better long term survival.
Squamous-cell carcinoma accounts for about 30% of lung cancers. They typically occur close to large airways. A hollow cavity and associated cell death are commonly found at the center of the tumor. About 9% of lung cancers are large-cell carcinoma. These are so named because the cancer cells are large, with excess cytoplasm, large nuclei and conspicuous nucleoli.
Small-cell lung carcinonia-In small-cell lung carcinoma (SCLC), the cells contain dense neurosecretory granules (vesicles containing neuroendocrine hormones), which give this tumor an endocrine/paraneoplastic syndrome association. Most cases arise in the larger airways (primary and secondary bronchi). These cancers grow quickly and spread early in the course of the disease. Sixty to seventy percent have metastatic disease at presentation.
In one embodiment the cancer is non-small lung carcinoma. In one embodiment there is provided treatment of renal cancer, for example renal cell carcinoma and/or urothelial cell carcinoma. Other examples of renal cancer include squamous cell carcinoma, juxtaglomerular cell tumor [reninoma), angiomyolipoma, renal oncocytoma, Bellini duct carcinoma, clear-cell sarcoma of the kidney, mesoblastic nephroma, Wilms' tumor, mixed epithelial stromal tumor, clear cell adenocarcinoma, transitional cell carcinoma, inverted papilloma, renal lymphoma, teratoma, carcinosarcoma, and carcinoid tumor of the renal pelvis. Renal cancer as employed herein refers to cancer of the kidney
In one embodiment the cancer is bladder cancer, for example is any of several types of malignancy arising from the epithelial lining (i.e., the urothelium) of the urinary bladder. About 90% of bladder cancers are transitional cell carcinoma. The other 10% are squamous cell carcinoma, adenocarcinoma, sarcoma, small cell carcinoma, and secondary deposits from cancers elsewhere in the body. The staging of is given below.
T (Primary tumour)
TX Primary tumour cannot be assessed; TO No evidence of primary tumour; Ta Non-invasive papillary carcinoma; Tis Carcinoma in situ ('flat tumour'); Tl Tumour invades subepithelial connective tissue; T2a Tumour invades superficial muscle (inner half); T2b Tumour invades deep muscle (outer half); T3 Tumour invades perivesical tissue; T3a Microscopically; T3b Macroscopically (extravesical mass); T4a Tumour invades prostate, uterus or vagina; T4b Tumour invades pelvic wall or abdominal wall;
N (Lymph nodes)
NX Regional lymph nodes cannot be assessed; NO No regional lymph node metastasis; Nl Metastasis in a single lymph node 2 cm or less in greatest dimension; N2 Metastasis in a single lymph node more than 2 cm but not more than 5 cm in greatest dimension, or multiple lymph nodes, none more than 5 cm in greatest dimension; N3 Metastasis in a lymph node more than 5 cm in greatest dimension
M (Distant metastasis)
MX Distant metastasis cannot be assessed; M0 No distant metastasis; Ml Distant metastasis.
The current disclosure extends to any stage of bladder cancer.
There are more than 30 different types of ovarian cancer which are classified according to the type of cell from which they start. Cancerous ovarian tumors can start from three common cell types: Surface Epithelium - cells covering the lining of the ovaries; Germ Cells - cells that are destined to form eggs; and Stromal Cells - Cells that release hormones and connect the different structures of the ovaries.
The present disclosure relates to treatment of ovarian cancer from any source, for example as described herein, in particular epithelium cells. Epithelial ovarian carcinomas (EOCs) account for 85 to 90 percent of all cancers of the ovaries.
Common Epithelial Tumors - Epithelial ovarian tumors develop from the cells that cover the outer surface of the ovary. Most epithelial ovarian tumors are benign (noncancerous). There are several types of benign epithelial tumors, including serous adenomas, mucinous adenomas, and Brenner tumors. Cancerous epithelial tumors are carcinomas - meaning they begin in the tissue that lines the ovaries. These are the most common and most dangerous of all types of ovarian cancers. Unfortunately, almost 70 percent of women with the common epithelial ovarian cancer are not diagnosed until the disease is advanced in stage.
There are some ovarian epithelial tumors whose appearance under the microscope does not clearly identify them as cancerous. These are called borderline tumors or tumors of low malignant potential (LMP tumors). The present disclosure includes treatment of the latter.
Germ Cell Tumors - Ovarian germ cell tumors develop from the cells that produce the ova or eggs. Most germ cell tumors are benign (non-cancerous), although some are cancerous and may be life threatening. The most common germ cell malignancies are maturing teratomas, dysgerminomas, and endodermal sinus tumors. Germ cell malignancies occur most often in teenagers and women in their twenties. Today, 90 percent of patients with ovarian germ cell malignancies can be cured and their fertility preserved.
Stromal Tumors - Ovarian stromal tumors are a rare class of tumors that develop from connective tissue cells that hold the ovary together and those that produce the female hormones, estrogen and progesterone. The most common types are granulosa-theca tumors and Sertoli- Ley dig cell tumors. These tumors are quite rare and are usually considered low-grade cancers, with approximately 70 percent presenting as Stage I disease (cancer is limited to one or both ovaries).
Primary Peritoneal Carcinoma-The removal of one's ovaries eliminates the risk for ovarian cancer, but not the risk for a less common cancer called Primary Peritoneal Carcinoma. Primary Peritoneal Carcinoma is closely rated to epithelial ovarian cancer (most common type). It develops in cells from the peritoneum (abdominal lining) and looks the same under a microscope. It is similar in symptoms, spread and treatment.
Stages of Ovarian Cancer
Once diagnosed with ovarian cancer, the stage of a tumor can be determined during surgery, when the doctor can tell if the cancer has spread outside the ovaries. There are four stages of ovarian cancer - Stage I (early disease) to Stage IV (advanced disease). The treatment plan and prognosis (the probable course and outcome of your disease) will be determined by the stage of cancer you have.
Following is a description of the various stages of ovarian cancer:
Stage I - Growth of the cancer is limited to the ovary or ovaries.
Stage IA - Growth is limited to one ovary and the tumor is confined to the inside of the ovary. There is no cancer on the outer surface of the ovary. There are no ascites present containing malignant cells. The capsule is intact.
Stage IB - Growth is limited to both ovaries without any tumor on their outer surfaces. There are no ascites present containing malignant cells. The capsule is intact.
Stage IC - The tumor is classified as either Stage IA or IB and one or more of the following are present: (1) tumor is present on the outer surface of one or both ovaries; (2) the capsule has ruptured; and (3) there are ascites containing malignant cells or with positive peritoneal washings.
Stage II - Growth of the cancer involves one or both ovaries with pelvic extension.
Stage HA - The cancer has extended to and/or involves the uterus or the fallopian tubes, or both. Stage IIB - The cancer has extended to other pelvic organs. Stage IIC - The tumor is classified as either Stage HA or IIB and one or more of the following are present: (1] tumor is present on the outer surface of one or both ovaries; (2} the capsule has ruptured; and (3) there are ascites containing malignant cells or with positive peritoneal washings.
Stage III - Growth of the cancer involves one or both ovaries, and one or both of the following are present: (1) the cancer has spread beyond the pelvis to the lining of the abdomen; and [2) the cancer has spread to lymph nodes. The tumor is limited to the true pelvis but with histologically proven malignant extension to the small bowel or omentum.
Stage IIIA - During the staging operation, the practitioner can see cancer involving one or both of the ovaries, but no cancer is grossly visible in the abdomen and it has not spread to lymph nodes. However, when biopsies are checked under a microscope, very small deposits of cancer are found in the abdominal peritoneal surfaces.
Stage IIIB - The tumor is in one or both ovaries, and deposits of cancer are present in the abdomen that are large enough for the surgeon to see but not exceeding 2 cm in diameter. The cancer has not spread to the lymph nodes.
Stage IIIC - The tumor is in one or both ovaries, and one or both of the following is present: (1) the cancer has spread to lymph nodes; and/or (2) the deposits of cancer exceed 2 cm in diameter and are found in the abdomen.
Stage IV - This is the most advanced stage of ovarian cancer. Growth of the cancer involves one or both ovaries and distant metastases [spread of the cancer to organs located outside of the peritoneal cavity) have occurred. Finding ovarian cancer cells in pleural fluid [from the cavity which surrounds the lungs) is also evidence of stage IV disease. In one embodiment the ovarian cancer is: type I, for example IA, IB or IC; type II, for example IIA, IIB or IIC; type III, for example IIIA, IIIB or IIIC; or type IV.
Thyroid cancer as employed herein refers to cancer of the thyroid originating from follicular or parafollicular thyroid cells and includes papillary thyroid cancer [75% to 85% of cases); follicular thyroid cancer [10% to 20% of cases); medullary thyroid cancer [5% to 8% of cases)- cancer of the parafollicular cells, often part of multiple endocrine neoplasia type 2; poorly differentiated thyroid cancer; anaplastic thyroid cancer [less than 5% of cases) is not responsive to treatment and can cause pressure symptoms, thyroid lymphoma, squamous cell thyroid carcinoma, sarcoma of thyroid. The present disclosure extends to treatment of thyroid cancer.
Bladder cancer as employed herein refers to cancer of the bladder including transitional cell bladder cancer, carcinoma in situ, papillary cancer and rarer types of bladder cancer such as squamous cell cancer and adenocarcinoma. The present disclosure extends to treatment of bladder cancer.
Esophageal cancer as employed herein refers to cancer of the oesphagus including esophageal squamous-cell carcinomas, esophageal adenocarcinomas, and variants of squamous-cell carcinoma, and non-epithelial tumors, such as leiomyosarcoma, malignant melanoma, rhabdomyosarcoma, lymphoma, among others. The present disclosure relates to treatment of the esphageal cancer. Head and neck cancer as employed herein refers to cancer of the neck and/or head, including mouth cancer, nasopharyngeal cancer, oropharyngeal cancer, paranasal sinus cancer and salivary gland cancer. The present disclosure includes treatment of head and neck cancer.
In one embodiment the treatment of the present disclosure is neo-adjuvant therapy, for example to shrink the tumour/carcinoma before surgery to remove the cancerous tissue or before chemotherapy to improve the chances of success of the latter or to reduce the severity of the treatment required.
In one embodiment the treatment of the present disclosure is adjuvant therapy, for example following surgery to remove the cancerous tissue.
In one embodiment the treatment of the present disclosure is adjuvant therapy, for example following chemotherapy
In patients where not all the cancerous tissue is removed by surgery then the patient may benefit from adjuvant therapy which is monotherapy employing a compound of formula [I), such as Varlitinib.
In patients where not all the cancerous tissue is removed by surgery then the patient may benefit from combination adjuvant therapy comprising a compound of formula (I) and chemotherapy or radiotherapy.
First line therapy as employed herein is the first therapy employed for the treatment of the cancer and in some instances the first line therapy may be neo-adjuvant therapy, in this context surgery will generally be considered a treatment.
Second line therapy as employed herein is treatment following first line therapy and may be adjuvant therapy. Thus, in the context of the present specification second line therapy is simply therapy other than first line therapy and includes, third line therapy, fourth line therapy etc.
Monotherapy as employed herein is wherein the compound of formula (I) an enantiomer thereof and/or a pharmaceutically acceptable salt thereof, is the only active agent being administered to the patient for the treatment of cancer.
In one embodiment there is provided treatment according to the present disclosure with a combination therapy comprising Varlitinib, for example Varlitinib and an anti-cancer therapy, such as chemotherapy.
Combination therapy as employed herein refers to wherein the compound of formula [I) an enantiomer thereof or a pharmaceutically acceptable salt thereof is employed for the treatment of the cancer in conjunction with one or more further anticancer treatments, for example where the treatment regimens for the two or more active anticancer agents overlap or where the two or more anticancer agents are administered concomitantly.
In one embodiment the combination therapy according to the present disclosure comprises a RON inhibitor, for example as disclosed WO2008/058229, incorporated herein by reference.
In one embodiment the combination therapy comprises a checkpoint inhibitor, such as a CTLA4 inhibitor, a PD-1 inhibitor or a PD-Ll inhibitor, in particular an antibody or binding fragment thereof.
Examples of pharmaceutically acceptable salts include but are not limited to acid addition salts of strong mineral acids such as HC1 and HBr salts and addition salts of strong organic acids, such as a methansulfonic acid salt, tosylates, furoates and the like, including di, tri salts thereof, such as ditosylates.
Analysis of patients to profile their cancer, for example to establish if their cancer is EGFR and HER2 positive is known and is routine in the art. Establishing if a cancer is HER2 amplified is also routine in the art
Chemotherapeutic Agents
Chemotherapeutic agent and chemotherapy or cytotoxic agent are employed interchangeably herein unless the context indicates otherwise.
Chemotherapy as employed herein is intended to refer to specific antineoplastic chemical agents or drugs that are "selectively" destructive to malignant cells and tissues, for example alkylating agents, antimetabolites including thymidylate synthase inhibitors, anthracyclines, anti- microtubule agents including plant alkaloids, taxanes, topoisomerase inhibitors, parp inhibitors and other antitumour agents. Selectively in this context is used loosely because of course many of these agents have serious side effects.
The preferred dose may be chosen by the practitioner, based on the nature of the cancer being treated.
Examples of alkylating agents, which may be employed in the method of the present disclosure include a platinum alkylating agent, nitrogen mustards, nitrosoureas, tetrazines, aziridines, platins and derivatives, and non-classical alkylating agents (such as procarbazine, altretamine and dacarbazine).
Examples of platinum containing chemotherapeutic agents (also referred to as platins], include cisplatin, carboplatin, oxaliplatin, satraplatin, picoplatin, nedaplatin, triplatin and lipoplatin (a liposomal version of cisplatin), in particular cisplatin, carboplatin and oxaliplatin. Others include eptaplatin, lobaplatin, miriplatin and dicycloplatin.
The dose for cisplatin ranges from about 20 to about 270 mg/m2 depending on the exact cancer. Often the dose is in the range about 70 to about 100mg/m2.
Nitrogen mustards include mechlorethamine, cyclophosphamide, melphalan, chlorambucil, ifosfamide and busulfan.
Nitrosoureas include V-Nitroso-N-methylurea (MNU), carmustine (BCNU), lomustine (CCNU] and semustine (MeCCNU), fotemustine and streptozotocin. Tetrazines include dacarbazine, mitozolomide and temozolomide.
Aziridines include thiotepa, mytomycin and diaziquone (AZQ).
Examples of antimetabolites, which may be employed in the method of the present disclosure, include anti-folates (for example methotrexate and pemetrexed), purine analogues (for example thiopurines, such as azathiopurine, mercaptopurine, thiopurine, fludarabine (including the phosphate form), pentostatin and cladribine), pyrimidine analogues (for example fluoropyrimidines, such as 5-fluorouracil (5-FU) and prodrugs thereof such as capecitabine [Xeloda®]), floxuridine, gemcitabine, cytarabine, decitabine, raltitrexed (tomudex) hydrochloride, cladribine and 6-azauracil.
Examples of anthracyclines, which may be employed in the method of the present disclosure, include daunorubicin (Daunomycin), daunorubicin (liposomal), doxorubicin (Adriamycin), doxorubicin [liposomal), epirubicin, idarubicin, valrubicin currently are used only to treat bladder cancer and mitoxantrone an anthracycline analog, in particular doxorubicin.
Examples of anti-microtubule agents, which may be employed in the method of the present disclosure, include include vinca alkaloids and taxanes.
Vinca alkaloids include completely natural chemicals for example vincristine and vinblastine and also semi-synthetic vinca alkaloids, for example vinorelbine, vindesine, and vinflunine
Taxanes include paclitaxel, docetaxel, abraxane, carbazitaxel and derivatives of thereof. Derivatives of taxanes as employed herein includes reformulations of taxanes like taxol, for example in a micelluar formulations, derivatives also include chemical derivatives wherein synthetic chemistry is employed to modify a starting material which is a taxane.
Topoisomerase inhibitors, which may be employed in a method of the present disclosure include type I topoisomerase inhibitors, type II topoisomerase inhibitors and type II topoisomerase poisons. Type I inhibitors include topotecan, irinotecan, indotecan and indimitecan. Type II inhibitors include genistein and ICR ng structure:
Figure imgf000019_0001
Type II poisons include amsacrine, etoposide, etoposide phosphate, teniposide and doxorubicin and fluoroquinolones.
In one embodiment the chemotherapeutic is a PARP inhibitor.
In one embodiment a combination of chemotherapeutic agents employed is, for example a platin and 5-FU or a prodrug thereof, for example cisplatin or oxaplatin and capecitabine or gemcitabine, such as FOLFOX.
In one embodiment the chemotherapy comprises a combination of chemotherapy agents, in particular cytotoxic chemotherapeutic agents.
In one embodiment the chemotherapy combination comprises a platin, such as cisplatin and fluorouracil or capecitabine.
In one embodiment the chemotherapy combination is capecitabine and oxaliplatin [XELOX].
In one embodiment the chemotherapy is a combination of folinic acid and 5-FU, optionally in combination with oxaliplatin (FOLFOX).
In one embodiment the chemotherapy is a combination of folinic acid, 5-FU and irinotecan (FOLFIRI), optionally in combination with oxaliplatin (FOLFIRINOX). The regimen, for example includes: irinotecan (180 mg/m2 IV over 90 minutes) concurrently with folinic acid (400 mg/m2 [or 2 x 250 mg/m2] IV over 120 minutes); followed by fluorouracil (400-500 mg/m2 IV bolus) then fluorouracil (2400-3000 mg/m2 intravenous infusion over 46 hours). This cycle is typically repeated every two weeks. The dosages shown above may vary from cycle to cycle.
In one embodiment the combination therapy employs a microtubule inhibitor, for example vincristine sulphate, epothilone A, N-[2-[(4-Hydroxyphenyl)amino]-3-pyridinyl]-4- methoxybenzenesulfonamide (ABT-751), a taxol derived chemotherapeutic agent, for example paclitaxel, abraxane, or docetaxel or a combination thereof. In one embodiment the combination therapy employs an mTor inhibitor. Examples of mTor inhibitors include: everolimus (RAD001), WYE-354, KU-0063794, papamycin (Sirolimus), Temsirolimus, Deforolimus(MK-8669), AZD8055 and BEZ235(NVP-BEZ235).
In one embodiment the combination therapy employs a MEK inhibitor. Examples of MEK inhibitors include: AS703026, CI-1040 (PD184352), AZD6244 (Selumetinib), PD318088, PD0325901, AZD8330, PD98059, U0126-EtOH, BIX 02189 or BIX 02188.
In one embodiment the combination therapy employs an AKT inhibitor. Examples of AKT inhibitors include: MK-2206 and AT7867.
In one embodiment the combination therapy employs an aurora kinase inhibitor. Examples of aurora kinase inhibitors include: Aurora A Inhibitor I, VX-680, AZD1152-HQPA (Barasertib], SNS- 314 Mesylate, PHA-680632, ZM-447439, CCT129202 and Hesperadin.
In one embodiment the chemotherapy combination employs a p38 inhibitor, for example as disclosed in W02010/038086, such as N-[4-({4-[3-(3-teri-Butyl-l-p-tolyl-ltf-pyrazol-5- yl)ureido]naphthalen-l-yloxy}methyl)pyridin-2-yl]-2-methoxyacetamide.
In one embodiment the combination therapy employs a Bcl-2 inhibitor. Examples of Bcl-2 inhibitors include: obatoclax mesylate, ABT-737, ABT-263(navitoclax) and TW-37.
In one embodiment the combination therapy comprises an antimetabolite such as capecitabine (xeloda), fludarabine phosphate, fludarabine (fludara), decitabine, raltitrexed (tomudex], gemcitabine hydrochloride and/or cladribine.
In one embodiment the combination therapy comprises ganciclovir, which may assist in controlling immune responses and/or tumour vasculation.
In one embodiment after combination therapy a monotherapy comprising a compound of formula (I), such as Varlitinib (as defined herein including doses described above) is employed, for example a maintenance therapy.
In one embodiment the combination therapy comprises a checkpoint inhibitor, for example selected from the group comprising:
Figure imgf000020_0001
r (ATP-
AZD-7762 a CHK1 and CHK2 inhibitor (with competitive CHK1 inhibitor with a Ki of 0.49 nM an IC50 of about 5 nM in a cell-free assay, with in a cell free assay. The molecule also VEGFR2, less potent activity against CAM, Yes, Fyn, Lyn, Aurora-A, FGFR3, Flt3, Fms (CSF-1R], Ret and Hck an Lck) Yes. It shows approximately 100 fold selectivity for CHK1 over CHK2)
Figure imgf000021_0001
Figure imgf000021_0002
V£fc0015S 22 fffcWissssi
hkl ICjp■ 0.017 μΜ Chkl !CM« 0.0076 μΜ
11= §.15 LE-0.37
Figure imgf000021_0003
V155422, V155991 and V158411 are CHKl and CHK2 inhibitors,
Figure imgf000021_0004
LY2606368 (prexasertib);
LY2603618 is a selective potent CHKl inhibitor MK
Figure imgf000022_0001
-8776 is a selective CHKl inhibitor
(with an IC50 of 3 nM in a cell-free assay. CHIR124 is CHKl inhibitor (is a novel and potent The molecule shows 500-fold selectivity inhibitor with IC50 of 0.3 nM in a cell-free assay. against CHK2) It shows 2,000-fold selectivity against CHK2 and
500 to 5000 fold less activity against CDK2/4 & Cdc2);
Figure imgf000022_0002
Debromohymenialdisine (SKF 108753) is a
natural product derived from sponges that in
Figure imgf000022_0003
TCS 2312 GNE-900 a CHKl inhibitor
CHKl inhibitor KI 0.38 nM, EC50 60 nM
Figure imgf000023_0001
CCT244747 a CHKl inhibitor
GDC-0575 (structure not given);
Figure imgf000023_0002
Figure imgf000024_0001
CP-466722 is a potent ATM inhibitor, which does not inhibit ATR. The molecule and PIKK
Figure imgf000024_0002
KU55933 is a potent and selective ATM inhibitor (vis-a-vie DNA-PK, Pi3K/Pi4K, ATR and mTOR) with an IC50 of 13 nM and KI of 2.2 mM
Figure imgf000024_0003
is an ATM inhibitor
KU60019 is an ATM inhibitor
Figure imgf000025_0001
N
Figure imgf000025_0002
U6027 is a potent ATR inhibitor
VE-821 is a potent and selective ATP competitive inhibitor of ATR with a Ki of 13 nM and an IC50 of 26 nM
itor
Figure imgf000025_0003
CCT241533 is a CHK2 inhibitor with about 63 fold selectivity over CHK1
Figure imgf000026_0001
AG-014699 (Rucaparib) a PARP-1 inhibitor (currently in Phase III trials for ovarian and pancreatic cancer) AZD-2281 (Olaparib) a PARP-1 and PARP-2
inhibitor (which is the first PARP inhibitor to
US FDA approval)
Figure imgf000027_0001
inhibitor (currently in Phase III trials for ovarian,
breast cancer and Ewing sarcoma) BMN-673 (Talazoparib) a PARP-1 and
PAR-2 inhibitor (which is currently in Phase III trials for ovarian, breast and other solid cancers)
Figure imgf000027_0002
(which is currently being evaluated in Phase III CEP-9722 a PARP-1 and PARP-2 inhibitor studies for breast, pancreatic, non-small cell
Figure imgf000027_0003
H
BGP-15 a PARP inhibitor (which has been E7016 (previously known as GPI-21016) a shown to protect against ischemia-reperfusion PARP inhibitor (undergoing Phase I trials in injury) combination with temozolomide for
advanced solid tumours and gliomas)
Figure imgf000028_0001
inhibitor
INO-1001 (3-aminobenzamide) a potent inhibitor of PARP [with IC50 of <50 nM in CHO cells and a mediator of oxidant-induced myocyte dysfunction during
reperfusion)
Figure imgf000028_0002
BGB-290 a PARP-1 and PARP-2 inhibitor (structure not shown); MP-124 a PARP-1 inhibitor (structure not shown);
or a pharmaceutically acceptable salt or solvate of any one of the same.
The following articles disclose certain checkpoint inhibitors: Drugs Fut 2003, 28(9): 881 Cell cycle inhibitors for the treatment of cancer Kong, N., Fotouhi, N., Wovkulich, P.M., Roberts, J; Novel pyrrole derivatives as selective CHK1 inhibitors: design, regioselective synthesis and molecular modelling Med. Chem. Commun., 2015,6, 852-859; PLoS One 2010; 5(8): el2214. Binding of protein kinase inhibitors to synapsin I inferred from pair-wise binding site similarity measurements. In one embodiment the checkpoint inhibitor is a PARP inhibitor. Examples of PARP inhibitors are disclosed in US7,449,464 and US8,071,623. The compounds disclosed in this paragraph are incorporated herein by reference. In one embodiment the checkpoint inhibitor is an antibody or binding fragment specific to a checkpoint protein, in particular one disclosed herein.
In one embodiment the checkpoint kinase inhibitor is independently selected from: 3- [(Aminocarbonyl)amino]-5-(3-fluorophenyl)-N-(3S)-3-piperidinyl-2-thiophenecarboxamide hydrochloride; (3R,4S)-4-[[2-(5-Fluoro-2-hydroxyphenyl)-6,7-dimethoxy-4-quinazolinyl]amino]- a,a-dimethyl-3-pyrrolidinemethanol dihydrochloride; 4,4'-diacetyldiphenylurea bis(guanylhydrazone) ditosylate; 9-Hydroxy-4-phenyl-pyrrolo[3,4-c]carbazole-l,3(2H,6H)-dione; (R)- -Amino-N-[5,6-dihydro-2-(l-methyl-lH-pyrazol-4-yl)-6-oxo-lH-pyrrolo[4,3,2- ef] [2,3]benzodiazepin-8-yl]-cyclohexaneacetamide; 9,10,11,12-Tetrahydro- 9,12-epoxy-lH- diindolo[l,2,3-fg:3',2',l'-kl]pyrrolo[3,4-i] [l,6]benzodiazocine-l,3(2H)-dione; 4'-[5-[[3- [(Cyclopropylamino)methyl]phenyl]amino]-lH-pyrazol-3-yl]-[l,l'-biphenyl]-2,4-diol; and (R)-5- ((4-((Morpholin-2-ylmethyl)amino)-5-(trifluoromethyl)pyridin-2-yl)amino)pyrazine-2- carbonitrile (CCT245737).
In one embodiment the check point inhibitor is prexasertib. Checkpoint kinase inhibitor as employed herein refers to an inhibitor that reduces or eliminates the biological activity of a cell regulatory checkpoint kinase 1 and/or 2. Cells that suffer DNA damage activate the checkpoint kinases CHKl and CHK2, which signal to initiate the DNA repair processes, limit cell-cycle progression and prevent cell replication, until the damaged DNA is repaired.
"Comprising" in the context of the present specification is intended to mean "including". Where technically appropriate, embodiments of the invention may be combined.
Embodiments are described herein as comprising certain features/elements. The disclosure also extends to separate embodiments consisting or consisting essentially of said features/elements.
Technical references such as patents and applications are incorporated herein by reference.
Any embodiments specifically and explicitly recited herein may form the basis of a disclaimer either alone or in combination with one or more further embodiments.
The present specification claims priority from SG10201708782U filed 25 October 2017, SG10201802791Q filed 3 April 2018, SG10201807632Q filed 5 September 2018, and SG10201808393X filed 26 September 2018, each incorporated herein by reference. The priority documents may be employed as basis for corrections to errors in the present specification.
The invention will now be described with reference to the following examples, which are merely illustrative and should not in any way be construed as limiting the scope of the present invention.
BRIEF SUMMARY OF THE FIGURES
Figure 1 ErbB family expression and its phosphorylation in representative HCC PDXs and the anti-tumour effects of Varlitinib on 3 PDXs with activated ErbB2/3. (A and B) Western blot shows the expression of EGFR, ErbB2, ErbB3, and its phosphorylation in representative HCC PDXs. (C-E) Growth curves and tumour weights of three high p-ErbB2/p-ErbB3 expressing PDXs, (C) HCCOl-0708, (D) HCC07-0409, and (E) HCC29-0909A, treated with three different doses of Varlitinib, 25, 50, and lOOmg/kg BID. Treatment was started when the tumours reached the size of approximately 120-150 mm3. The tumour volumes were measured every 2- 3 days and the tumour weights were measured at the end of the experiments. Shown are mean and SD [n=8-10).
Figure 2 Dose- and time-dependent ErbB family pathway inhibition by Varlitinib in
HCC29-0909A PDX model. Mice bearing HCC29-0909A PDX were treated with vehicle control, 25, 50, or lOOmg/kg BID. Tumours were collected at day 2 (A) and day 14 (B) post-Varlitinib treatment. Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis. Blots were incubated with indicated antibodies. Representative blots were shown.
Figure 3 Effects of Varlitinib on tumour cell proliferation, tumour cell death, and vessel normalisation in HCC29-0909A PDX model. Mice bearing HCC29-0909A PDX were treated with vehicle control and lOOmg/kg BID. Tumours collected at day 14 were processed for paraffin [for cleaved PARP and p-histone 3 SerlO staining] or Tissue-Tek embedding (for CD31 staining). (A) Representative images of tumour sections from vehicle-treated and Varlitinib-treated mice stained for p-Histone H3 (SerlO), cleaved PARP, and CD31. (B) Representative images of tumour sections from vehicle-treated (top) and Varlitinib-treated (bottom} mice perfused with biotinylated Lycopersicon Esculentum (Tomato) lectin and pimonidazole hydrochloride, followed by immunohistochemical staining. Representative photographs were shown.
Figure 4 Transcriptome analysis of Varlitinib-treated HCC29-0909A and HCCOl-0708
HCC PDX models. (A) The area-proportional Venn diagram analysis showing the common activated genes (left) and the common repressed genes (right in lOOmg/kg BID and 50mg/kg BID Varlitinib-treated HCC29-0909A model for 14 days. (B) The heatmap showing the common Varlitinib-dysregulated 2331 genes in HCC29-0909A model. (C) The area-proportional Venn diagram analysis showing the common activated genes (left) and the common repressed genes (right) in lOOmg/kg BID and 50mg/kg BID Varlitinib-treated HCCOl-0708 model. (D) The heatmap showing the common Varlitinib-dysregulated 457 genes in HCCOl-0708 model. (E) The 3-way area-proportional Venn diagram analysis showing the common activated genes (left) and the common repressed genes (right) in lOOmg/kg BID and 50mg/kg BID Varlitinib-treated HCC29-0909A model as well as lOOmg/kg BID Varlitinib-treated HCCOl-0708 model.
Figure 5 Dose-dependent β-catenin pathway inhibition and membrane translocation of β-catenin by Varlitinib in the tested HCC29-0909A and HCCOl-0708 PDX models. Mice bearing HCC29-0909A (A) or HCCOl-0708 (B) with three different doses of Varlitinib, 25, 50, and lOOmg/kg BID for indicated time. Treatment was started when the tumours reached the size of approximately 120-150 mm3. Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis for the β-catenin, its related signalling molecules, and downstream targets of β-catenin. (C) Representative images of tumour sections from vehicle-treated (top) and Varlitinib-treated (bottom) HCC29- 0909A, HCC07-0409, and HCCOl-0708 PDXs stained for β-catenin.
Figure 6 Hypothetic model of the Varlitinib-mediated tumour regression and vessel normalisation in HCC.
Figure 7 ErbB family expression in additional 28 hepatocellular carcinoma (HCC) patient-derived xenografts (PDXs) and the anti-tumour effects of Varlitinib on the high ErbB2 /3-expressing PDXs. (A and B) Western blot shows the expression of EGFR, EbB2, ErbB3, and its phosphorylation in 28 HCC PDXs. (C-D) Growth curves and tumour weights of HCC21-0208 (C) and HCC16-1014 (D) treated with Varlitinib at lOOmg/kg BID. Treatment was started when the tumours reached the size of approximately 120-150 mm3. The tumour volumes were measured every 2-3 days and the tumour weights were measured at the end of the experiments. Shown are mean and SD (n=5).
Figure 8 Dose- and time-dependent ErbB family pathway inhibition by Varlitinib in
HCCOl-0708 PDX model. Mice bearing HCCOl-0708 PDX were treated with vehicle control, 25, 50, or lOOmg/kg BID. Tumours were collected at day 2 (A) and day 11 (B) post-Varlitinib treatment. Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis. Blots were incubated with indicated antibodies. Representative blots were shown.
Dose-dependent inhibition of tumour cell proliferation, induction of apoptosis and formation of capillary-like blood vessels by Varlitinib in HCC29-0909A PDX model. Mice bearing HCC29-0909A PDX were treated with vehicle control, 25, 50, or lOOmg/kg BID. Tumours collected at day 14 were processed for paraffin [for cleaved PARP and p-histone 3 SerlO staining] or Tissue-Tek embedding (for CD31 staining). Representative images shown of tumour sections from vehicle-treated and Varlitinib-treated (25, 50, or lOOmg/kg BID) mice stained for p-Histone H3 (SerlO) (upper), cleaved PARP (middle), and CD31 (bottom).
Tumour regression by Varlitinib treatment in three other PDX models, HCC01- 0708, HCC07-0409, and HCC21-0208. Mice bearing indicated PDX were treated with vehicle control or lOOmg/kg BID. Tumours collected at indicated day were processed for paraffin [for cleaved PARP and p-histone 3 SerlO staining] or Tissue- Tek embedding (for CD31 staining). Representative images of tumour sections from vehicle-treated and Varlitinib-treated mice stained for p-Histone H3 (SerlO) (upper), cleaved PARP (middle), and CD31 (bottom).
Effect of Varlitinib on vessel normalisation in HCC29-0909A PDX model. Mice bearing HCC29-0909A PDX were treated with vehicle control and lOOmg/kg BID. Tumours collected at day 14 were processed for immunohistochemistry. Representative images of tumour sections from vehicle-treated and Varlitinib- treated mice perfused with biotinylated Lycopersicon Esculentum (Tomato) followed by immunohistochemical staining. Representative photographs were shown. Magnification 100X (Top), 200X (Middle) and 400X (Bottom).
Identification of Varlitinib-resistant gene signature by global gene expression analysis. (A) Heat-map showing the identified differentially expressed genes from the quantile normalised data of four HCC PDXs, HCC21-0208, HCCOl-0708, HCC07- 0409, and HCC29-0909A with high stringent cut-off threshold of >2 and <-2 fold- change and FDR adjusted p-value < 0.0001. (B) The expression of NRG1 in the four analysed PDXs indicated by normalised probe intensity. Order of data points from left to right are HCC21-0208, HCCOl-0708, HCC07-0409 and HCC29-0909A. See also Table 12. (C) The expression of ErbB family in the four analysed PDXs indicated by normalised probe intensity. For each gene, the order of the data points from left to right are HCC21-0208, HCCOl-0708, HCC07-0409 and HCC29-0909A.
Identification of β-catenin mutation and the constitutively active β-catenin pathways in the Varlitinib-sensitive PDXs. (A) The expression of β-catenin and its downstream targets in the four analysed PDXs indicated by normalised probe intensity. (B) The expression of Wnt/TGF -related targets in the four analysed PDXs indicated by normalised probe intensity. For each gene, the order of the data points from left to right are HCC21-0208, HCCOl-0708, HCC07-0409 and HCC29-0909A. ** indicates p-value < 0.01, *** indicates p-value < 0.001, **** indicates p-value < 0.0001. See also Table 13.
Identification of Varlitinib-potency gene signature by transcriptome analysis.
Heat-map showing the identified differentially expressed genes from the normalised RNA-Seq data of HCCOl-0708 and HCC29-0909A with cut-off threshold of >2 and <- 2 fold-change and FDR adjusted p-value < 0.05.
Genetic alterations of CTNNB1, EGFR, ERBB2, ERBB3, and ERBB4 in TCGA HCC data. Oncoprint displaying the genetic alterations of the indicated genes in the TCGA HCC dataset (n=360) (A] and the subsets of HCC which harbour any types of β- catenin mutation (n-96) (B). Colour coding indicates types of genetic alterations: dark grey, homozygous amplification; light grey, heterozygous amplification; grey, heterozygous deletion; and black, mutation. Left, genetic alteration percentage, shows dose-dependent tumour volume growth inhibition in HCC patient derived xenograft model of HCC29-0909A after administration of 25mg/kg BID, 50 mg/kg BID or lOOmg/kg BID of varlitinib.
shows dose-dependent tumour volume growth inhibition in HCC patient derived xenograft model of HCCOl-0708 after administration of 25mg/kg BID, 50 mg/kg BID or lOOmg/kg BID of varlitinib.
shows the results of an in vitro experiment to investigate the induction of apoptosis by varlitinib in HCC cell lines, including sorafenib-resistant cell lines,
shows the apoptosis (AnnexinV) profile for PLC/PRF/5 cells after 48 hour culture in the presence of varlitinib.
Frequent upregulation of ErbB3 in HCC and in vitro effect of Varlitinib in high ErbB3-expressing liver cancer cell lines. (A) ERBB3 expression in various publicly available HCC dataset. T: HCC tumours, NT: non-cancerous tissues; *p < 0.05, ***p < 0.001, ****p < 0.0001, paired or unpaired t test was used if data passed normality test, Wilcoxon matched paired signed rank test or unpaired Mann-Whitney test was used if data did not pass normality test. (B) Dose-dependent inhibition of Varlitinib in representatives of ERBB3\iig and ERBB3Low liver cancer cell lines. PLC/PRF/5, Hep3B, HepG2/C3A, and Huh7 are ERBB3H\gh cell lines and SNU182, SNU423, SNU449, and SNU475 are ERBB3Low cell lines. Different doses of Varlitinib in normal culture medium with 1% DMSO were added into the 6-well plates seeded with different cell lines. The cell lines were cultured for 2-4 weeks, followed by fixation and Giemsa staining. (C) Inhibition of ErbB receptor family and their downstream signalling molecules by Varlitinib. The £7?Bi?J?High PLC/PRF/5 cell line was treated with 3 μΜ of Varlitinib for indicated time. Blots were incubated with indicated antibodies. a-Tubulin was used as equal loading marker. See also Figure 25.
Identification of Varlitinib responder and non-responder PDXs and the importance of ErbB2/3 dependence in HCC. (A) In vivo effect of Varlitinib in treated PDXs. The 3 Varlitinib responders, HCCOl-0708, HCC07-0409, and HCC29- 0909A models, were treated with vehicle control, 25, 50, and lOOmg/kg BID (in the red box), whereas the 3 Varlitinib non-responders, HCC21-0208, HCC16-1014, and HCC26-0808A, were treated with vehicle control and lOOmg/kg BID [in the yellow box). Tumour volume was measured every 2-3 days. Data represent mean ± SEM (n = 5-10). **p < 0.01, ****p < 0.0001, ns, not significant. Comparisons with vehicle control-treated PDXs (unpaired t test). (B) Western blot analysis of 3 responders and 2 non-responders. a-Tubulin was used as equal loading marker. (C) Gene expression analysis oiNRGl, EGFR, ERBB2, and ERBB3 between responders and non-responders. Data represent mean ±SEM (responders n=9 and non-responders n=27), *p < 0.05, ****p < 0.0001 (unpaired t test). (D) Survival analysis of ERBB2, ERBB3, and NRG1 in HCC. Cutoff finder (Budczies et al., 2012) was used to identify the best cutoff gene expression value for ERBB2 (left), ERBB3 (middle), and NRG1 (right) in the Singapore HCC dataset[44] (n=48). HR represents hazard ratio. See also Figure 26.
Figure 22 (A) Enrichment plots of the two highly enriched and significant gene sets in Varlitinib responders and one in Varlitinib non-responders. (B) Expression of the reported G6 signature-related genes in Varlitinib responders and nonresponders. Data represent mean ± SEM (responders n=9 and non-responders n=27), **p <0.01, ****p < 0.0001 (unpaired t test). (C) Correlation analyses between ERBB3 and G6-related genes [LEF1 and AXIN2) and between ERBB3 and IGF1R. Pearson correlation analysis is used when both data passed normality test, whereas Spearman correlation analysis is used when one of the data did not pass normality test. See also Figure 27.
Figure 23 Transcriptomic analysis identifies dose-dependant responses and the potential inhibition mechanisms in responder PDXs. (A) Three-way area proportional Venn diagram analysis of differential expressed gene (DEG) analysis from RNA-Seq analysed Varlitinib-treated PDXs. (B) The normalised read counts were analysed based on gene sets of glycolysis and gluconeogenesis (left) and hypoxia (right). Data from control (indicated in blue box), 50mg/kg (indicated in red box), and lOOmg/kg BID (indicated in green box) ofVarlitinib-treated HCC29-0909A are shown. Data from control (indicated in blue box) and lOOmg/kg BID (indicated in green box) of Varlitinib-treated HCC01-0708 are shown. (C) Ingenuity Pathway Analysis (IPA) predicted HIF1A as upstream regulator in the treated PDXs. (D) Repressed expression of HIF1A in treated PDXs. Data represent mean ± SEM (n = 3). **p < 0.01. Comparisons with vehicle control-treated PDXs (unpaired t test). (E) Repressed expression of LGR5 and upregulated CDH1 in treated PDXs. Data represent mean ± SEM (n = 3). **p < 0.01, ****p < 0.0001, ns, not significant. Comparisons with vehicle control-treated PDXs (unpaired t test). See also Figure 28.
Figure 24 Varlitinib-mediated β-catenin pathway inhibition in the treated responder
PDXs. (A-B) Dose-dependent β-catenin pathway inhibition by Varlitinib. Western blot analysis of the protein lysates collected on day 14 post-treatment of Varlitinib in HCC29-0909A (A) and on day 11 post-treatment of Varlitinib in HCCOl-0708 (B). Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis. Blots were incubated with indicated antibodies. a-Tubulin was used as equal loading marker. Representative blots were shown. See also Figure 30. Analysis of ERBB3m and ERBB3Low liver cancer cell lines. (A) Identification of ERBB3m and ERBBS^ liver cancer cell lines. Expression of ERBB3 was compared among 27 liver cancer cell lines in Cancer Cell Line Encyclopedia (CCLE) dataset. Median expression of ERBB3 in these 27 cell lines was used as as cut-off for ERBB3m and ERBB3h0W cell lines. *"*p < 0.0001, unpaired t test. (B) Expression of phospho- and total ErbB family receptors in selected ERBB3m and ERBB3Lov! cell lines. Blots were incubated with indicated antibodies. a-Tubulin was used as equal loading marker. (C) 72-hour cytotoxicity test of Varlitinib in ERBB3m cell lines. Different concentrations of Varlitinib in 1%DMS0 was added into the cells for 72 hours. CyQUANT NF Cell Proliferation Assay Kit was used to measure the absorbance and GraphPad Prism version 7.00 was used to determine the cytotoxicity.
Varlitinib treatment in HCC PDXs. (A) ErbB family expression in 56 HCC (PDXs. Western blot shows the expression of EGFR, ErbB2, ErbB3, and its phosphorylation in 56 HCC PDXs. Blots were incubated with indicated antibodies. α-Tubulin was used as equal loading marker. (B) In vivo effect of Varlitinib in treated non-responder PDXs. Additional 11 HCC PDXs, HCCOl-0909, HCC06-0606, HCC09-0913, HCC13- 0109, HCC13-0212, HCC15-0114, HCC17-0211, HCC25-0705A, HCC26-0808B, HCC29-1104, and HCC30-0805B models, were treated with vehicle control and lOOmg/kg BID Tumour volume was measured every 2-3 days. Data represent mean ± SEM (n = 5-10). ns, not significant. Comparisons with vehicle control-treated PDXs (unpaired t test). (C) Principal component analysis ofquantile normalised gene expression dataset Smaller circle indicates Varlitinib responders, larger circle circle indicates non-responders. Triplicate samples of 3 responder PDXs and 9 non- responder PDXs were analysed by Affymetrix Human Genome U133 Plus 2.0 microarray. (D) Heatmap of the DEG.
Analysis of responders and non-responders. Gene expression analysis of TGFBR1, TGFBR2, TEAD1, IGF2, IGF1R, IRS2, NOTCH1, N0TCH2, and JAG1 between responders and non-responders. Triplicate samples of 3 responder PDXs and 9 non-responder PDXs were analysed by Affymetrix Human Genome U133 Plus 2.0 microarray. Probe intensities in each sample were quantile normalised. Data represent mean ± SEM (responders n=9 and non-responders n=27), *p < 0.05, **p < 0.01, "*p < 0.001, "**p < 0.0001 (unpaired t test).
Transcriptome analysis of Varlitinib treatment in HCC29-0909A, HCCOl-0708, and HCC16-1014. (A-B) Area proportional Venn diagram analysis of differential expressed gene (DEG) analysis from RNA-Seq analysed Varlitinib-treated responder PDXs, HCC29-0909A (A) and HCCOl-0708 (B). (C-E) Downregulation of pro- angiogeneic factors by Varlitinib in treated responder PDXs. (C) Inhibition of EPO, PDGFA, NRP2, BMP2 in both Varlitinib-treated HCC29-0909A and HCCOl-0708. (D) Inhibition of ANG, PGF, VEGFB in Varlitinib-treated HCC29-0909A. (E) Inhibition of PDGFC in Varlitinib-treated HCCOl-0708. Data represent mean ± SEM (n = 3). *p < 0.05, "p < 0.01, "*p < 0.001, **"p < 0.0001. Comparisons with vehicle control-treated PDXs (unpaired t test). (F-H) Downregulation of β-catenin- and YAPl-related genes by Varlitinib in treated responder PDXs. (F) Inhibition of YAP1, IRX3, and MYC in both Varlitinib-treated HCC29-0909A and HCCOl-0708. (G) Inhibition of LEF1 in Varlitinib-treated HCC29-0909A. (H) Inhibition of SOX9, HEY1, CTGF, CYR61 in Varlitinib-treated HCCOl-0708. Data represent mean ± SEM (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001. Comparisons with vehicle control-treated PDXs (unpaired t test]. (I) Pathways enriched in HCCOl-0708 when compared to HCC29-0909A by Ingenuity Pathway Analysis (IPA). (J) CTNNB1 expression in control of HCC29- 0909A, control of HCCOl-0708, and lOOmg/kg Varlitinib-treated HCCOl-0708. "*p < 0.001. Comparison with vehicle control-treated HCCOl-0708 PDX (unpaired t test]. (K) Heatmap of the inhibition of embryonic stem cell core gene set in Varlitinib- treated HCCOl-0708. Control of HCC29-0909A, Control of HCC16-1014, and Varlitinib-treated HCC16-1014 are included for the comparison. Data from control (indicated in dark grey box) of HCC29-0909A, from control (indicated in dark grey box) and lOOmg/kg BID (indicated in light grey box) of Varlitinib-treated HCCOl- 0708, and from control (indicated in dark grey box) and 75mg/kg BID (indicated in light grey box) of Varlitinib-treated HCC16-1014 are shown. (L & M) Heatmap of the inhibition of YAP signalling in Varlitinib-treated PDXs. YAP conserved signature (L) and GO Hippo signalling (M) gene sets were used. Data from control (indicated in dark grey box), 50mg/kg (indicated in black box), and lOOmg/kg BID (indicated in light grey box) of Varlitinib-treated HCC29-0909A, from control (indicated in dark grey box) and lOOmg/kg BID (indicated in light grey box) of Varlitinib-treated HCCOl-0708, and from control (indicated in dark grey box) and 75mg/kg BID (indicated in light grey box) of Varlitinib-treated HCC16-1014 are shown.
Figure 29 Dose-dependent ErbB family pathway inhibition by Varlitinib in responder
PDX models. Western blot analysis of the protein lysates collected on day 14 post- treatment of lOOmg/kg QD Varlitinib in HCC07-0409 PDX model. Two tumours from each condition were lysed and equal amount of the protein lysates were used. Blots were incubated with indicated antibodies. a-Tubulin was used as equal loading marker. Representative blots were shown.
Figure 30 Varlitinib-mediated β-catenin pathway inhibition in the treated responder
PDXs. (A-B) β-catenin pathway inhibition by Varlitinib. Western blot analysis of the protein lysates collected on day 14 post-treatment of Varlitinib in HCC29-0909A (A) and on day 11 post-treatment of Varlitinib in HCC07-0409 (B). Two tumours from each condition were lysed and equal amount of the protein lysates were used for Western blot analysis. Blots were incubated with indicated antibodies. α-Tubulin was used as equal loading marker. Representative blots were shown.
Figure 31 shows hypothetical model of the Varlitinib-mediated tumour growth inhibition and vessel normalisation in HCC.
EXAMPLES
Example 1
This study demonstrates Varlitinib-mediated tumour regression and vascular normalisation in ErbB-dependent and β-catenin mutated hepatocellular carcinoma. Materials and Methods
Reagents
Antibodies against EGFR, ErbB2, ErbB3, Akt, β-catenin, Axin2, Axinl, c-met, survivin, N- cadherin, E-cadherin, Dvl3, Dvl2, Cdc25C, p27, p21, E2F1, Rb, Cleaved caspase 3, cleaved caspase 7, cleaved PARP, and phosphorylation-specific antibodies against Akt Ser473, LRP6 Serl490, RanBP3 Ser58, β-catenin Tyrl42, β-catenin Tyr654, Non-p (Active) β-catenin Ser33/Ser37/Thr41, EGFRTyrl068, ErbB3 Tyrl289, ErbB2 Tyrl221/1222, Rb Ser780, p90RSK Thr359/363, mTOR Ser2448, p70S6K Thr421/Ser424, S6R Ser235/236, 4EBP1 Thr70, Cdc2 Tyrl5, c-Jun Thr73, Histone 3 SerlO, Cdk Thrl4/Serl5, and ERK1/2 were obtained from Cell Signalling Technology, Beverly, MA. The antibodies against ERK1/2 and a-tubulin were from Santa Cruz Biotechnology Inc, Santa Cruz, CA, USA. Anti-mouse CD31 antibody was from BioLegend, San Diego, CA, USA. Varlitinib was obtained from ASLAN Pharmaceuticals Ltd, Singapore.
Xenograft models
This study received ethics board approval at the SingHealth. All mice were maintained according to the Guide for the Care and Use of Laboratory Animals published by the National
Institutes of Health, USA.
HCC tumours have previously been used to create patient-derived xenograft models(Huynh et al., 2006) of which the 56 models were used to screen for the expression of EGFR, ErbB2, ErbB3, p-EGFR, p-ErbB2 and p-ErbB3 by Western blot analysis.
Three phospho-ErbB2/ErbB3-positive lines (HCC29-0909A, and HCC07-0409, HCCOl-0708 and two phospho-ErbB2/ErbB3-negative models (HCC21-0208 and HCC16-1014) as determined by
Western blot analysis were used to establish tumours in male SCID mice (In Vivo, Singapore) aged 9 to 10 weeks.
For dose-response experiment, mice bearing the HCC29-0909A, HCC07-0409, and HCC01- 0708 xenografts (8-10 mice per group) were orally given vehicle [3 parts Polyethylene glycol) average Mn 300 (PEG, Aldrich Cat#202371) and 7 parts 30% w/v Research Grade Captisol (Ligand Pharmaceuticals, San Diego, CA) or 3 doses of Varlitinib (25 or 50, 75 and 100 mg/kg BID) for indicated days. Treatment started when the tumours reached the size of approximately 120-150 mm3.
To investigate the antitumor effects of Varlitinib, mice bearing HCC21-0208 and HCC16-
1014 tumours were orally administered either vehicle (n=5) or 100 mg/kg BID Varlitinib (n=5) for indicated days. Treatment started when the tumours reached the size of approximately 120-150 mm3. Tumour growth was monitored and tumour volume was calculated as described (Huynh et al., 2006, 2010). At the end of the study, the mice were sacrificed with body and tumour weights recorded and the tumours harvested for later analyses.
Vessel perfusion study
Each mouse bearing tumour xenografts received intravenously with 100 mg of Biotinylated Lycopersicon Esculentum (Tomato) Lectin (VectorLabs #B-1175) prepared in 100 μΐ of 0.9% NaCl. The tumours were harvested 10 minutes after lectin perfusion, fixed in 10 % formalin buffer solution, embedded in paraffin. Five μηι sections were prepared. After blocking endogenous peroxidase activity and nonspecific staining, the sections were incubated 1 hour at room temperature with Streptavidin Peroxidase (Lab Vision Corporation, Fremont, CA). To visualize productive microvessels, immunohistochemistry was performed using the streptavidin-biotin peroxidase complex method, according to the manufacturer's instructions (Lab Vision, Fremont, Calif.). For the quantification of mean microvessel density in sections, 10 random 0.159 mm2 fields at lOOx magnification were captured for each tumour.
To determine the extent of hypoxia in tumour tissues, mice bearing indicated tumours were treated with vehicle, or 100 mg/kg Varlitinib BID for indicated days. Mice were i.p. injected with pimonidazole hydrochloride (60 mg/kg, 2.5 μΐ/g of mouse body weight) 1 hour before tumours harvested. Hypoxic regions of tumour were identified by staining the sections with Hypoxyprobe plus Kit HP2 (Chemicon) as described by the manufacturer.
Western blot analysis
3-4 independent tumours from vehicle and drug-treated mice were homogenized separately in lysis buffer and 80 μg of proteins per sample were analysed by Western blot analysis as described (Huynh et al., 2006, 2010).
Immunohistochemistry
5-μηι sections were stained with CD31, p-Histone 3 SerlO and cleaved PARP antibodies to assess microvessel density, cell proliferation, and apoptosis respectively, as described previously (Huynh et al., 2006, 2010).
Transcriptome sequencing analysis, global gene expression analysis, and bioinformatic analysis
A total of 200 ng of QC-qualified total RNA was used for TruSeq mRNA library prep, followed by 150bp paired-end sequencing in HiSeq4000 platform by BGI HK, Ltd. Since the RNA samples would be contaminated with mouse cells due to the growth of PDXs in immunocompromised SCID mouse, an extra filtering step to remove mouse component was introduced. The raw sequencing reads were aligned to hgl9_mml0 mixed reference by Burrows-Wheeler Aligner (BWA) and the read pairs were removed as long as any of the paired reads mapped to mmlO chromosome and/or rRNA sequence to filter mouse contamination and rRNA reads, respectively. The detected mouse read rates at genome and gene levels ranged from 0.68% to 11.2% and 0.21% to 5.12%, respectively. In average, 74 million to 90 million clean reads per sample were obtained. The BAM files were then uploaded to Partek Flow for further analysis. Aligned reads were quantify to Partek E/M annotation model in Partek Flow (Partek Inc. St. Louis, MO, USA), followed by total count and add 0.0001 normalisation of gene counts and GSA differential expression detection from the comparisons of 50mg/kg varlitinib treatment vs control and lOOmg/kg varlitinib treatment vs control in both models using the cut-off threshold of >2 and -2 fold-change and FDR adjusted p-value < 0.05, followed by heatmap generation. The global gene expression analysis was done by quantile normalized data from Affymetrix GeneChip Human Genome U133 Plus 2.0 Array with high stringent cut-off threshold of >2 and -2 fold-change and FDR adjusted p-value < 0.0001. The identified gene lists were then compared using BioVenn online tool for the area-proportional Venn diagram analysis (Hulsen et al., 2008). The KEGG Pathway Enrichment analysis was carried out in Partek Genomics Suite (Partek Inc. St. Louis, MO, USA) with the cut-off threshold as FDR adjusted p-value <0.05. The differentially expressed gene lists were analysed through the use of Ingenuity Pathway Analysis (IPA)(Kramer et al., 2014) (QIAGEN Inc., https://www.qiagenbioinformatics om/products/ingenuitypathway-analysis TCGA data access and Oncoprint analysis ware carried out with the use of cBioPortal (Cerami et al., 2012; Gao et al., 2013).
Results
Varlitinib suppresses tumour growth through dose-dependent inhibition of ErbB family pathways in HCC PDXs
Previous clinical trials have shown that targeting one or two ErbB members concurrently had subtle clinical benefit. To have a meaningful clinical benefit, inhibition of all ErbB family simultaneously may be needed. Varlitinib is a best-in-class reversible pan-HER inhibitor with a good clinical safety profile.
To investigate the anti-tumour activity of Varlitinib, we first determine the protein expression and activation of EGFR, ErbB2, and ErbB3 in 56 HCC PDXs (Huynh et al., 2006) by Western blot analysis. As shown in Figure 1A and IB and Figure 7A and 7B, 91% (51/56), 82% (46/56) and 93% (52/56) of the tested PDX expressed high levels of EGFR, ErbB2, and ErbB3, respectively. ErbB4 was undetectable in all PDX models tested. Low but detectable levels of phosphorylated EGFR at Tyr 1068 were detected in 89% (50/56) of the PDX models. Phospho-ErbB2 at Tyrl221/1222 and p-ErbB3 at Tyrl289 were detected in 10.7% (6/56) and 37.5% (21/56) of PDX models tested, respectively. The data suggests that ErbB2 and ErbB3 were activated in small subset of HCC. Since HCCOl-0708, HCC07-0409, and HCC29-0909A PDX models expressed EGFR, ErbB2, and ErbB3 with the highest levels of activated ErbB2 and ErbB3 among all the tested PDXs, as indicated by the levels of phosphorylation, they were chosen for further Varlitinib testing (Figure 1A). HCC21-0208 was selected to serve as negative control because it has the very low level of ErbB2 and undetectable level of ErbB3 (Figure IB). Mice bearing HCCOl-0708, HCC07-0409, and HCC29- 0909A tumours were treated with three different doses of Varlitinib (25, 50, and lOOmg/kg BID). Figure 1C-1E showed that Varlitinib inhibited tumour growth in a dose-dependent manner. This correlated well with tumour weight at harvest. HCCOl-0708 is a fast-growing PDX that tolerated Varlitinib at low dosing, but the tumour growth and weight of that were significantly suppressed at the highest tested dosing, suggesting that higher dosing of Varlitinib is able to overcome the drug tolerance (Figure 1C). Comparatively, HCC07-0409 and HCC29-0909A were more sensitive to Varlitinib and showed better dose-dependent inhibition effect (Figure ID and IE). Statistically significant growth inhibition is also observed among all different dosing in these two models, meaning that these models are sensitive and respond to Varlitinib very well. The latter one was dramatically suppressed even at 50mg/kg BID, indicating that Varlitinib is highly potent on HCC29- 0909A model, which has the highest level of p-ErbB3 among all PDXs (Figure IE). On the other hand, Varlitinib at the dose of 100 mg/kg BID had no significant anti-tumour activity in HCC21-0208, which expresses undetectable level of ErbB3. Similarly, the growth of HCC16-1014 was unaffected by Varlitinib at the dose of 75mg/kg BID. Correlative study reveals that the levels of phosphorylated (activated) ErbB2 and ErbB3 but not total ErbB are indicative of Varlitinib sensitivity (Figure 7C and 7D).
To study the effects of Varlitinib on the ErbB family and their downstream targets, Western blot was carried out. The protein lysate was prepared at equal concentrations of the treated tumour samples collected on day 2-post treatment and/or the last day of the experiments for Western blot with the use of a panel of antibodies detecting the ErbB members and their downstream targets. In HCC29-0909A model, Varlitinib effectively inhibited p-EGFR and p-ErbB2, and p-ErbB3 on day 2 and 14 post-drug treatment (Figure 2). The suppression of phosphorylated and total ErbB3 became more significant on day 14. While p27 level was elevated, p-AKT, p-mTOR, p-p90RSK, p-S6R, p-Cdc2, and p-Rb were decreased dose-dependently. At the later time point, p-ERKl /2, p-p70S6K, p-4EBPl, CDC25C, and E2F1 were significantly decreased, while p21 was increased. These findings suggested the dose- and time-dependent inhibition of ErbB downstream targets in this model. Furthermore, in the highest dosing of Varlitinib-treated HCC07-0409 model (Figure 29), inhibition of p-EGFR, p- ErbB2, p-ErbB3, p-ERKl/2, and p-Rb were detected. Interestingly, the Varlitinib-mediated ErbB pathway blocking was different in the HCCOl-0708 model (Figure 8). There was mild inhibition found in p-EGFR, p-ErbB2, p-ErbB3, p-AKT, p-mTOR, p-p90RSK, and p-S6R on day 2 post-treatment (Figure 8A). However, significant suppression of p-Cdc2 and p-Rb as well as activation of p27 was found in the highest dosing of Varlitinib-treated HCCOl-0708 model. On day 12 post-treatment, marked decrease of p-ERKl/2 and p-p70S6K were observed (Figure 8B). The data of Varlitinib- treated HCCOl-0708 model provided molecular insights to explain the tolerance ability of HCCOl- 0708 at low dosing treatment. Better treatment outcome was achieved only at the dose, where Varlitinib effectively inhibits its target activation.
Subsequently, immunohistochemical staining of formalin-fixed paraffin-embedded (FFPE) samples was carried out to determine the changes of proliferation and apoptosis in the Varlitinib- treated PDXs. It clearly demonstrates that Varlitinib effectively inhibited tumour cell proliferation, induction of apoptosis in HCC29-0909A dose-dependently as indicated by reduced p-Histone H3 (SerlO) staining as well as enhanced cleaved PARP staining (Figure 3A and Figure 9}. Similar phenotypic changes were observed in other two Varlitinib-sensitive PDXs (HCCOl-0708 and HCC07- 0409) as well, but not in the Varlitinib-resistant PDX, HCC21-0208. It demonstrated that Varlitinib effectively and molecularly suppressed cell cycle progression and induced apoptosis in the ErbB- dependent PDXs, as a result tumour shrinkage is observed (Figure 10).
Varlitinib promotes vessel normalisation and tumour perfusion in HCC PDX
More than a decent ago, vessel normalisation concept which recalculates the imbalance between pro- and anti-angiogenesis was introduced (Goel et al., 2012; Jain, 2001, Cerniglia et al., 2009; Izumi et al, 2002). To determine whether Varlitinib is able to normalize the blood vessels in HCC, we performed the IHC on vehicle- and Varlitinib-treated tumours using CD31 antibody. Surprisingly, the density of blood vessels was significantly increased following Varlitinib treatment as determined by CD31 immunostaining (Figure 3A and Figure 9). The blood vessel diameter was significantly reduced and more capillary-like blood vessels were found in the tumours treated with Varlitinib.
To determine whether the capillary-like blood vessels induced by Varlitinib were functional, the experiment of biotinylated tomato lectin perfusion was performed to label the vascular endothelium and to detect perfused vasculature structure, followed by pimonidazole HC1 infusion to measure the hypoxic microenvironment in HCC29-0909A PDX tumours as direct evidence of vessel normalisation. Figure 3B and Figure 11 clearly showed that very little or no lectin was detected in blood vessels of vehicle-treated tumours suggesting that most of the blood vessels in vehicle-treated tumours are not functional. Large regions of the tumour section were stained positively with hypoxyprobe suggesting the hypoxic regions. In contrast, majority of capillary-like blood vessels induced by Varlitinib were lectin immunostaining suggesting that they were well perfused and functional. In addition, no hypoxyprobe were detected across the large section of the tumour indicating the region was well oxygenated. These data suggest that inhibition of HER family members by Varlitinib results in blood vessel normalisation. These capillary-like blood vessels are well-perfused leading to reduction of hypoxia in the tumour microenvironment.
Transcriptome analysis reveals the molecular basis of Varlitinib-mediated tumour suppression and vessel normalisation
Although both HCCOl-0708 and HCC29-0909A PDX models are ErbB3-highly expressed and activated, their responses to Varlitinib are quite distinct that highest tested dosing of Varlitinib is needed to suppress the growth of HCCOl-0708, whereas lower dosing is sufficient to suppress that of HCC29-0909A [Figure 1C and IE]. Therefore, to further investigate the similarities and differences between two models and between Varlitinib treatments at different doses, total mRNA sequencing (RNA-Seq) was used to analyse the triplicate samples of vehicle-, 50mg/kg Varlitinib, and lOOmg/kg Varlitinib-treated PDX samples which were collected on Day 11 and Day 14 post- treatment for HCCOl-0708 and HCC29-0909A, respectively.
There were 3422 differentially expressed genes identified (1720 downregulated and 1702 upregulated) in lOOmg/kg Varlitinib-treatment and 2778 differentially expressed genes identified (1377 downregulated and 1401 upregulated) in 50mg/kg Varlitinib-treatment in HCC29-0909A (Figure 4A). In total, 1122 and 1209 commonly downregulated and upregulated genes, respectively, were identified in Varlitinib-treated HCC29-0909A (Figure 4B). The top 20 activated and repressed genes in HCC29-0909A are shown in Tables 1A and IB, respectively. According to pathway enrichment analysis, the top 6 significantly suppressed KEGG pathways are ribosome pathway, RNA transport, steroid biosynthesis, central carbon metabolism in cancer, and HIFl signalling pathway (Table 1C). The Ingenuity Pathway Analysis (IPA) demonstrated the suppression of EIF2, eIF4, p70S6K, and mTOR signalling with predicted HIF1A as the inhibited upstream regulator as well as activation of ILl-mediated inhibition of RXR function, FXR/RXR and LXR/RXR pathway with predicted HNF1A and PPARA activated upstream regulators (Table ID), showing that the repressed ErbB downstream pathways correlates with hepatic lipid differentiation. In the contrast, there are 2151 differentially expressed genes identified (1167 repressed and 984 activated) in lOOmg/kg Varlitinib-treatment and 1842 differentially expressed genes identified (931 repressed and 911 activated) in 50mg/kg Varlitinib-treatment in HCCOl-0708 (Figure 4C).
The top 20 activated and repressed genes in high dosing treated HCCOl-0708 are shown in Tables 2A and 2B, respectively. Interestingly, the number of common differentially expressed genes reduced significantly. In total, only 265 and 192 commonly downregulated and upregulated genes, respectively, are identified in Varlitinib-treated HCCOl-0708 (Figure 4D). Two of the significantly repressed KEGG pathways in Varlitinib-treated HCCOl-0708 PDXs are central carbon metabolism in cancer (p-value = 2.40 x 10"3) and HIF-1 signalling pathway (p-value = 1.11 x 10"2). IPA analysis suggested that high dosing of Varlitinib suppressed Wnt/^-catenin pathway (p-value = 2.41 x 10 3) with predicted NUPR1, TGFB1, and Raf as the inhibited upstream regulators (p-value = 1.90 x 10-13, 3.10 x 10 12, and 1.21 x 10 11, respectively). The numbers of differentially expressed genes identified from different dosing in both models indicate that higher dosing of Varlitinib used, more genes will be affected. As the area-proportional Venn diagrams showed [Figure 4A and 4C), there are much more common genes identified in Varlitinib-treated HCC29-0909A than those in treated HCC01- 0708 [2331 genes vs 457 genes}, suggesting that similar genes and pathways were modulated by different dosing of Varlitinib in the former model. However, different dosing of Varlitinib has distinct effects on the latter model. These phenotypes molecularly align with the dose-dependent Varlitinib- mediated growth suppression in PDXs [Figure 1C and IE).
Furthermore, to interrogate the Varlitinib inhibition gene signature from the treated PDX, three-way area-proportional Venn diagram analysis was used to analyse the gene lists generated from lOOmg/kg BID and 50mg/kg BID treatment in HCC29-0909A and lOOmg/kg BID treatment in HCCOl-0708 (Figure 4E]. The top 20 activated and repressed genes are shown in Table 3A, respectively. The gene list and heatmap with 433 genes [195 downregulated and 238 upregulated) was identified and generated, followed by IPA and KEGG pathway enrichment analysis. Top 2 KEGG pathways, central carbon metabolism (p-value = 1.46 x 10 s) and HIFl-related pathways [p-value = 6.44 x 10"7), overrepresented in both treated PDXs reveals the core pathways that are effectively targeted by Varlitinib in HCC [Table 3B). IPA analysis further supported that glycolysis I pathway [p-value = 1.28 x 10 3) with HIF1A and PKM predicted as upstream targets [p-value = 2.50 x 10-15 and 1.44 x 10 7, respectively] were suppressed, whereas LPS/IL1 mediated inhibition of RXR function [p-value = 7.74 x 10 7) and FXR/RXR activation [p-value = 1.30 x 10 5) were elevated in the Varlitinib-treated PDXs [Table 3C).
Jain's group pointed out that imbalance of pro- and anti-angiogenic factors in the tumour microenvironment contributes to angiogenesis and tumour progression [Goel et al., 2012; Jain, 2005). As the strong evidence from Figure 3B proved the effective vessel normalisation was induced by Varlitinib, we further hypothesise that there is significant repression of pro-angiogenic factors and other factors in the tested PDXs, contributing to the normalisation. HIF1A expression is found to be significantly downregulated in the high dosing of Varlitinib-treated group in HCCOl-0708 model [-2.05 fold-change, FDR-adjusted p-value = 3.05 x 10 4), which demonstrates the consistency of the cellular phenotype and gene expression level. We then analysed the angiogenesis-related genes in the treated PDXs. The data showed that Varlitinib significantly repressed gene expression of the key pro-angiogenetic factors, PDGFA [-299.46 fold-change, FDR-adjusted p-value = 6.12 x 10 11), EPO [- 157.64 fold-change, FDR-adjusted p-value = 8.78 x 10 9), VEGFB [-15.68 fold-change, FDR-adjusted p-value = 2.31 x 10 10), and PGF (-4.33 fold-change, FDR-adjusted p-value = 4.18 x 10 8), and BMP2 (- 2.99 fold-change, FDR-adjusted p-value = 4.48 x 10 5), NRP2 (-2.23 fold-change, FDR-adjusted p- value = 0.030) in the high dosing of Varlitinib-treated HCC29-0909A model. For the Varlitinib- treated HCCOl-0708 model, pro-angiogenetic factors, TM4SF1 (-216.53 fold-change, FDR-adjusted p-value = 8.68 x 10-n), NRP2 (-11.04 fold-change, FDR-adjusted p-value = 1.44 x 10-e), EPO (-5.19 fold-change, FDR-adjusted p-value = 3.9 x 10 3), BMP2 (-4.34 fold-change, FDR-adjusted p-value = 2.15 x 10-6), BMP2-target gene FST (-4.16 fold-change, FDR-adjusted p-value = 0.010) and PDGFC (- 2.19 fold-change, FDR-adjusted p-value = 1.47 x 10 5) were identified.
The gene expression analysis demonstrated here that Varlitinib could inhibit similar EPO- related pathway and distinct VEGF-dependent [PDGFA, VEGFB, PGF, and NRP2) and VEGF- independent [PDGFC and BMP2/FST) anti-angiogenic pathways in two different PDX models to facilitate the vessel normalisation (Andrae et al., 2008; Kertesz et al., 2004; Krneta et al., 2006; Li et al., 2010; Lin et al., 2014; Zuo et al, 2016). Furthermore, 48.82-fold activation [FDR-adjusted p-value = 4.25 x 10-11) of HRG expression is found in Varlitinib-treated HCCOl-0708. HRG which encodes histidine-rich glycoprotein was found to suppress tumour growth by inducing vessel normalisation and macrophage polarisation in the immunocompetent tumour mouse model (Rolny et al., 2011). Recently, HRG has been further proved to function as a tumour suppressor in HCC through inhibiting FGF to Erkl/2 pathway and diminishing FGFR activation (Zhang et al., 2015). Therefore, HRG gene activation is accounted for another mean of vessel normalisation. Taken together, Varlitinib potently and effectively inhibits expression of anti-angiogenic factors as well as elevates expression of vessel normalisation factor, resulting in vessel normalisation and potentially improving drug delivery and efficacy.
Varlitinib enhances immune infiltration in the treated tumours
Recently, a study has suggested the mutual regulation of vessel normalisation and immune infiltration in multiple human cancers (Tian et al., 2017). Therefore, we sought to understand if immune infiltration would be affected in the Varlitinib-treated PDXs, following vessel normalisation. The RNA-Seq reads that align to mouse genome reference, mmlO were used to analyse the stromal components in the PDX models. Myeloid cell-related markers were specifically analysed. Interestingly, much higher expression of Cd68, Itgax (Cdl lc), Itgam (Cdl lb), /li¾rrei (F4/80), Ly6cl (Ly6C), I14ra (IL-4Rct), Lgals3 (MAC2), Csfl r (Cdl l5), and major histocompatibility complex (MHC)- class II related genes, H2-Aa, H2- Abl, H2-DMa, H2-DMbl, H2-DMb2, H2-Ea-ps, and H2-Ebl have been detected in the high dosing Varlitinib-treated HCCOl-0708 PDX (Table 11A), suggesting the presence of higher infiltration of monocytes, tumour-associated macrophages (TAMs), and Cdllb+MHC-class 11+ dendritic cells (DCs) in the treated tumours. Undetectable Ly6G (Grl) suggests the absence of tumour-associated neutrophils (TANs) in tumours. TAMs could indicate a wide spectrum of activations towards anti-tumoural Ml or pro-tumoural M2 states (Murray et al., 2014). Accordingly, a much higher M(LPS+IFNgamma)-associated gene expression, including elevated expression oiStatl, Tnf, Nos2, Irf5, and Nfkbiz, has been detected in Varlitinib-treated PDX, revealing the infiltration of highly activated Ml macrophage (Table 11B). Separately, the high expression of Socs3, Tgfbl, Sbno2, Stat6, CcI24, II4ra, and Id3 observed suggested that the presence of M2-like macrophage without defined activation status was found in the treated PDX (Table 11B). Also, the undetectable expression II10 and Arg 1 suggested that the M2-like macrophages might not be highly activated. Taken together, significantly higher myeloid immune gene signature was found in the Varlitinib-treated HCC PDX, possibly facilitated by vessel normalisation.
Varlitinib effectively inhibits mutated β-catenin pathways and mediates membrane translocation of mutated β-catenin
Due to the difference of Varlitinib efficacy in the tested PDXs, we have further investigated the gene signature that represents Varlitinib efficacy (Figure 12). Whole exome sequencing (WES) was further performed for all five PDXs to determine if these PDXs contain any mutation on β- catenin and ErbB family members (Table 9). Interestingly, the WES data demonstrated that all three Varlitinib-sensitive PDXs contain β-catenin T41A mutation, whereas ErbB family members exhibit different mutational patterns. Lower CTNNB1 expression but very high expression of β-catenin- interacting co-activators LEF1 and TCF4 as well as β-catenin-target genes, CCND1, MMP2, TBX3, EPHB2, and SPARCL1 were found in Varlitinib-sensitive PDXs (Figure 13A and Table 13A). In the contrast, the Varlitinib-resistant PDX, HCC21-0208, contains wild-type CTNNB1 but with much higher TGFBR1 /2, IGF2/IGFR/IRS1, NOTCH1 /JAG1, TEAD1 /2, and CYR61 expression and lower MST1 expression comparatively from the gene expression microarray data (Figure 13B and Table 13B]. The data aligns with pathway analysis and suggests that Varlitinib-resistant PDX exhibited highly activated TGF-β, NOTCH, and Hippo pathways. Compared to the proposed molecular classes of HCC (Lachenmayer et al., 2012; Sia et al., 2017], the Varlitinib-resistant PDX, HCC21-0208 is related to Wnt/TGF-β class, whereas the Varlitinib-sensitive PDXs, HCCOl-0708, HCC07-0409, and HCC29- 0909 are related to CTNNB1 class or more specifically CTNNB1 mutation class. Furthermore, the gene signature of Varlitinib-potency of the two RNA-sequenced PDXs was determined and shown in Figure 14.
Given that HCCOl-0708, HCC07-0409 and HCC29-0909A harbour T41A β-catenin mutation and the Wnt/^-catenin-related genes/pathways are inhibited according to WES and RNA-Seq analysis separately we sought to further investigate if the mutated β-catenin and its related pathway members are suppressed at protein level and whether the localisation of mutated β-catenin is affected by the Varlitinib treatment. Figure 5 and Figure 30B displayed marked inhibition of β- catenin and its related pathways in the lOOmg/kg BID Varlitinib-sensitive PDXs, HCC29-0909A, HCCOl-0708, and HCC07-0409 models, β-catenin upstream regulators, p-LRP6 and DVL3, as well as downstream targets, Axin2, survivin, c-Jun, c-Met, and N-cadherin were suppressed and E-cadherin expression was elevated by Varlitinib in the treated HCC29-0909A (Figure 5A). c-Jun was reported to physically interact with β-catenin and TCF4 and to stabilise
Figure imgf000043_0001
interaction for the transcription of β-catenin-target genes and cancer development (Gan et al., 2008; Nateri et al., 2005]. The inhibition of p-c-Jun by Varlitinib suggested the transcription repression of β-catenin. In addition, the β-catenin-transcriptional marks, ρ-β-catenin (Tyrl42 and Tyr654) (Brembeck et al., 2004; van Veelen et al., 2011), active β-catenin mark, ηοη-ρ-β-Catenin (S33/S37/T41), and total β- catenin were repressed (Figure 5B). Phosphorylated-LRP6, active β-catenin mark, survivin, and p=- c-Jun were found to be suppressed in HCCOl-0708 model and reduced transcriptional mark and active mark of β-catenin, Axin2, survivin, DVL3, and c-Met were also observed in HCC07-0409 model (Figure 30B). Interestingly, degradation mark, ρ-β-Catenin (S33/S37/T41) and p-RanBP3 Ser58were reduced in Varlitinib-treated HCC29-0909A (Figure 30A). The data suggested that reduction of β-catenin protein expression is not caused by phosphorylation of S33/S37/T41- mediated β-TrCP-ubiquitination and β-catenin may be exported from nucleus through phosphorylation of β-catenin at Tyrl42 (Krejci et al., 2012} and RanBP3-dependent mechanism (Hendriksen et al., 2005; Spiegelman et al., 2000; Yoon et al., 2008). Immunohistochemistry clearly demonstrated that β-catenin in vehicle-treated tumours was mainly located in the cytoplasm and nucleus (Figure 5C). However, expression of β-catenin in Varlitinib-treated tumours was indeed located in the membrane, indicated by the diminished nuclear staining of β-catenin and the enhanced membrane staining. Taken together, these results strongly support that Varlitinib targets mutated β-catenin and its related pathways in ErbB-dependent tumours by inhibition of p-LRP6, p- β-catenin at Tyrl42, and p-RanBP3 Ser58, resulting in β-catenin membrane translocation and inhibiting downstream targets of β-catenin.
NRG1/ERBB3 pathway dependence and differentiation status correlate with Varlitinib treatment efficacy in PDXs To reveal the differences in gene expression and signalling pathways between varlitinib- sensitive and varlitinib-resistant PDX models, gene expression microarray was performed on four treatment-naive PDXs (HCCOl-0708, HCC07-0409, HCC29-0909A, and HCC21-0208), followed by gene set and pathway enrichment analyses (Figure 12). The differentially expressed genes were identified from the quantile normalised data of Affymetrix GeneChip Human Genome U133 Plus 2.0 Array with cut-off threshold of 2 fold-change and high stringent FDR adjusted p-value < 0.0001, followed by heatmap generation (Figure 12A). In total, 2591 differentially expressed genes were detected, in which 1035 genes were highly expressed and 1556 genes were lowly expressed in the varlitinib-sensitive group. The gene lists with top 20 highly and lowly expressed genes are summarised in Table 4A and Table 4B, respectively. The most abundantly expressed gene in varlitinib-sensitive PDXs was NRG1 (fold-change > 500; Figure 12B). It encodes neuregulin 1, which is well-documented as ErbB family receptor ligand(Hsieh et al., 2011). Moreover, higher EGFR and ERBB3 expression was also detected in the varlitinib-sensitive group (Figure 12C and Table 12). The data is in line with the Western blot data shown in Figure IB that HCC21-0208 has undetectable ErbB3 protein, suggesting that ErbB3 pathway dependence is important for varlitinib treatment. Subsequently, Ingenuity Pathway Analysis (IPA) was used to analyse the differentially expressed gene sets. The analysis revealed that the varlitinib-sensitive PDXs have much higher glutathione metabolism and FXR/RXR and PXR/RXR activation with higher HNF4A and HNF1A activity predicted (Table 4C). In the contrast, the varlitinib-resistant PDX has statistically significantly higher pathway activities of angiogenesis, epithelial adherent junction signalling, epithelial-mesenchymal transition signalling, embryonic stem cell pluripotency, RhoA signalling, and Hippo signalling with higher CD24 signalling predicted. Studies have demonstrated that HNF4A is a key liver differentiation driver and CD24 is a tumour-initiating cell marker in HCC (Enane et al., 2017; Lee et al., 2011). In summary, varlitinib sensitivity correlates with ErbB3 expression and differentiation status in in HCC PDXs.
Sternness/differentiation status correlates with Varlitinib potency in high-ErbB3 expressing and β-catenin mutated PDXs
According to the Western blot analysis shown in Figure 2 and Figure 8, there was lesser degree of ErbB member suppression detected in Varlitinib-treated HCCOl-0708 when compared to the treated HCC29-0909A. However, the ErbB downstream targets such as p-Erkl/2, p-p70S6K, p- Cdc2, and p-Rb were suppressed. Also, functionally Varlitinib is able to enhance vessel normalisation in HCCOl-0708 model. As HRG was heavily upregulated in the treated HCCOl-0708 and a study reported that HRG is able to inhibit FGFR pathway (Zhang et al., 2015), we then postulated that other receptor tyrosine kinases (RTKs) including FGFR, which act upstream of Erkl/2 and Akt, could be also inhibited by Varlitinib in the HCCOl-0708 model. Subsequently, we re-analysed the RNA-Seq data with the focuses of RTKs. We discovered that IGF2, FGFR1, and FGFR2 were downregulated in 25.63-fold, 24.49-fold, and 4.46-fold, respectively. Conversely, this observation was not found in the HCC29-0909A model. Therefore, this led us to explore the global gene expression difference between the two Varlitinib-sensitive PDX models. RNA-Seq data of the vehicle-treated PDXs were then compared to elucidate the differences of Varlitinib potency in gene expression level. In total, 1615 differentially expressed genes are identified, in which 1069 genes and 546 genes are highly and lowly expressed in HCCOl-0708, respectively, when compared to HCC29-0909A [Figure 14). Among the top highly expressed genes, HEY1 and SOX9 were showed significantly higher expression in HCCOl-0708 (48.88 and 28.90-fold, respectively) than those in HCC29-0909A (Table 5A). Furthermore, the pathway enrichment analysis also demonstrated (Table 5C) that HCCOl-0708 has much higher Hippo, Wnt, IGFR, and FGFR pathway activities, whereas HCC29-0909A retains stronger metabolic activities, revealing the impaired differentiation in the former model (Table 5D). The RNA-Seq comparison further showed that β-catenin, Notch, and Hippo targets, MYC, CTGF, CYR61, and YAPl were highly expressed in HCCOl-0708, but were significantly inhibited by high dosing of Varlitinib (Table 5E and Table 5F).
Activated Wnt, Hippo, Notch, IGF, and FGFR are related the tumour-initiating cells and chemoresistance in HCC (Lau et al., 2016; Liu et al., 2016; Martinez- Quetglas et al., 2016; Villanueva et al., 2012). Accordingly, our study demonstrated that Varlitinib potency correlates with a spectrum of sternness/differentiation level in ErbB3-expressing and β-catenin mutated HCC PDXs.
Discussion
We have found that Varlitinib treatment potently inhibited tumour growth of ErbB- dependent HCC and effectively promoted vascular normalisation. Better efficacy was identified in HCC29-0909A, HCC07-0409, and HCCOl-0708 HCC PDX models, which display high levels of p- ErbB2 and p-ErbB3. Among these PDXs, HCC29-0909A possesses the highest p-ErbB3. HCC21-0208 model has undetectable p-ErbB2 and p-ErbB3 demonstrated poor efficacy in the Varlitinib treatment. This observation suggests that Varlitinib sensitivity is determined by the levels of p- ErbB3 and/or pErbB2 and raises the possibility that p-ErbB3 and p-ErbB2 are more reliable biomarkers for patient selection than total ErbB2 or ErbB3. Among the group of Varlitinib-sensitive PDXs, higher Wnt, Hippo, Notch, IGF, and FGFR pathway signalling were observed in the HCCOl- 0708 PDX, which required high dosing of Varlitinib to achieve better tumour growth inhibition. Pathway analyses revealed that Varlitinib suppressed tumour cell proliferation and promoted apoptosis by inhibiting phosphorylation of ErbBl-3, RAS/RAF/MEK/MAPK, p70S6K, S6 ribosomal, 4EBP1, Cdk-2, Cdc-2, and retinoblastoma pathways in dose- and time-dependent manners. Transcriptome and gene expression analyses have further revealed the Varlitinib-efficacy, -potency, and -inhibition gene signatures. Our findings also suggested that NRGl/ErbB3-dependant HCC tumours, displaying low sternness properties and high hepatic differentiation markers correlates with better efficacy and potency in Varlitinib-treatment. The findings are summarised in the hypothetic model shown in Figure 6.
In the present study, we also found that while angiogenic tumour vasculature from vehicle- treated tumours are hyperdilated and distorted, the majority of vessels from Varlitinib-treated tumours were slim, elongated, and regularly shaped resemble to blood vessels in normal liver. Varlitinib-treated tumours are well vascularized with a rather normal and more functional vasculature as determined by lectin perfusion than distorted and sparse tumour vasculature of vehicle-treated tumours. The decrease in tumour hypoxia in Varlitinib-treated tumours suggests that the dense capillary-like network of vessels was well perfused. Varlitinib induced vessel normalisation possibly through inhibiting HIF/VEGF-depending and -independent angiogenesis. Our hypothesis is supported by early studies showing that poor oxygen, concomitant with HIF-la induction, is associated with a more aggressive tumour phenotype, genetic ablation of HIF-Ια in various tumours results in reduced tumour mass and increased apoptosis. HIF-1 activity stimulates neovascularization by enabling tumour and host cells to produce a variety of proangiogenic factors like VEGF-A, PDGF-B, FGF-2, and angiopoietins that stimulate new blood vessel formation within hypoxic areas (Calvani et al., 2006; Okuyama et al., 2006). Also, support to our observation that Varlitinib induces blood vessel normalization by previous study showing that anti-HER2 therapies (trastuzumab) promote normalisation in HER2-positive breast cancer (Goel et al., 2011).
T41A mutations have been identified from 2% to 18.8% in analysed HCC and the β-catenin mutations are found to be associated with low-stage of HCC (Austinat et al., 2008; Boyault et al., 2007; Cleary et al., 2013; Enane et al., 2017; Guichard et al., 2012; Hoshida et al., 2009; Kan et al., 2013; Legoix et al., 1999; Nault et al., 2013; Nhieu et al., 1999; Waisberg and Saba, 2015; Wong et al., 2001). Molecularly, upon phosphorylation at serine 45, followed by serines 37/33 and threonine 41, β-catenin is ubiquitinated by β-TrCP ubiquitin ligases, followed by proteasomal degradation. However, mutations at T41 is found to prevent GSK3-mediated p-phosphorylation at serines 37 and 33, further avoiding β-TrCP recognition (Aberle et al., 1997; Liu et al., 2002; MacDonald et al., 2009; Orford et al., 1997). T41A mutation is further shown to be constitutively active mutant, which enhances nuclear localisation of TCF4, resulting in elevated expression of β-catenin-targets (Hsu et al., 2006). Since phosphorylation of β-catenin at tyrosine 654 and at tyrosine 142 is essential for binding to E-cadherin (Roura et al., 1999) and a-catenin (Aberle et al., 1996; Piedra et al., 2003; Pokutta and Weis, 2000) respectively, inhibition of tyrosine phosphorylation at these sites by Varlitinib would facilitate the interaction of E-cadherin and a-catenin with β-catenin. This could lead to the assembly of the E-cadherin- a-catenin^-catenin complex at the plasma membrane and to decreased β-catenin-dependent transcription.
In HCC, there are several molecular classifications proposed, describing that Wnt/^-catenin is one of the main classes. The first FDA-approved targeted therapy in HCC, Sorafenib, is found to modulate Wnt^-catenin signalling in in vitro and in vivo CTNNBl-class liver cancer models, demonstrated by reduced TCF/LEF luciferase-reporter activity and β-catenin expression (Boyault et al., 2007; Hoshida et al., 2009; Lachenmayer et al., 2012; Sia et al., 2017). Targeting Wnt^-catenin has been comprehensively discussed (Pez et al, 2013; Vilchez et al., 2016). Wnt/^-catenin pathway inhibitor ICGOOl, FH535, and other small inhibitors have been tested in HCC in different in vitro and in vivo models (Delgado et al., 2014; Gedaly et al., 2014; Handeli and Simon, 2008). Cheng's group and Gedaly's group further proved that the combination of sorafenib and ICGOOl as well as the combination of sorafenib and FH535 had better treatment outcome in their experimental models (Galuppo et al, 2014; Lin et al, 2016). However, studies thus far displayed that the promoted downregulation or degradation of wild-type β-catenin is the mean of β-catenin pathway inhibition in HCC and there is no small inhibitor available and proven to be mutant β-catenin inhibitor. Furthermore, according to the latest HCC TCGA data (Ally et al., 2017), 46% of HCC patients contain CTNNBl genetic alteration, including missense mutations and amplification, while 26%-38% of patients carry ErbB family genetic alteration (n=360, Figure 15A). Figure 15B specifically demonstrated the ErbB family genetic alterations (23%-36%) in the subset of HCC patients with CTNNBl mutation (n=96). It suggests that CTNNBl mutation and ErbB family genetic alteration are independent but not mutual exclusive events in HCC patients. However, our study suggests that Varlitinib would show high efficacy and potency in the specific subset of HCC which depends on both ErbB pathway and mutated β-catenin. Molecularly, Varlitinib effectively targeted β-catenin with T41A mutation in ErbB-dependent HCC by inhibiting ρ-β-catenin Tyrl42 and p-RanBP3 Ser58 as well as downregulated p-LRP6 and DVL3. As a result, the mutated β-catenin is transported from nucleus and cytoplasm to membrane. Several clinical trials of Varlitinib in different cancer types, including cholangiocarcinoma and gastric cancer have been initiated and presented its safety profile with greatly improved toxicity profile compared to the irreversible pan-Her inhibitors such as Neratinib and Dacomitinib. Therefore, Varlitinib is a clinically available pan-HER and mutated β- catenin inhibitor, that can be used to target p-ErbB2 and p-ErbB3-highly expressing and CTNNB1 subset of HCC. The robust and tolerate anti-tumour activity of Varlitinib in HCC PDX models and its safety profile warrants the development of clinical trial with trial enrichment for Varlitinib in HCC. Example 2 - Varlitinib induces apoptosis in HCC cell lines
An in vitro experiment was conducted to investigate whether varlitinib induced apoptosis in HCC cell lines.
HCC cells were grown in cell culture medium with 10% FBS and with varying concentrations of varlitinib. Apoptosis profiles were analysed at 24 hour and 48 hour timepoints, using Muse Annexin V & Dead Cell Assay kit and Muse Cell Analyser. Early apoptosis cells (identified as Annexin V-PE positive and Dead Cell Marker negative} were measured and plotted.
Figure 18 shows the results of the experiment. As can be seen, varlitinib was able to induce early apoptosis in all of the HCC cell lines tested after 48 hours of incubation. Varlitinib was particularly effective in sorafenib resistant cells (Huh7-SorR) where almost 70% early apoptosis was observed when high dose of varlitinib was used, suggesting that varlitinib could be effective in patients who progress on sorafenib.
Figure 19 shows the apoptosis profile for the PLC/PRF/5 (PLC) cells after 48 hour culture in the presence of varlitinib. Note the increase in percentage of apoptotic cells correlates with increasing varlitinib concentration.
In summary, the data presented demonstrates that Varlitinib can induce apoptosis in HCC cell lines, including sorafenib resistant cell lines.
Example 3 - Varlitinib-mediated tumour growth inhibition and vascular normalisation in activated ErB2/3-dependent and mutated β-catenin hepatocellular carcinoma
In this study, we evaluated the biological effects and efficacy of Varlitinib in HCC cell lines and patient-derived xenograft (PDX) models, and studied the molecular mechanisms of Varlitinib- mediated tumour suppression and anti-angiogenic potential. We sought to identify gene signatures associated with efficacy of Varlitinib, and the responsive subclass of HCC - by transcriptome and gene expression analyses.
Results:
Upregulation of ERBB3 is frequent in HCC and Varlitinib effectively inhibits high ErbB3- expressing liver cancer cell lines
To evaluate the gene expression profile of the ErbB family in HCC, we analysed 10 publicly available gene expression datasets. Interestingly, all 10 datasets displayed a statistically significant upregulation of ERBB3 in the HCC tumours versus the experimental controls, whereas expression of other ErbB family members showed no significant change (Figure 2 OA). Next, we sought to understand the gene expression profile of the ErbB family in a panel of 27 liver cancer cell lines from the Cancer Cell Line Encyclopedia (CCLE) gene expression dataset. Using the mean expression of ERBB3, 27 liver cancer cell lines were grouped in ERBB3msh and ERBB3Low accordingly [Figure 25A and Table 7). We then tested the protein expression of EGFR, ErbB2, and ErbB3 in selected liver cancer cell lines by Western blot (Figure 25B). The Western blot data showed that ERBB3msh cell lines (Hep3B, HepG2/C3A, Huh7, and PLC/PRF/5) had strong ErbB3 expression, whereas ERBB3Low cell lines (SNU398, SNU387, SNU182, SNU423, and SNU475) had undetectable ErbB3 protein expression, with the exception of SNU449 cell line. The Hep3B Sorafenib-resistant cell line (Hep3B- SorR), which we established by step-wise exposure to increasing amounts of Sorafenib until the maximal dose that the cell lines could tolerate, showed higher phosphorylation status of EGFR, ErbB2, and ErbB3. Upregulated ErbB pathway activity could therefore be an acquired Sorafenib resistance mechanism.
We tested the cytotoxicity of Varlitinib in representative ERBB3Hl&h and ERBB3Low cell lines (Figures 20B and 25C). Colony formation assay in Figure 20B showed that Varlitinib-mediated growth inhibition was more potent in ERBB3msh cell lines in a dose-dependent manner. Furthermore, Varlitinib effectively inhibited ErbB receptors and their downstream signalling in the treated PLC/PRF/5 cell line (Figure 20C), suggesting that Varlitinib could effectively and potently inhibit ERBB3™& HCC.
We next determined the protein expression of EGFR, ErbB2, and ErbB3 in 56 previously established HCC PDXs (Huynh et al 2006) by Western blot analysis. As shown in Figure 26A, 91% (51/56), 82% (46/56), and 93% (52/56) of the tested PDX expressed high levels of EGFR, ErbB2, and ErbB3, respectively. ErbB4 was undetectable in all PDX models.
Varlitinib effectively suppresses in vivo tumour growth in high p-ErbB2/3 and NRG1 expressing HCC patient-derived xenografts (PDXs)
To investigate the in vivo anti-tumour activity of Varlitinib, we then randomly selected 17 PDXs, with a range of expression of the ErbB family, to be used for Varlitinib treatment. The mice bearing PDX tumours were treated with Varlitinib (lOOmg/kg BID). Additional dosing of 25 and 50mg/kg BID were adopted to treat three PDXs, HCCOl-0708, HCC07-0409, and HCC29-0909A (Figures 21A and 26B). Surprisingly, Varlitinib showed effective tumour growth inhibition in only three PDXs (HCCOl-0708, HCC07-0409, and HCC29-0909A), at the dose of lOOmg/kg BID. HCC07- 0409 and HCC29-0909A were more sensitive to Varlitinib and showed superior dose-dependent inhibition (Figure 21A). Statistically significant growth inhibition was also observed among all three doses in the two models (p-122 value <0.0001). The latter model exhibited dramatic tumour suppression even when treated with Varlitinib at 50mg/kg BID, signifying that Varlitinib is highly potent in the HCC29-0909A model. HCCOl-0708 was a fast-growing PDX mouse model that tolerated Varlitinib at low dosing. Tumour growth was significantly suppressed at the highest dose, suggesting that a higher dosing of Varlitinib may overcome intrinsic drug resistance. Conversely Varlitinib at the dose of 100 mg/kg BID had no significant anti-tumour activity in the rest of the tested PDX models (Figures 21A and 26B).
The PDXs were categorised into responders and non-responders according to their treatment responses. Three responders and two non-responders were reassessed by Western blot (Figure 21B). As shown in Figure 21B, the three responders showed much higher p-ErbB2/3 and ErbB2/3 compared to the two non-responders. The expression of p-EGFR, total EGFR, p-Erkl/2, and p-Akt were similar in all tested PDXs, suggesting that responders are highly dependent on activated ErbB2/3 signalling. As Erkl/2 and Akt coud be activated by ErbB receptors, other growth receptors (GRs) and receptor tyrosine kinases (RTKs], simultaneously (Turke et al), our data suggests that the high expression of p-Erkl/2 and p-Akt in non-responders could be driven by other GRs or RTKs.
We further sought to understand the molecular differences between Varlitinib responders and non-responders. Three responders (HCCOl-0708, HCC07-0409, and HCC29-0909A) and 9 non- responders (HCCOl-0909, HCC06-0606, HCC13-0109, HCC21-0208, HCC25-0705A, HCC26-0808A, HCC26-0808B, HCC29-1104, and HCC30-0805B) were analysed by gene expression microarray, followed by gene set enrichment analysis. Interestingly, all 3 responders grouped together based on the principal component analysis (Figure 26C). 46 differentially expressed probes were identified by analysis between responders and non-responders with the threshold of fold-change=2 and FDR adjusted p-value=0.0001. The heatmap was generated accordingly (Table 8 and Figure 26D). NRG1 ranked the top differentially expressed gene (Figures 21C and Table 8). NRG1 is recognised as an ErbB3 ligand. The binding of NRG1 to ErbB3 facilitates the heterodimersation of ErbB3 and ErbB2 to activate the downstream signalling pathways. The gene expression dataset also validated that responders possessed statistically higher expression of EGFR, ERBB2, and ERBB3 (Figure 21C]. As illustrated in Figure 21D, poorer prognosis was correlated to high ERBB2 and ERBB3 expression in HCC tumours in the Singapore HCC dataset, using the cutoff finder for survival analysis. High NRG1 expression also reflected a similar trend (insignificant p-value).
Responders belong to G6 subgroup and CTNNB1 subclass of HCC regardless of Hepatitis B status
Gene set enrichment analysis (GSEA) was performed to better understand the difference between responders and non-responders against all gene sets in the Molecular Signatures Database (MSigDB) v6.2 (Table 6 and Figure 22A). The GSEA identified that the gene sets which were related to HCC CTNNB1 subclass, HCC G6 subclass, lower interferon signalling, Wnt pathway, Myc signalling, and lower YAP1 signalling, were highly enriched in responders. In contrast, gene sets of CTNNB1 phosphorylation cascade (beta-catenin degradation), Notch signalling, and TGF-beta signalling pathways were highly enriched in non-responders (Table 6 and Figure 22A). As described by Boyault and co-workers (Boyault et al), G6 subgroup had much higher CTNNB1 pathway activities. This is similar to the CTNNB1 subclass described by Llovet's group (Lachenmayer et al, Chiang et al). We then focused on several G6- and CTNNB1 subclass-specific genes in responders and non- responders (Figure 22B). Responders had significantly higher expression of FGF13, AXIN2, LEF1, EPHB2, GLUL, LAMA3, TBX3, SPARCL1, and LGR5. Conversely, non-responders had higher expression of TGFRB1, TGFRB2, TEAD1, IGF2, IGF1R, IRS2, NOTCH1, N0TCH2, and JAG1 (Figure 27). In addition, expression of ERBB3 positively correlated with LEF1 and AXIN2 and negatively correlated with IGF1R (Figure 22C). These suggest that non-responders could potentially belong to G1-G4 subgroups (Boyault et al). Taken together, the gene expression analysis of responders and non-responders suggests that responders possess high ERBB2/3 and NRG1 expression, correlating to the G6 subgroup and CTNNB1 subclass of HCC. This correlation was not related to viral hepatitis status. Additionally, the mutations in EGFR, ERBB2, and CTNNB1 were mainly identified in responders (Table 9).
Transcriptome analysis reveals the molecular basis of Varlitinib-mediated tumour suppression Although HCCOl-0708 and HCC29-0909A PDX models highly expressed activated ErbB3, they had distinct, differentiated responses to Variitinib. Maximal dosing (lOOmg/kg BID) of Variitinib was needed to suppress HCCOl-0708 tumour growth, whereas a lower dose [50mg/kg BID) was sufficient to suppress HCC29-0909A tumour growth (Figure 21A). To further investigate the similarities and differences between sample variation and dose-dependent Variitinib treatments, total mRNA sequencing (RNA-Seq) was used to analyse samples of vehicle-, 50mg/kg BID Variitinib-, and lOOmg/kg BID Varlitinib-treated PDX samples. The samples were collected on Day 11 and Day 14 post-treatment for HCCOl-0708 and HCC29-0909A, respectively (Figures 23 and 28A-B). The samples of vehicle- and 75mg/kg BID Varlitinib-treated non-responder HCC16-1014 were used as negative control of the treatment for RNA-Seq analysis (Table 10).
There were 3728 differentially expressed genes identified (1904 upregulated and 1824 downregulated) in lOOmg/kg Varlitinib-treatment and 3023 differentially expressed genes identified (1522 upregulated and 1501 downregulated) in 50mg/kg Varlitinib-treatment in HCC29- 0909 A (Figure 28A). When comparing between the two treatments in Varlitinib-treated HCC29- 0909A, there were 1337 and 1243 commonly upregulated and downregulated genes, respectively. In contrast, there were 2474 differentially expressed genes identified (836 activated and 1638 repressed) in lOOmg/kg Varlitinib-treatment and 1947 differentially expressed genes identified (1024 activated and 923 repressed) in 50mg/kg Varlitinib-treatment in HCCOl-0708 (Figure 28B). Also, there were very few differentially expressed genes detected in the treated non-responder HCC16-1014 (Table 10). The numbers of differentially expressed genes identified from the different dosing in both HCC29-0909A and HCCOl-0708 models indicated that the higher the dose of Variitinib used, the higher the number of changes in gene expression. The area-proportional Venn diagrams showed (Figures 28A and 28B) that there were more common genes identified in the Varlitinib-treated HCC29-0909A than those in the treated HCCOl-0708 (2580 genes vs 496 genes), suggesting that similar genes and pathways were modulated by different dosing of Variitinib in the former model. However, different dosing of Variitinib had distinct effects on the latter model. These phenotypes molecularly aligned with the dose-dependent Varlitinib-mediated growth suppression in PDXs (Figure 21A).
The top 20 commonly activated and repressed genes in HCC29-0909A are shown in Tables 1A and IB, respectively. The top 6 significantly suppressed KEGG pathways were the ribosome pathway, RNA transport, steroid biosynthesis, central carbon metabolism in cancer, and HIF1 signalling pathway (Table 1C). The Ingenuity Pathway Analysis (IPA) demonstrated the suppression of EIF2, eIF4, p70S6K, and mTOR signalling with predicted HIF1A as the inhibited upstream regulator, as well as activation of ILl-mediated inhibition of RXR function, FXR/RXR and LXR/RXR pathway with predicted HNF1A and PPARA activated upstream regulators (Table ID), showing that the repressed ErbB downstream pathways correlate with hepatic lipid differentiation.
The top twenty commonly activated and repressed genes in the Varlitinib-treated HCCOl- 0708 are shown in Tables 2A and 2B, respectively. In the Varlitinib-treated HCCOl-0708 model, there were 270 and 226 commonly upregulated and downregulated genes, respectively (Figure 28B). Interestingly, there were fewer number of common differentially expressed genes in HCCOl- 0708 than in HCC29-0909A. Two of the significantly repressed KEGG pathways in Varlitinib-treated HCCOl-0708 PDXs were involved in the central carbon metabolism in cancer (p-value = 2.40 x 10- 3) and the HIF-1 signalling pathway (p-value = 1.11 x 10-2}. IPA analysis suggested that high dosing of Varlitinib could suppress the Wnt/p-catenin pathway (p-value = 2.41 x 10-3) with predicted NUPR1, TGFB1, and Raf as the inhibited upstream regulators (p-value = 1.90 x 10-13, 3.10 x 231 10- 12, and 1.21 x 10-11, respectively).
To examine the Varlitinib inhibition gene signature from the treated PDXs, differentially expressed genes were compared between lOOmg/kg BID and 50mg/kg BID treatment in HCC29- 0909A, and lOOmg/kg BID treatment in HCCOl-0708 (Figure 23A). The overlapped 546 genes (259 upregulated and 287 downregulated) were investigated using IPA and KEGG pathway enrichment analysis. The top 20 activated and repressed genes are shown in Table 1A. The top 2 KEGG pathways include the central carbon metabolism (p-value = 1.46 x 10-8) and the HIFl-related pathways (p- value = 6.44 x 10-7). The two overrepresented pathways reveal the core mechanisms that are effectively targeted by Varlitinib in HCC (Table 4A). IPA analysis demonstrated a suppressed glycolysis pathway (p-value = 1.28 x 10-3) with HIF1A (p-value = 2.50 x 10-15) and PKM (p-value= 1.44 x 10-7) predicted as upstream targets. LPS/IL1 mediated inhibition of RXR function (p-value = 7.74 x 10-7) and FXR/RXR activation (p-value = 1.30 x 10-5) were elevated in the Varlitinib-treated PDXs (Table 3C).
Varlitinib inhibits glycolysis, HIF1A pathway, and angiogenesis in the treated PDXs
We then specifically analysed glycolysis and HIFlA-related pathways that were highly enriched and significantly inhibited in the treated PDXs. Figure 23B demonstrated that Varlitinib treatment was able to suppress selected glycolytic genes, as well as hypoxia- and angiogenesis- related genes. HIF1A was predicted as the key upstream regulator of Varlitinib inhibition (Figure 23C). The expression of HIF1A was significantly inhibited in both the treated HCC29-0909A and HCCOl-0708 models (Figure 23D). Figures 28C-E further revealed that Varlitinib could inhibit the EPO pathway, VEGF-dependent {PDGFA, VEGFB, PGF, and NRPZ) and VEGF-independent {ANG, PDGFC and BMP2) pro-angiogenic pathways to facilitate vessel normalisation (Krneta et al, Kertesz et al and Andrae et al). Expression of LGR5 (β-catenin upstream regulator), YAP1, IRX3, and MYC were downregulated in both treated HCC29-0909A and HCCOl-0708 models. Conversely, expression of CDH1 was enhanced (Figures 23E and 28F). LEF1 was reduced in the treated HCC29- 0909A model (Figure 28G). On the contrary, higher expression of YAPl-related genes as well as repression of these genes were observed in the treated HCCOl-0708 (Figure 28H). By comparing HCC29-0909A and HCCOl-0708, we have further identified that stem cell related pathways are enriched in the HCCOl-0708, which displayed instrinsic resistance to Varlitinib when treated at lower dosing (Figure 281). Given that both HCC29-0909A and HCCOl-0708 contain mutated CTNNB1 (Table 9), it should also be noted that the expression of CTNNB1 in HCCOl-0708 was higher than in HCC29-0909A (Figure 28J), suggesting that CTNNB1 is a high-expressing mutant in HCCOl-0708. Heatmap analysis clearly indicated that HCCOl-0708 had significantly more embryonic stem cell core genes and Hippo/YAPl-related genes expressed than HCC29-0909A and the gene expression could be effectively repressed by high Varlitinib treatment. Comparatively, the non-responder HCC16-1014 did not show any inhibition in these pathways (Figures 28K, 28L and 28M). As a result, these analyses suggest that higher dosing of Varlitinib is required for optimal effect against high ErbB2/3-expressing HCC with higher instrinsic sternness-related gene expression. Time- and dose-dependent Varlitinib treatment inhibits ErbB receptor pathways and promotes vessel normalisation and tumour perfusion in HCC PDX
The effects of Varlitinib on the ErbB family and their downstream targets were studied using Western blot (Figures 2A, 2B, and 29). Protein levels in PDXs were investigated, using a panel of antibodies detecting the ErbB members and their downstream targets. Protein lysates were prepared at equal concentrations from the treated tumour samples collected on day 2 post treatment and/or the last day of the experiments. In the HCC29-0909A model, Varlitinib effectively inhibited p-EGFR, p-ErbB2, and p-ErbB3 on days 2 and 14 post-drug treatment (Figures 2A and 2B). The suppression of phosphorylated and total ErbB3 became more prominent on day 14 post treatment (Figure 2B). While p27 level was elevated, p-AKT, p-mTOR, p-p90RSK, p-S6R, p-Cdc2, and p-Rb levels were decreased in a dose-dependent manner. At a later time point, p-ERKl/2, p-p70S6 , p-4EBPl, CDC25C, and E2F1 expression were significantly decreased, while p21 expression was increased. These findings demonstrated dose- and time-dependent inhibition of ErbB downstream targets in this model. Furthermore, at the highest dose of Varlitinib-treated HCC07-0409 model (Figure 29), expression of p-EGFR, p-ErbB2, p-ErbB3, p-ERKl/2, and p-Rb were all inhibited.
Immunohistochemistry staining of formalin-fixed paraffin-embedded (FFPE) samples was carried out to determine changes in proliferation, apoptosis, and angiogenesis in the Varlitinib- treated PDXs (Figures 3A and 3B, 9, and 10). The analysis demonstrated that dose-dependent Varlitinib effectively inhibited tumour cell proliferation and induced apoptosis in HCC29-0909A. This was indicated by reduced p-Histone H3 (SerlO) staining and enhanced cleaved PARP staining (Figures 3A and 3B and 9). Similar phenotypic changes were observed in the other two Varlitinib responder PDXs (HCCOl-0708 and HCC07-0409), but not in the Varlitinib non-responder PDX, HCC21-0208 (Figure 10). This demonstrated that Varlitinib effectively suppressed cell cycle progression and induced apoptosis in the ErbB-dependent PDXs, which corresponded to tumour shrinkage. In addition, the density of blood vessels was significantly increased following Varlitinib treatment, as demonstrated by positive CD31 immunostaining (Figure 3A and 3B). There was a significant reduction in the diameter of the blood vessels and more capillary-like blood vessels were found in Varlitinib responder PDXs. To determine whether Varlitinib-induced capillary-like blood vessels were functional, biotinylated tomato lectin was injected intravenously into vehicle- and Varlitinib-treated tumour-bearing mice for labelling of the vascular endothelium. This allowed the detection of perfused vasculature structure. Additionally, pimonidazole HC1 infusion was used to measure the hypoxic microenvironment in HCC29-0909A PDX. These provided direct evidence of vessel normalisation as a treatment effect. Figure 3A and 3B clearly depicts that little or no lectin was detected in blood vessels of vehicle-treated PDXs, suggesting that most of the blood vessels in vehicle-treated PDXs were non-functional. Large regions of the tumour section stained positively with hypoxyprobe, indicating regions of hypoxia. In contrast, the majority of capillary-like blood vessels, in the HCC PDX exposed to Varlitinib treatment, stained positive for biotinylated lectin, suggesting that these blood vessels were well-perfused and functional. Furthermore, hypoxyprobe staining was negative across the large section of the Varlitinib-treated responder PDXs, showing that the region was well-oxygenated. The data suggests that the inhibition of ErbB family members by Varlitinib results in blood vessel normalisation. This was supported by the observation of well- perfused capillary-like blood vessels that reversed hypoxia in the tumour microenvironment. This IHC analysis was concordant to the gene set enrichment analysis findings that angiogenesis- and hypoxia-related gene sets were highly enriched in the Varlitinib-treated PDXs. (Figures 23B-D) Varlitinib effectively inhibits mutated β-catenin pathways and mediates membrane translocation of mutated β-catenin
According to the RNA-Seq analysis (Figures 23E and 28G], β-catenin-related genes, such as
LGR5 and LEF1, were inhibited by Varlitinib. We then sought to further investigate if the β-catenin and its related pathway members would be suppressed at the protein level, and whether the localisation of mutated β-catenin was affected by the Varlitinib treatment.
Figures 24A, 24B, 30A and 30B displayed marked inhibition of β-catenin and its related pathways in the Varlitinib responder PDXs, HCC29-0909A, HCCOl-0708, and HCC07-0409 models. β-catenin upstream regulators, p-LRP6 and DVL3, as well as downstream targets, Axin2, survivin, c- Jun, c-Met, and N-cadherin, were suppressed. In contrast, E-cadherin expression was elevated by Varlitinib in the treated PDXs (Figure 24A). c-Jun was reported to physically interact with β-catenin and TCF4 to stabilise β-ΰ3ΐβηϊη/ ΓΐΡ4 interaction for the transcription of β-catenin target genes and cancer development (Nateri et al). The inhibition of p-c-Jun by Varlitinib suggested the transcriptional repression of β-catenin. In addition, factors associated with active β-catenin export, ρ-β-catenin at Tyrl42 (van Veelen et al) and p-RanBP3 at Ser58 (Hendriksen et al), active β-catenin indicator (ηοη-ρ-β-Catenin at S33/S37/T41), β-catenin and E-Cadherin interrupted signal (ρ-β- catenin at Tyr654), and total β-catenin were repressed in the treated HCC29-0909A (Figures 24A and 30A). Interestingly, degradation indicator, ρ-β-Catenin at S33/S37/T41 was also reduced, even as there is a missense mutation at β-Catenin, i.e. Threonine to Alanine (T41A, Figure 24A and Table 9). Phosphorylated-LRP6, active β-catenin, Survivin, and p-c-Jun (Ser73) were found to be suppressed in the HCCOl-0708 model (Figure 24B). The reduction of transcriptional and active sites of β-catenin, Axin2, Survivin, DVL3, and c-Met were also observed in the HCC07-0409 model (Figure 30B). The data suggests that reduction of β-catenin phosphorylation at Tyrl42 (Krejci et al) and the RanBP3-dependent mechanism are responsible for its nuclear export (Yoon et al). Also, suppression of β-catenin phosphorylation at Tyr654 might facilitate the binding of β-catenin to E-Cadherin. Immunohistochemistry clearly demonstrated that β-catenin in vehicle-treated tumours was mainly located in the cytoplasm and nucleus (Figure 5C). However, expression of β-catenin in Varlitinib- treated PDXs localised to the membrane, indicated by the diminished nuclear staining of β-catenin and the enhanced membrane staining. The combined results strongly support our hypothesis that Varlitinib targets mutated β-catenin and its related pathways in ErbB-dependent tumours - by inhibiting β-catenin upstream regulator LGR5 and p-LRP6, ρ-β-catenin at Tyrl42 and Tyr654, nuclear export regulator p-RanBP3 at Ser58, as well as upregulating E-Cadherin expression. These, in turn, result in β-catenin membrane translocation, thereby inhibiting downstream targets of β- catenin. T41A β-catenin mutation is thought to be constituitively active. However, our study demonstrates that its nuclear localisation and β-catenin-driven genes could be inhibited by Varlitinib in high p-ErbB2/3 expressing HCC.
Discussion
We have demonstrated that among the ErbB receptor family, ERBB3 is upregulated in a subset of HCC. We established that Varlitinib, a small molecule pan-ErbB inhibitor, effectively inhibited high £7?£?£fJ?-expressing HCC in vitro. Additionally, in vivo efficacy was demonstrated in the responder PDXs, HCC29-0909A, HCC07-0409, and HCCOl-0708 in immune-deficient NOD/SCID mice, which displayed high levels of p-ErbB2/3, total ErbB2/3, and NRG1 gene expression. The non- responders, HCC21-0208 and HCC16-1014 models expressed lower p-ErbB2 and p-ErbB3 levels, and were found to have poor efficacy to Varlitinib treatment. This observation suggests that Varlitinib sensitivity is determined by the quantitative expression of p-ErbB3 and/or pErbB2 and raises the possibility that the presence of p-ErbB2/3 could potentially be specific for patient treatment stratification. Furthermore, expression of ERBB2 and ERBB3 are known to have prognostic value, whereas that of NRG1 has the same trend in our Singapore HCC dataset (Figure 21D).
Gene expression analysis demonstrated that the three responders, with established higher
ERBB2/3 expression, belong to both the G6 subgroup and CTNNB1 subclass in the non-proliferation class of HCC, whereas the 14 tested non-responder PDXs are related to the HCC subgroups with TGF and IGF/IGF1R pathways and S1/S2 tumours in the proliferation class of HCC (Llovet et al 2017,Boyault et al 2007,Lachenmayer et al ). It indicates the strong correlation of the ErbB family and β-catenin pathways in the specific subclass of HCC PDXs, that could be effectively inhibited by Varlitinib. It also reveals the diverse oncogenic pathway dependence in HCC and that alternative targeted treatments such as the TGF , IGFIR, mTOR, and FGFR inhibitors are likely needed to treat Varlitinib non-responders. Among the group of high ErbB2/3-expressing responder PDXs, Varlitinib was especially potent against HCC29-0909A. It may be due to the low activity of mutated β-catenin in the model, as indicated by the IPA pathway enrichment analysis (Figure 281). Conversely, a higher dose is required in the HCCOl-0708 model, which has higher mutated CTNNB1 expression, YAPl signalling activity, and sternness, β-catenin-driven cancers require YAPl for tumour progression (Rosenbluh et al). YAPl and its related sternness properties are also associated with drug resistance in multiple cancers (Zanconato et al). Lin and colleagues previously identified through genetic screen that YAPl is the key resistance driver of RAF- and MEK- targeted therapies (Lin L et al). However, our pre-clinical in vivo data showed that the intrinsic resistance, possibly contributed by mutated β-catenin and YAPl, in the HCCOl-0708 PDX model could be overcome by a higher dose of Varlitinib, further supported by the significant suppression of β-catenin/YAPl -driven genes, SOX9, HEY1, CTGF, and CYR61 (Figure 28H).
Pathway analyses further revealed that Varlitinib suppressed tumour cell proliferation and promoted apoptosis in the responder PDXs by inhibiting phosphorylation of EGFR, ErbB2/3, MEK/MAPK, AKT/mTOR, p70S6K, S6R, 4EBP1, Cdk-2, Cdc-2, and retinoblastoma pathways in a dose- and time-dependent manner. Subsequently, hypoxia reduction, vessel normalisation, β- catenin membrane translocation, and β-catenin pathway inhibition were induced in the Varlitinib- treated PDXs. Taken together, our findings suggest that NRGl/ErbB2/3-dependent and β-catenin- mutated HCC tumours are most vulnerable to Varlitinib. These findings are summarised in the hypothetical model shown in Figure 31.
In this study, we found that while tumour vasculature from vehicle-treated tumours were hyper-dilated and distorted, the majority of blood vessels from Varlitinib-treated tumours were instead slim, elongated, and regularly-shaped, bearing resemblance to physiologically healthy blood vessels in a normal liver. The reduction in tumour hypoxia likely resulted from better perfusion of dense capillary-like blood vessels in the tumour microenvironment following Varlitinib exposure. Varlitinib possibly induced vessel normalisation through inhibiting HIF/VEGF-independent and HIF/VEGF-dependent angiogenesis. Our data also shows direct evidence that pan-ErbB inhibitor is able to repress HIF1A expression in a dose-dependent manner. Previous studies have reported that poor oxygen supply to the tumour bed, concomitant with HIF-Ια induction, is associated with a more aggressive tumour phenotype. Genetic ablation of HIF-Ια in various tumours results in reduced tumour mass and increased apoptosis. HIF-1 activity stimulates neovascularisation by enabling tumour and host cells to produce a variety of proangiogenic factors like VEGF-A, PDGF-B, FGF-2, and angiopoietins that stimulate new blood vessel formation within hypoxic areas (Okuyama et al, Calvani et al). A previous study in other cancer models showed that Erlotinib and Gefitinib could inhibit pro-angiogenic factor VEGF expression by HIFlA-dependent and independent mechanisms (Pore et al). Another study showed that anti-ErbB2 therapies (Trastuzumab) promote normalisation in ErbB2-positive breast cancer by inhibiting HIF-Ια and pro-angiogenic factors (Goel et al]. Recently, the synergistic anti-tumour role of anti-angiogenesis in reprogramming the immune and metabolic tumour microenvironment is clinically proven in the July 2018 Federal Drug Agency (FDA) granting of breakthrough therapy designation for the combination of atezolizumab (anti- PDL1 antibody) and bevacizumab (anti-angiogenesis inhibitor) in the first and second line treatment of advanced HCC following highly compelling Phase lb clinical trial results. A phase 3 trial using the same combination as first line treatment for advanced HCC patients has been started (ClinicalTrials.gov Identifier: NCT03434379). In our study, we have clearly shown that Varlitinib can induce vessel normalisation in HCC. Thus, it would be rational to combine Varlitinib and an immune checkpoint inhibitor in HCC, firstly tested in humanised HCC model (Zhao et al) and/or immunocompetent HCC model, then in human studies.
β-catenin mutations are found to be associated with the earlier stages of HCC and the T41A mutation has been identified in up to 18.8% of analysed HCC (Guichard et al). Upon phosphorylation at serine 45, followed by serines 33/37 and threonine 41, wild-type β-catenin is ubiquitinated by β- TrCP ubiquitin ligases, destined for proteasomal degradation. However, mutations at T41 prevent GSK3-mediated phosphorylation at serines 33/37, avoiding β-TrCP recognition (Liu et al). T41A mutation is also found to be a constitutively activated mutant, which enhances nuclear localisation of TCF4, resulting in elevated expression of β-catenin-targets (Hsu et al H-T et al). Since phosphorylation of β-catenin at tyrosines 142 and 654 is essential for binding to E-cadherin (Roura et al) and oc-catenin (Pokutta et al), inhibition of tyrosine phosphorylation at these sites by Varlitinib would facilitate the interaction of E-cadherin and a-catenin with β-catenin. This could lead to the assembly of the E-cadherin-a-catenin^-catenin complex at the plasma membrane, and the decrease of β-catenin-dependent transcription. Our findings are consistent with early studies in other cancer models that inactivation of ErbB2 reduces phosphorylation of β-catenin at tyrosine 654, which faciliates β-catenin-E-cadherin interaction (Shibata et al). It was previously reported that EGFR is the direct target of Wnt^-catenin in liver (Tan et al), Still, we are the first group to show that pan-ErbB inhibition effectively represses mutated β-catenin in in vivo models, linking ErbB receptor activities to mutated β-catenin localisation and functions in HCC. Further experiments will be needed to further confirm the proposed molecular mechanisms. Also, the effect of Varlitinib on other β-catenin mutations would need to be mechanistically explored. The Wnt/^-catenin has been described as one of the main functional classes of molecular classification in HCC. Sorafenib - a BRAF, C-RAF and VEGF-R inhibitor - can modulate Wnt/ -catenin signalling in in vitro and in vivo CTNNBl-class liver cancer models. This was demonstrated by showing reduced TCF/LEF luciferase-reporter activity and β-catenin expression following Sorafenib treatment (Boyault et al, Lachenmayer et al). There has been considerable exploration of targeting Wnt/ -catenin against HCC. Wnt/ -catenin pathway inhibitors ICGOOl, FH535, and other small inhibitors have been tested in HCC in different in vitro and in vivo models. Several groups showed that the combination of Sorafenib and ICGOOl, and the combination of Sorafenib and FH535, resulted in better treatment outcomes in experimental models respectively (Galuppo et al, Gedaly et al).
However, studies have shown that inhibition of the β-catenin pathway in HCC is associated with the downregulation or degradation of wild-type β-catenin. Currently, no small molecule inhibitor that directly targets mutant β-catenin exists. Llovet and coauthors recently reviewed that immune exclusion and so resistance to immunotherapy is contributed by activation of the β-catenin pathway in HCC - consistent with findings in melanoma (Llovet et al 2018). Thus, Varlitinib could potentially reverse immunosuppression via vessel normalisation and targeting the activating β- catenin pathway in HCC. Combining Varlitinib with an immune checkpoint inhibitor in HCC would also be rational. To date, there are limited enrichment strategies for HCC clinical trials with any molecular targeted inhibitor (Llovet et al 2018).
Our present study demonstrates that Varlitinib shows good efficacy in the specifically defined subset of HCC that is dependent on both the ErbB pathway and β-catenin mutation. We also demonstrate that Varlitinib promotes apoptosis and vessal normalisation with reduction of tumour progression and hypoxia in this subset of HCC. Varlitinib effectively inhibits ErbB receptors and their downstream oncogenic signalling such as MEK/Erk and AKT/mTOR pathways. In addition, it also targets β-catenin with T41A mutation in ErbB-dependent HCC by inhibiting ρ-β-catenin Tyrl42 /654 and p-RanBP3 Ser58, as well as inducing LGR5, p-LRP6, and DVL3 downregulation. As a result, mutated β-catenin is transported from the nucleus and cytoplasm to the cell membrane, reducing the expression of β-catenin-driven genes.
Several clinical trials of Varlitinib in different cancer types, including cholangiocarcinoma and gastric cancer, have been initiated. Varlitinib has a comparably favorable safety profile compared to irreversible pan-ErbB inhibitors such as Neratinib and Dacomitinib. The recommended dose of Varlitinib for clinical trials is between 200mg BID to 500mg BID, which is equivalent to 41mg/kg BID to 102.5mg/kg BID in mouse (calculation is based on Nair and Jacob (Hulsen et al). Our study reveals that 50mg/kg BID in mice (~250mg BID in human) showed efficient growth supression in high p-ErbB2/3 expressing PDXs. Intrinsic resistance demonstrated by high β- catenin/YAPl-related gene signature in high p-ErbB2/3 expressing PDX could be overcome by increasing the dose to lOOmg/kg BID in mice (~490mg BID in human). In contrast, even with very high dose of Varlitinib given to the mice (lOOmg/kg BID), there is no anti-tumour benefit in the non- responders. Expression of p-ErbB2/3 and/or ErbB2/3 could be a predictive biomarker in the clinical treatment of HCC with Varlitinib.
In summary, Varlitinib is a pan-ErbB and, as we report here, an inhibitor of mutated β- catenin - that can selectively target the p-ErbB2/3 highly-expressing and CTNNB1 subset of HCC. Our data suggests that the selection of HCC patients with high p-ErbB2/3 expression could be a useful target and predictive biomarker for Varlitinib clinical efficacy. As the molecular and immune classification of HCC is further evolved and expanded, the role of tailoring systemic therapies, whether molecular or immune modulating, can potentially optimise treatment efficacy based on HCC subtypes.
Materials and Methods
Reagents, Xenograft models, Vessel perfusion study, Western blot analysis and Immunohistochemistry were the same as described in Example 1.
Transcriptome sequencing analysis, global gene expression analysis, and bioinformatic analysis
Three tumours from each condition (control, 50mg/kg, and lOOmg/kg Varlitinib treatment) of 95 HCC29-0909A and HCCOl-0708 and two tumours from each condition (control and lOOmg/kg treatment) of HCC16-1014 were used for transcriptome analysis. A total of 200 ng of total RNA was used for illumina TruSeq mRNA library prep, followed by 150bp paired-end sequencing on an illumina HiSeq4000 platform by BGI HK, Ltd. The raw sequencing reads were aligned to mouse mmlO genome reference and human hg38 genome reference by Spliced Transcripts Alignment to a Reference (STAR) aligner version 2.5.3a, separately using Partek Flow (Partek Inc. St. Louis, MO). In average, 74 million to 90 million clean reads per sample were obtained. The total human alignments ranged from 70 million to 87 million per samples. Aligned reads were then quantify to hg38 - RefSeq Transcripts 85 -2018-05-02 annotation model using Partek E/M algorithm, followed by low expressed gene filtering and total count and add 0.0001 gene counts normalisation. GSA differential expression detection model was used to compare 50mg/kg Varlitinib treatment vs control and lOOmg/kg Varlitinib treatment vs control in PDX models using the cut-off threshold of >2 and -2 fold-change and FDR adjusted p-value < 0.05, followed by heatmap generation.
The gene expression analysis was done by quantile normalised data from Affymetrix GeneChip Human Genome U133 Plus 2.0 Array with cut-off threshold of >2 and -2 fold-change and high stringent FDR adjusted p-value < 0.0001. The identified gene lists were compared using BioVenn online tool for the area-proportional Venn diagram analysis (Hulsen et al). The KEGG Pathway Enrichment analysis was carried out in Partek Genomics Suite (Partek Inc. St Louis, MO) with the cut-off threshold of FDR adjusted p-value <0.05. The differentially expressed gene lists were analysed through the use of Ingenuity Pathway Analysis (IPA) (Kramer et al) (QIAGEN Inc., https://www.qiagenbioinformatics.com/products/ingenuitypathway-analysis). Gene Set Enrichment Analysis (GSEA) was carried out in Partek Genomics Suite (Partek Inc. St. Louis, MO) or javaGSEA with 100000 number of permutations. Molecular Signature Database (MSigDB) v6.2 was used for the analysis (Liberzon et al, Subramanian et al).
Statistical analysis
Statistical analysis was performed in GraphPad Prism version 7.00 for Mac OS X (GraphPad Software, La Jolla California USA, www.graphpad.coin)
Tables B
Gene Symbol Fold-change FDR adjusted p -value Gene Symbol Fold-change FDR adjusted p -value
F 03 309.49 1.09E-11 PDGFA -299.455 6. 12E-11
FOXG1 241.30 9.10E-14 CERSl -298.654 7.96E-12
CLDN4 226.57 7.12E-15 GDF1 -248.957 3.14E-12
TBC1D9 198.24 1.47E-09 LY6E -224.589 9.46E-15
ΛΤΡ2Β4 166.61 5.23E-10 CD81 -212.999 7.12E-15
ERVH48-1 141.06 3.58E-04 KCTD15 -209.852 9.76E-13
RORC 139.28 l.OOE-ll ZNF853 -208.827 2.28E-06
IRF6 109.17 2.86E 12 T E 132A -176.207 1.04E-10
KIF26B 94.93 8.61E-09 EPO -157.635 8.78E-09
KCNK1 90.20 1.84E-10 COTtl -127.493 2.21E-11
EVPL 86.83 1.12E-14 ARTS -125.85 9.50E-13
CYP1A1 76.02 8.12E-12 ADGRD1 -120.298 2.65E-12
ENPP4 71.01 2.72E-10 TUSC3 -95.4738 1.08E-10
GUCY2C 63.91 6.40E-11 GAA -91.7216 1.31E-13
NAGS 60.31 2.45E-10 FAM 19A5 -81.2813 2.42E-05
MUC13 57.24 1.08E-09 TB BP1 -72.0157 4.34E-10
MAGEA6 56.06 6.11E-10 PNMA6A -59.9852 7.12E-15
DSC3 52.33 4.75E-08 NAT8L -59.4931 3.99E-14
RNASEL 51.30 2.21E-10 DBN1 -58.5321 1.76E-04
S10OA1 47.57 2.02E-14 ASNS -51.5197 4.37E-11
Pathway Name En richmerit Score Enrichment p-val ye
Ribosome 23.5353 6.00E-11
RNA transport 8.54517 0.000194482
Steroid biosynthesis 0.000942895
Central carSon metabolism In cancer 6.08887 C.O0226796
Ceil cycle 5.65022 0.00351675
HlF-1 signaling pathway 5.56328 0.00383619
Biosynthesis of amino acids 5.29439 0.00501967
Protein processing in endoplasmic reticulum 5.18743 0.00558633
Fatty acid metabolism 4.82731 0.00800803
Biosynthesis of unsaturated fatty acids 4.4S445 0.0112831
Predicated repressed pathways by IPA analysis Enricnment p -value
EIF2 Signaling 3.34 x lO45
Regulation of elF4 and p70S6 Signaling 1.92 x 10"'
mTOK Signaling 4.57 x 10"a
Pyrine Nucleotides Oe Novo Biosynthesis 11 3,04 x lO'7
Suserpathway of Cholesterol Biosynthesis S.25 x 10"s
Predicted activated pathways b IPA analysis Enrichment p-¥alye
LPS/iL-1 Mediated Inhibition of RXR Function 2.78 x 10"5
Acute Phase Response Signaling 1.43 x Iff*
FXR/RXR Activation 4.71 lO"
LXR/RXR Activation 9.31 x 10 4
Table 1 - Transcriptome analysis of Varlitinib treatment in HCC29-0909A. (A) Top 20 common activated genes and (B) top 20 common repressed genes in lOOmg/kg and 50mg/kg BID Varlitinib- treated HCC29-0909A. FDR adjusted p-values are shown. (C) KEGG pathway analysis showing the enriched pathways generated from the gene lists of significant differentially expressed genes with the threshold of >2 and <-2 fold-change with FDR adjusted p-values < 0.05. (D) Ingenuity Pathway Analysis showing the predicted repressed and activated pathways with enrichment FDR adjusted p- values <0.05.
A B
Gene Symbol Fold-change FOR adjusted p -value
Gene Symbol Fold-change FDR adjusted p -value
CYP4A11 55059.40 1.98E- 16
TM4SF1 -216.53 8.68E-11
GSTA1 90.87 5.16E-14
CPLX2 -44,9391 5.64E-07
HPD 57.74 8.72E-12
CDKN1C -43.3875 8.12E-07
NPW 48.32 4.55E-13 STRA6 -33.1811 3.77E-08
CFHR2 36.68 2.64E-04 0KK1 -31.3167 4.25E-11
GNMT 33.12 1.45E-08 LYZ -28.5924 6.06E-10
C1DEC 28.84 2.84E-11 NT5E -24.7924 4.30E-12
CD4 28.58 1.01E-13 FHL2 -22.5319 7.16E-10
CFHR1 28.63 4.23E- 12 SLC6A8 -21.7241 4.66E-07
ITIH4 27.51 1.45E-09 BG -21.1068 2.07E-09
SLC04C1 24.90 4.55E-13 CD8B -19.9052 1.26E-06
SERPIN61 24.00 1.17E-12 FRZB -18.9016 2.77E-10
H GCS2 23.49 8.72E-12 PLEKHA2 -15,0221 2.40E-07
TMPRSS6 22.13 2.87E-12 LlNGOl -13.8232 2.95E-11
ABLIM3 21.92 1.77E-09 8NIP3L -13.5145 9.07E-12
AGPAT9 21.47 2.93E-09 FAM 19A5 -13.4038 3.03E-02
A R1C1 20.72 2.83E-09 TNFRSF25 -11.8909 4.60E-09
AKR1D1 20.56 1.71E-10 TFEB -11.5725 8.86E-09
CDHR5 18.36 5.64E-13 RNASET2 -11.3201 6.06E-10
LBP 18.31 1.96E-08 F AM 43 A -11.0796 1.79E-07
Predicated repressed pathways by IPA analysis Enrichment p -value
Axonal Guidance Signaling 6.00 x 10"
Adipogenesis pathway 1.09 x 10'
Osteoarthritis Pathway 2.08 x 10 :
Wnt/P-catenin Signaling 2.41 x 10 :
-3
Pyridoxal 5'-phosphate Salvage Pathway 2.43 x 10'
Predicted activated pathways by IPA analysis Enrichment p -va
29
Superpathway of Cholesterol Biosynthesis 1.53 x 10"
23
LX /RXR Activation 7.14 x 10
21
FXR/RXR Activation 2.65 x 10
15
Cholesterol Biosynthesis I 1.04 x 10"
15
Cholesterol Biosynthesis II (via 24,25-dihydrolartosterol) 1.04 x 10"
Table 2 - Transcriptome analysis of Varlitinib treatment in HCCOl-0708. (A) Top 20 common activated genes and (B) top 20 common repressed genes in lOOmg/kg and 50mg/kg BID Varlitinib- treated HCCOl-0708. FDR adjusted p-values are shown. (C) Ingenuity Pathway Analysis showing the predicted repressed and activated pathways with enrichment FDR adjusted p-values based on the gene lists of significant differentially expressed genes with the threshold of >2 and <-2 fold- change with FDR adjusted p-values < 0.05.
Activated Genes Repressed Genes
F 03 T EM 132A
RORC EPO NAGS FAM19A5
ADIRF ASNS
CES3 H 2
FAT4 TTB 1
S 10OA14 NSMF
SIC 16 A 14 HAL
AZGP1 OBSL1
PPL LOCIOO 507194
LOC200772 GPC4
TNNIi CERCA B
SIC 1A7 ADRA2C
Pathway Name Enrichment Score Enrichment p-value BCHE SERP1N11 Central carbon metabolism in cancer 1.80E+01 146E-08 CYP4A11 SPARC H!F-l signaling pathway 14.2551 6 4E-07 NPY6R SLC6A8 Glycolysis / Glucorieogenesis 7.61446 0.000493269
ENPEP SCARF2 Alanine, aspartate and glutamate metabolism 5.33005 0.00484382
HAOl HS6ST2 Biosynthesis of amino acids 5.11366 0.00601403
CYP3A5 AOM TGF-beta signaling pathway 4.67044 0.00936819
BHMT I X3 Ribosome 4.58088 0.0102459
Giycosamirtoglycan biosynthesis - heparan sulf; 3.76538 0,0231589 Carbon metabolism 3.64968 0.0259994
Giycosp ingolipid biosynthesis - tacto and neol 3.61461 0,0269275 Valine, leucine and isoleucine biosynthesis 3.2439 0.0390115 Fructose and mannose metabolism 3.17348 0.0418576 Neomycin, kanamycin and gentamicin biosyntr 3.02562 0.0485276
Predicated repressed pathways by IPA analysis Enrichment p -value
Glycolysis I L28 x i£f
HistHine Degradation 111 2,19 M ΜΓ
Purine Nucleotides De Novo Biosynthesis I 4,23 x 10' 3
Asparagine Biosynthesis 1 9.03 x 10
EIF2 Signaling 1,27 x 10'
Predicted activated pathways by IPA analysis Enrichment p -value
LPS/IL-l Mediated In hibit ton of RXi f nction 7.74 x Iff7
Glycine Betaine Degradation 3.21 x iff6
FXR/RXR Activation 1,30 x 10"s
Acute Phase Response Signaling 2,33 x iff5
Tryptophan Degradation III (Eukaryottc) 1.21 x 10 * Table 3 - The 3-way Venn diagram analysis revealing the Varlitinib-inhibition gene signature.
(A) The top 20 activated and repressed genes commonly identified in 50 and lOOmg/kg BID Varlitinib-treated HCC29-0909A as well as lOOmg/kg BID Varlitinib-treated HCCOl-0708. (B) KEGG pathway analysis showing the enriched pathways with FDR adjusted p-values <0.05 generated from the lists of commonly dysregulated genes. (C) Ingenuity Pathway Analysis showing the predicted repressed and activated pathways with enrichment FDR adjusted p-values < 0.05.
A Gene Symbol Fold-change FDR adjusted p
CNTN3 -148.708 1.76E-12
Gene Symbol Fold-change FDR adjusted p -v B
ADH1B -138.106 5.74E-09
NRG1 551.22 1.38E-11
ZNF667-AS1 -92.1645 8.56E-13
H19 /// M1R675 495.52 6.54E 39
CYP26B1 -83.2785 4.45E-11
MEG 3 201.28 9.79E-10
ADH1B -66.5402 3.76E-07
DIOl 194.99 1.21E-09
PCDHB5 -62.4209 1.18E-11
FCER1G 186.30 3.08E-13
IGFBP2 -57.8871 4.50E-09
H19 /// IR675 158.20 2.23E-07
TUBB2B -53.0594 8.98E-10
DNAJC1S 147.41 7.73E-13
ZNF667 -52.277 2.23E-11
FCER1G 102.28 l.lZE-ll
CADPS -47.7625 6.26E-07
MEG 3 101.27 5.99E-09
LI MS3-LOC440895 -43.4773 1.24E-09
BC034313 /// RP1U1N9.4 93.70 l.OOfc lO
ZNF667 -42.0682 4.63E-14
MEG 3 93.43 3.73E-07
DIP2C -40.777 2.14E-13
CD36 87.01 5.75E-09
PDGFA -38.5822
EPDR1 78.33 6.7OE-10 1.41E-10
KRT23 74.41 9.83E-07 PCDHB14 -37.4345 1.20E-11
CCDC169-SOHLH 2 /// SOHLH2 68.48 1.9 IE -08 GABRB2 -37.1071 2.87E-14
NARR /// RA834 68.14 1.93E-14 IMAP1L3 -34.2161 5.92E-08
FM03 67.05 8.78Ε-Π TDRP -33.0318 9.52E-10
NARR /// RAB34 63.52 1.27E-11 G LYR1 /// SEPT6 -32.8903 5.86E-12
PLAGL1 59.78 2.34E-11 PRDM 16 -31.3786 1.32E-12
CD36 58.08 3.48E-08
Predicated repressed pathways by IPA analysis Enrichment p -value
C •5
Inhibition of Angiogenesis by TSP 1 1.44 x 10'
4
Epithelial Adherens Junction Signaling 4.77 x 10"
Regulation of the Epithelial-Mesenchymal Transition Pathway
Human Embryonic Stem Cell Pluripotency
Predicted activated pathways by IPA analysis Enrichment p -value
Glutathione Redox Reactions I 1.10 x 10 ;
Glutathione-mediated Detoxification 2.10 x Iff3
FXR/RXR Activation 6.62 x Iff4
Tetrahydrofolate Salvage from 5,10-metheny:tetrahydrofolate 6.78 x Iff4
PXR/RXR Activation 1.01 x Iff3
Table 4 - Identification of Varlitinib-resistant gene signature by global gene expression analysis. (A) The top 20 highly expressed genes and (B) the top 20 lowly expressed genes identified in Varlitinib-sensitive PDXs, HCC29-0909A, HCCOl-0708, and HCC07-0409 models. (C) Ingenuity Pathway Analysis showing the predicted repressed and activated pathways with enrichment FDR adjusted p-values < 0.05. ** indicates p-value < 0.01, *** indicates p-value < 0.001, **** indicates p- value < 0.0001. Gene Symbol Fold- change DR adjusted p -value
Gene Symbol :o(d change FDR adjusted p aiue
DUSP6 52.90 5.03E-11
A CYP4A11 -21309.6 3.05E-15
SE A6B 51.47 6.41E-11
GSTA1 -43.6589 2.46E-12
PPP1R14C 47.44 2.5OE-07
SAA4 -38.6979 5.03E-11
HEY1 45,08 2.72E-12
CFHR2 -28.458 6.48E-04
NFE2 40.87 1.67E-09
HRG -27.1965 2.08E-09
T 4SF1 38.23 8.17E-08
CI DEC -22.4431 2.52E-10
COL9A3 37,07 6.22E-09
NPW -21.3434 4.41E-11 pec 35.26 3.5OE-03
U6T1A4 -20.9517 1.29E-10
§0X3 29.47 3J5E-11
CFHR1 -20.8533 5.96E-11
IGF2 24,55 7.12E-09
AGPAT9 16.4033 2.36E 08
FGFR 1 24,15 1.0 E-10
AKR1C1 -15,85 2.86E-08
C8orf 23.70 4.43E-06 PON1 -15 5251 2.58E-06
FRZB 21,88 3.8OE-10 SERPING 1 -15.1625 4.41E-11
TNFRSF25 19.76 8.58E-10 HULC -13.6682 1.35E-07
GPX2 19.70 1.18E-09 CD4 -13.1305 2.24E-11
FREM1 17.61 1.35E-05 ALDOB -13.0788 6.22E-09
TFEB 17,51 2.17E-09 WNT11 -12.6081 1.25E-07
CPLX2 17,23 2.28E-05 HPD -12.5572 2.04E-08
SPHK1 15.38 4.39E-08 AKRIDi -11.9887 7.12E-09
MUC13 15.19 9.81E-07 FETUB -11.8688 1.81E-10
Pathways activated in HCCOl-0708 i nrichment Score Enrichment p-value
Proteoglycans in cancer 9.25E+00 9.65E-05
ECM-receptor interaction 7.62369 0.000488737
Hippo signaling pathway 6.49953 0.00150415
Central carbon metabolism in cancer 6.46148 0.00156248
Small cell iung cancer 5.93642 0.00264148
Arrhythmcgenic right ventricular cardtomycpa 5.76271 0.00314259
Focal adhesion 5.55414 0.0038714
Hippo signaling pathway -multiple species 5.27186 0.00513406
Wnt signaling pathway 5.20257 0.00550239
Hypertrophic cardiomyopathy (HCM) 4.99425 0.00677678
Table 5 - Identification of Varlitinib-potency gene signature by transcriptome analysis. (A)
The top 20 highly expressed genes identified in higher Varlitinib-potent PDX HCC29-0909A when compared to lower Varlitinib-potent HCCOl-0708. (B) the top 20 lowly expressed genes identified in higher Varlitinib-potent PDX HCC29-0909A when compared to lower Varlitinib-potent HCCOl- 0708. (C) Ingenuity Pathway Analysis showing the predicted activated pathways with enrichment FDR adjusted p-values < 0.05. Pathways activated in HCC29-0909A Innchment Score Enrichment p-value
PPAR signaling pathway 37.5943 4.71E-17
Metabolic pathways 30.2388 7.37E-14
Complement and coagulation cascades 19.2842 4.22 E-09
Fatty acid metabolism 17.9968 1.53 E-08
Steroid biosynthesis 17.8928 1.7QE-08
Steroid hormone biosynthesis 15.5438 1.78E-07
Fatty acid degradation 14.5233 4.93 E-07
Fatty acid biosynthesis 12.4488 3.92 E-06
Terpenoid backbone biosynthesis 11.4075 1.11E-05
Pyruvate metabolism 11.3895 1.13E-05
HCCOl-0708 iOOmg/kg Varlitinib BID
HCCOl-0708 vs HCC29-0909A
value
Gene ID Fold-change FDR p -value
CYR61 -49.52
CTGF 13.54 9.75 Iff8
9.17 x 10'*
CYR61 12.53 CTGF -43.59
1.09 x 10* 1.81 x ι 11
MYC 3.68 2.68 x 10 s MYC -9.88 4.63 x 10"3
YAPl 2.57 2.62 x 10* YAPl -3.84 8.43 x Iff9
Table 5 - Identification of Varlitinib-potency gene signature by transcriptome analysis. (D)
Ingenuity Pathway Analysis showing the predicted repressed pathways with enrichment FDR adjusted p-values < 0.05. (E) The expression of CTGF, CYR61, MYC, and YAPl in HCCOl-0708 when compared to those in HCC29-0909A. (F) The expression of CTGF, CYR61, MYC, and YAPl in lOOmg/kg BID Varlitinib-treated HCCOl-0708 when compared to those in vehicle control-treated HCCOl-0708.
Gene eel enriched itt Re spoode f · vs N ΟΙΜΓ Θ S ponde re N . of rotM' S ES NES p-value q-valn
SANSO _W T_PATHWAY_REQUIRE_MYC 148 0.5724 2.1125 < 0.O001 0.0119
CH lA GjrVER_CANCER_SUBCLASS_t lTERFERON_D 122 0.6764 2.1088 < 0.0001 0.0107
CHIANG LIVER CANCER SUBCLASS CTNNB1 JJP 375 0.6835 2.0936 < 0.0001 0.0130
BOYAULT_L(VER_CANCER_SUBCLASS_Ge_UP 162 0.5793 2.0333 < 0.0001 0.0245
YAP1_DN 92 0.5640 2.0263 < 0.0001 0.0239
SANSOM_APC_TARGETS_UP 289 0.4634 1.9458 « ο.οοοι 0.0427
HALLMARK_MYC_TARGETS_V2 130 0.5947 1-7684 0.0119 0.0427
REACTOME_CTNNB1 _PHOSPHORYLAT >N_CASCAOE 51 0.7036 -2.0401 0.0012 0.0468
HALL ARK_NOTCH_SGNALWG 88 0.5550 -15475 < 0.0001 0.0028
KEG&TGF-beti signaling pathway 219 0.4275 -13420 < 0.0001 0.0480
Table 6 - Gene enrichment analysis in Varlitinib responder and non-responder PDXs. GSEA of Varlitinib responders vs non-responders with the use of MSigDB v6.2 [Liberzon et al, Enane et al]. The enriched gene sets with q-value < 0.05 are shown. ES refers to enrichment scores, NES refers to normalised enrichment scores. ERBBf £f?8B3L0W Table 7 - Analysis of ERBB3™ and
SNU878 JHH6
SNU761 SNU423 ERBB3Low liver cancer cell lines. The list of
Hu 1 SNU475 ERBB3Hi and ERBB3^>W liver cancer cell lines. Li7 SNU387
Hu 7 SNU398
Alexander SNU182
JHH1 JHH2
PLC HLF
Hu 6 HLE
JHH7 JHH4
Hep3B SNU449
JHH5 SNU886
HepG2
NCIH6S4
C3A
Gene Symbol Fold-Change FDR-adlusieet p-value Table 8
NRG1 3.199 3.998E-19
BC034319///RP11-11N9.4 2.945 1J86E-1 Varlitinib
_ 2.884 8.527E-22 treatment in
CRISP2 2.536 7.433E-15
FGF13 2.511 2891E-30 HCC PDXs. The
L1NC01419 2.489 6.242E-1
CDH1 2.427 1 ,991 E-12 differential
MME 2.393 1. 56E-13 expressed gene FCER1G 2,393 4.980E-10
SLC1A2 2.348 8.165E-14 (DEG) analysis of
OLR1 2.326 1.156E-1
FSTL5 2.318 2.313E-10 Varlitinib
CCND2 2.316 8.88ΘΕ-09
COL 1A2 2.300 1,634E-12 responders vs
ERVW-1 2.228 1. 27E-26 non-responders.
RR11 -382F24.1 ///RP11 -382F24.2 2.224 5.033E-09
— 2.217 1.752E-13 The DEG was
FMN2 2.200 2.281E-14
PIN 2.173 2.777E-17 done with
RBK446 2.163 1.060E 12 microarray data
FMN2 2.132 2.537E-12
CD36 2.127 1.227E-08 and threshold of
DCAF4L2 2.113 6.228E-08
TMEM52B 2.113 1.149E-09 fold-change > 2
NFE4 2.078 1.934E-32 and FDR-adjusted
GPX8 2.069 3.985E-11
FBXL13 2.067 1J32E-29 p-value < 0.0001
APCDD1 2.065 1.062E-11
EDDM3A 2.049 2535E-1 in Partek
CD36 2.04Θ 1 10BE-08 Genomics Suite.
BAGE 2.039 3.454E-21
COLEC12 2.036 6.134E-21
ABC A3 2.027 3.800E-17
RNASE2 2.020 4.932E-17
PTN 2.016 1.463E-16
PAG1 -2,005 7.201 E-12
GLYR1///SEPT6 -2.020 4046E-18
SEPT6 -2,049 1.636Ξ-18
SUA RCA 2 -2,098 4.456E-16
MSP -2.116 3.778E-22
RASEF -2,161 6.082E-09
LIMS3-L OC440895 ///LOC100288570
// L OC 100507334/// L OC440895 -2.222 1.178E-11
STOX2 -2,272 6.801 E-11
DIP2C -2.320 9.043E-09
SMARCA2 -2.360 7.745E-20
S MA RCA 2 -2.602 1 58SE-19 POX models 1 wspofises to wwmvn mm ERBB2 ERBB3 ERBB4 CTNNBl Hepatitis ¾ iitts tccoi-roos Its ponder R521K P1170A, I655¥ Nit Nil T41A Hepatitis B positive
HC.C07-O409 Respenier RS21K P1170A, I6SS¥ Nit Ni! T41A Hepatitis i positive
MCC29-0909A Responds r R5J 1 PliTOA, I6S5¥ Nit NI! T41A Hepatitis B positive
HCCOl 0909 Non-responder Ni Nil Nit Ni! Nil Hepititis B positive
HCC06-0606 Non-res ponder Kit Nil Nit Nil Nil Hepatitis B positive
HCOT-0913 Non- res ponder Nit Nil Nit Nil Nit Negative hCCn-0109 Non-responder Nit Nil Nit Nil Ni! Hepatitis B positive
HCC13-0212 Non-responder Wt m Nit Ni! Nit Hepititis 8 positive
HCC15-0114 Non-responder NI Nil Nit Nil Nil Hepatitis C positive
HCC1S-1014 Non-responder Nit m Nil Mutated Nil Negative
HCC17-0211 Mon-responder Nit Nil Nil Ni! Ni! Vsg.tive
HCQ1-020S Non-responder R521K Nil Nit Mutated Ni! Hepatitis B positive
HCC2S-0705A Non-responder Nit Nil Nit Nil Nit Hepititis B positive
HCC26-0808A Non-responder Nil NI Nil Nil Ni! Hepatitis B positive
HCC26-0908B Non-responder Nit Nil Nil Nil Nil Negative
HCQ3-1104 Non-responder Nit NI Nil Ni! Ni! Hepatitis S positive
HCC30-060SB Non-responder NI NI Nil Nit Nil Hepatitis B posi ive
Table 9 - Analysis of responders and non-responders. Mutations and Hepatitis statuses of the treated PDXs. Mutations at EGFR, ERBB2, ERBB3, and CTNNBl were determined by whole exome sequencing. Nil represents no mutations being detected.
HCC16-1014
Differentially Expressed Gene Fold-change FDR-adjusted p-value
SDS 3.83 6.64 x 10 "
CYP3A4 3 65 4.00 x 10
UGT1A1 2.85 2.00 x 10
FXYD2 -2 88 2.00 x 10 "
Table 10 - Transcriptome analysis of Varlitinib treatment in HCC29-0909A, HCCOl-0708, and HCC16-1014. DEG analysis from RNA-Seq analysed Varlitinib-treated non-responder PDX, HCC16- 1014. Four genes passed the threshold of fold change >2 and p-value < 0.05.
Gent symbol Common name Fold-change FDR adjusted p-value
Cd68 CD68 10.27 1.03 St 10
Itgax CDllc 80.33 1,30 x 10"J
itgom 3
CDllb 5.29 5.S2 x - 10
Adgrel F4/80 8.96 4.43 X 10
Ly6el Ly6C 8.93 7.12 x 10 *
Mm IL-4Ra 5.09 8.43 x 10"*
Lgais3 MAC2 22.18 3.09 x 10"*
Csflr CD115 10.48 2.37 x 10 *
H2-Aa HC class II 35.69 1.94 x 10"*
H2-AU HC class il 40.00 7,18 x 10"'
H2-DMO MHC class 11 20 62 5.46 10
-I
H2-DMU MHC class II 49.56 1 64 10
H2-Ea-ps MHC class il 22.40 2,28 x 10 ;
H2-EU MHC class II 38.84 1.08 x 10 * Table 11 - Immune gene signature in high dosing of Varlitinib-treated HCCOl-0708 PDX model. (A) Differentially expressed myeloid-related genes. Gene symbol Fold-change FDR adjusted rvalue
B Statl 5.16 3.48 x 107
39.39 1.32 X 10J
Nos2 29.52 4.20 x 10"*
IrfS 20.79 7.05 M Ιθ"
Nfkbii 2.30 0.02
So€s3 9.03 1.16X 10*
Tgfil 7.» 5.20 X 10*
Sbno2 6.84 1.74 x 10*
StatS 6.44 2.31 x 106
Cc!24 5.18 8.11x 103
i #.14 6.71x10*
Table 11 - Immune gene signature in high dosing of Varlitinib-treated HCCOl-0708 PDX model. (B) Differentially expressed Ml-like and M2-like tumour-associated macrophage (TAM)- related genes.
Gene symbol' FeltJ-chan§e PDR-aifusted p-nilye
Figure imgf000066_0001
ERB82 1.66 6.7$ x 10-*
ΕΛΒΒ3 2.41 1..011 tH
ERB84 1.13 0.l3(n.s.J
Table 12 - Higher EGFR and ERBB3 expression detected in Varltinib sensitive group
CTNNB1 -8.01 1.81x10*
LEF1 33.00 8.52x 10-"
TCF4 14.86 9.93 x 10 0
CCND1 3.21 6.01 xlO'7
MMP2 210 4.04 x10"3
TBX3 2.27 7.46 0·5
EPHB2 3.23 2.53 x10-7
SPARCL 1 290 2.05 x 10 *
Gene symbol Fold-change FDR-adjusted p- value
B TGF8R1 -283 4.61 x 10*
TGFBR2 •6.51 1,02x10*
IGF2 -42.64 1.04x 10-*
IRS1 4.35 1.63 10-'
NOTCH! -1161 5.06 XIO*
JAG1 -7.10 2.86x10*
TEAD1 -2.35 1.24x10*·
TEAD2 -358 1.48x10*
Table 13 - Identification of β-catenin mutation and the constitutively active β-catenin pathways in the Varlitinib-sensitive PDXs. (A) Expression of β-catenin and its downstream targets and (B) expression of Wnt/TGF -related targets, in the four analysed PDXs indicated by normalised probe intensity.

Claims

Claims
1. A method of treating a patient for cancer (for example a solid tumour) by administering a com ound of formula (I):
Figure imgf000067_0001
an enantiomer thereof or a pharmaceutically acceptable salt of any one of the same, for example wherein the compound of formula (I) is Varlitinib or a pharmaceutically acceptable salt thereof, wherein the patient has been identified as having a beta-catenin pathway mutation (in particular a mutation in beta-catenin).
2. The method according to claim 1, wherein the mutation is at a serine residue.
3. The method according to claim 1 or 2, wherein the mutation is at a threonine residue.
4. The method according to any one of claims 1 to 3, wherein the mutation is in the GSK 3β region of the beta-catenin.
5. The method according to any one of claims 1 to 4, wherein the mutation is T41A.
6. The method according to any one of claims 1 to 5, wherein the patient has a cancer selected from: liver cancer (such as hepatocellular carcinoma), biliary tract cancer, gall bladder cancer, breast cancer (such as none ER+ breast cancer), prostate cancer, colorectal cancer, ovarian cancer, cervical cancer, lung cancer, gastric cancer, pancreatic, bone cancer, bladder cancer, head and neck cancer, thyroid cancer, skin cancer, renal cancer, and oesophagus cancer and combinations of two or more of the same.
7. The method according to claim 6, wherein the cancer is hepatocellular carcinoma.
8. The method according to any one of claims 1 to 7, wherein the compound of formula (I) is
Figure imgf000067_0002
or a pharmaceutically acceptable salt thereof.
9. A method according to any one of claims 1 to 8, wherein the compound of formula (I) is provided as the free base.
10. A method according to any one of claims 1 to 9, wherein the compound of formula [I) is administered as a pharmaceutical formulation.
11. A method according to any one of claims 1 to 10, wherein the compound of formula (I) or a pharmaceutical formulation comprising same is administered orally.
12. A method according to any one of claims 1 to 11, wherein the compound of formula (I) or a pharmaceutical formulation comprising the same is administered bi-daily.
13. A method according to any one of claims 1 to 12, wherein each dose of the compound of formula (I) is in the range 100 to 900mg, for 100, 200, 300, 400, 500, 600, 700, 800 or 900mg.
14. A method according to claim 13, wherein each dose of the compound of formula (I) is in the range 200 to 500mg, for example 250 mg or 490 mg.
15. A method according to claim 14, wherein each dose is 300g or 400mg.
16. A method according to any one of claims 1 to 15, wherein the compound of formula (I) or formulation comprising the same is employed as a monotherapy.
17. A method according to any one of claims 1 to 16, wherein the compound of formula [I] is employed in a combination with another anti-cancer agent.
18. A method according to claim 17, wherein the combination therapy comprises a chemotherapeutic agent.
19. A method according to claim 18, wherein the chemotherapeutic agent is independently selected from the group comprising a platin [such as cisplatin or oxaliplatin), gemcitabine, capecitabine, 5-FU, FOLFOX, FOLFIRI and FOLFIRINOX.
20. A compound of formula (I), as defined in claim 1, an enantiomer thereof or a pharmaceutically acceptable salt of any one of the same, for example wherein the compound of formula [I) is Varlitinib or a pharmaceutically acceptable salt thereof, for use in treating a cancer patient (for example a solid tumour) wherein the patient has been identified as having a beta-catenin pathway mutation (in particular a mutation in beta-catenin.
21. Use of a compound of formula (I), as defined in claim 1, an enantiomer thereof or a pharmaceutically acceptable salt of any one of the same, for example wherein the compound of formula (I) is Varlitinib or a pharmaceutically acceptable salt thereof, for the manufacture of a medicament for treating a cancer patient (for example a solid tumour) wherein the patient has been identified as having a beta-catenin pathway mutation (in particular a mutation in beta- catenin).
PCT/SG2018/050537 2017-10-25 2018-10-25 Varlitinib for use in treating cancer in a patient identified as having a beta-catenin pathway mutation WO2019083457A1 (en)

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
SG10201708782U 2017-10-25
SG10201708782U 2017-10-25
SG10201802791Q 2018-04-03
SG10201802791Q 2018-04-03
SG10201807632Q 2018-09-05
SG10201807632Q 2018-09-05
SG10201808393X 2018-09-26
SG10201808393X 2018-09-26

Publications (1)

Publication Number Publication Date
WO2019083457A1 true WO2019083457A1 (en) 2019-05-02

Family

ID=64607282

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/SG2018/050537 WO2019083457A1 (en) 2017-10-25 2018-10-25 Varlitinib for use in treating cancer in a patient identified as having a beta-catenin pathway mutation

Country Status (1)

Country Link
WO (1) WO2019083457A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10682353B2 (en) 2015-09-04 2020-06-16 Aslan Pharmaceuticals Pte Ltd Varlitinib for use in the treatment of resistant or refractory cancer
WO2021185234A1 (en) * 2020-03-16 2021-09-23 正大天晴药业集团股份有限公司 Combined pharmaceutical composition of compound as c-met kinase inhibitor and use thereof

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008058229A1 (en) 2006-11-08 2008-05-15 Bristol-Myers Squibb Company Pyridinone compounds
US7449464B2 (en) 2003-03-12 2008-11-11 Kudos Pharmaceuticals Limited Phthalazinone derivatives
WO2010038086A2 (en) 2008-10-02 2010-04-08 Respivert Limited Novel compounds
WO2010127417A2 (en) * 2009-05-05 2010-11-11 Katholieke Universiteit Leuven Hepatocellular carcinoma
US8071623B2 (en) 2007-01-10 2011-12-06 Instituto Di Ricerche Di Biologia Molecolare P. Angeletti Spa Amide substituted indazoles as poly(ADP-ribose)polymerase(PARP) inhibitors
EP2963114A1 (en) * 2013-02-27 2016-01-06 Daiichi Sankyo Company, Limited Method for predicting responsiveness to compound inhibiting mapk signal transduction pathway
WO2017037298A1 (en) * 2015-09-04 2017-03-09 Aslan Pharmaceuticals Pte Limited A combination therapy comprising varlitinib and an anticancer agent
WO2017184086A1 (en) * 2016-04-21 2017-10-26 Aslan Pharmaceuticals Pte. Ltd. Method of treating liver cancer

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7449464B2 (en) 2003-03-12 2008-11-11 Kudos Pharmaceuticals Limited Phthalazinone derivatives
WO2008058229A1 (en) 2006-11-08 2008-05-15 Bristol-Myers Squibb Company Pyridinone compounds
US8071623B2 (en) 2007-01-10 2011-12-06 Instituto Di Ricerche Di Biologia Molecolare P. Angeletti Spa Amide substituted indazoles as poly(ADP-ribose)polymerase(PARP) inhibitors
WO2010038086A2 (en) 2008-10-02 2010-04-08 Respivert Limited Novel compounds
WO2010127417A2 (en) * 2009-05-05 2010-11-11 Katholieke Universiteit Leuven Hepatocellular carcinoma
EP2963114A1 (en) * 2013-02-27 2016-01-06 Daiichi Sankyo Company, Limited Method for predicting responsiveness to compound inhibiting mapk signal transduction pathway
WO2017037298A1 (en) * 2015-09-04 2017-03-09 Aslan Pharmaceuticals Pte Limited A combination therapy comprising varlitinib and an anticancer agent
WO2017184086A1 (en) * 2016-04-21 2017-10-26 Aslan Pharmaceuticals Pte. Ltd. Method of treating liver cancer

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
"Novel pyrrole derivatives as selective CHK1 inhibitors: design, regioselective synthesis and molecular modelling", MED. CHEM. COMMUN., vol. 6, 2015, pages 852 - 859
DATABASE EMBASE [online] ELSEVIER SCIENCE PUBLISHERS, AMSTERDAM, NL; 1 July 2018 (2018-07-01), SHUEN W H ET AL: "Varlitinib demonstrates tumor regression and vessel normalization in ErbB-dependent and mutated beta-catenin hepatocellular carcinoma patient-derived xenograft model", XP002787978, Database accession no. EMB-623508221 *
KIM J ET AL: "664P:Phase IIa study to evaluate the biological activity of ASLAN001 in HER-1/2 co-expressing or HER-2 amplified advanced gastric cancer", vol. 25, no. Suppl 4, 1 September 2014 (2014-09-01), pages iv226, XP002764105, ISSN: 0923-7534, Retrieved from the Internet <URL:https://academic.oup.com/annonc/article/25/suppl_4/iv226/2241465/664PPHASE-IIA-STUDY-TO-EVALUATE-THE-BIOLOGICAL> DOI: 10.1093/ANNONC/MDU334.49 *
KONG, N.; FOTOUHI, N.; WOVKULICH, P.M.; ROBERTS, J: "Cell cycle inhibitors for the treatment of cancer", DRUGS FUT, vol. 28, no. 9, 2003, pages 881, XP002503863, DOI: doi:10.1358/dof.2003.028.09.761413
PLOS ONE, vol. 5, no. 8, 2010, pages e12214
SHUEN W H ET AL: "Varlitinib demonstrates tumor regression and vessel normalization in ErbB-dependent and mutated beta-catenin hepatocellular carcinoma patient-derived xenograft model", CANCER RESEARCH 20180701 AMERICAN ASSOCIATION FOR CANCER RESEARCH INC. NLD, vol. 78, no. 13, Supplement 1, 1 July 2018 (2018-07-01), ISSN: 1538-7445 *
SUSAN ELLARD ET AL: "Abstract #3603:ARRY-334543 in ErbB2 positive metastatic breast cancer and other ErbB expressing-cancers: experience from expansion cohorts on a phase I study", vol. 50, no. 9, 1 May 2009 (2009-05-01), pages 869 - 870, XP002764103, ISSN: 0197-016X, Retrieved from the Internet <URL:http://cancerres.aacrjournals.org/content/69/9_Supplement/3603> *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10682353B2 (en) 2015-09-04 2020-06-16 Aslan Pharmaceuticals Pte Ltd Varlitinib for use in the treatment of resistant or refractory cancer
US10849899B2 (en) 2015-09-04 2020-12-01 Aslan Pharmaceuticals Pte Ltd Combination therapy comprising Varlitinib and an anticancer agent
WO2021185234A1 (en) * 2020-03-16 2021-09-23 正大天晴药业集团股份有限公司 Combined pharmaceutical composition of compound as c-met kinase inhibitor and use thereof

Similar Documents

Publication Publication Date Title
WO2020150656A1 (en) Methods for assessing and treating cancer
Mazzoccoli et al. Biology, epidemiology, clinical aspects of hepatocellular carcinoma and the role of sorafenib
US10682353B2 (en) Varlitinib for use in the treatment of resistant or refractory cancer
Damia et al. The pharmacological point of view of resistance to therapy in tumors
US20220313700A1 (en) Methods for treating map3k8 positive cancers
US20210316003A1 (en) Biomarkers for sacituzumab govitecan therapy
Klapp et al. The DNA damage response and inflammation in cancer
Zhang et al. Molecular mechanisms of hepatoblastoma
WO2018136010A1 (en) Combination therapy
WO2019083457A1 (en) Varlitinib for use in treating cancer in a patient identified as having a beta-catenin pathway mutation
WO2019083456A1 (en) Varlitinib for use in treating cancer to normalise angiogenesis in a cancer mass
WO2019083458A1 (en) Varlitinib for use in treating cancer to reduce hypoxia
US20210251998A1 (en) Systems and methods for treating cancer
WO2019083455A1 (en) Varlitinib for use in treating cancer in a patient identified as having activated her1, her2 and/or her3 receptor cancerous cells
WO2013166366A1 (en) Cul4b as predictive biomarker for cancer treatment
US20220003770A1 (en) Classification and Treatment of Gastric Cancer
Ellis et al. Clinically actionable insights into initial and matched recurrent glioblastomas to inform novel treatment approaches
Gonçalves et al. Mechanisms of aggressiveness in glioblastoma: Prognostic and potential therapeutic insights
Udagawa et al. Targeted sequencing reveals genetic variants associated with sensitivity of 79 human cancer xenografts to anticancer drugs
Interpret et al. Panel Analysis: TruSight™ Oncology 500
US20180112258A1 (en) Medical uses and methods for treating cancer using monopolar spindle 1 (mps1) kinase inhibitors
TW201909916A (en) Use of C-MET inhibitors for the treatment of cancers with MET mutations
Sahu et al. Multidrug resistance, a major obstacle in hepatocellular carcinoma treatment: challenges and future perspectives
Sun et al. Rational Design of PARP1/c-Met Dual Inhibitors for Overcoming PARP1 Inhibitor Resistance Induced by c-Met Overexpression
Wang et al. Advancements in the Understanding of Small-Cell Neuroendocrine Cervical Cancer: Where We Stand and What Lies Ahead

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18815049

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18815049

Country of ref document: EP

Kind code of ref document: A1