WO2021001431A1 - Use of pi3ka-selective inhibitors for treating metastatic disease in patients suffering from pancreatic cancer - Google Patents

Abstract

Pancreatic ductal adenocarcinoma is a paradigmatic model of undetected micrometastatic disease. This clinical situation being poorly investigated, we devised a nove preclinical protocol using circulating cell-free DNA (cDNA) as a micrometastatic disease biomarker. Amongst actionable markers of disease progression, a novel PI3Kα activation signature specifically correlated with poor prognosis independently of patient tumour staging as confirmed in patient-derived early metastatic cultures. Tumour-restricted genetic or pharmacological PI3Kα inhibition reduced micro-metastatic disease by acting on both tumoural cell migratory behaviour independently of genetic alterations and on the protumoural CD206- positive stroma. PI3Kα therefore drives pro-inflammatory metastatic features that could be pharmacologically targeted to delay macro-metastatic dissemination. Thus the results herein disclosed justify that patients with high cDNA levels should be treated with PI3Kα-selective inhibitors.

Description

USE OF PI3KA-SELECTIVE INHIBITORS FOR TREATING METASTATIC DISEASE IN PATIENTS SUFFERING FROM PANCREATIC CANCER

FIELD OF THE INVENTION:

The present invention is in the field of oncology.

BACKGROUND OF THE INVENTION:

In humans, PI3Ks are composed of 8 isoforms distributed into 3 classes. Each class I PI3K dimers called RI3Ka, RI3Kb, RI3Kg and PI3K6 are composed of a catalytic subunit (pi 10a, pi 10b, pi lOy and p 11 Od) and a regulatory subunit (p85 for a, b and g, and pl01/p87 for g). While PI3Ka and RI3Kb are ubiquitously expressed, RI3Kg and RI3Kd are restricted to cardiovascular system and leukocytes in normal tissues but can be found overexpressed in solid tumours (1). PI3Ks are lipid kinases which phosphorylate phosphatidyl inositol 4,5 - biphosphate (PIP2) into PIP3 that acts as a second messenger and regulates various functions in normal and tumour cells via the PI3K/Akt/mTOR pathway. The PI3K/Akt axis is frequently hyper-activated in cancers and has been tested as a clinical target in the recent years (2,3). PI3K inhibitors are currently described as cytostatic agents, since PI3K activity is critically driving oncogenesis in a cell-autonomous manner. However, the clinical importance of other cell functions regulated by this pathway in a non-cancer cell autonomous manner, in particular on myeloid-derived suppressor cells (MDSC) or macrophages, has been underestimated. Direct inactivation of this pathway in tumoural macrophages was recently shown as a good target to prevent solid tumour progression (4). It is unclear if tumoural PI3K activity by itself contributes to the rewiring of tumoural immune microenvironment. This demonstration could pave the road for the use of PI3K-targeted therapies in combination with immunotherapies and/or chemotherapies (3,5).

Pancreatic ductal adenocarcinoma (PDAC) is a cancer with dramatic prognostic (6) where activation of class I PI3K is high and linked with a poor prognostic (7). Localised, locally advanced and metastatic PDAC is characterised by early relapse of surgery and failure of long term disease control with chemotherapies. Molecular characterisation of large cohorts of PDAC patients failed to identify actionable pathways associated with mutations except for oncogenic Kras, found in more than 80% of all patients. There are many altered signalling pathways downstream oncogenic Kras such as mitogen-activated protein kinase (MAPK) and PI3K pathways, but also Braf signalling, transforming growth factor (TGFP), Notch and DNA repair pathways and among them, the PI3K/Akt is one of the most critically affected (8). The lipid kinase PI3Ka was shown by us and others to drive initiation of pancreatic cancer downstream of oncogenic Kras (9, 10). However, little is known concerning the importance of this PI3K isoform in the progression of existing tumours towards a metastatic disease.

Cell free DNA (cfDNA) and more precisely circulating tumour DNA (ctDNA) appears in clinical oncology as an attractive biomarker for early cancer detection, diagnosis, prognosis, predictive factor of relapse to therapies (11). In cancer patients, ctDNA represents a variable fraction of cfDNA (12) and can be distinguished by the presence of specific cancer-associated mutations. The exact biological mechanisms underlying the release of cfDNA remain unclear: it can be due to apoptosis and necrosis of cancer cells (or healthy cells), and it can be secreted directly by tumour or micro-environment cells such as immune and inflammatory cells (13). Response to systemic therapy has been associated with low or undetectable ctDNA levels in several cancer types, suggesting that monitoring ctDNA could help to predict response or resistance to therapies (14). In neo-adjuvant setting, a clinical setting where maximal tumour shrinkage is needed in a curative attempt, signal-targeting therapies need to be developed especially in the aim to stop tumour cell invasion and micrometastatic dissemination. Since cfDNA had been studied as an exploratory biomarker of micrometastatic disease in PD AC (15), we propose that cfDNA as a sign of micro-metastatic disease could predict signal targeted efficiency towards metastatic dissemination and have chosen pancreatic ductal adenocarcinoma (PD AC) as a paradigmatic model of inflammatory micrometastatic disease to demonstrate our hypothesis.

SUMMARY OF THE INVENTION:

As defined by the claims, the present invention relates to methods for treating metastatic disease in patients suffering from pancreatic cancer.

DETAILED DESCRIPTION OF THE INVENTION:

Pancreatic ductal adenocarcinoma is a paradigmatic model of undetected micrometastatic disease. This clinical situation being poorly investigated, we devised a novel preclinical protocol using circulating cell-free DNA (cDNA) as a micrometastatic disease biomarker. Amongst actionable markers of disease progression, a novel PI3Ka activation signature specifically correlated with poor prognosis independently of patient tumour staging as confirmed in patient-derived early metastatic cultures. Tumour-restricted genetic or pharmacological RI3Ka inhibition reduced micro-metastatic disease by acting on both tumoural cell migratory behaviour independently of genetic alterations and on the protumoural CD206- positive stroma. RI3Ka therefore drives pro-inflammatory metastatic features that could be pharmacologically targeted to delay macro-metastatic dissemination. Thus the results herein disclosed justify that patients with high cDNA levels should be treated with PI3Ka-selective inhibitors.

The first object of the present invention relates to a method of treating micrometastatic disease in a patient suffering from pancreatic cancer comprising administering to the patient a therapeutically effective amount of a PI3Ka- selective inhibitor.

As used herein the term "pancreatic cancer" or "pancreas cancer" as used herein relates to cancer which is derived from pancreatic cells. In particular, pancreatic cancer included pancreatic adenocarcinoma (e.g., pancreatic ductal adenocarcinoma) as well as other tumors of the exocrine pancreas (e.g., serous cystadenomas), acinar cell cancers, and intraductal papillary mucinous neoplasms (IPMN).

In some embodiments, the pancreatic cancer is KRAS mutated. As used herein, "KRAS" refers to v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog. KRAS is also known in the art as NS3, KRAS1, KRAS2, RASK2, KI-RAS, C-K-RAS, K-RAS2A, K-RAS2B, K-RAS4A and K-RAS4B. This gene, a Kirsten ras oncogene homolog from the mammalian ras gene family, encodes a protein that is a member of the small GTPase superfamily. A single amino acid substitution can be responsible for an activating mutation. The transforming protein that results can be implicated in various malignancies, including lung cancer, colon cancer and pancreas cancer. KRAS mutations are well known in the art and are frequently found in neoplasms include those at exon 1 (codons 12 and 13) and exon 2 (codon 61) (e.g., the 34A, 34C, 34T, 35 A, 35C, 35T or 38A mutations). Other examples of KRAS mutations include, but are not limited to, G12C, G12D, G13D, G12R, G12S, and G12V. Somatic KRAS mutations are found at high rates in leukemias, colorectal cancer (Burmer et al. Proc. Natl. Acad. Sci. 1989 86: 2403- 7), pancreatic cancer (Almoguera et al. Cell 1988 53 : 549-54) and lung cancer (Tam et al. Clin. Cancer Res. 2006 12: 1647-53). Methods for identifying KRAS mutations are well known in the art and are commercially available (e.g. In Therascreen (Qiagen) assay, Taqman® Mutation Detection Assays powered by castPCR™ technology (Life Technologies)).

As used herein, the term“metastasis” has its general leaning in the art and refers to the condition of spread of cancer from the organ or tissue of origin to additional distal sites in the patient. The process of tumor metastasis is a multistage event involving local invasion and destruction of intracellular matrix, intravasation into blood vessels, lymphatics or other channels of transport, survival in the circulation, extravasation out of the vessels into secondary sites and growth in the new location(s). Increased malignant cell motility has been associated with enhanced metastatic potential in animal as well as human tumors. A used herein, the term “micrometastatic disease” has its general meaning in the art and refers to a locally invasive cancer from the organ or tissue of origin, for example, to proximal tissues or sentinel lymph nodes. Therefore the RI3Ka- selective inhibitor is thus particularly suitable for reducing metastatic dissemination and progression.

As used herein, the term "treatment" or "treat" refer to both prophylactic or preventive treatment as well as curative or disease modifying treatment, including treatment of subject at risk of contracting the disease or suspected to have contracted the disease as well as subjects who are ill or have been diagnosed as suffering from a disease or medical condition, and includes suppression of clinical relapse. The treatment may be administered to a subject having a medical disorder or who ultimately may acquire the disorder, in order to prevent, cure, delay the onset of, reduce the severity of, or ameliorate one or more symptoms of a disorder or recurring disorder, or in order to prolong the survival of a subject beyond that expected in the absence of such treatment. By "therapeutic regimen" is meant the pattern of treatment of an illness, e.g., the pattern of dosing used during therapy. A therapeutic regimen may include an induction regimen and a maintenance regimen. The phrase "induction regimen" or "induction period" refers to a therapeutic regimen (or the portion of a therapeutic regimen) that is used for the initial treatment of a disease. The general goal of an induction regimen is to provide a high level of drug to a subject during the initial period of a treatment regimen. An induction regimen may employ (in part or in whole) a "loading regimen", which may include administering a greater dose of the drug than a physician would employ during a maintenance regimen, administering a drug more frequently than a physician would administer the drug during a maintenance regimen, or both. The phrase "maintenance regimen" or "maintenance period" refers to a therapeutic regimen (or the portion of a therapeutic regimen) that is used for the maintenance of a subject during treatment of an illness, e.g., to keep the patient in remission for long periods of time (months or years). A maintenance regimen may employ continuous therapy (e.g., administering a drug at a regular intervals, e.g., weekly, monthly, yearly, etc.) or intermittent therapy (e.g., interrupted treatment, intermittent treatment, treatment at relapse, or treatment upon achievement of a particular predetermined criteria [e.g., disease manifestation, etc.]).

As used herein, the term“PI3K” has its general meaning in the art and refers to a phosphoinositide 3-kinase. PI3Ks belong to a large family of lipid signaling kinases that phosphorylate phosphoinositides at the D3 position of the inositol ring (Cantley, Science, 2002, 296(5573): 1655-7). PI3Ks are divided into three classes (class I, II, and III) according to their structure, regulation and substrate specificity. Class I PI3Ks, which include RI3Ka, RI3Kb, RI3Kg, and PI3K5, are a family of dual specificity lipid and protein kinases that catalyze the phosphorylation of phosphatidylinosito-4,5-bisphosphate (PIP2) giving rise to phosphatidylinosito-3,4,5-trisphosphate (PIP3). PIP3 functions as a second messenger that controls a number of cellular processes, including growth, survival, adhesion and migration. All four class I PI3K isoforms exist as heterodimers composed of a catalytic subunit (pi 10) and a tightly associated regulatory subunit that controls their expression, activation, and subcellular localization. RI3Ka, RI3Kb, and PI3K5 associate with a regulatory subunit known as p85 and are activated by growth factors and cytokines through a tyrosine kinase-dependent mechanism (Jimenez, et al, J Biol Chem., 2002, 277(44):41556-62) whereas RI3Kg associates with two regulatory subunits (plOl and p84) and its activation is driven by the activation of G-protein- coupled receptors (Brock, et al., J Cell Biol., 2003, 160(1): 89-99).

Non-limiting examples of PI3Ka- selective inhibitors are disclosed in Schmidt-Kittler et al, Oncotarget (2010) l(5):339-348; Wu et al., Med. Chem. Comm. (2012) 3 :659-662; Hayakawa et al, Bioorg. Med. Chem. (2007) 15(17): 5837-5844; and PCT Patent Application Nos. WO2013/049581 and WO2012/052745, the contents of which are herein incorporated by reference in their entireties. In particular non-limiting embodiments, the PI3Ka- selective inhibitor is derived from imidazopyridine or 2-aminothiazole compounds. Further non-limiting examples include those described in William A Denny (2013) Phosphoinositide 3-kinase a inhibitors: a patent review, Expert Opinion on Therapeutic Patents, 23:7, 789-799. Further non limiting examples include BYL719, INK-11 14, INK-1117, NVP-BYL719 (Alpelisib), SRX2523, LY294002, PIK-75, PKI-587, A66, CH5132799 and GDC-0032 (taselisib). One inhibitor suitable for the present invention is the compound 5-(2,6-di-morpholin-4-yl- pyrimidin-4-yl)-4-trifluoromethyl-pyridin-2-ylamine that is described in W02007/084786, which is hereby incorporated by reference in its entirety hereto. Another inhibitor suitable for the present invention is the compound (S)-Pyrrolidine-l,2-dicarboxylic acid 2-amide l-({4- methyl-5-[2-(2,2,2-trifluoro-l,l-dimethyl-ethyl)-pyridin-4-yl]-thiazol-2-yl}-amide) that is described in WO 2010/029082, which is hereby incorporated by reference in its entirety hereto.

In some embodiments, the PI3Ka- selective inhibitor is an inhibitor of PI3Ka expression. An“inhibitor of expression” refers to a natural or synthetic compound that has a biological effect to inhibit the expression of a gene. In some embodiments, said inhibitor of gene expression is a siRNA, an antisense oligonucleotide or a ribozyme. For example, anti-sense oligonucleotides, including anti-sense RNA molecules and anti-sense DNA molecules, would act to directly block the translation of PI3KA mRNA by binding thereto and thus preventing protein translation or increasing mRNA degradation, thus decreasing the level of PI3KA, and thus activity, in a cell. For example, antisense oligonucleotides of at least about 15 bases and complementary to unique regions of the mRNA transcript sequence encoding PI3KA can be synthesized, e.g., by conventional phosphodiester techniques. Methods for using antisense techniques for specifically inhibiting gene expression of genes whose sequence is known are well known in the art (e.g. see U.S. Pat. Nos. 6,566,135; 6,566,131; 6,365,354; 6,410,323; 6, 107,091; 6,046,321; and 5,981,732). Small inhibitory RNAs (siRNAs) can also function as inhibitors of expression for use in the present invention. PI3KA gene expression can be reduced by contacting a patient or cell with a small double stranded RNA (dsRNA), or a vector or construct causing the production of a small double stranded RNA, such that PI3KA gene expression is specifically inhibited (i.e. RNA interference or RNAi). Antisense oligonucleotides, siRNAs, shRNAs and ribozymes of the invention may be delivered in vivo alone or in association with a vector. In its broadest sense, a "vector" is any vehicle capable of facilitating the transfer of the antisense oligonucleotide, siRNA, shRNA or ribozyme nucleic acid to the cells and typically cells expressing PI3KA. Typically, the vector transports the nucleic acid to cells with reduced degradation relative to the extent of degradation that would result in the absence of the vector. In general, the vectors useful in the invention include, but are not limited to, plasmids, phagemids, viruses, other vehicles derived from viral or bacterial sources that have been manipulated by the insertion or incorporation of the antisense oligonucleotide, siRNA, shRNA or ribozyme nucleic acid sequences. Viral vectors are a preferred type of vector and include, but are not limited to nucleic acid sequences from the following viruses: retrovirus, such as moloney murine leukemia virus, harvey murine sarcoma virus, murine mammary tumor virus, and rous sarcoma virus; adenovirus, adeno-associated virus; SV40-type viruses; polyoma viruses; Epstein-Barr viruses; papilloma viruses; herpes virus; vaccinia virus; polio virus; and RNA virus such as a retrovirus. One can readily employ other vectors not named but known to the art. In some embodiments, the inhibitor of expression is an endonuclease. In a particular embodiment, the endonuclease is CRISPR-cas. In some embodiment, the endonuclease is CRISPR-cas9, which is from Streptococcus pyogenes. The CRISPR/Cas9 system has been described in US 8697359 B1 and US 2014/0068797. In some embodiment, the endonuclease is CRISPR-Cpfl, which is the more recently characterized CRISPR from Provotella and Francisella 1 (Cpfl) in Zetsche et al. (“Cpfl is a Single RNA- guided Endonuclease of a Class 2 CRISPR-Cas System (2015); Cell; 163, 1-13). As evidenced in the EXAMPLE of the present specification, the micrometastatic disease correlated with high cfDNA levels in the patients. Therefore qualifying the cfDNA level in sample obtained from the patient is suitable for determining whether the patient is eligible to the treatment with the RBKa-selective inhibitor. Thus in some embodiments, the method of the present invention comprises the steps of i) quantifying the cfDNA level in a sample obtained from the patient ii) comparing said level with a predetermined reference level and iii) administering to the patient the therapeutically effective amount of the a PI3Ka-selective inhibitor when the level determined at step i) is higher than the predetermined reference value.

As used herein, the term“cell free DNA” or“cfDNA” has its general meaning in the art and refers to the DNA nucleic acid is released by the cell and present in the sample. Methods for determining the total concentration of cell free nucleic acids are well known in the art. For example, the method is described in WO2012/028746. Q-PCR is thus the preferred method for determining said concentration. In some embodiment, the method comprises the step of amplifying and quantifying a nuclear target nucleic acid sequence According to the invention, the nuclear target nucleic acid sequence is a sequence which is located in the nucleus human genome. The skilled person can thus easily select the appropriate nuclear target nucleic acid sequences. Cell free nucleic acid in a patient suffering from a cancer is constituted of nucleic acids of tumor and non-tumor origin. Thus in some embodiments, the nuclear target nucleic sequence is a mutant target nucleic acid sequence. As used herein the term“mutant nucleic acid” refers to a nucleic acid bearing a point mutation of interest. It is thus important to select a mutation which has a tumor origin to quantify only the nucleic acids which derives from cancer cells. In some embodiments, the mutation is located in a the KRAS gene or TP53 gene., For example, the mutation is located in a gene selected from the group consisting of TP53 (394, 395, 451, 453, 455, 469, 517, 524, 527, 530, 586, 590, 637, 641, 724, 733, 734, 743, 744, 817, 818, 819, 820, 839, 844, 916) or PIK3CA (1530, 1624, 1633, 1634, 1636, 1656, 3140, 3140, 3140). KRAS mutation may include any mutation as described above. Typically, the target nucleic acid sequences have a length between 160 and 210 base pairs. In some embodiments, the target nucleic acid sequence has a length of 160; 165; 170; 175; 180; 185; 190; 195; 200; 205; or 210 base pairs.

As used herein the term“sample” refers to any biological sample obtained from the subject that is liable to contain cell free nucleic acids. Typically, samples include but are not limited to body fluid samples, such as blood, ascite, urine, amniotic fluid, feces, saliva or cerebrospinal fluids. In some embodiments, the sample is a blood sample. By "blood sample" it is meant a volume of whole blood or fraction thereof, e.g., serum, plasma, etc. Any methods well known in the art may be used by the skilled artisan in the art for extracting the free cell nucleic acid from the prepared sample.

A "therapeutically effective amount" refers to an amount effective, at dosages and for periods of time necessary, to achieve a desired therapeutic result. A therapeutically effective amount of drug may vary according to factors such as the disease state, age, sex, and weight of the individual, and the ability of drug to elicit a desired response in the individual. A therapeutically effective amount is also one in which any toxic or detrimental effects of the antibody or antibody portion are outweighed by the therapeutically beneficial effects. The efficient dosages and dosage regimens for drug depend on the disease or condition to be treated and may be determined by the persons skilled in the art. A physician having ordinary skill in the art may readily determine and prescribe the effective amount of the pharmaceutical composition required. For example, the physician could start doses of drug employed in the pharmaceutical composition at levels lower than that required in order to achieve the desired therapeutic effect and gradually increase the dosage until the desired effect is achieved. In general, a suitable dose of a composition of the present invention will be that amount of the compound which is the lowest dose effective to produce a therapeutic effect according to a particular dosage regimen. Such an effective dose will generally depend upon the factors described above. For example, a therapeutically effective amount for therapeutic use may be measured by its ability to stabilize the progression of disease. A therapeutically effective amount of a therapeutic compound may decrease tumor size, or otherwise ameliorate symptoms in a subject. One of ordinary skill in the art would be able to determine such amounts based on such factors as the patient's size, the severity of the patient's symptoms, and the particular composition or route of administration selected. An exemplary, non-limiting range for a therapeutically effective amount of drug is about 0.1-100 mg/kg, such as about 0.1-50 mg/kg, for example about 0.1-20 mg/kg, such as about 0.1-10 mg/kg, for instance about 0.5, about such as 0.3, about 1, about 3 mg/kg, about 5 mg/kg or about 8 mg/kg. An exemplary, non-limiting range for a therapeutically effective amount of an antibody of the present invention is 0.02-100 mg/kg, such as about 0.02-30 mg/kg, such as about 0.05-10 mg/kg or 0.1-3 mg/kg, for example about 0.5-2 mg/kg. Administration may e.g. be intravenous, intramuscular, intraperitoneal, or subcutaneous, and for instance administered proximal to the site of the target. Dosage regimens in the above methods of treatment and uses are adjusted to provide the optimum desired response (e.g., a therapeutic response). For example, a single bolus may be administered, several divided doses may be administered overtime or the dose may be proportionally reduced or increased as indicated by the exigencies of the therapeutic situation. In some embodiments, the efficacy of the treatment is monitored during the therapy, e.g. at predefined points in time. As non-limiting examples, treatment according to the present invention may be provided as a daily dosage of the agent of the present invention in an amount of about 0.1-100 mg/kg, such as 0.2, 0.5, 0.9, 1.0, 1.1, 1.5, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1 1, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 40, 45, 50, 60, 70, 80, 90 or 100 mg/kg, per day, on at least one of days 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, or 40, or alternatively, at least one of weeks 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1 1, 12, 13, 14, 15, 16, 17, 18, 19 or 20 after initiation of treatment, or any combination thereof, using single or divided doses every 24, 12, 8, 6, 4, or 2 hours, or any combination thereof.

Typically, the RI3Ka- selective inhibitor is administered to the patient in the form of a pharmaceutical composition which comprises a pharmaceutically acceptable carrier. Pharmaceutically acceptable carriers that may be used in these compositions include, but are not limited to, ion exchangers, alumina, aluminum stearate, lecithin, serum proteins, such as human serum albumin, buffer substances such as phosphates, glycine, sorbic acid, potassium sorbate, partial glyceride mixtures of saturated vegetable fatty acids, water, salts or electrolytes, such as protamine sulfate, di sodium hydrogen phosphate, potassium hydrogen phosphate, sodium chloride, zinc salts, colloidal silica, magnesium trisilicate, polyvinyl pyrrolidone, cellulose-based substances, polyethylene glycol, sodium carboxymethylcellulose, polyacrylates, waxes, polyethylene-polyoxypropylene- block polymers, polyethylene glycol and wool fat. For use in administration to a subject, the composition will be formulated for administration to the patient. The compositions of the present invention may be administered orally, parenterally, by inhalation spray, topically, rectally, nasally, buccally, vaginally or via an implanted reservoir. The used herein includes subcutaneous, intravenous, intramuscular, intra-articular, intra- synovial, intrasternal, intrathecal, intrahepatic, intralesional and intracranial injection or infusion techniques. Sterile injectable forms of the compositions of this invention may be aqueous or an oleaginous suspension. These suspensions may be formulated according to techniques known in the art using suitable dispersing or wetting agents and suspending agents. The sterile injectable preparation may also be a sterile injectable solution or suspension in a non-toxic parenterally acceptable diluent or solvent, for example as a solution in 1,3-butanediol. Among the acceptable vehicles and solvents that may be employed are water, Ringer's solution and isotonic sodium chloride solution. In addition, sterile, fixed oils are conventionally employed as a solvent or suspending medium. For this purpose, any bland fixed oil may be employed including synthetic mono-or diglycerides. Fatty acids, such as oleic acid and its glyceride derivatives are useful in the preparation of injectables, as are natural pharmaceutically-acceptable oils, such as olive oil or castor oil, especially in their polyoxyethylated versions. These oil solutions or suspensions may also contain a long-chain alcohol diluent or dispersant, such as carboxymethyl cellulose or similar dispersing agents that are commonly used in the formulation of pharmaceutically acceptable dosage forms including emulsions and suspensions. Other commonly used surfactants, such as Tweens, Spans and other emulsifying agents or bioavailability enhancers which are commonly used in the manufacture of pharmaceutically acceptable solid, liquid, or other dosage forms may also be used for the purposes of formulation. The compositions of this invention may be orally administered in any orally acceptable dosage form including, but not limited to, capsules, tablets, aqueous suspensions or solutions. In the case of tablets for oral use, carriers commonly used include lactose and corn starch. Lubricating agents, such as magnesium stearate, are also typically added. For oral administration in a capsule form, useful diluents include, e.g., lactose. When aqueous suspensions are required for oral use, the active ingredient is combined with emulsifying and suspending agents. If desired, certain sweetening, flavoring or coloring agents may also be added. Alternatively, the compositions of this invention may be administered in the form of suppositories for rectal administration. These can be prepared by mixing the agent with a suitable non-irritating excipient that is solid at room temperature but liquid at rectal temperature and therefore will melt in the rectum to release the drug. Such materials include cocoa butter, beeswax and polyethylene glycols. The compositions of this invention may also be administered topically, especially when the target of treatment includes areas or organs readily accessible by topical application, including diseases of the eye, the skin, or the lower intestinal tract. Suitable topical formulations are readily prepared for each of these areas or organs. For topical applications, the compositions may be formulated in a suitable ointment containing the active component suspended or dissolved in one or more carriers. Carriers for topical administration of the compounds of this invention include, but are not limited to, mineral oil, liquid petrolatum, white petrolatum, propylene glycol, polyoxyethylene, polyoxypropylene compound, emulsifying wax and water. Alternatively, the compositions can be formulated in a suitable lotion or cream containing the active components suspended or dissolved in one or more pharmaceutically acceptable carriers. Suitable carriers include, but are not limited to, mineral oil, sorbitan monostearate, polysorbate 60, cetyl esters wax, cetearyl alcohol, 2- octyldodecanol, benzyl alcohol and water. Topical application for the lower intestinal tract can be effected in a rectal suppository formulation (see above) or in a suitable enema formulation. Patches may also be used. The compositions of this invention may also be administered by nasal aerosol or inhalation. Such compositions are prepared according to techniques well- known in the art of pharmaceutical formulation and may be prepared as solutions in saline, employing benzyl alcohol or other suitable preservatives, absorption promoters to enhance bioavailability, fluorocarbons, and/or other conventional solubilizing or dispersing agents. For example, an antibody present in a pharmaceutical composition of this invention can be supplied at a concentration of 10 mg/mL in either 100 mg (10 mL) or 500 mg (50 mL) single-use vials. The product is formulated for IV administration in 9.0 mg/mL sodium chloride, 7.35 mg/mL sodium citrate dihydrate, 0.7 mg/mL polysorbate 80, and Sterile Water for Injection. The pH is adjusted to 6.5. An exemplary suitable dosage range for an antibody in a pharmaceutical composition of this invention may between about 1 mg/m² and 500 mg/m². However, it will be appreciated that these schedules are exemplary and that an optimal schedule and regimen can be adapted taking into account the affinity and tolerability of the particular antibody in the pharmaceutical composition that must be determined in clinical trials. A pharmaceutical composition of the invention for injection (e.g., intramuscular, i.v.) could be prepared to contain sterile buffered water (e.g. 1 ml for intramuscular), and between about 1 ng to about 100 mg, e.g. about 50 ng to about 30 mg or more preferably, about 5 mg to about 25 mg, of the inhibitor of the invention.

A further object of the present invention relates to a method of determining whether a patient achieve a response with a PI3Ka-selective inhibitor that is used for the treatment of the micrometastatic disease in patient suffering from pancreatic cancer comprising determining the cfDNA level in a sample obtained from the patient during the course of the treatment wherein a decrease in said level indicates that the patient achieves a response or wherein a stable level or an increase level indicates that the patient does not achieve a response.

The method is thus particularly suitable for discriminating responder from non responder. As used herein the term“responder” in the context of the present disclosure refers to a patient that will achieve a response, i.e. a patient where the micrometastatic disease is eradicated, reduced or improved. According to the invention, the responders have an objective response and therefore the term does not encompass patients having a stabilized cancer such that the disease is not progressing after the therapy with the PI3Ka- selective inhibitor. A non responder or refractory patient includes patients for whom the micrometastatic disease does not show reduction or improvement after the therapy with the PI3Ka- selective inhibitor. According to the invention the term“non-responder” also includes patients having a stabilized cancer. Typically, the characterization of the patient as a responder or non-responder can be performed by reference to a standard or a training set. The standard may be the profile of a patient who is known to be a responder or non-responder or alternatively may be a numerical value. Such predetermined standards may be provided in any suitable form, such as a printed list or diagram, computer software program, or other media. When it is concluded that the patient is a non responder, the physician could take the decision to stop the therapy to avoid any further adverse sides effects or to adjust the dose for improving the response.

The invention will be further illustrated by the following figures and examples. However, these examples and figures should not be interpreted in any way as limiting the scope of the present invention.

FIGURES:

Figure 1: PI3Ka inhibition delays the rapid progression of PD AC in KPC mice. A,

Schematic representation of drug dosage: KPC mice diagnosed with aggressive carcinoma (6 mice per group) were given daily oral doses of vehicle or BYL-719 (50 mg/kg) for 10 days. BYL-719 reduces the phosphorylation of Akt and ERK in pancreatic tissues as analysed by WB in whole tissue lysates 6h after gavage. B, Representative images and C, quantification of Ki67 positive cells in pancreatic tumours and in metastasized liver sections. D, Evaluation of PI3Ka inhibition on tumour growth rate. E, Development of ascites in KPC mice. F, Percentage of micro metastases in liver, lung and spleen in KPC mice. G, Survival curve of KPC mice presenting different levels of cfDNA (n=43, Logrank test: p=0.0257). H, Quantification of cfDNA in KPC mice at different stages of the disease. I, Quantification of cfDNA in KPC mice after PI3Ka inhibition. J, Quantification of the 160 - 210bp fragment in KPC mice. Mean +/- SEM (* p<0.05, ** p<0.005, *** pO.0001)

EXAMPLE:

Material & Methods

Human pancreatic samples

Human pancreatitis sample were collected according French and European legislation, and stored accordingly in CRB, Toulouse, IUCT-O.

Transcriptomics and bioinformatic analysis

Amongst the publicly available micro-array repositories, we selected transcriptional profiling datasets of normal pancreas, chronic pancreatitis and pancreatic primary tumoural tissues from localised or metastatic patients. Published data on human samples were retrieved from public databases E EMBL 6 (40) from compatible platforms, normalized using RMA method (R 3.2.3, bioconductor version 3.2), collapsed (collapse microarray), filtered (SD>0.25), and statistically tested using an ANOVA test corrected with Benjamini & Hochberg method (BH). For each sample, individual scoring for Hallmarks or Reactome (actualised list of genes downloaded from MSigDB version (software. broadinstitute. org/gsea/msigdb) and reactome (www.reactome.org)) was performed using the Autocompare SES software (available at https://sites.***.com/site/fredsoftwares/products/autocompare_ses) using the “greater” (indicating an enriched gene set) Wilcoxon tests with frequency-corrected null hypotheses (41), then values in each group of patients compared using an ANOVA test. PI3Ka activation signature was designed as the intersection of genes up- and down-regulated by shRNA against PIK3CA in a human PIK3CA mutated breast cancer cell line as well as PI3ka_human_LINCS_CMAP and PBKa human mTOR CMAP LINCS gene signatures (42,43). Hierarchical clustering between patients was performed using PI3Ka activation signature. Confirmed PD AC samples from public data bases were selected for further bioinformatical analysis. In details, mRNA expression data and clinical data from confirmed PDAC patients of PAAD (TCGA) (175 patients), PACA-AU (267 patients) and GSE21501 (102 patients - cohort enriched in locally advanced PDAC) cohorts were retrieved. Amongst the 175 well annotated TGCA patients, 21 patients were considered as localised according to their UICC staging (T=0, 1 or 2, N=0, M=0). For each patient, a PI3Ka activation signature score was given using SES auto compare software, and patients were hierarchically clusterised in 3 groups corresponding to high, medium or low scoring. Overall survival of patients in each cluster was plotted and statistical differences calculated using log rank test. Inhibitors and chemotherapies

All PI3K inhibitors were purchased (Clinisciences) and dissolved in DMSO to obtain a final concentration of 10 mM. PI3K inhibitors we stored in 10 mM stock solution in dimethyl sulfoxide (DSMO):

Table

vitro on recombinant proteins of all the tested PI3K inhibitors for pi 10a, rΐΐqb, rΐΐqd, rΐΐqg and mTOR. NA: not available. Cell culture

R211, PDAC8661, DT4994, 10158, 10593, R6344, R6430, R6065, R6141 are murine pancreatic cell lines obtained from KPC mice. PANC-1 (ATCC CRL-1469), Capan-2 (ATCC HTB-80) and AsPC-1 (ATCC CRL-1682) are human pancreatic cell lines. HL-60 (ATCC CCL- 240) and NOMO-1 (CVCL 1609) are acute myeloid leukaemia cell lines. PC3 (ATCC CRL- 1435) is a prostate cancer cell line. MDA-MB-231 (HTB-26) and MDA-MB-468 (HTB-132) are breast cancer cell lines. All the cell lines were obtained from the American Type Culture Collection (ATCC, Manassas, VA).

R211, PDAC8661, DT4994, 10158, 10593, R6344, R6430, R6065, R6141, PANC-1, MDA-MB-231 and MDA-MB-468 were cultured in DMEM (Dulbecco's Modified Eagle Medium) 4.5g of glucose supplemented with 10% foetal bovine serum, 1% L-Glutamine, 1% penicillin/streptomycin and 0.01% plasmocin. Capan-2, AsPC-1, HL-60, NOMO-1 and PC3 were cultured in RPMI (Roswell Park Memorial Institute) supplemented as described previously. Cells were maintained in culture at 37°C in a humidified 5% CO2 atmosphere.

Table 2. Cell ines and associated mutations. AML: Acute myeloid leukaemia; WT:

Wild type.

Cytotoxicity assay PANC-1, R6344, R6430, R6065, R6141, MDA-MB-231, MDA-MB-468, PC3 (2.5xl0³), R211 (2xl0³), PDAC8661 (2xl0³), DT4994 (2xl0³) 10158 (6xl0²) and 10593 (6xl0²), cells were seeded in 96-well plates. Twenty-four hours later, cells were treated with dose range of a-selective (A66, BYL719), b-selective (TGX-221, TGX-155), b/d-selective (TGX-115, AZD8186), g-selective (AS252424) or pan PI3K inhibitors (LY294002, BKM120, GDC0941, PI- 103) at 0.1, 1 and 10 mM in complete medium. Three days after treatment, living cells were coloured by a colorimetric assay using MTT (3-(4, 5-dimethylthiazolyl-2)-2, 5- diphenyltetrazolium bromide) (Euromedex) for adherent cell lines or MTS (3-(4,5- dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) for acute myeloid cell lines. Control cells were treated with 0.1% DMSO and considered as 100% living cells. Each condition was done in triplicates and each experiment was performed at least three times.

Motility assay

R211 and PDAC8661 cells (lxlO⁵ cells) were seeded in 24-well plates. When cells reached confluence, a scratch was realised using a pipette tip and cells were treated or not with a-selective (A66, BYL719) and pan-PI3K inhibitors (LY294002, BKM120, GDC0941) at 0.01, 0.1 or 1 pM. Cells were observed after 8h and 24h by microscopy and the scratch surface was analysed using ImageJ software. Control cells were treated with 0.01% DMSO. 5 pictures were taken by condition and each experiment was performed at least three times.

Migration assay

R211, PDAC8661, PANC-1, 10158, 10593, R6344, R6430, R6065, R6141, MDA-MB- 231, MDA-MB-468 and PC3 cells (5xl0⁴ cells) were seeded in the upper compartment of Boyden chamber inserts in serum-free DMEM. Four hours after seeding, complete DMEM was added in the lower compartment to trigger migration. Cells were treated with a-selective (A66, BYL719), b- selective (TGX-221, TGX-155), b/d-selective (TGX-115, AZD8186), g-selective (AS252424) and pan-PI3K inhibitors (LY294002, BKM120, GDC0941, PI-103) at 0.1 and 1 pM in the lower and upper compartment. Twenty-four hours after treatment, the medium was aspirated and the inserts were fixed with formaldehyde 3.7% during 20 minutes then washed three times with EEO. Cells were stained with crystal violet during 20 minutes and washed three times with EEO. A cotton swab was used to remove non-migrated cells from the insert. Migrated cells were analysed using ImageJ software. Control cells were treated with 0.01% DMSO. Ten pictures per condition were taken and each experiment was performed at least three times. siRNA

Cells were transfected with Lipofectamine 2000 (ThermoFisher Scientic) transfection reagent in OptiMEM medium with SMARTpool ON-TARGETplus mouse siRNA (Dharmacon) targeting: Pik3ca, Pik3cb, Pik3cg, Pik3cd according to the manufacturer’s protocols. ON-TARGETplus Non-targeting control siRNAs (Dharmacon) were used as negative controls. Twenty-four hours after transfection, cells were used for migration or qPCR experiments.

Western blot

Cells were washed twice with cold PBS and scraped in cold lysis buffer (50mM Tris at pH 7.4, 150 mM NaCl, Triton 1%, 1 mM EDTAMgC12, 2 mM DTT, 2 mM NaF, 4 mM Na orthovanadate and a protease inhibitors cocktail from Roche). The concentration of proteins was determined using the BCA protein assay kit (Interchim). Fifty micrograms of proteins were separated by SDS-PAGE in an 8% or 12% polyacrylamide gel and transferred on a PVDF membrane with the Trans-Blot Turbo transfer system (Bio-Rad). Membranes were then saturated 45 minutes in TBS (50 mM Tris, 150 mM NaCl)/0.1% Tween 20 (TBST)/5% milk and incubated overnight at 4°C under agitation with the corresponding primary antibody (see below). Membranes were washed 3 times with TBST and incubated lh30 with the corresponding secondary antibody coupled with horseradish peroxidase (see below). Membranes were washed 3 times with TBST and immunocomplexes were visualized using ECL RevelBlot Plus (Ozyme).

Immunofluorescence

PDAC8661 (5xl0⁴ cells) were seeded in 6-well plates containing glass slides. Twenty- four hours after seeding, cells were treated with a-selective (A66, BYL719) PI3K inhibitors at 1 or 10 mM. Fifteen minutes or 8 hours after treatment, cells were washed twice with PBS then fixed 10 minutes with paraformaldehyde (PFA) 4%/PBS. Cells were washed three times with PBS during 5 minutes then permeabilised 5 minutes with Triton X-100 1 %/PBS. Cells were washed three times with PBS during 5 minutes then non-specific antibody fixation was prevented with Bovine Serum Albumin (BSA) 1 %/PBS during 30 minutes. Actin was marked with 1 :200 phalloidin Texas-red (Sigma) during 30 minutes in the dark. Cells were washed three times with PBS during 5 minutes then marked one minute with 0.1 pg/mL DAPI. Cells were washed three times with PBS during 5 minutes then mounted with Mowiol on a glass slide. Cells were observed with a Cell Observer video microscope (Zeiss) with a 63x objective and the number of podosomes was counted using Zen software (Zeiss). Around 10 pictures were taken per slide. Control cells were treated with 0.1% DMSO.

RNA extraction

Cells were cultured in 100 mm petri dishes until they reach 80% confluence. They were washed twice with cold PBS, scraped in cold PBS then centrifuged 5 minutes at 300g at 4°C. Cell pellet was resuspended and homogenized in 1 mL Trizol (Invitrogen). Chloroform was added (1/5 of Trizol volume), the suspension was vortexed then incubated 3 minutes at room temperature. The suspension was centrifuged 10 minutes at 18,000g at 4°C. The aqueous phase was isolated and completed with isopropanol (equal volume). The suspension was vortexed, incubated 10 minutes at 4°C then centrifuged 10 minutes at 18,000g at 4°C. The supernatant was discarded and the pellet washed with cold EtOH 80%. The suspension was vortexed then centrifuged 5 minutes at 18,000g at 4°C. The supernatant was discarded, the pellet dried then washed with cold EtOH 83%. The suspension was vortexed then centrifuged 5 minutes at 18,000g at 4°C. The supernatant was discarded, the pellet dried then resuspended in RNAse free water. RNA concentration was determined with the NanoDrop (Thermofischer).

RT-qPCR 1 pg of RNA was used to obtain cDNA using the RevertAid H minus reverse transcriptase, random hexamers and the corresponding mix (ThermoScientific). Primers were all designed with Primer-BLAST (NCBI). qPCRs were realized using the SsoFast EvaGreen supermix (Bio-Rad). Actin was used as a housekeeping gene.

Mice

The LSL-Kras^G12D and LSL-p53^R172H knock-in (from D. Tuveson, Mouse Models of Human Cancers Consortium repository, National Cancer Institute-Frederick), Pdxl-Cre (from D.A. Melton, Harvard University, Cambridge, MA) and pi 10a^lox/lox (from B. Vanhaesebroeck, University College London, London) strains were interbred on a mixed background (CD1/SV129/C57B16) to obtain the compound mutants LSL-Kras^G12D-LSL-p53^R172H-Pdxl-Cre (KPC), LSL-Kras^G12D-Pdxl-Cre (KC), LSL-Kras^G12D-Pdxl-Cre-pl 10a^+/lox (KC;pl l0a^+/lox), Kras^G12D-Pdxl-Cre-pl 10a^+/lox (KC; pl l0a^lox/lox), LSL-Kras^G12D-LSL-p53^R172H-Pdxl-Cre- pl l0a^+/lox (KPC; pl l0a^+/lox), LSL-Kras^G12D-LSL-p53^R172H-Pdxl-Cre-pl l0a^+/lox (KPC; pl l0a^lox/lox). Littermates not expressing Cre were used as controls. The Ptfla^Cre/+-LSL- Kras^G12D/+ (KC; r110a^+/+) and Ptfla^Cre/+-LSL-PIK3CA^H1047R/+ (pl lOa^H1047R) strains were obtained from D. Saur (Technische Universitat Miinchen, Munich). Genotyping was performed as described in Hingorani et al (2005) and Baer et al (2014). All animal procedures and were conducted in compliance with the Ethical committee according to European legislation translated to French Law as Decret 2013-118 1st of February 2013 (APAFIS 3601- 2015121622062840).

Tumour detection by ultrasound imaging

Ultrasound imaging was performed using the VisualSonics Vevo2100 High Resolution System equipped with an ultrasound transducer in the 25-55MHz range. Animal preparation and imaging procedures were performed as described in Sastra et al (54). KPC mice were monitored once per week from 12 weeks old onwards; when a tumour was detected, the ultrasounds were performed every other day. Tumour area was measured by delimiting the tumour border and then obtaining the major axis, at least 5 replicates were performed per mice per ultrasound. Tumour growth rate corresponds to the increase (%) of the tumour size after the onset of the treatment.

Blood collection and plasma separation

Blood samples were collected every two weeks via retro-orbital collection after the mice turned 12 weeks. A maximum volume of 100 - 150pL was collected every time. The blood was collected using glass Pasteur pipettes and then transferred to Eppendorf tubes containing 20pl 0.5M EDTA. Blood counts were performed using Yumizen H500 hematology analyzer (HORIBA), calibrated for murine blood. Plasma was separated by centrifuging blood at 4000 rpm for 20 min at 4°C within 3h of blood collection. The plasma was stored at -80°C until further use. cfDNA extraction from blood plasma

Before cfDNA extraction, blood plasma was re-centrifuged at 14000 rpm for 10 min at room temperature to reduce debris contamination. cfDNA was extracted from blood plasma using the QIAmp DNA Mini Kit (QIAGEN) according to the kit’s protocol except for eluting the cfDNA in 50pL of elution buffer. cfDNA samples were stored at -20°C until further use.

Quantification of cfDNA by qPCR

For quantifying the relative cfDNA in blood plasma we firstly established a standard curve using extracted DNA from a murine cell line R211 (without LSL cassette, expressing mutated kras and tp53) and from pancreatic extracts from mice expressing the LSL cassette (not recombined). The qPCR was performed using l pL of DNA and SsoFast Eva Green Supermix (BioRad®). For the standard curve, we did serial dilutions up until 1/1000 of the two DNA extracts and a qPCR of two different genes, p53 and GAPDH. We designed and selected the most specific and efficient primers (Sigma- Aldrich®) using Primer-Blast (NCBI). The obtained standard curves permitted the further relative quantification of cfDNA in mouse plasma samples.

cfDNA thresholds were calculated by compiling all cfDNA measurements from mice with normal pancreas, high grade Panins, localised and metastatic PD AC. For the survival curve (Kaplan-Meier), mice were assigned as normal, medium or high level of cfDNA according to the average of all their cfDNA values.

Characterization of cfDNA The Fragment Analyzer™ was used to determine the size of DNA in blood plasma. For characterizing the cfDNA, the DNF-474 High Sensitivity NGS Fragment Kit was used. Three different profiles of cfDNA fragmentation were obtained: for normal and healthy mice, the electropherogram did not present any fragment; for mice bearing high grade Panins and localised PD AC, a 160-210bp fragment was always found; for mice with metastatic PD AC, the electropherogram presented the aforementioned 160-210bp fragment, in addition to larger fragments but at lower concentrations.

PI3Ka inhibitor oral treatment

We purchased the PBKa inhibitor BYL-719 from ApexBio. BYL-719 was dissolved in 0.5% methyl cellulose with 0.2% Tween-80 and administered by oral gavage at 50mg/kg daily.

Histology and immunohistochemistry

Tissues were fixed in 10% neutral-buffered formalin (Sigma HT501128) and embedded in paraffin. For pathological analysis, tissues were serially sectioned (4pm), and then stained with haematoxylin eosin (H&E). All tissues were analysed in blinded fashion. Histopathological scoring of pancreatic lesions was performed on sections 100 pm apart, 3 sections per pancreas.

Immunostaining was conducted using standard methods on formalin-fixed, paraffin- embedded tissues. After rehydration, slides were permeabilised for 10 min in 0.1% Triton/1% PBS; they were later subjected to heat-antigen retrieval in citrate buffer. Slides were incubated 20 min with Protein Block (Dako X0909) for preventing non-specific binding. Subsequently, sections were incubated overnight with primary antibodies (See Table below), washed, incubated 15 min with 3% H2O2 and then washed. All antibodies were revealed using the HRP- Detection Reagent Signal Stain®Boost (CST 8114) and developed through AEC (Dako K3464) incubation. Slices were counterstained with haematoxylin and mounted using Glycergel Mounting Medium (Dako C0563).

Animal sample size was calculated using a power analysis in order to obtain statistically significant results from ki67 (n=5, p<0.02) and CD206 (n=6, p<0.05) immune staining (Supplementary Table 9).

Luminex® assay

A MILLIPLEX®_MAP (Merc Millipore # MCYTMAG70PMX32BK) assay was performed using 25 pL of non-diluted mouse blood plasma. The assay was performed according to the manufacturer’s protocol. We tested for a panel of 32 mouse chemokines and cytokines: Eotaxin, G-CSF, GM-CSF, PTNGg, IL-la, IL-Ib, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-9, IL-10, IL-12 (p40), IL-12 (p70), IL-13, IL-15, IL-17, IP-10, KC, LIF, LIX, MCP-1, M-CSF, MIG, MIP-la, MIR-Ib, MIP-2, RANTES, TNFa, and VEGF. Only the statistically significant results were presented. Statistical analysis

Experimental data provided at least 3 experimental replicates and five measurements replicates. Statistical analyses were performed with GraphPad Prism using T-tests (non- parametric Mann- Whitney): *p<0.05, **p<0.01, ***p<0.001. Results

A RI3Ka specific transcriptomic signature inversely correlates with patients’ prognosis and correlates with pancreatic tumour cells aggressiveness

Here, we first sought to determine in an unbiased manner which signalling pathway was associated with aggressive features in PD AC.

We designed a cluster of publically available patient databases to compare primary tumours or pancreas from metastatic PDAC (mPDAC), localised PDAC (1PDAC), chronic pancreatitis (CP) or normal (normal) pancreas (data not shown). We searched for hallmarks of biological and signalling pathways as well as designed a PI3Ka activation gene signature, based on expression levels of PI3Ka-regulated curated genes. RI3Ka is a key enzyme necessary for insulin signalling (16), angiogenesis (17) and PDAC initiation (9). Amongst all the mRNA expression-based hallmarks of biological pathways, PI3K/Akt/mTOR hallmark was significantly increased in mPDAC patients (data not shown). Three reactome signatures of PI3K/Akt activation were found significantly or nearly significantly increased in mPDAC as opposed to 1PDAC (data not shown). PI3Ka activation scoring allowed to cluster 8/9 mPDAC patients (data not shown). We next extended our finding to larger cohorts of PDAC patients (data not shown). High scoring of PI3Ka activation was significantly increased in patients with poorest prognosis regardless of their stage (data not shown); patient cohorts enriched in locally advanced cancers had mostly a low scoring level of PI3Ka activation (data not shown). Finally, PI3Ka activation signature can discriminate, amongst patients diagnosed as localised, those with better overall survival (OS) (data not shown). A gene subset of PI3K activation signature was found increased in metastatic patients (data not shown); we confirmed increased expression of these genes in all pancreatic cell lines derived from metastatic sites as well as primary cultures of locally invasive patient-derived cells as compared to the primary tumour cell lines (data not shown). Thus, even if oncogenic mutation of PI3Ka is rare in PDAC (18), high scoring of PI3K activation was a worse prognostic factor, disregarding the stage of the disease. Only few localised PDAC presented a medium scoring level of PI3Ka activation, associated with a worse OS. In conclusion, activation of non-mutated PI3Ka is a strong prognosis factor on survival.

PI3Ka is necessary for pancreatic cancer cell migration and survival in a Kras mutated context We next investigated why this aggressive phenotype could be specifically dependent on PI3Ka activity. We used a panel of mutant Kras murine (isolated from genetically engineered mouse models GEMMs of pancreatic cancer) and human pancreatic tumour cell lines combined with various mutational status on p53 and pl6 encoding genes to mimic the PD AC genetic landscape. We compared two different pharmacological strategies of PI3K inhibition using either isoform-selective or global PI3K inhibitors, also called pan-PI3K inhibitors (data not shown). In all tested cell lines from human or murine origin, a-selective inhibitors A66 and BYL-719 presented a concentration-dependent capacity to inhibit pancreatic cancer cell F-actin positive podosomes, cell motility and directed cell migration with strong effects observed at the low concentration of 1 mM, independently from effects on cell proliferation at the same concentration. All a-selective were able to strongly inhibit numbers of viable cells and Akt phosphorylation on Ser473, main downstream target of PI3K, at high concentrations (data not shown). As expected, pan-PI3K inhibitors were more efficient for decreasing cell numbers, however, for motility, migration and levels of p-Akt, the efficiency varied depending on the inhibitors. Of note, a-selective inhibitors specifically reduced the phosphorylation of a protein of a lower molecular weight, possibly corresponding to Akt2. To confirm our observation with a genetic approach, we treated murine pancreatic tumour cells PDAC8661 with siRNA targeting pi 10a or a combination of siRNA targeting each class I PI3K catalytic subunit, mimicking a pan-PI3K inhibitor (data not shown). Inhibiting pi 10a expression or the expression of all class I PI3K catalytic subunits induced the same inhibition of PDAC8661 migration, confirming the specific role of PI3Ka in pancreatic tumour cell migration (data not shown). In order to find a pattern that could explain differences observed with pan-PI3K inhibitors, we tested the correlation between the in vitro IC50 of PI3K inhibitors (determined on recombinant proteins) for each class I PI3K isoform, and their capacity to inhibit migration in R211, PDAC8661 and PANC-1 cells (data not shown). We showed that, at 1 mM, the ability of all PI3K inhibitors to regulate cell migration strongly depends on their capacity to target PI3Ka, but not RI3Kb (data not shown). We reported the p-value of effect vs. IC50 correlation test for each class I isoform in each cell line individually and found that only PI3Ka regulates pancreatic directed cell migration (data not shown), cell motility (data not shown), cytotoxicity (data not shown) and pSer473 Akt phosphorylation (data not shown). We confirm this analysis by showing that, at a concentration where an a-selective or a global inhibitor induced the same effect on Akt phosphorylation, an a-selective compound presented a moderate action on the number of living cells while it drastically affected cell migration (data not shown). Biologically efficient PI3K inhibition annihilates PI3Ka strongly, but not necessarily specifically amongst the four class I PI3K isoforms, resulting in a PI3Ka-specific motility and migration inactivation.

PI3Ka regulates pancreatic adenocarcinoma cell motility and migration regardless the driving oncogenic mutation

We next used murine pancreatic tumour cell lines (10158 and 10593) induced by oncogenic PIK3CA (gene encoding for PI3Ka catalytic domain). Most pancreatic cancer bearing patients harbour a Kras oncogenic mutation. Less of 5% patients present a PIK3CA oncogenic mutation; however, this mutation mimics Kras oncogenic pathway (18). We found a higher Akt phosphorylation on Ser473 in PIK3CA oncogenic cell lines compared to the Kras mutated cell lines, not associated with an increased basal migration (data not shown). Besides, a-selective inhibitors similarly inhibited pancreatic tumour cell migration independently of KRAS/PIK3CA mutation (data not shown) but reduced significantly more the number of oncogenic PIK3CA living cells compared to Kras-mutated cell lines (data not shown). A basal oncogenic Kras-PI3Ka coupling is sufficient to drive pancreatic cancer cell migration, while hyper-active oncogenic PI3Ka further regulates cell proliferation/survival. PI3Ka oncogenicity is also commonly described as directly coupled only to oncogenic Kras or tyrosine kinase receptors (19). To challenge this concept, we finally studied the impact of the genetic alterations (on KRAS , PIK3CA , PTEN) and of the organ of origin (pancreas, myeloid lineage, prostate and breast) in determining the role of each class I PI3K in tumour cell migration and cytotoxic sensitivity in response to PI3K inhibitors (data not shown). Strikingly, all pancreatic cancer cell lines depended on PI3Ka activity to regulate their motility and cell viability despite the genetic context, which was not the case in other organ backgrounds (data not shown). Our results show that PI3Ka activation drives in vitro features associated with pancreatic tumour cell progression regardless of the Kras mutational status. Moreover, we confirm the implication of other class I PI3K in other organ context (e.g. PI3Ry in myeloid lineage).

PI3Ka inhibition delays the rapid progression of PD AC and prevents protumoural M2 macrophages infiltration

We next sought to test the effects of PI3Ka- selective pharmacological inhibition on aggressive established tumours in vivo. The KPC mouse model (LSL-KrasG12D/+; LSL- Trp53R172H/+; Pdx-l-Cre mice) where aggressive pancreatic tumours spontaneously develop under the action of Kras and p53 oncogenic mutations, is one of the most widely used in preclinical testing of new therapeutic agents for pancreatic cancer (20). KPC mice diagnosed with an aggressive carcinoma, detected through high-resolution ultrasound (US) imaging (data not shown), were treated with the PI3Ka-selective inhibitor, BYL-719, or vehicle (Figure 1A). In pancreas and not in spleen or lung, BYL-719 drastically reduces pS473-Akt and pT202/Y204-Erkl/2 phosphorylation (Figure 1A). BYL-719 line of treatment significantly reduced tumour cell proliferation in both primary and metastatic sites (Figure 1 B, CT resulting in a delay in tumour growth rate, in ascites development as a marker of peritoneal dissemination, and in the development of distant metastases in liver, lung and spleen (Figure 1D-F). While PDAC patients with increased levels of circulating tumoural DNA (ctDNA) present a worse prognostic (15), cell free circulating DNA (cfDNA) is a recognized marker of inflammation (11). We thus decided to test the correlation between the latter two markers and the development of metastasis in the KPC preclinical model. We quantified the cfDNA in blood plasma samples by longitudinally measuring the relative concentration of p53 and GAPDH genes and then correlated those findings with the anatomo-pathological results from the pancreas and metastatic site organs (data not shown). We found that the longitudinal average levels of cfDNA correlated with disease progression and mouse lethality (Figure 1G). The cfDNA level was significantly increased in mice prior tumour detection and at sacrifice. High cfDNA level at detection thus appears as an indirect early reflect of an undetected micrometastatic disease by US echography (Figure 1H). Indeed, these mice were found to develop metastasis in at least one organ site. In a second cohort of KPC mice from a different animal house, we were able to predict the pathology (primary tumour vs micrometastatic dissemination) by the level of cfDNA at sacrifice (data not shown). Analysis of cfDNA integrity showed the presence of a distinct 160-210 bp fragment selectively increased in mice which developed a metastatic PDAC as compared with mice with high grade PanlNs, localised PDAC (data not shown). This short fragment was shown to be specific to tumour cells (21). Analysing the effect of BYL-719 on cfDNA, we found that it tended to decrease the overall concentration of cfDNA (Figure II). as well as the concentration of the 160-210 bp DNA fragment (Figure IJ). suggesting that BYL-719 could reduce the tumour burden.

Because detectable levels of cfDNA are also associated with an increased inflammatory pro-tumoural immune reaction, we performed a complete blood count to assess inflammation indicators. We observed that metastatic KPC mice also presented a significantly increased count of white blood cells (data not shown), as well as increased monocyte (data not shown) and granulocyte (data not shown) numbers. Considering that monocytes are the precursors of macrophages, which play an important role in pro-tumourigenic immune response, we analysed the presence of macrophages, as well as of the pro-tumoural macrophages (22,23) by IHC staining directed against F4/80 and CD206. We found that BYL-719 significantly decreased the differentiation into pro-tumourigenic CD206-positive macrophages (M2 macrophages) in tumour adjacent tissues (data not shown). Moreover, increased peritumoural CD206 staining predicted further development of metastatic foci and ascites (data not shown). We compared the global pharmacological inhibition of PI3Ka with the selective genetic inactivation of the PI3Ka in the epithelium, and we found that the inactivation only in pancreatic epithelial cells was sufficient to completely prevent M2 macrophage infiltration selectively around high grade lesions (data not shown). Expression of hyper-activated oncogenic PI3Ka also increased the number of infiltrating macrophages around tumours as compared to oncogenic Kras-induced tumours (data not shown). BYL-719 tended to increase IL-6-positive cells in pancreatic sections (data not shown); the genetic inactivation of half the expression of PI3Ka significantly increased IL-5 levels, decreased IL-6 production and CDCL5/LIX levels in plasma (data not shown). CXCL5/LIX levels was found responsible of CD206-polarization and prostate cancer progression in metastatic stage (24). The in vivo inhibition of PI3Ka delays the rapid progression of aggressive cfDNA-positive PD AC by preventing the infiltration of M2 pro- tumoural macrophages in peritumoural tissue.

Discussion:

In light of the apparent failure of signal targeted therapies in advanced pancreatic metastatic patients as compared to other solid tumours addicted to oncogenic pathways (25), other approaches should be attempted to prevent their rapid progression; we demonstrate that targeting the mutated Kras-PI3Ka-driven signal that critically sustains all aspects of oncogenicity, and in particular those at the origin of tumour-induced rewiring of the environment driving early metastatic dissemination, is amongst them.

In the clinic, single agents hitting PI3K have a limited action (26) most likely due to a non-optimized selection of patients with advanced disease to be targeted (27). Interestingly, in breast cancer, clinical results with pan-PI3K inhibitor were not always correlated to mutational status of PIK3CA in primary tumour biopsies, however when PIK3CA mutation was detected in circulating tumour DNA, PFS improvement became largely significant (7 months versus 3.2 months) (28). Our data suggest that this concept could be extended to select PD AC patients for PI3K inhibitor treatment. In this cancer setting, high levels of circulating DNA as a probable marker of micrometastatic disease was found correlated with a poor prognosis (15). Experimental data from others also argue that disseminating cells were detected very early on in the disease history (29). The longitudinal analysis of circulating DNA levels and fragments demonstrates an early increase of these parameters in an experimental model of pancreatic cancer; the measure of the concentration of the 160-210 bp fragment increases the prediction rate as showed in patients by others (30). In terms of toxicity, the main concern for using potent RI3Ka inhibitors remains the induction of insulin feedback which could feed the tumours (2); those concerns could be resolved by a clinical management of hypoglycemia upon treatment (2,31).

Our data also demonstrate that oncogenic Kras-PI3Ka coupling leads to specific functions of relevant clinical importance in pancreatic organ context. In mutant Kras-driven lung cancers, inactivation of PI3Ka showed only partial responses (32), MAPK pathway activity being important in this context. From our results, there are three non-exclusory explanations of the importance of Kras-PI3Ka coupling in this organ setting. First, in pancreas, PI3Ka selective inactivation induced a decreased phosphorylation of Erkl/2. Our data are not in line with data from others in this matter (33-35); these later authors used pan-PI3K inhibitors. Inhibition of MEK has a minimal anti-tumoural action due the induction of a strong activatory feedback on Akt (36). Selectivity towards PI3Ka could prevent the induction of compensatory signals towards MAPK pathway. Second, the action of PI3Ka on cytoskeleton remodelling in pancreatic cancer cells is very sensitive to its pharmacological inhibition regardless the genetic background, and remodelling of actin cytoskeleton appears to be key in PD AC progression so that tumour cells can extrude themselves from the strongly desmoplastic and lowly vascularised pancreatic primary tumours. Very early on in the discovery of oncogenic Kras properties, it was demonstrated that PI3K signal towards actin remodelling was the key first signalling event leading to the transformation of cells (37). Later, it was also found that actin remodelling was linked to the PI3Ka-driven glucose metabolism regulation (38). We find that to have a strong effect on all protumoural properties, PI3Ka should be potently inhibited; besides, the actin cytoskeleton remodelling and associated cellular functions such as migration, motility is very sensitive to PI3Ka inhibition. This event is not as drastically needed in other solid tumour locations which do not have desmoplastic features, explaining why PI3Ka activity is critically involved in all genetic context of PD AC cell lines, but not in other organ context. Third, protumoural macrophages by secreting chemo-attractive chemokines such as CXCL5 in the organ around the tumour favour dissemination (24).

Finally, one could wonder given our results, why oncogenic mutations on PIK3CA are not found more frequently in PD AC patients. Genetically engineered mouse models and study of subclonal populations in cancer evolution clearly argue that oncogenic PI3Ka are weak oncogenic drivers (39). Hence, even if pancreatic oncogenic PI3Ka cell lines harboured increased PI3K signalling, they did not present increased migratory or proliferative properties. Our data find instead that the absence of oncogenic mutation of RBKa, which allows to be activated by its natural upstream components (such as oncogenic Kras, increased insulin signalling), is a more efficient way to trigger proinflammatory protumourigenic action of PI3K signalling.

In conclusion, our data demonstrate that PI3K-targeting agents could be efficient on micrometastatic disease (measured by cfDNA) notably for PD AC patients resulting in delay on macrometastatic dissemination.

REFERENCES:

Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.

1. Vanhaesebroeck B, Guillermet-Guibert J, Graupera M, Bilanges B. The emerging mechanisms of isoform-specific PI3K signalling. Nat Rev Mol Cell Biol. 2010; 11 :329-41.

2. Goncalves MD, Hopkins BD, Cantley LC. Phosphatidylinositol 3-Kinase, Growth Disorders, and Cancer. N Engl J Med. 2018;379:2052-62.

3. Pons-Tostivint E, Thibault B, Guillermet-Guibert J. Targeting PI3K Signaling in Combination Cancer Therapy. Trends Cancer. 2017;3 :454-69.

4. Ali K, Soond DR, Pineiro R, Hagemann T, Pearce W, Lim EL, et al. Inactivation of PI(3)K pi 105 breaks regulatory T-cell-mediated immune tolerance to cancer. Nature. 2014;510:407-11.

5. Okkenhaug K, Graupera M, Vanhaesebroeck B. Targeting PI3K in Cancer: Impact on Tumor Cells, Their Protective Stroma, Angiogenesis, and Immunotherapy. Cancer Discov. 2016;6: 1090-105.

6. Neoptolemos JP, Kleeff J, Michl P, Costello E, Greenhalf W, Palmer DH. Therapeutic developments in pancreatic cancer: current and future perspectives. Nat Rev Gastroenterol Hepatol. 2018;15:333-48. 7. Schlieman MG, Fahy BN, Ramsamooj R, Beckett L, Bold RJ. Incidence, mechanism and prognostic value of activated ART in pancreas cancer. Br J Cancer. 2003;89:2110-5.

8. Witkiewicz AK, McMillan EA, Balaji U, Baek G, Lin W-C, Mansour J, et al. Whole-exome sequencing of pancreatic cancer defines genetic diversity and therapeutic targets. Nat Commun. 2015;6:6744.

9. Baer R, Cintas C, Dufresne M, Cassant-Sourdy S, Schonhuber N, Planque L, et al. Pancreatic cell plasticity and cancer initiation induced by oncogenic Kras is completely dependent on wild-type PI 3-kinase pi 10a. Genes Dev. 2014;28:2621-35.

10. Wu C-YC, Carpenter ES, Takeuchi KK, Halbrook CJ, Peverley LV, Bien H, et al. PI3K regulation of RAC 1 is required for KRAS-induced pancreatic tumorigenesis in mice. Gastroenterology. 2014; 147: 1405-1416. e7.

11. Alix-Panabieres C, Pantel K. Clinical Applications of Circulating Tumor Cells and Circulating Tumor DNA as Liquid Biopsy. Cancer Discov. 2016;6:479-91.

12. Diehl F, Schmidt K, Choti MA, Romans K, Goodman S, Li M, et al. Circulating mutant DNA to assess tumor dynamics. Nat Med. 2008; 14:985-90.

13. Schwarzenbach H, Hoon DSB, Pantel K. Cell-free nucleic acids as biomarkers in cancer patients. Nat Rev Cancer. 2011;11 :426-37.

14. Dawson S-J, Tsui DWY, Murtaza M, Biggs H, Rueda OM, Chin S-F, et al. Analysis of circulating tumor DNA to monitor metastatic breast cancer. N Engl J Med. 2013;368: 1199-209.

15. Pietrasz D, Pecuchet N, Garlan F, Didelot A, Dubreuil O, Doat S, et al. Plasma Circulating Tumor DNA in Pancreatic Cancer Patients Is a Prognostic Marker. Clin Cancer Res Off J Am Assoc Cancer Res. 2017;23 : 116-23.

16. Vanhaesebroeck B, Ali K, Bilancio A, Geering B, Foukas LC. Signalling by PI3K isoforms: insights from gene-targeted mice. Trends Biochem Sci. 2005;30: 194-204.

17. Graupera M, Guillermet-Guibert J, Foukas LC, Phng L-K, Cain RJ, Salpekar A, et al. Angiogenesis selectively requires the pl lOalpha isoform of PI3K to control endothelial cell migration. Nature. 2008;453 :662-6.

18. Eser S, Reiff N, Messer M, Seidler B, Gottschalk K, Dobler M, et al. Selective requirement of PI3K/PDK1 signaling for Kras oncogene-driven pancreatic cell plasticity and cancer. Cancer Cell. 2013;23 :406-20.

19. Thorpe LM, Yuzugullu H, Zhao JJ. PI3K in cancer: divergent roles of isoforms, modes of activation and therapeutic targeting. Nat Rev Cancer. 2015; 15:7-24. 20. Hingorani SR, Wang L, Multani AS, Combs C, Deramaudt TB, Hruban RH, et al. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell. 2005;7:469-83.

21. Thierry AR, Mouliere F, Gongora C, Ollier J, Robert B, Ychou M, et al. Origin and quantification of circulating DNA in mice with human colorectal cancer xenografts. Nucleic Acids Res. 2010;38:6159-75.

22. Griesmann H, Drexel C, Milosevic N, Sipos B, Rosendahl J, Gress TM, et al. Pharmacological macrophage inhibition decreases metastasis formation in a genetic model of pancreatic cancer. Gut. 2017;66: 1278-85.

23. Zhu Y, Herndon JM, Sojka DK, Kim K-W, Knolhoff BL, Zuo C, et al. Tissue- Resident Macrophages in Pancreatic Ductal Adenocarcinoma Originate from Embryonic Hematopoiesis and Promote Tumor Progression. Immunity. 2017;47:323-338. e6.

24. Roca H, Jones ID, Purica MC, Weidner S, Koh AJ, Kuo R, et al. Apoptosis- induced CXCL5 accelerates inflammation and growth of prostate tumor metastases in bone. J Clin Invest. 2018; 128:248-66.

25. Cintas C, Douche T, Therville N, Arcucci S, Ramos-Delgado F, Basset C, et al. Signal-Targeted Therapies and Resistance Mechanisms in Pancreatic Cancer: Future Developments Reside in Proteomics. Cancers. 2018; 10.

26. Shapiro GI, Rodon J, Bedell C, Kwak EL, Baselga J, Brana I, et al. Phase I safety, pharmacokinetic, and pharmacodynamic study of SAR245408 (XL 147), an oral pan-class I PI3K inhibitor, in patients with advanced solid tumors. Clin Cancer Res Off J Am Assoc Cancer Res. 2014;20:233-45.

27. Le Toumeau C, Delord J-P, Gonqalves A, Gavoille C, Dubot C, Isambert N, et al. Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial. Lancet Oncol. 2015;16: 1324-34.

28. Baselga jose. BELLE-2. San Antonio Breast Cancer Abstr S6-01.

29. Rhim AD, Mirek ET, Aiello NM, Maitra A, Bailey JM, McAllister F, et al. EMT and dissemination precede pancreatic tumor formation. Cell. 2012; 148:349-61.

30. Mouliere F, Chandrananda D, Piskorz AM, Moore EK, Morris J, Ahlbom LB, et al. Enhanced detection of circulating tumor DNA by fragment size analysis. Sci Transl Med. 2018; 10.

31. Rodon J, Dienstmann R, Serra V, Tabemero J. Development of PI3K inhibitors: lessons learned from early clinical trials. Nat Rev Clin Oncol. 2013; 10: 143-53. 32. Castellano E, Sheridan C, Thin MZ, Nye E, Spencer-Dene B, Diefenbacher ME, et al. Requirement for interaction of PI3-kinase pi 10a with RAS in lung tumor maintenance. Cancer Cell. 2013;24:617-30.

33. Alagesan B, Contino G, Guimaraes AR, Corcoran RB, Deshpande V, Wojtkiewicz GR, et al. Combined MEK and PI3K inhibition in a mouse model of pancreatic cancer. Clin Cancer Res Off J Am Assoc Cancer Res. 2015;21 :396-404.

34. Baer R, Cintas C, Therville N, Guillermet-Guibert J. Implication of PI3K/Akt pathway in pancreatic cancer: When PI3K isoforms matter? Adv Biol Regul. 2015;59: 19-35.

35. Junttila MR, Devasthali V, Cheng JH, Castillo J, Metcalfe C, Clermont AC, et al. Modeling targeted inhibition of MEK and PI3 kinase in human pancreatic cancer. Mol Cancer Ther. 2015; 14:40-7.

36. Kong B, Wu W, Cheng T, Schlitter AM, Qian C, Bruns P, et al. A subset of metastatic pancreatic ductal adenocarcinomas depends quantitatively on oncogenic Kras/Mek/Erk-induced hyperactive mTOR signalling. Gut. 2016;65:647-57.

37. Rodriguez-Viciana P, Warne PH, Khwaja A, Marie BM, Pappin D, Das P, et al. Role of phosphoinositide 3 -OH kinase in cell transformation and control of the actin cyto skeleton by Ras. Cell. 1997;89:457-67.

38. Hu H, Juvekar A, Lyssiotis CA, Lien EC, Albeck JG, Oh D, et al. Phosphoinositide 3 -Kinase Regulates Glycolysis through Mobilization of Aldolase from the Actin Cytoskeleton. Cell. 2016; 164:433-46.

39. Hanker AB, Kaklamani V, Arteaga CL. Challenges for the Clinical Development of PI3K Inhibitors: Strategies to Improve Their Impact in Solid Tumors. Cancer Discov. 2019;9:482-91.

40. Abdollahi A, Schwager C, Kleeff J, Esposito I, Domhan S, Peschke P, et al. Transcriptional network governing the angiogenic switch in human pancreatic cancer. Proc Natl Acad Sci U S A. 2007; 104: 12890-5.

41. Tosolini M, Algans C, Pont F, Ycart B, Fournie J-J. Large-scale microarray profiling reveals four stages of immune escape in non-Hodgkin lymphomas. Oncoimmunology. 2016;5:el 188246.

42. Bosch A, Li Z, Bergamaschi A, Ellis H, Toska E, Prat A, et al. PI3K inhibition results in enhanced estrogen receptor function and dependence in hormone receptor-positive breast cancer. Sci Transl Med. 2015;7:283ra51. 43. Zhang Y, Kwok-Shing Ng P, Kucherlapati M, Chen F, Liu Y, Tsang YH, et al. A Pan-Cancer Proteogenomic Atlas of PI3K/AKT/mTOR Pathway Alterations. Cancer Cell. 2017;31 :820-832. e3.

44. Jamieson S, Flanagan JU, Kolekar S, Buchanan C, Kendall ID, Lee W-J, et al. A drug targeting only pi 10a can block phosphoinositide 3 -kinase signalling and tumour growth in certain cell types. Biochem J. 2011;438:53-62.

45. Fritsch C, Huang A, Chatenay-Rivauday C, Schnell C, Reddy A, Liu M, et al. Characterization of the novel and specific PI3Ka inhibitor NVP-BYL719 and development of the patient stratification strategy for clinical trials. Mol Cancer Ther. 2014; 13 : 1117-29.

46. Jackson SP, Schoenwaelder SM, Goncalves I, Nesbitt WS, Yap CL, Wright CE, et al. PI 3-kinase pi lObeta: a new target for antithrombotic therapy. Nat Med. 2005; 11 :507— 14.

47. Ali K, Camps M, Pearce WP, Ji H, Riickle T, Kuehn N, et al. Isoform-specific functions of phosphoinositide 3 -kinases: pi 10 delta but not pi 10 gamma promotes optimal allergic responses in vivo. J Immunol Baltim Md 1950. 2008; 180:2538-44.

48. Knight ZA, Chiang GG, Alaimo PJ, Kenski DM, Ho CB, Coan K, et al. Isoform- specific phosphoinositide 3 -kinase inhibitors from an arylmorpholine scaffold. Bioorg Med Chem. 2004; 12:4749-59.

49. Barlaam B, Cosulich S, Degorce S, Fitzek M, Green S, Hancox U, et al. Discovery of (R)-8-(l-(3,5-difluorophenylamino)ethyl)-N,N-dimethyl-2-morpholino-4-oxo- 4H-chromene-6-carboxamide (AZD8186): a potent and selective inhibitor of RI3Kb and PI3K5 for the treatment of PTEN-deficient cancers. J Med Chem. 2015;58:943-62.

50. Pomel V, Klicic J, Covini D, Church DD, Shaw JP, Roulin K, et al. Furan-2- ylmethylene thiazolidinediones as novel, potent, and selective inhibitors of phosphoinositide 3- kinase gamma. J Med Chem. 2006;49:3857-71.

51. Burger MT, Pecchi S, Wagman A, Ni Z-J, Knapp M, Hendrickson T, et al. Identification of NVP-BKM120 as a Potent, Selective, Orally Bioavailable Class I PI3 Kinase Inhibitor for Treating Cancer. ACS Med Chem Lett. 2011;2:774-9.

52. Folkes AJ, Ahmadi K, Alderton WK, Alix S, Baker SJ, Box G, et al. The identification of 2-(lH-indazol-4-yl)-6-(4-methanesulfonyl-piperazin-l-ylmethyl)-4- morpholin-4-yl-thieno[3,2-d]pyrimidine (GDC-0941) as a potent, selective, orally bioavailable inhibitor of class I PI3 kinase for the treatment of cancer. J Med Chem. 2008;51 :5522-32.

53. Raynaud FI, Eccles SA, Patel S, Alix S, Box G, Chuckowree I, et al. Biological properties of potent inhibitors of class I phosphatidylinositide 3-kinases: from PI-103 through PI-540, PI-620 to the oral agent GDC-0941. Mol Cancer Ther. 2009;8: 1725-38. 54. Sastra SA, Olive KP. Quantification of murine pancreatic tumors by high- resolution ultrasound. Methods Mol Biol Clifton NJ. 2013;980:249-66.

Claims

1. A method of treating micrometastatic disease in a patient suffering from pancreatic cancer comprising administering to the patient a therapeutically effective amount of a PI3Ka- selective inhibitor.

2. The method of claim 1 wherein the pancreatic cancer is pancreatic ductal adenocarcinoma.

3. The method of claim 1 wherein the pancreatic cancer is KRAS mutated.

4. The method of claim 1 wherein the PBKa-selective inhibitor is alpelisib.

5. The method of claim 1 wherein the PBKa-selective inhibitor is an inhibitor of PBKa expression.

6. The method of claim 1 which comprises the steps of i) quantifying the cfDNA level in a sample obtained from the patient ii) comparing said level with a predetermined reference level and iii) administering to the patient the therapeutically effective amount of the a PBKa-selective inhibitor when the level determined at step i) is higher than the predetermined reference value.

7. The method claim 6 wherein the sample is a blood sample.