The present invention relates to methods and compositions for use for preventing or treating skin precancerous lesions and/or skin carcinoma. More particularly, a composition of the invention comprises the combination of at least one Pattern Recognition Receptor (PRR) agonist with an anti-ILlO Receptor antibody.

Background of the invention:

The skin is an immune competent tissue. Tumor progression in epithelial tissues is controlled by immune mechanisms in place within the epithelium itself (Hagerling et al, Trends Cell Biol. 2015, 25:214-220). Indeed, it has been well established that a strong lymphocytic infiltration with cytotoxic T lymphocytes (CTL) is associated with a good clinical outcome (Pages et al, N Engl J Med. 2005, 353:2654-2666). Dendritic cells (DCs) are key actors of antitumoral immunity as they are in charge of educating CTL. Nevertheless, antitumoral DCs are rare cells in progressing tumors, as factors expressed by the tumoral environment inhibit their function and favor the recruitment or the local shaping of immunosuppressive DCs and other immunosuppressive immune cells. Dysfunctional DCs are a cause of immune system failure to detect and reject transformed cells and tumor rejection.

In particular, DCs, also called "professional Antigen Presenting cells", have strong ability at antigen presentation and thus at inducing T cell adaptive immunity (Banchereau J and Palucka AK, Nat Rev Immunol. 2005, 5:296-306). DCs are also crucial actors for controlling humoral immunity and can interact with innate immune system cells, including natural killer (NK) cells, phagocytes and mast cells.

DCs appear like key targets for the generation of protective anti-tumor immunity because DCs allow the activation of the adaptive immunity (cellular and humoral) and also the innate immune system (Steinman and Banchereau, Nature 2007. 449:419-26).

In the art, therapeutic strategies based on DCs are known for treating cancer. As example, DC-based vaccination approaches have been developed regarding the noteworthy ability of these cells in coordinating innate and adaptive immune responses. DC vaccination promotes tumor-specific effector T cells to induce tumor mass reduction and promotes immunological memory for controlling tumor relapse (Sabado and Bhardwaj, Immunotherapy. 2010, 2: 37- 56).

Another promising cancer therapy based on DCs relies on targeting specific innate sensors expressed by DCs named Pattern Recognition Receptors (PRR) in order to activate them. Importantly, DC PRRs that can be targeted for cancer immunotherapy include receptors that recognize microbes and more particularly members of the Toll Like Receptor (TLR) family. Ten different TLRs have been identified in humans and thirteen in mice (Beutler, Nature 2004, 430:257-263). TLRs represent the best-characterized family of PRRs, which recognize pathogen-associated molecular patterns (PAMPS) derived from bacteria or virus, expressed by many cell types, and that induce signaling cascades leading to expression of proinflammatory molecules (West et al., Ann Rev Cell Dev Biol. 2006, 22:409-37).

In parallel to TLRs family, other key PRRs families that can be targeted for cancer immunotherapy comprise: the RNA helicase RIG-I-Like Receptor (RLR) family, the C-Type Like Receptor (CLR) family and the NOD-like Receptor (NLR) family.

PRRs expression profile, including TLRs, appears to be one of the guidelines for diagnosing and treating tumors. The analysis of the PRRs distribution in different types of tumors and of the cellular response to PRRs agonists or antagonists in different types of tumors is herein revealed by the inventors.

Using a stepwise mouse model of skin carcinogenesis, it has been demonstrated that DCs infiltrate pre-cancerous skin lesions and carcinomas but display altered phenotype and functions. They also express specific PRRs at pre-cancerous and cancerous stages that offer the opportunity to identify rationale targets for the local restoration of their antitumoral function (Figure 1). Consequently, PRRs including TLRs represent potential candidates to stimulate the antitumor immunity locally at the site of carcinogenesis and skin lesions.

PRR-based cancer therapy and more particularly TLR-based cancer therapy strategies are known in the art.

As example, WO2007063421 relates to methods for modulating an immune response by administering a composition comprising a Toll-like receptor agonist and an immune mediator which, in particular, downregulates the expression of the anti- inflammatory cytokine IL-10. However, the targeted immune mediators are not modulators directed to IL-10 or its receptor such as an IL10R antagonist. These methods can be used to provide therapeutic treatment for cancerous conditions and infectious diseases. TLR9 agonists such as CpG ODN are mentioned but no specific CpG-ODN, in particular Class C CpG ODN, is identified. Cancers may be treated in accordance with the invention disclosed in this document, in particular skin squamous cell carcinoma and skin basal cell carcinoma, but the treatment of precancerous lesions is not suggested.

WO03045431 relates to methods for treating disease states, including cancer, by activating dendritic cells from the host which are rendered hyporesponsive to activation stimuli by the disease. These methods comprise a step of administering the host with an effective amount of a tumor-derived DC inhibitory factor antagonist in combination with an effective amount of a Toll-like receptor (TLR) agonist. TLR9 agonists are specifically cited such as CpG ODN but, the specific use of Class C CpG-ODN is not suggested. The DC inhibitory factor antagonist is for example an antagonist of IL-10 or an antagonist of the IL-10 receptor. Types of cancer which could be treated include but are not limited to melanoma. Other skin cancers such as skin squamous cell carcinoma and skin basal cell carcinoma or skin conditions such as precancerous lesion are not identified.

Inventors herein disclose a new therapeutic approach. Inventors have discovered that compositions comprising the combination of at least one PRR agonist with an anti-IL-10 Receptor antibody are effective for treating precancerous lesions such as those observed in actinic keratosis or Bowen disease and cancers such as skin squamous cell carcinoma (SCC) and/or skin basal cell carcinoma (BCC). This therapeutic strategy for treating precancerous lesions and/or skin carcinoma is neither described nor suggested in the art. Precancerous lesions are skin conditions in need of treatment. Targeting precancerous lesions enables to treat cancer at an early stage before the tumor becomes aggressive.

Triggering at least one PRR by an agonist according to the present invention leads to the potentiation of IL-12 secretion by DCs. IL-12 is an interleukin involved in the generation of adaptive Thl responses, 11-12 is particularly involved in maturation of naive T cells into cytotoxic CD8 ⁺ T cells and IFNy producing helper CD4 ⁺ T cells, in promotion of IFN-γ and TNF-a secretion by NK cells or in the promotion of anti-angiogenic activity. For all these aspects, enhancing IL-12 secretion through PRRs activation is a promising approach to induce anti-tumor immunity. However, intratumoral DCs and other immune cells produced huge amounts of IL-10 in skin tumors. IL-10 is a pro-tumor interleukin involved in immunosuppression by eliciting Treg expansion. IL-10 has also been described as a strong modulator of DCs function and as example, IL-10 negatively regulates IL-12 production and inhibits the T-cell costimulatory potential of DCs. IL-10 is a major factor involved in immune escape of tumors. There is a need in antagonizing IL-10 secretion to promote efficacy of tumor treatment. Inventors herein disclose a new efficient strategy to reach a high-efficacy tumor and precancerous lesion treatment. Preliminary results indicate that the treatment according to the invention induces at least the generation of CD8 ⁺ T cells and IFN-Y cytokine to achieve its therapeutic function.

In parallel, inventors have demonstrated that the anti-IL-lOR antibody can be administered locally and more particularly, peritumorally and/or intratumorally in a low amount in combination with at least one PRR agonist in the same vehicle or separately. This new therapeutic approach enables an effective and concomitant action of the compounds at the specific site of tumor in skin carcinoma or skin precancerous lesions at lower doses. Moreover, the low amount of anti-IL-lOR antibody decreases the collateral systemic toxicity of anti-IL-lOR antibodies in patients.

Finally, dendritic cells from most solid tumors are inflammatory DCs, but nevertheless there is a specific imprinting of each tissue (the skin is different from the lung as an example) that probably influences their exact phenotype and responsiveness. This invention was specifically shaped to be effective in treating skin carcinoma and skin pre-cancerous lesions.

Summary of the invention:

According to a first aspect of the present invention a composition for use for preventing or treating precancerous lesions and/or skin carcinoma is herein provided. This composition comprises:

(i) at least one Pattern Recognition Receptor (PRR) agonist which potentiates IL-12 secretion, and

(ii) an anti-ILlO Receptor (IL10R) antibody which antagonizes an IL-10-mediated protumor immunosuppression.

In one aspect, the PRR agonist of the composition is an infiammosome molecule agonist, preferably an infiammosome molecule agonist selected from a TLR agonist, a CLR agonist, a NLR agonist, and a RLR/RNA and DNA sensors agonists.

In a particular aspect, the TLR agonist is selected from a TLR9 agonist, a TLR7 agonist and/or a TLR8 agonist.

In a more particular aspect the TLR7 and/or TLR8 agonist is Resiquimod (R848).

In a more particular aspect the TLR8 agonist is selected from Resiquimod (R848), ssRNA40, ORN02, ORN06, ssPolyU Naked, ssPolyU, ssRNA40, ssRNA41, ssRNA-DR, CL075, CL097, Poly(dT), TL8-506 and a combination thereof. In another more particular aspect the TLR9 agonist is an oligodeoxynucleotide sequence containing one or more CpG (CpG ODN).

In another more particular aspect the CpG ODN is a Class C CpG ODN.

In another more particular aspect the Class C CpG ODN is selected from ODN 2395, ODN M362 and DSL03.

In another aspect, the CLR agonist is selected from a Dectin-1 agonist, Dectin-2 agonist and/or Mincle agonist.

In a more particular aspect, the Dectin-1 agonist is selected from β(1→4,1→3, l→6)-glucan peptide (BGP) from Trametes versicolor; Beta-l,3-glucan from Alcaligenes faecalis (Curdlan AL); Heat-killed preparation of Candida albicans; Heat-killed S. cerevisiae; Soluble β(1 ^→ 3, 1 ^→ 6)-glucan from Laminaria digitata; β(1→3, l ^→ 4)-glucan from Cetriana islandica; β(1→6)- glucan from Lasallia pustulata; β(1→3, l→6)-glucan from Schizophyllum commune; β(1→3, l ^→ 6)-glucan from Sclerotium rolfsii; and Cell wall preparation of S. cerevisiae (Zymozan). In another more particular aspect, the Dectin-2 agonist is Malassezia furfur cell wall preparation (furfurman).

In another more particular aspect, the Mincle agonist is selected from heat killed Mycobacterium tuberculosis and/or Trehalose-6,6-dibehenate (TDB), a major synthetic immunostimulatory component of Mycobacterium tuberculosis.

In another aspect, the RLR/RNA and DNA sensors agonist is a RIG-1 agonist and/or a MDA5 agonist.

In a more particular aspect, the RIG-1 agonist is selected from 5 'triphosphate DsRNA, short polyLC and/or a RNA virus.

In another more particular aspect, the MDA5 agonist(s) is(are) long poly I:C and/or EMCV. In another aspect, the NLR and/or inflammosome molecule agonist is a Nodi agonist and/or a Nod2 agonist.

In a more particular aspect the Nodi agonist is a synthetic derivative of peptidoglycan (PGN) of a gram positive or gram-negative bacteria.

In another more particular aspect, the Nod2 agonist is a bacterial muramyl dipeptide (MDP) or a derivative thereof.

In another aspect, the anti-IL-10 Receptor (IL10R) antibody according to the invention is specific for the IL10R1 chain of the IL10 receptor (also called CD210).

In a particular and preferred aspect, the composition according to the invention is for local administration. In a more particular aspect, the composition according to the invention is for intratumoral and/or peritumoral administration.

In another aspect, the anti-ILlOR antibody and the at least one PRR agonist are for simultaneous co-administration in a separate vehicle or in a same vehicle.

In a further aspect, the composition according to the invention is for use for preventing or treating actinic keratosis and/or Bowen disease.

In another aspect, the composition according to the invention is a composition for use for preventing or treating skin squamous cell carcinoma (SCC) and/or skin basal cell carcinoma (BCC).

Also herein described is the use of at least one Pattern Recognition Receptor (PRR) agonist which potentiates IL-12 secretion, and of at least one anti-ILlO receptor antibody which antagonizes IL-10-mediated protumor immunosuppression for preparing a composition for use as a medicament, in particular a composition for preventing or treating actinic keratosis and/or Bowen disease or a composition for treating skin squamous cell carcinoma (SCC) and/or skin basal cell carcinoma (BCC).

Further herein described is a kit including a composition according to the present invention, typically a combination of at least one PRR agonist as herein described and of at least one anti-ILlO receptor antibody as herein described, and uses thereof in prophylaxis or therapy. Also herein described is a method for preventing or treating actinic keratosis, Bowen disease, skin squamous cell carcinoma (SCC) and/or skin basal cell carcinoma (BCC), wherein the method comprises a step of administering a composition according to the invention to a subject in need thereof.

Detailed description of the invention:

According to a first aspect of the present invention, a composition for use for treating a precancerous lesion and/or skin carcinoma is herein provided. This composition comprises:

(i) (an effective amount of) at least one Pattern Recognition Receptor (PRR) agonist which potentiates IL-12 secretion, and

(ii) (an effective amount of) at least one anti-ILlO Receptor (IL10R) antibody which antagonizes ILlO-mediated protumor immunosuppression.

In one particular aspect, the PRR agonist of the present invention is selected from a TLR agonist, a CLR agonist, a NLR, and a RLR/RNA and/or DNA sensors agonist. Preferably, the PRR agonist is a TLR agonist and/or a CLR agonist. In one aspect, the TLR agonist of the present invention is selected from TLR1 , TLR2, TLR4, TLR6, TLR7, TLR8, TLR9, TLR11, TLR12, TLR 13 and a combination thereof Preferably, the TLR agonist is a TLR9 agonist, TLR7 and/or TLR8 agonist. More Preferably, the TLR agonist is a TLR9 agonist.

Examples of TLR8 agonists suitable for the invention are Resiquimod (R848), ssRNA40, ORN02, ORN06, ssPolyU Naked, ssPolyU, ssRNA40, ssRNA41 and ssRNA-DR, CL075, CL097, Poly(dT) and TL8-506. When the oligonucleotides ssRNA40, ORN06, ssPolyU, ssRNA41 and ssRNA are used, they can be in a form complexed with a transfection reagent as for example the cationic lipid known under the name LyoVec™.

An example of TLR8 and/or TLR7 agonist is Resiquimod (R848).

Examples of TLR9 agonists suitable for the invention are oligodeoxynucleotides sequences containing one or more CpG motifs (CpG ODNs). Preferably, TLR9 agonists are CpG ODN sequences. More preferably, the CpG ODNs are Class C CpG ODNs. Examples of Class C CpG ODNs suitable for the invention are ODN 2395, ODN M362 and D-SL03. Other examples of Class C CpG ODNs suitable for the invention can be found in Jurk et al, 2004, Vollmer et al, 2004 and Yang L. et al, 2013.

As used herein, the terms "Class C CpG ODN" consists of a stimulatory hexameric CpG motif positioned at or near the 5' end and linked by a T spacer to a GC-rich palindromic sequence.

CpG ODNs suitable for the invention can be of any length greater than 18 nucleotides and can contain modification, such as a modification of the 3ΌΗ or 5ΌΗ group, modification of a nucleotide base, modification of the sugar component, and modification of the phosphate ring. The prior list of modifications only gives examples and is not limitative to appreciate the scope of the invention. CpG ODN may be single or double stranded DNA sequence or other modified polynucleotides. CpG ODNs sequences may or may not include one or more palindromic sequences.

The CpG ODNs can be isolated using conventional polynucleotides isolation procedures, or can be synthetized using techniques and nucleic acid synthesis equipment which are well known, in the art including, but not limited to, enzymatic methods, chemical methods and the degradation of larger oligonucleotide sequences (see, for example, Ausubel et al, 1987 and Sambrook et al, 1989).

In another embodiment, the RLR/RNA and/or DNA sensors, preferably RLR/RNA and DNA sensors, agonist of the present invention is selected from Ddx58/RIG-I, Ifihl/MDA5/RLR2, Dhx58/LGP2/RLR3, Dhx30, Dhxl6, Dhx8, Dhx9, Dhx40, Dhx30, Dhx35, Dhx34, Dhx33, Dhxl5, Dhx38, Dhx36, Dhx57, Dhx32, Dhx29,Dhx37, Lrrfipl, Ddxl, Ddx3, Ddx4, Ddx5, Ddx6, DdxlO, Ddxl l, Ddxl7, Ddxl8, Ddxl9, Ddx20, Ddx21, Ddx23, Ddx24, Ddx25, Ddx26, Ddx27, Ddx28, ddx31, Ddx39, Ddx41, Ddx42, Ddx43, Ddx46, Ddx47, ddx49, Ddx50, Ddx51, Ddx52, Ddx54, Ddx55, ddx56 and Ddx59. Preferably, the RLR/RNA and DNA sensors agonist is a RIG-1 agonist and/or a MDA5 agonist, even more preferably a RIG-1 and MDA5 agonist.

Examples of RIG-I agonists suitable for the invention are 5 'triphosphate DsRNA, short polyLC and/or RNA virus.

Examples of MDA5 agonists suitable for the invention are long poly I:C and EMCV.

In one aspect, the NLR and/or inflammosome molecule(s) agonist of the present invention is selected from Nlrpl, Nlrp3, Nlrp6, Nlrp9, NlrplO, Nlrpl2, Nlrc3, Nlrc4, Nlrc5, Nlrxl, Aim2, Nodi, Nod2, P2rxl, P2rx2, P2rx4, P2rx5 and P2rx7, preferably from a Nodi and/or Nod2 agonist.

An example of a Nodi agonist suitable for the invention is a synthetic derivative of peptidoglycan (PGN) of a gram positive or gram negative bacteria.

An example of a Nod2 agonist suitable for the invention is a bacterial muramyl dipeptide (MDP) or a derivative thereof.

In another aspect, the C-Type Lectin receptor agonist is selected from Clec7a/dectinl, Clecl0a/MGL/CD301, Clec4a/DCIR, Clec4b/DCAR, Clec4d/dectin3, Clec4e/Mincle, Clec4g, Clec4n/Dectin2, DC-SIGN/CD209 and MRC1/CD206. Preferably, the CLR agonist is selected from a Dectin 1 agonist, a Dectin-2 agonist and/or a Mincle agonist. Examples of Dectin 1 agonists suitable for the invention are selected from β(1 ^→ 4,H 3, l ^→ 6)-glucan peptide (BGP) from Trametes versicolor; Beta-l,3-glucan from Alcaligenes faecalis (Curdlan AL); Heat-killed preparation of Candida albicans; Heat-killed S. cerevisiae; Soluble β(1→3, 1 →6)-glucan from Laminaria digitata; β(1→3, l ^→ 4)-glucan from Cetriana islandica; β(1→6)- glucan from Lasallia pustulata; β(1→3, l→6)-glucan from Schizophyllum commune; β(1→3, l ^→ 6)-glucan from Sclerotium rolfsii; and Cell wall preparation of S. cerevisiae (Zymozan). An example of Dectin-2 agonist suitable for the invention is Malassezia furfur cell wall preparation (furfurman).

Examples of Mincle agonists suitable for the invention are preferably selected from heat killed Mycobacterium tuberculosis and/or Trehalose-6,6-dibehenate (TDB), a major synthetic immunostimulatory component of Mycobacterium tuberculosis. PRR agonists listed herein above can be in a modified form. Further suitable agonists different to the PRR agonists listed above may also be used, wherein said agonists function to activate at least one PRR listed above.

Antibodies according to the invention are anti-ILlOR antibodies. The epitope targeted by the antibodies according to the invention can be linear or conformational. Preferably, the anti- IL10R antibody is specific for the ILIORI chain of IL-10 receptor (also called CD210), specifically expressed by immune cells.

In summary, the IL-10 receptor is composed of two different chains: ILIORI and IL10R2. ILIORI is encoded by illOra gene and is expressed at low level and only in immune cells. IL10R2 encoded by illOrb gene is widely, strongly and constitutively expressed in most cells and tissues (Sabat et al, Cytokine and growth factor reviews, 2010). There is a strategic interest to target the ILIORI chain due to its specific expression on immune cells. Indeed, the use of an ILIORI neutralizing antibody enables to block IL-10 immunosuppressive signaling in immune cells.

Anti-ILlOR antibodies can be of the isotype IgG, IgA or IgD. In further embodiments, the antibody is selected from the group consisting of IgGl, IgG2, IgG3, IgG4, IgM, IgAl, IgA2, IgAsec, IgD, IgE or has immunoglobulin constant and/or variable domain of IgGl, IgG2, IgG3, IgG4, IgM, IgAl, IgA2, IgAsec, IgD or IgE. In other aspects, the anti-ILlOR antibody is a recombinant antibody, polyclonal antibody, monoclonal antibody, human antibody, humanized antibody, chimeric antibody, single domain antibody, bispecific antibody, diabody, multispecific antibody, nanobody, IL10R binding protein or a mixture of these. In some embodiments the chimeric antibody is a genetically engineered fusion of parts of a non- human (e.g., mouse, rat, rabbit) antibody with parts of a human antibody.

The anti-ILlOR antibody can comprise one or more mutation(s) and/or one or more modification(s) such as glycosylation (fucosylation, mannosylation, galactosylation, etc.) or phosphorylation to increase its activity and/or stability.

The anti-ILlOR antibody according to the invention can be an antibody with enhanced ADCC or an antibody with low immunogenicity.

The anti-ILlOR antibody of the invention can be produced according to methods and techniques well known to a person having ordinary skills in the art. For example but not limitatively, a humanized antibody can be generated through standard molecular biology techniques. This can comprise grafting of the rodent complementarity-determining regions (CDRs) into a human framework. However, this technique is mostly an iterative process and a number of elements come into play when designing a humanized antibody such as the length of the CDRs, the human frameworks and the substitution of residues from the rodent mAb into the human framework regions (back mutations). As another example but not limitatively, a human antibody can be generated using transgenic mice carrying parts of the human immune system rather than the mouse system. A fully human monoclonal antibody also can be prepared by immunizing mice transgenic for large portions of human immunoglobulin heavy and light chain loci.

In a particular aspect, the composition according to the invention is delivered to the subject in need thereof by local administration. The subject is typically a mammal, in particular a human being. The subject is for example a human patient.

Examples of local administration routes suitable for the invention are but not limitatively transdermal, subcutaneous, intradermal, intratumoral injections and topical administration. Preferably, the composition according to the invention is delivered by peritumoral and/or by intratumoral administration(s). More preferably, the composition is administered by peritumoral or intratumoral administration in small lesions. But, a person having ordinary skilled in the art can easily adapt the composition for another administration route such as non-local administration route. Examples of non-local administration routes are but not limitatively intramuscular, intravenous and transmucosal.

In another aspect, the composition contains the anti-ILlOR antibody in a low amount comprised between 0.25mg/kg and 5mg/kg. Preferably, the anti-ILlOR antibody is used in an amount comprised between 0.25mg/kg and 1.25mg/kg. More preferably, the anti-ILlOR is used in an amount comprised between 0.5 and 1.25mg/kg or 0.75 and 1.25mg/kg. Most preferably, the anti-ILlOR is used in an amount comprised between 1.00 and 1.25mg/kg. In a further embodiment, the composition contains the anti-ILlOR antibody at 0.25, 0.30 0.35, 0.40, 0.45, 0.5, 0.55, 0.60, 0.65, 0.70, 0.75, 0.80, 0.85, 0.90, 0.95, 1.00, 1.05, 1.10, 1.15, 1.20, 1.25, 1.30, 1.35, 1.40, 1.45, 1.50, 1.55, 1.60, 1.65, 1.70, 1.75, 1.80, 1.85, 1.90, 1.95, 2.00, 2.05, 2.10, 2.15, 2.20, 2.25, 2.30, 2.35, 2.40, 2.45, 2.50, 3.00, 3.50, 4.00, 4.50, 5.00 mg/kg. Preferably, the composition contains the anti-ILlOR antibody at 1.25 mg/kg. The precise dose will depend upon a number of factors such as administration route and physiological state of the patient.

In a further aspect, the composition contains the at least one PRR agonist in a same amount than the amount of the at least one anti-ILlOR antibody, possible of the several anti-ILlOR antibodies, in the composition. The composition preferably contains at least one PRR agonist in an amount between 1.5 times and 150 times higher than the amount of anti-ILlOR antibody(ies) in the composition. The composition preferably contains at least one TLR and/or at least one CLR agonist in an amount 1.5 times, 2 times, 3 times, 4 times, 5 times, 6 times, 7 times, 8 times, 9 times, 10 times, 15 times, 20 times, 30 times, 40 times, 50 times, 100 times, 150 times higher than the amount(s) of anti-ILlOR antibody(ies) in the composition. The composition contains at least one TLR and/or at least one CLR agonist in an amount between ½ times and 1/150 times lower than the amount of anti-ILlOR antibody(ies) in the composition. The composition preferably contains at least one TLR and/or at least one CLR agonist in an amount of ½ times, 1/3 times , ¼ times , 1/5 times, 1/6 times, 1/7 times, 1/8 times, 1/9 times, 1/10 times, 1/20 times, 1/30 times, 1/40 times, 1/50 times, 1/100 times, 1/150 times, lower than the amount of anti-ILlOR antibody(ies) in the composition. Preferably, the composition contains at least one PRR agonist in an amount between 0.1 mg/kg and 1.5 mg/kg, whatever the amount of anti-ILlOR antibody(ies) in the composition. More preferably, the composition contains at least a PRR agonist in an amount of 1.5 mg/kg, whatever the amount of anti-ILlOR antibody(ies) in the composition. As a reminder, the precise dose will depend upon a number of factors such as administration route and physiological state of the patient.

In a further aspect, the anti-ILlOR antibody and the PRR agonist(s) of the composition of the invention are for simultaneous co-administration in a separate vehicle or in a same vehicle. When the anti-ILlOR antibody and the PRR agonists are for simultaneous co-administration in a separate vehicle, they can be administered according to a same administration route or according to different administration routes. Examples of administration routes suitable for the invention have been already cited herein above. When the anti-ILlOR antibody and the PRR agonists are simultaneously co-administered in a separate vehicle according to a same administration route, the administration is at the same region or a different region of the subject body. Preferably, the anti-ILlOR antibody(ies) and the PRR agonist(s) the composition of the invention are simultaneously co-administered in a same vehicle and peritumorally.

In another aspect, the antibody(ies) and the PRR agonist(s) of the composition of the invention are administered according to a sequential administration. More particularly, either the anti-ILlOR antibody(ies) is(are) administered before the PRR agonist(s) or the PRR agonist(s) are administered before the anti-ILlOR antibody(ies). They can be administered according to a same administration route or according to different administration routes. Examples of administration routes suitable for the invention have been already identified herein above. In the context of the present invention, slow release formulation, or a slow release apparatus may be used for continuous administration of the herein described composition.

Also herein described is a composition for use for treating skin precancerous lesions, for example a composition for use for treating actinic keratosis and/or Bowen disease. Preferably, the skin precancerous lesions are actinic keratosis. Skin precancerous lesions targeted by the invention may also be actinic cheilitis or premalignant cutaneous horn.

In another aspect, a composition as disclosed herein above is described for use for treating of skin carcinoma such as skin squamous cell carcinoma (SCC) and/or skin basal cell carcinoma (BCC).

Also herein described is a method for preventing or treating a precancerous lesion and/or cancer comprising a step of administering a composition according to the invention to a subject in need thereof.

Preferably, the precancerous lesion is selected from actinic keratosis and Bowen disease. Preferably, the cancer is a skin carcinoma, preferably a skin carcinoma selected from a skin squamous cell carcinoma (SCC) and a skin basal cell carcinoma (BCC).

Typically, a method for preventing or treating actinic keratosis, Bowen disease, skin squamous cell carcinoma (SCC) and/or skin basal cell carcinoma (BCC) is herein described. This method comprises a step of administering a composition according to the invention to a subject in need thereof.

Further herein described is a kit including a therapeutic combination according to the invention, that-is-to-say a composition comprising (an effective amount of) at least one PR agonist(s) and (an effective amount of) at least one anti-IL-lOR antibody.

Unless otherwise defined, all technical and scientific terms used herein have the meaning commonly understood by a person who is skilled in the art in the field of the present invention.

The invention is further illustrated below by the following Examples, which are not intended to limit its scope. Brief description of the drawings

Figure 1 shows the TLR (A) and CLR (B) transcripts levels expressed by skin-associated dendritic cells and intralesional dendritic cells. Transcriptomic analyses were performed on sorted dendritic cells from skin of control skin (naive skin, n=2), PMA treated-mice (PMA skin, n=3), and from skin surrounding papillomas (Papilloma skin, n=4), skin surrounding tumors (Tumor skin, n=4), papillomas (n=6) and tumors (n=5) of pooled DMBA/PMA treated FVB/N mice. The data correspond to the mean values of raw expression of selected TLR and CLR transcripts in the different DC samples.

Figure 2 shows the in vitro experimental approach disclosed in example 1.

Figure 3 shows the stimulation of dendritic cells sorted from TCI cutaneous tumors with different TLR agonists. The TLR9 agonist, the ODN CpG class C 2395 as well as other nucleic acids types including Imiquimod and Resiquimod (TLR7L), ORNSA19 (TLR13L) and zymozan (TLR2/Dectinl agonists) induce the production of IL-12p40 by intratumoral DCs contrary to other TLR ligands that induce the secretion of IL-6 and TNF but no or low levels of IL-12p40. The intratumoral DCs also produce high levels of IL-10 compared to FLT3L-derived DCs from naive C57/B6 mice after stimulation with several TLR agonists. Cytokine concentration (IL10, IL-6, TNF and IL-12p40) were measured in culture supernatants of intratumoral DCs and control FLT3L-derived DCs after in vitro stimulation during 22 hours with indicated TLR and CLR agonists using the CBA technology. Bars represent the mean value of cytokine concentration ± SEM for tumor DC (n= between 4 and 6), control DC (n=l). The cytokine production by non-stimulated intratumoral DCs was substracted.

Figure 4 shows the cytokine pattern secreted by total intratumoral CD45 ⁺ leukocytes and leukocytes from draining lymph nodes after in vitro stimulation. CD45 ⁺ Leukocytes were isolated from 14 days and 21 days tumors (A) or from the lymph nodes draining 21 days tumors or draining wild-type control mice (B). 100 000 CD45 ⁺ leukocytes were stimulated with PMA/ionomycin for 4 hours. Culture supernatants were collected and frozen. Cytokines measured using the CBA technology.

Figure 5 shows the cytokine pattern produced by intratumoral CD45 ⁺ leukocytes after in vitro stimulation with several TLR and CLR agonists. Intratumoral leukocytes were isolated from 14 days tumors and CD45+ cells sorted. 100 000 CD45 ⁺ intratumoral leukocytes were stimulated with different TLR and CLR ligands in presence of a blocking antibody specific for the IL10R alpha chain/CD210 (Blocking IL10R) or an isotypic antibody (IC). Culture supernatants were collected after 72h and cytokines measured using the CBA technology. Data correspond to a first experiment.

Figure 6 shows the cytokine pattern produced by intratumoral CD45 ⁺ leukocytes after in vitro stimulation with selected TLR agonists. Intratumoral leukocytes were isolated from 14 days tumors and CD45 ⁺ cells sorted. 100 000 CD45 ⁺ intratumoral leukocytes were stimulated with different TLR and CLR ligands in presence of a blocking antibody specific for against the IL10R alpha chain/CD210 (Blocking ILIOR) or an isotypic antibody (IC). Culture supernatants were collected after 72h and cytokines measured using the CBA technology. Data correspond to a second experiment.

Figure 7 shows the impact of local treatments on tumor sizes and leukocyte infiltration.

(A) The mice were sacrificed at day 22 and the tumoral volumes measured after the indicated in vivo treatment using the elipsoid formula 7i/6*H*W*D. (B) Tumors were dissociated by enzymatic digestion and the percentage of tumor-associated CD45 ⁺ leukocytes determined by flow cytometric analysis.

Figure 8 shows the impact of local TLR treatments on dendritic cells and neutrophils infiltration and on DC maturation. The mice were sacrificed at day 22 and the tumors were dissociated by enzymatic digestion. (A) Gating strategy for flow cytometric analysis of myeloid cell subsets. (B) Percentages of tumor-associated immature and mature dendritic cells and neutrophils and (C) expression of the maturation markers CD40 and CD86 by mature dendritic cells (black bars) and immature dendritic cells (grey bars) were determined by flow cytometric analysis.

Figure 9 shows the impact of local treatments on the proportion of tumor-associated CD4 ⁺ T cells, CD8 ⁺ T cells and NKp46 ⁺ ILC. (A) Gating strategy for lymphoid cell subsets analysis by flow cytometry. (B) Mean values ± SD of the frequency of the indicated subsets in two independent experiments.

Figure 10 shows the impact of local treatments on the proportion of tumor-associated FoxP3 ⁺ CD4 ⁺ T cells. A) Gating strategy used for the flow cytometric analysis of FoxP3 CD4 ⁺ T cells. (B) Mean values ± SD of the frequency of the indicated subsets in one experiment. Figure 11 shows the impact of in vivo local treatments on T cell polarization. Tumors and draining lymph nodes from control and treated mice were collected and dissociated 3 days post-treatment. Sorted CD45 ⁺ tumor cell suspensions (100 000/w) (A) and CD45 ⁺ lymph nodes cells (100 000/w) (B) were stimulated in vitro with anti CD3/CD28 coated Dynabeads and culture supernatants collected after 24h. The cytokines produced were measured by the CBA technology. Data correspond to mean values ± SD (n=2).

Figure 12 shows the concentration of the IL10R blocking antibody needed in the ODN2395/IL10R local treatment to induce the regression of tumors and the reprogramming of T cell immunity after treatment. Fourteen-days tumor-bearing mice (n=4 per group) received one peritumoral injection of ODN2395 (25μg) or control ODN (25μg) with the indicated quantity of IL10R blocking antibody. (A) Tumoral volumes were calculated the day of (before) the treatment (circles), 7 days post-treatment (squares) and 15 days post-treatment (triangles) using the elipsoid formula 7i/6*H*W*D. (B) D29 Tumors were dissociated and total tumor cell suspensions (500 000 cells/well) from mice bearing control tumors (IC/ODNc) and from treated mice bearing residual tumors were stimulated with anti CD3/CD28 coated Dynabeads and culture supernatants collected after 24h and frozen. The cytokines produced were measured by the CBA technology.

Figure 13 shows the impact of injection routes of IL10R antibody on tumor outgrowth.

Mice bearing 12-day s old TCI intradermal tumors received the treatment as indicated and the tumor outgrowth was followed by daily measurement.

Figure 14 shows that the intradermal injection of ODN2395 and IL10R antibody in non lesional skin does not induce any inflammation or necrotic lesions. Cryosections (20μιη) were performed on the back skin of control mice and mice treated with 25 μg of ODN2395 together with 50μg of IL10R antibody and were stained with hematoxylin/eosin coloration. Magnification (upper panels: xlO, lower panels: x40).

Figure 15 shows that the regression of primary tumors induced a memory-type response that delays the growth of secondary tumors. Mice that rejected primary skin tumors and naive control mice were injected intradermally with 20 000 TCI tumor cells. The tumoral volumes were measured 11 days (circles), 18 days (squares) and 27 days (triangles) post-grafting using the elipsoid formula 7i/6*H*W*D.

Figure 16 shows that TLR9L/CD210 treatment controls the tumor growth of intradermal 38T1 tumors and potentiates the production of IFN-Y and TNF antitumoral cytokines. FVB/N mice were injected intradermally with 200 000 38T1 DMBA/PMA derived epithelial tumor cell line. Mice were treated 30 days after grafting with either 25μg of IL10R antibody (ODN2395/IL10R group) or with control molecules used at the same concentration (ODNc/IC group). A) Tumoral volumes were determined before (circles) and 12 days after treatment (rectangles) for control (ODNc/IC) and treated mice (ODN2395/IL10R). B) Cytokine profiles were determined by CBA technology on culture supernatants obtained after 24 hour stimulation of total tumor cells from control mice (circles) and treated mice (rectangles) with anti CD3/CD28 coated Dynabeads. Example 1: In vitro consequences of TLRs stimulation on DCs phenotype and polarization of intratumoral leukocytes in mice.

The aim of this example is to determine the in vitro consequences of TLRs stimulation on dendritic cells (DCs) phenotype/function and on the cytokines secreted by intratumoral leukocytes in mice.

• Methodology

Specific TLRs or CLRs were shown to be expressed by the dendritic cells sorted from papillomas and carcinomas (Figure 1). Among them, several agonists of selected TLRs and CLRs were used to stimulate sorted intratumoral DCs. These agonists include: acetylated lipopeptides (a TLR1/2 agonist, Pam2Csk4 used at ^g/mL; a TLR2/6 agonist, Pam3Csk4 used at 2μg/mL), Imiquimod ^® used at l(^g/mL (a TLR7 agonist) and Resiquimod ^® used at l(^g/mL (re cognized by TLR7 in mice and TLR7 and TLR8 in humans), 23S rRNA oligoribonucleotide from salmonella used at 2μg/mL (TLR13 agonist) and protozoan profilin- like proteins such as toxoplasma gondii profilin protein used at 3μg/mL (recognized by TLR11 and TLR12) and CpG ODNs (TLR9 agonists) including class A, B and C CpG. The CpG ODN 2395 has been used in the following experiments at a concentration of 5μΜ. CpG ODN 2395 is a CpG ODN of class C with the broadest range of applications and usable in both mice and humans. Zymosan and Curdlan directed against the CLR dectin-1 were used at l(^g/mL. TDB (Trehalose-6,6-dimycolate from M. tuberculosis), a Mincle agonist, was used at l(^g/mL.

The effect of these TLR ligands was first tested using mouse dendritic cells freshly isolated from skin tumors by cell sorting. These DCs were isolated from skin TCI tumors obtained by the syngeneic grafting of the epithelial TCI tumor cell line by intradermal injection in C57BL/6 mice. Flt3-ligand derived DC subsets (CD1 lb ⁺ FLDC and CD24 ⁺ FLDC) were used as control DCs. TLR and CLR agonists from Invivogen were used at the optimal concentration recommended by the manufacturer (CpG ODN, LPS, Imiquimod, Resiquimod, Zymosan and TDB) or were titrated on DCs to determine the optimal concentration in dose/response experiments (Pam3Csk4 used at 20, 2 and 0.2 μg/mL; Pam2CSK4 used at 1 and O. ^g/mL; ORNSA19 used at 2 and O^g/mL; Profilin used at 15μg/mL and 3μg/mL; Curdlan used at 100 and 10μg/mL). Secondly, their immunostimulatory potential was evaluated using intratumoral CD45 ⁺ leukocytes as illustrated in Figure 2. Functions of DC subsets after in vitro TLR/CLR stimulation was assessed by measuring the cytokine levels produced by sorted intratumoral DCs (100 000 DCs/well) cultured in vitro during 22 hours with the proposed TLR/CLR ligands. The following cytokines were quantified in DCs culture supernatants using the CBA technology (BD Biosciences): IL6, IL10, TNF, ILlb, IL12p40, IL12p70 and IL23pl9.

The polarization of intratumoral leukocytes was measured in vitro by analyzing the cytokines produced by sorted CD45 ⁺ leukocytes (100 000 leukocytes/well) isolated from TCI tumors after 4h stimulation with a mix of PMA (100 ng/mL; Sigma- Aldrich) and ionomycin (^g/mL; Sigma- Aldrich) or after 24h or 72 hours of stimulation with the above mentioned TLR or CLR ligands. Cytokines including Thl (IFNY, IL12), Th2 (IL4, IL5 and IL13), Thl7 (IL17A), regulatory (IL10) and pro -inflammatory (TNF, ILip, IL6) cytokines were quantified in culture supernatants using the CBA (technology). In some experiments, the intratumoral leukocytes were stimulated with the indicated TLR ligand in presence of the neutralizing CD210/IL10R1 antibody (BioXcell, clone lB1.3a) or its isotypic control at the indicated concentrations.

• Results

DC functional maturation through TLR/CLR stimulation

Phenotypic analyses revealed that most intratumoral DCs expressed intermediate levels of MHCII as well as CD40 and CD86 co-stimulatory molecules when freshly isolated from tumors. Their in vitro stimulation with the different TLR agonists increased MHCII and CD86 expression without notable differences between agonists indicating that TLR stimulation induced the maturation of intratumoral DCs. The TLR agonists used included Imiquimod (TLR7L), Resiquimod (TLR7L/TLR8L), Class C CpG ODN (ODN 2395) (TLR9L) and its control, Staphylococcus A. ORN (ORN SA19) (TLR13L) and its ORN control, Pam2Csk4 (TLR2/6L), Pam3Csk4 (TLR1/2L), Toxoplasma gondii profilin (TLR11/12L), ultrapure E. coli LPS (strain 0111 :B4) (TLR4L) and zymosan (TLR2 and Dectin-1 agonist) used at optimal concentration.

In a second step, sorted intratumoral DCs were stimulated in vitro with the above mentioned TLR and CLR agonists and the cytokines produced in culture supernatant were quantified (Figure 3). Importantly, intratumoral DCs produced huge amounts of IL10 in response of most stimuli used contrary to control DCs (Figure 3A). Furthermore, intratumoral DCs produced substantial levels of IL12p40 in response to ODN2395 (TLR9L), as well as in response to other nucleic acids types including Imiquimod and Resiquimod (TLR7L), ORNSA19 (TLR13L) and in response to zymozan (TLR2/Dectinl agonists). They did not secrete any IL-12 after stimulation with TLR1/2 and TLR2/6 agonists (Pam2Csk4 and Pam3Csk4, respectively) or LPS contrary to control non tumoral CD24 Flt3L-derived DCs (Figure 3B). Both intratumoral DCs and Flt3L-derived DCs secreted huge amounts of inflammatory cytokines IL-6 and TNF in response of most stimuli (Figure 3C and 3D).

In summary, these biased cytokine responses of intratumoral DCs after in vitro stimulation with stimuli that activate control DCs indicate they have a specific cytokine pattern related to a protumoral profile. Such profile is expected to be associated with low efficacy to activate anti-tumoral CD8 ⁺ cytotoxic T lymphocytes (CTL). Consequently, these results indicate that the control of such immunosuppressive cytokines could be a central key to efficiently induce the reprogramming of antitumoral immunity.

Reprogramming of the antitumoral immunity

Sorted intratumoral CD45 ⁺ total leukocytes have been used to screen more ligands and combination of ligands in presence or not of blocking antibodies. Such CD45 based assay has the advantage to take into account the impact of stimulation on all the immune cells present in the tumoral environment and consequently reflects more the in vivo situation.

As illustrated in Figure 4, tumor leukocytes produced huge amount of IL10 and TNF, as well as low amounts of IFN-Y (also herein identified as IFNg) and IL-17A in response to broad unspecific PMA/ionomycin stimulation. Furthermore, the leukocytes from lymph nodes draining the tumor produced less IFN-Y than leukocytes from a wild type mouse. This indicates that the tumoral environment is immunosuppressive and that the IL-10 immunosuppressive cytokine plays a role in the repression of the protective antitumoral Thl immunity.

The next step was to determine whether selected TLR/CLR agonists alone or in combination with an anti-ILlOR antibody blocking the signaling through the IL-10 immunosuppressive cytokine were able to induce the IL-12 antitumoral cytokine as well as Thl cytokines (IFNy) in the tumoral environment. As shown in figure 5 (first upper graph) and figure 6 (first below graph), the stimulation with a Class C CpG ODN (ODN2395) was the best stimulus (either alone or in association with an anti-ILlOR antibody) that induced amounts of IL-12 as shown in two independent experiments (figures 5 and 6). Imiquimod, resiquimod and zymozan were also able to induce the secretion of IL12, when used in association with an anti-ILlOR antibody. Imiquimod, resiquimod and zymozan were not able to induce the secretion of IL12 when each used in the absence of an anti-ILlOR antibody. More importantly, the blockade of soluble IL10 by adding an anti-ILlO receptor alpha chain antibody (CD210) during the stimulation dramatically increased the production of IL12 after TLR9 stimulation (25-fold increase) and after resiquimod (TL7L) and zymozan stimulations (10-fold increase in both conditions). In comparison, Imiquimod which was used as the reference immunostimulatory treatment against basal cell carcinoma did not induce the production of IL12 when used in the absence of an anti-ILlOR antibody. Further, the same stimuli also induced the production the IFN-γ antitumoral cytokine when used in combination with CD210 which can be explained by the type of stimulation.

In summary, TLR9L stimulation alone or in combination with the neutralizing CD210/IL10R1 antibody restored the production of IL12 by intratumoral leukocytes. To a less extent, imiquimod, resiquimod and zymosan also restored the production of IL12 when used in combination with CD210. These TLR agonists were the most promising treatments that inventors selected for in vivo studies.

Example 2: In vivo validation of molecule selected in vitro using an orthotopic model of carcinoma based on the intradermal grafting of TC-1 epithelial tumor cells.

The aim of this example is to validate the molecules selected in vitro using a mouse model of carcinoma based on the intradermal grafting of TC-1 epithelial tumor cells.

• Methodology

A mouse model of carcinoma based on the intradermal grafting of the epithelial TCI tumor cell line was used. 20 000 tumor cells were injected intradermally and tumoral volumes were measured 22 days after grafting using the ellipsoid formula (tumoral volume=7i/6.L.W.D). In absence of treatment, mice developed tumors with volumes comprised between 500 and 600 mm ³ . When the tumors reached a volume comprised between 30 and 50 mm ³ (day 7 post- injection), mice were homogeneously distributed in the indicated groups (between n= 3 to 7 mice per group) for treatments at D7, D14 and D18 by peritumoral injection: PBS and ODN control (control groups), Imiquimod, Resiquimod, ODN 2395 alone. Due to its efficacy, ODN2395 with antilLlOR antibody at indicated concentrations was injected peritumorally only once at D7 or D14 as well as its control (ODNc/IC). A group of mice were treated with the reference treatment used for BCC which is Aldara® cream (imiquimod containing cream). Aldara® was applied topically every other day as recommended in published setting. The impact of the different treatments on tumor growth, immune cell infiltrates and T cell immunity was analyzed as detailed in result section. • Results

First, the impact of local treatments as detailed above on tumor outgrowth was assessed. As shown in Figure 7 A, ODN2395 treatment significantly delayed the tumor growth as well as the topical treatment with Aldara® but these two treatments were not sufficient to induce complete tumor regression. The most potent treatment was the one combining ON2395 (TLR9L) and 50 μg of IL10R blocking antibody that led to a total regression of tumors with only a single treatment at day 7 whereas the antibody alone had no effect on tumor growth. Due to the dramatic effect of the local treatment with TLR9 agonist/ILlOR antibody on tumor regression, inventors secondly tested the impact of a treatment combining ODN2395 and 5μg of IL10R antibody given at day 14 on tumors that reached approximately 100 mm ³ . Such treatment is efficient to induce the total regression of the tumors in 3 of the 4 treated mice. The fourth tumor had considerably regressed and had a volume of 28mm ³ . Interestingly, as shown in Figure 7B with the analysis of the frequency of infiltrating leukocytes after treatment, the regressing tumor of the ODN2395 +IL10R 5μg group had a threefold increase in infiltrating leukocytes which suggests that such treatment stimulates the antitumoral immunity.

Secondly, the impact of local treatments on the frequency of infiltrating myeloid cells and on DCs maturation was analyzed. Tumors were resected at day 22 and cells dissociated enzymatically. Cell suspensions were stained with fluorescent antibodies to quantify the frequency of DCs and neutrophils as well as DCs maturation based on MHCII, CD40 and CD86 expression by flow cytometric analysis (Figure 8). As shown in Figure 8B, all the treatments increased the frequency of mature DCs and reduced the proportion of neutrophils that are supposed to be immunosuppressive cells in the tumor environment. Interestingly, the ODN2395/IL10R treatment is the one that induced the best CD86 increase on immature and mature DCs, which supports that tumor regression is associated with improved immune response.

In a third step, the impact of local treatments on frequency and phenotypes of tumor- associated lymphocytes was assessed. To that end, the following lymphocyte types were quantified by flow cytometric analysis using classical markers:

- CD4 ⁺ CD3 ⁺ and CD8 ⁺ CD3 ⁺ T lymphocytes as well NKp46+CD3neg innate lymphoid cells including natural killer cells (Figure 9);

Foxp3+ regulatory CD4 ⁺ T lymphocytes (Figure 10)

As shown in Figure 9, the combined ODN2395/IL10R treatment induced the best increase in CD8 ⁺ T lymphocytes (14-fold increase compared to control group) indicative of a role of the adaptive immunity in tumor regression. Nevertheless, treatment with ODN2395 alone increased the proportion of CD8 ⁺ T lymphocytes. This combined treatment reduced by threefold the frequency of FoxP3 ⁺ regulatory CD4 ⁺ T lymphocytes, as well as the treatment with ODN2395 (Figure 10). Altogether, these data strongly suggest that the combined ODN2395/IL10R treatment, and to a lesser extent the ODN2395 treatment, induced the reprogramming of adaptive antitumor immunity. The ODN2395-based treatment did not seem to modulate the proportion of NKp46 innate lymphoid cells at the time point studied.

In contrast, Aldara, Imiquimod and resiquimod that stimulates the immune system through TLR7 in mice increased the proportion of NKp46 ⁺ innate lymphoid cells, as well as the proportion of CD8 ⁺ T cells and did not affect the proportion of CD4 ⁺ T lymphocytes (Figures 9-10). This suggests that these TLR7 agonists mainly stimulated the innate immune compartment.

Finally, the impact of local treatments on in vivo T cell polarization was studied. To that end, the intratumoral CD45 ⁺ leukocytes were isolated from 22-days tumors after the indicated treatments and sorted tumoral CD45 ⁺ cells or cells from draining lymph nodes (100 000 cells per well) were stimulated with anti CD3/CD28 coated Dynabeads. Cell supernatants were collected 24 hours after stimulation for cytokine assessment.

As shown in Figure 11 A, ODN2395 treatment is the only one to induce the production of massive amounts of intratumoral IFNy (IFNg), suggesting a reprogramming of the adaptive antitumoral immunity. The absence of tumors after ODN2395/IL10R treatment or the too smallest size of the residual tumor precluded the analysis of cytokine profiles. Nevertheless, in order to have an idea of the immunity induced in ODN2395/IL10R experimental groups, the cytokine profiles from draining lymph nodes leukocytes were measure after anti CD3/CD28 stimulation (Figure 11B). Interestingly, we observed huge amounts of IFNy and TNF in all experimental groups including TLR9L except for the group treated with ODN2395 and IL10R at day 7. This can be explained by the fact that the analysis was done at day 22, i.e. 15 days after the in vivo treatment. At this time point, most primed lymphocytes were not awaited in draining lymph nodes but rather in peripheral tissues.

In summary, in vivo peritumoral treatments with ODN2395 and IL10R antibody were able to induce the total regression of existing tumors in a mouse model of grafted skin tumor. In comparison, the topical treatment with Aldara® alone (imiquimod containing cream) induced only a delay in tumor growth like the peritumoral administration of ODN2395 alone.

The ODN2395/IL10R combination was efficient after only one local treatment with 5 μg amounts of blocking antibody given on established tumors. The originality of this approach relied on the peritumoral delivery of a TLR ligand that potentiates IL12 production with low amounts of anti-ILlOR blocking antibody. Such combined local treatments considerably reduced adverse effects.

From a mechanistic point of view, ODN2395 alone or in combination with IL10R antibody induced the production of IL12 and IFNy by intratumoral leukocytes as well as the expansion of CD8 ⁺ T cells, in favor of a restoration of the antitumoral adaptive immunity. In contrast, in inventors' in vivo model, Imiquimod seemed to stimulate more efficiently the innate immunity (reduction of tumor-associated granulocytes and expansion of tumor-associated ILC) than the adaptive immunity (CD8 ⁺ T cell expansion and regulatory CD4 ⁺ T cell contraction).

Finally, these results are in accordance with the in vitro data disclosed in example one.

Example 3: Dose/response efficacy of anti IL10R blocking antibody combined with ODN2395 on tumor growth and antitumor immunity reprogramming.

The aim of this example is to determine the minimal efficacy dose of peritumoral anti IL10R blocking antibody needed in the ODN2395/IL10R combined treatment to induce the tumor regression of intradermal TCI tumors. · Methodology

The minimal dose of IL10R required in the ODN2395/IL10R combined treatment to induce the regression of 14 days TCI tumors was evaluated. To that end, 14-days tumor-bearing mice (4 mice per group) were injected with 25μg of ODN2395/class C ODN CpG as done in previous experiment or with the isotype control (ODNc) together with either 25μg, 5μg or ^g of anti-ILlOR blocking antibody. The impact of the treatment on tumor growth and cytokine profiles was analyzed as previously described (Figure 12).

• Results

First, tumoral volumes were measured 7 days and 15 days after the treatment corresponding to tumors of 21 and 29 days, respectively (Figure 12A).

The ODN2395 treatment with 25μg of anti IL10R blocking antibody induced a total regression of all the tumors as observed macroscopically at day D21. Nevertheless, at day 29 one of the mice had a small residual tumor. The ODN2395 treatment with 5μg of anti-ILlOR blocking antibody induced the total regression of tumors in 2 mice from 4 (50% total regression) and a delay in the growth in the two other mice. The ODN2395 treatment with lμg of anti-ILlOR blocking antibody induced a delay in tumor growth in the four treated mice but not a total regression.

Second, the impact of the doses of anti-ILlOR blocking antibody in the combined ODN2395/IL10R treatment on the reprogramming of the antitumoral immunity was measured by comparing cytokine profiles. To that end, at day 29 (15 days post-treatment), control mice and mice treated with ODN2395 and different doses of anti IL10R blocking antibody, but which did not completely reject the tumor, were sacrificed and their cytokine profiles analyzed. Total tumor cell suspensions (5000 000 per well) were stimulated with anti CD3/CD28 Dynabeads during 24 hours and culture supernatants collected for cytokine analysis with CBA technology (Figure 12B). All the tumors treated with ODN2395 and anti IL10R antibody independently of the doses of blocking antibody except one were infiltrated with IFN-Y producing immune cells. Importantly, the treated tumor that produced very low levels of IFNy is a tumor that did not respond to the treatment at all (the one with the highest tumoral volume in group ODN2395/IL10R). In contrast, control tumors produced either very low levels of IFNy or no IFNy. ¾ treated tumors produced substantial amount of IL17A compared to only ¼ non treated tumors. Moreover, all the tumors either from control and treated groups produced substantial amounts of IL10 and TNF which is in accordance with the inflammatory and immunosuppressive T cell environment of TCI tumors (Figure 12B). The levels of ILlb, IL4 and IL12p40 were very modest in all groups after CD3/CD28 stimulation which reflects the absence of Th2 effectors (IL4) and the absence of indirect in vitro stimulation of myeloid infiltrating cells (ILlb, IL12) in inventors' experimental setting (data not shown).

In summary, these data demonstrated that ODN2395/IL10R treatment induced the generation of IFNy producing T cell effectors in the majority of mice at the lowest doses of anti-ILlOR blocking antibody. Nevertheless, their direct role in tumor regression remains to be established.

Example 4: The aim of this example is to determine the impact of different routes of injection of the treatment on the tumor outgrowth.

· Methodology

The best route of administration of the anti IL10R antibody required in the ODN2395/IL10R combined treatment to induce the regression of 12 days TCI tumors was evaluated. To that end, 12-days tumor-bearing mice (3 mice per group) were injected as followed: i) 25μg of ODN2395/class C ODN CpG with 25 μg of antilLlOR antibody both by peritumoral injection; ii) 25μg of ODN2395/class C ODN CpG by peritumoral route together with 25μg of anti- IL10R antibody by intraperitoneal route; ii) 5μg of ODN2395/class C ODN CpG by peritumoral route together with 25(^g of anti-ILlOR antibody by intraperitoneal route to increase the tumoral impact of the antibody. The impact of the treatment on tumor growth was analyzed as previously described (Figure 12).

• Results

The ODN2395 treatment with 25 μg of anti IL10R blocking antibody, both administered peritumorally, induced a total regression of two of the three tumors as observed macroscopically between day 3 and 5 post-treatment. In contrast, both other treatments did not induce the total regression of any tumor.

In conclusion, the combined peritumoral administration of the anti-ILlOR antibody together with the TLR9L agonist improved the efficacy of the treatment compared to treatments in which the antibody was injected intraperitoneally. Example 5: Effect of injection of ODN2395 /IL10R treatment in non lesional skin

The aim of this example is to determine the toxicity of skin injection of ODN2395 with the highest dose of IL10R on non lesional skin. As ODN2395/IL10R peritumoral treatment was shown to induce necrotic lesions, inventors investigated the impact of an injection of ODN2395/IL10R in non lesional skin from naive C57BL/6 mice.

Interestingly, the TLR9L/CD210 treatment (25 μg ODN2395 and 25 μg of CD210) did not induce any macroscopically observable necrotic lesions or local inflammation (red inflamed skin) in all mice tested (n=10) when injected intradermally in the shaved back skin of naive mice. More, as illustrated in Figure 14, hematoxylin/eosin staining of microscopic sections of skin from control (naive or ODNc/IC) and treated mice (ODN2395/IL10R) revealed the absence of inflammatory infiltrates in treated skin compared to control skin.

In summary, the ODN2395/IL10R treatment is innocuous in non lesional skin. Example 6: Role of the adaptive immunity in the treatment-induced regression of tumors

The aim of this example is to assess whether the tumor regression induced after ODN2395/IL10R involved a memory adaptive immunity able to control the growth of secondary tumors.

The regression of primary TCI tumors after peritumoral in vivo treatment with ODN2395/IL10R was associated with the recruitment of IFN-Y producing CD8 T effectors (see examples 2 and 4). In order to assess whether the immune response induced during this tumor regression was sufficient to control the growth of secondary tumors, tumor-free mice after ODN2395/IL10R treatment were challenged by intradermal injection of TCI tumor cells as previously. Naive mice were also injected with tumor cells at the same time and tumoral volumes measured at days 11, 18 and 27 after grafting. As shown in Figure 15, the tumor growth in mice that rejected a primary tumor was delayed but not completely suppressed compared to control mice. This moderate effect can be due to the extremely rapid capacity of TCI tumor cells to grow in skin after grafting. Nevertheless, these data demonstrated that a memory-type immune response was induced during the treatment with ODN2395/IL10R of primary skin tumors.

Example 7: Efficacy of the TLR9L/CD210 treatment in the 3 mSCC-38 skin tumor model

The aim of this example is to evaluate the efficacy of ODN2395/IL10R treatment in another skin tumor model based on the orthotopic grafting of the mSCC-38 epidermal tumor cell line.

• Methodology

A mouse model of orthotopic skin tumor has been used in order to extend the preclinical data. The mSCC-38 epidermal tumor cell line was derived from a primary skin tumor generated in FVB/N mice after chemical treatment with the chemical agents DMBA and PMA. In this model, tumors developed slowly and were more difficult to measure. Tumors of sizes comprised between 10 mm ³ and 225 mm ³ develop in 30 days after intradermal injection of 200 000 cells from 38-T1 tumor. In absence of treatment, these tumors are more immunogenic than TCI skin tumors (more infiltrated with IFN-Y producing T cells) which can explain in part the size heterogeneity of tumors. 30-days tumors were treated with ODN2395 and IL10R at a dose of 5μg per mouse or with control ODNc and isotypic control antibody (ODNc+IC group). Twelve days later, tumor volumes were measured and the cytokines produced by intratumoral leukocytes measured after anti CD3/CD28 stimulation as described in example 3.

• Results

First, tumoral volumes were measured 12 days after treatment. As illustrated in Figure 16 A, ODN2395/IL10R combined treatment delayed the growth of mSCC-38 tumors compared to control (ODNc/IC). Five from seven tumors had regressed after TLR9L/CD210 treatment whereas all the tumors from control mice were progressing (n=4).

Second, the polarization of the T cell infiltrate was analyzed by quantification of the cytokines produced by culture supernatants 24 hours after anti CD3/CD28. The smallest tumors from ODN2395/IL10R combined group were pooled in order to have the numbers of cells required for each analysis. As shown in Figure 16B, the ODN2395/IL10R combined treatment increased the production of IFN-Y and TNF by T cell lymphocytes compared to control group, indicating that the tumor growth control induced by the treatment also involved the stimulation of antitumoral adaptive immunity in this second model.

In summary, ODN2395/IL10R combined treatment induced the partial control of 30-days mSCC-38 skin which is associated with a local increase in IFN-Y and TNF producing T cells that represent potential antitumoral T cell effectors.