NEW ONCO-IMMUNOLOGIC PROGNOSTIC AND THERANOSTIC MARKERS

Abstract

The present invention relates to the identification of a set of prognostic and theranostic markers enabling to assess the tumor microenvironment (immunogenic or immunosuppressive) of solid tumors and predict the responsiveness or the non-responsiveness to immunotherapy of the affected patient.

Claims

1. An assay to assess the immunogenic or immunosuppressive microenvironment of a tumor comprising the quantification in a bioptic sample of said tumor of: the expression of the hMENA11a marker in tumor cells and in the stromal cells of said sample; the expression of the FN1 marker in the stroma of said sample; the expression of the CD3, CD8 and CD20 markers in the cells of said sample; the expression of the PDL1 marker in the tumor cells of said sample wherein, the expression of hMENA11a higher than a cutoff value of 50 in the tumor cells of said sample; the absence of expression of hMENA11a in the stromal cells of said sample; the expression of the FN1 marker lower than an intensity value, called score, equal to 2 in the stroma of said sample and the expression of the CD3, CD8 and CD20 markers in at least 5 cells per mm.sup.2 in the tumor area or in the invasive margin of said sample; the presence of tertiary lymphoid structures (TLS) consisting of an aggregate of B lymphocytes within the structure and T cells outside of the same, assessed respectively by the expression of CD20, CD8 and CD3 markers, are representative of an immunogenic tumor microenvironment, the expression of hMENA11a lower than a cutoff value of 50 in the tumor cells of said sample; the expression of the FN1 marker equal to or greater than a cutoff value of score 2 in the stroma of said sample and the absence of expression of the CD3, CD8 and CD20 markers in cells within the tumor area or in the invasive margin of said sample are representative of an immunosuppressive tumor microenvironment.

2. The assay according to claim 1, wherein the expression of the marker hMena V6 and/or of the marker total hMena, wherein said total hMena marker comprises all the hMena protein isoforms, is further quantified in the tumor and stromal cells of said sample, wherein: the absence of hMena V6 expression in tumor cells of said sample and/or the total hMena expression lower than a cutoff value of 80 in the tumor cells of said sample; and the absence of any of said markers in the stroma of said sample are representative of an immunogenic tumor microenvironment; the presence of hMena V6 expression in tumor cells of said sample and an expression thereof in the stromal cells of said sample and/or the presence of total hMena expression greater than a cutoff value of 80 in the tumor cells and an expression thereof in the stromal cells of said sample are representative of an immunosuppressive tumor microenvironment.

3. The assay according to claim 1, wherein said expression is quantified by immunohistochemical techniques with specific antibodies that selectively bind the 11a isoform of hMENA of the hMena protein, specific antibodies that selectively bind CD3, specific antibodies that selectively bind CD8, specific antibodies that selectively bind CD20 and specific antibodies that selectively bind FN1 and, optionally, specific antibodies that selectively bind hMena V6 isoform of hMena protein and/or anti-total hMena-specific antibodies.

4. The assay according to claim 3, wherein said expression is quantified on consecutive sections of said sample, each for each marker, with detection techniques using secondary antibodies labeled with peroxidase-conjugated streptavidin, dextran polymers conjugated to peroxidase, or directly to dyes.

5. The assay according to claim 4, wherein said expression is quantified on one or more sections of said sample, with detection techniques using secondary antibodies labeled with a different fluorochrome for each secondary antibody.

6. The assay according to claim 1, wherein said expression is quantified by detecting mRNA of each of said markers.

7. The assay according to claim 1, wherein said biopsy sample is a sample included in paraffin.

8. A method for predicting the responsiveness or non-responsiveness, of patients with solid tumors, to treatment by immunotherapy, said method comprising analyzing the tumor microenvironment with the assay as described in claim 1, wherein patients presenting an immunogenic tumor microenvironment will be responsive to immunotherapy treatment while patients presenting an immunosuppressive tumor microenvironment will be non-responsive to treatment by immunotherapy.

9. The method according to claim 8, wherein said immunotherapy is carried out with one or more immunological checkpoint inhibitor drugs.

10. A kit for the assessment of the microenvironment of a tumor, comprising: reagents to quantify the expression of the hMENA11a marker in a biopsy sample; reagents to quantify the expression of the FN1 marker in a biopsy sample; reagents to quantify the expression of the CD3, CD8 and CD20 markers in a biopsy sample; and reagents to quantify the expression of the PDL1 marker in a biopsy sample.

11. A kit according to claim 10, further comprising reagents for quantifying the expression of the hMena V6 marker and/or reagents to quantify the expression of the total hMena marker.

12. The kit according to claim 10, wherein said reagents comprise specific antibodies that selectively bind the 11a isoform of hMENA of the hMena protein, specific antibodies that selectively bind CD3, specific antibodies that selectively bind CD8, specific antibodies that selectively bind CD20 and specific antibodies that selectively bind FN1 and, optionally, with specific antibodies that selectively bind the hMena V6 isoform of hMena protein and/or anti-total hMena-specific antibodies.

13. The kit according to claim 12, further comprising secondary antibodies labeled with peroxidase-conjugated streptavidin, dextran polymers conjugated to peroxidase, or directly to dyes.

14. The kit according to claim 12, further comprising secondary antibodies labeled with a different fluorochrome for each secondary antibody.

15. The kit according to claim 10, wherein said reagents are oligonucleotides specific for the detection of the mRNA of each of said markers.

Description

DETAILED DESCRIPTION OF THE FIGURES

[0028] FIG. 1. hMENA11a transfection significantly reduces extracellular matrix components in the secretoma of breast carcinoma cells, as demonstrated by mass spectrometry analysis. (a) Heat map of proteins secreted in different amounts in hMENAv6-expressing BT549 CNTR cells and hMENA11a-expressing clones (#14 and #104) (left). The analysis was conducted on three biological replicates. Quantification of the reduction of proteins associated with ECM composition is reported in the histograms (Right). (b) Heat map of proteins secreted differently between DAL CNTR cells and trasfectants stable for hMENA11a (left). The analysis was conducted on four biological replicates. Quantification of the reduction of proteins associated with ECM composition and also identified in BT549 cell line is reported in the histograms (Right). (c) Three networks identified by an algorithm of upstream regulators using Ingenuity Pathway Analysis (IPA).

[0029] In particular, the figure shows how the expression pattern of the two hMENA isoforms influences the composition of the extracellular matrix, which in turn plays a fundamental role in immune cell infiltration into the tumor, and therefore in response to therapy.

[0030] FIG. 2. hMENAv6 transfection increases, and hMENA11a transfection reduces fibronectin (FN1) expression of NSCLC cell lines. Low levels of FN1 in the stroma of tumors with a high expression of hMENA11a represent a favorable prognostic factor in early NSCLC patients. (a) Western blot of A549 cells transfected with empty vector (CNTR), hMENAv6 and hMENA11a with the indicated antibodies. The quantification of the increase or reduction of the expression of FN1 in cells transfected with hMENAv6 or hMENA11a is reported on the right (fold). (b) Consecutive sections of a representative NSCLC case labeled with Pan-hMENA, hMENA11a and FN1 antibodies. 20 enlargement. Scale Bar=30 m. (c) immunohistochemical characterization of 114 (lymph) node-negative NSCLCs, showing how the hMENA11a pos [high hMENA11a/low hMENA (t) as reported in Bria et al 2014] cases more frequently express low levels of FN1 in the stroma. (d-f) Kaplan-Meier evaluation of disease-free survival (DFS) of node-negative NSCLC patients according to dichotomized hybMENA11a and stromal FN1 expression.

[0031] The figure shows how the expression pattern of the two hMENA isoforms influences the expression of a fundamental component of the extracellular matrix which is fibronectin (FN1). Said data was obtained on cell lines and validated in tumor tissues, in which high levels of FN1 in the stroma surrounding the tumor are associated with low expression levels of hMENA11a in tumor cells. From a clinical standpoint, this association has a prognostic significance. Patients expressing high levels of hMENA11a in tumor cells and low levels of FN1 in the stroma have better prognosis.

[0032] FIG. 3. hMENA11a expression correlates with the presence of tertiary lymphoid structures (TLS) in the tumor area (AT) and low levels of stromal fibronectin in early (NO) NSCLC patients. Sections of a representative case decorated with antibodies toward CD3 (A), CD20 (B), hMENA11a (C) and Fibronectin (D). Magnification 100. (E) Schematic TLSs model (F) Histograms represent the percentage of tumors categorized for the expression pattern of the hMENA isoforms (hMENA11a pos: hMENA11a high/Pan-hMENA low; hMENA11a neg: all other 3 cases) and show the presence (TLS pos) or absence (TLS neg) of TLS in the tumor area (AT). (G) Histograms represent the percentage of tumors categorized by low-level (FN low) or high-level (FN high) expression of fibronectin in the stroma; and show the presence (TLS pos) or absence (TLS neg) of AT TLSs. (H) Histograms represent the percentage of tumors categorized by the expression pattern of hMENA isoforms and fibronectin expression in the stroma (HybMENA11a pos/FN low; HybMENA11a neg/FN high;) and show the presence (TLS pos) or absence (TLS neg) of TLSs AT.

[0033] The figure shows how the expression of hMENA11a is associated to low expression levels of FN1 and to the presence of tertiary lymphoid structures in the tumor area. Therefore, the evaluation of hMENA isoforms expression, in parallel with the presence of T and B cells (lymphocytes) can add information related to the functional organization of the lymphocyte infiltrate.

[0034] FIG. 4. hMENA11a expression inversely correlates with PD-L1 expression in (lymph) node-positive (N+) NSCLC and in squamous NSCLC. Two representative cases. A-B, weak immunoreactivity with anti-hMENA11a (A) and strong positivity for PD-L1 (Dako, clone 22C3). C-D, high hMENA11a expression (C), and absence of PD-L1. Alveolar macrophages are positive for PD-L1. (E-F). Histograms show inverse correlation between the expression of hMENA11a and of PD-L1 in node-positive NSCLC (E) and in squamous NSCLC (F) patients.

[0035] In a case record of lung carcinoma patients, hMENA11a expression inversely correlates with PD-L1 expression.

[0036] FIG. 5. Specific hMENA11a silencing induces increase of PD-L1 ligand at membrane protein level and of mRNA in lung carcinoma cells. EpT1Lu, H1650 e H1975 lung carcinoma cell lines were transfected with control siRNA (si-CNT), siRNAs specific for hMENA11a (si-11a) and siRNA for all hMENA isoforms (si-MENA). (A) PD-L1 levels analyzed by surface cytofluorimetric staining via specific monoclonal antibody. Bar graphs show percentage values of PD-L1 levels in EpT1Lu lines (left-side panel), H1650 (central panel) and H1975 (right-side panel). Each bar represents the mean percentage (SEM) of 5 different experiments. P value calculated by t test. (B) Analysis by Western blotting of hMENA11a and Pan hMENA expressions by the related specific antibodies. Anti-tubulin antibody was used to normalize the amount of gel-loaded protein. (C) Analysis by Real-Time PCR of mRNA levels of PD-L1. Gene GAPDH was used as endogenous control.

[0037] The figure shows that hMENA11a silencing in various lung carcinoma cell lines induces an increase of PD-L1 protein expression on cells surface (A), and an increase of its messenger RNA (C), indicating that the assessment of hMENA isoforms expression can represent not only an indication of the composition of the extracellular matrix and the lymphocyte infiltrate, and therefore of a tumor microenvironment, immunogenic or immunosuppressive, but can be correlatable to an on-tumor expression of PD-L1, ligand of the main target of immunomodulatory therapies, indicating that a patient whose tumor expresses hMENA11a, even though PD-L1-negative, might be responsive to treatment. The invention suggests the need of a method including, besides the expression of PD-L1currently the sole marker used in clinical practice to select patients to be treated with inhibitors of this pathwaythe expression of the above-cited microenvironment markers.

[0038] FIG. 6. hMENA/hMENAv6, but not hMENA11a, are expressed in the stroma of pancreatic adenocarcinoma, and in particular in CAFs, where hMENAv6 modulates their functional activity. A. Immunohistochemical analysis, with pan hMENA and hMENA11a antibodies, of consecutive sections of a representative case of pancreatic adenocarcinoma. B. Western blots of lysates of normal fibroblasts (NF) and CAFs obtained from pancreatic tissue (PDAC) or from distal fibroblasts and CAFs obtained from NSCLC lung carcinoma with the indicated antibodies. C. Matrigel invention assay on CAFs obtained from lung adenocarcinoma (L-CAF) silenced for all hMENA isoforms (left) or transfected with hMENAv6 (right). D. Collagen gel contraction assay with CAFs obtained from lung (L-CAF) or pancreatic (P-CAF) adenocarcinoma, silenced for all hMENA isoforms. E. Zymogram for MMP2 activity performed with the conditioned medium of CAFs obtained from pancreatic adenocarcinoma (P-CAF), silenced for all hMENA isoforms (left), or of distal fibroblasts obtained from lung adenocarcinoma (L-DF), transfected with hMENAv6 (right).

[0039] The figure shows that hMENA/hMENAv6, and not hMENA11a, is expressed in the stroma, and in particular in tumor-associated fibroblasts of lung and pancreatic tumors. In particular, the v6 isoform modulates the functional activity of tumor-associated fibroblasts, which in turn, by the production of extracellular matrix components and soluble factors, carry out an immunomodulating activity and therefore play a role in defining an immunosuppressive and/or non-permissive microenvironment. Said results suggest the need to introduce in the method of the invention the assessment of hMENA expression in the tumor stroma.

[0040] FIG. 7. The figure shows a graphic summary of the transition from the expression of the epithelial and anti-invasive 11a isoform of hMENA to the expression of the mesenchymal proinvasive hMENADv6 isoform. Said transition influences signaling pathways important in the dialogue among tumor cell, ECM, CAFs and immune cells, causing a change of the tumor microenvironment from immunogenic to immunosuppressive. In-tumor detection of hMENA isoforms expression pattern in association with other markers, such as fibronectin (FN1) in the stroma, the presence/absence of lymphocyte infiltrate, also organized in tertiary lymphoid structures (TLS) in the tumor area, enables to identify the immunogenic or immunosuppressive characteristics of the tumor microenvironment, providing predictive information on patients' responsiveness to immunotherapy.

GLOSSARY

[0041] For the purposes of the present description, the wording tumor microenvironment (TME) denotes, in accordance with the scientific literature, the environment in which the tumor develops and progresses, and is mainly comprised of extracellular matrix and cellular component, of which cancer-associated fibroblasts (CAFs), immune cells, among which T, B lymphocytes, tumor-associated macrophages, and endothelial cells and pericytes forming blood and lymph vessels are part. All these components cross-talk among them and with tumor cells, causing a highly dynamic situation.

[0042] The wording immune checkpoints in the present description denotes, in accordance with the scientific literature, proteins that control immune system activation, transmitting an inhibitory-type signal inside a T-lymphocyte (T-cell). Examples of these proteins include PD-1 and CTLA-4.

[0043] The wording immune checkpoint inhibitors in the present description denotes, in accordance with the scientific literature, inhibitory molecules tasked with eliminating restrictions to most of T-cell-mediated immune responses.

[0044] The wording target therapy in the present description denotes, in accordance with the scientific literature, a type of pharmacological therapy for the treatment of tumors addressed at opposing the specific mechanisms of the carcinogenesis process of the individual tumors, by interfering with molecules needed for tumor growth and progression. Drugs of this type are accordingly effective toward defined neoplasia types, and are therefore selective (hence, reference is also made to customized therapy). Target therapy intervenes on the mechanisms linked to oncogene and oncosuppressor gene expression, underlying the specific action of tumor promotion, entailing transformation of the cell from normal to pathological.

[0045] The wording invasive margin in the present description corresponds to that commonly accepted in oncologic histology and denotes the region of the histological section at the boundary between tumor cells and untransformed tissue. In the literature, by invasive margin it is meant a region of about 500 micrometers on each side of the boundary between tumor tissue and untransformed tissue cells.

[0046] The wording tertiary lymphoid structures (TLS) in the present description denotes, in accordance with the scientific literature, lymph node-like structures present in many solid tumors, comprised of T cells and mature dendritic cells surrounded by a B lymphocyte follicle.

[0047] For the purposes of the present invention [0048] hMENA alias ENAH, Gene ID: 55740 [0049] hMENA11a, Enah variant 1 ID: NM_001008493 splicing variant containing an additional exon (11a) [0050] hMENADv6, hMENA splicing variant lacking exon 6 ID: EU255274.1 [0051] hMENA11a: HybMENA11a hybrid variable pos=[high hMENA11a/low hMENA (t)]; HybMENA11a neg (all other 3 cases) [0052] FN1 fibronectin 1 Gene ID: 2335 [0053] PD-L1, alias CD274, Gene ID: 29126 [0054] CD3, T cell co-receptor, consisting of one CD3 chain, one CD3 chain, and two CD3 chains [0055] CD20 B lymphocyte marker, Gene ID: 931 [0056] CD4, Gene ID: 920 [0057] CD8, Gene ID: 925

[0058] In the present description, the terms total hMena or pan-hMena are used as synonyms.

DETAILED DESCRIPTION OF THE INVENTION

[0059] The present invention therefore enables to define the tumor microenvironment of solid tumors and to predict responsiveness of patients affected by said tumors to immunotherapy on the basis of said microenvironment.

[0060] Therefore, object of the invention is an assay to assess the microenvironment, immunogenic or immunosuppressive, of a tumor comprising the quantification in a bioptic sample of said tumor: [0061] the expression of the hMena11a marker in tumor cells and in the stromal cells of said sample; [0062] the expression of the FN1 marker in the stroma of said sample; [0063] the expression of the CD3, CD8 and CD20 markers in the cells of said sample; [0064] the expression of the PDL1 marker in the tumor cells of said sample
wherein,
the expression of hMena11a higher than a cutoff value of 50 in the tumor cells of said sample; the absence of hMena11a expression in the stromal cells of said sample; the expression of the FN1 marker lower than an intensity value (score) 2 in the stroma of said sample; the expression of the CD3, CD8 and CD20 markers in at least 5 cells per mm.sup.2 in the tumor area or in the invasive margin of said sample; the presence of tertiary lymphoid structures (TLS) consisting of an aggregate of B lymphocytes within the structure and T cells outside of the same, assessed respectively by the expression of markers CD20 and CD3, are representative of an immunogenic tumor microenvironment;
while the expression of hMena11a lower than a cutoff value of 50 in the tumor cells of said sample; the expression of the FN1 marker equal to or greater than a cutoff value of score 2 in the stroma of said sample and the absence of expression of the CD3, CD8 and CD20 markers in cells within the tumor area or in the invasive margin of said sample are representative of an immunosuppressive tumor microenvironment.

[0065] The presence/absence of TLSs (tertiary lymphoid structures) can be assessed by morphological analysis of the immunohistochemical staining of CD3, CD8 and CD20, in which clusters of CD20+ cells surrounded by CD3-positive areas are considered TLSs.

[0066] In one embodiment of the above-described assay, the expression in tumor and stromal cells of said sample of the hMena V6 marker and/or the total hMena marker expression can be further quantified, wherein said total hMena marker comprises all hMena protein isoforms, and wherein:

the absence of hMena V6 expression in tumor cells of said sample and/or the total hMena expression lower than a cutoff value di 80 in the tumor cells of said sample;
and the absence of any of said markers in the stroma of said sample are representative of an immunogenic tumor microenvironment; while
the presence of hMena V6 expression in tumor cells of said sample and an expression thereof in the stromal cells of said sample and/or
the presence of total hMena expression greater than a cutoff value of 80 in the tumor cells and an expression thereof in the stromal cells of said sample are representative of an immunosuppressive tumor microenvironment.

[0067] The quantifications in the above-described assay can be performed, for instance, by immunohistochemistry methods as explained in more detail hereinafter. The detection of the two different hMENA isoforms can therefore be performed with specific antibodies for each isoform, i.e. antibodies specifically binding only a single hMena isoform. In case of antibodies specific for the 11a isoform, and therefore binding exclusively said isoform, the antibodies can bind, e.g., epitopes present in exon 11a, and therefore absent in all other isoforms.

[0068] In case of antibodies specific for the v6 isoform, and therefore binding exclusively said isoform, the antibodies can bind epitopes astride exon 6 deletion sites, therefore absent in all other isoforms.

[0069] Antibodies specific for hMENA isoforms are described in the literature.

[0070] Antibodies specific for total hMena, i.e. antibodies binding specifically epitopes common to all hMENA isoforms, and therefore binding exclusively hMena protein in all his forms, are also described in the literature.

[0071] In the figures appended to the present description, data of experiments performed with antibodies specific for the 11a isoform of hMena and with antibodies specific for all hMENA isoforms (also called anti-total hMena, o pan-hMena antibodies) are depicted.

[0072] In this case, the data obtained with these two markers are also representative of the v6 isoform (as confirmed by analyses performed on mRNAs).

[0073] For instance, the staining for pan-hMENA and hMENA11a can be quantified both by an intensity value (score) from 0 to 3 and by the percentage of positive cells. A continuous variable is generated as product of intensity and percentage, obtaining a numerical value from 0 to 300. Then, a ROC analysis is performed in order to find the optimal cutoff of the two variables. The identified cutoff is different for the two variables (80 for pan-hMENA and 50 for hMENA11a).

[0074] By pan-hMENA low expression (low), for the purposes of the present invention a value of <80 is meant, whereas by pan-hMENA high expression (high) a value of >80 is meant. As to hMena11a, by low expression (low), for the purposes of the present invention a value of <50 is meant, whereas by hMENA11a high expression (high) a value of >50 is meant. Therefore, for equal values of pan-hMENA and hMENA11a (e.g., 60) pan-hMENA low (as <80) and hMENA11a high (as >50) will be had. As indicated above, the expression value is calculated as product between staining intensity (score 0, 1, 2, 3) and percentage of positive cells (0-100), whereas the optimal cutoff value for the two variables has been identified by a ROC analysis (Bria 2014, above). It is known that in tumor cell lines the expression of hMENA11a and hMENADv6 is alternative, and positivity for pan-hMENA is always accompanied by positivity for one or the other of the hMENA11a and hMENADv6 isoforms. These data were obtained by western blot, using antibodies specific for hMENA11a; hMENADv6 and pan-hMENA, which recognizes all hMENA isoforms. On the basis of these preclinical results on cell line models, it can be speculated that the group of tumors that are pan-hMENA high/hMENA11a low includes tumors expressing the hMENAv6 isoform, even though this is not directly assessable owing to the lack of an antibody specific for hMENAv6 that works in immunohistochemistry.

[0075] The computation of FN1 marker expression is performed by using an extremely simple staining intensity scale, from 0 to 3: score 0: no staining; score 1: weak, 2: moderate, and 3: strong. Scores 0-1 are merged and labeled as low, whereas scores 2-3 are combined and labeled as High. A person skilled in the field could evaluate staining intensity with respect to other markers used in histopathological practice as positive controls, indicating the occurred staining reaction; such controls, as is known, vary tumorwise and a person skilled in the field will be able to select from the literature the most suitable markers. As mentioned above, in one of the embodiments of the invention, the assay according to any one of the above-described embodiments, the expression of the various markers is quantified by immunohistochemistry techniques with specific antibodies that selectively bind each marker, i.e.: specific antibodies that selectively bind the 11a isoform of hMENA of the hMena protein, specific antibodies that selectively bind CD3, specific antibodies that selectively bind CD8, specific antibodies that selectively bind CD20 and specific antibodies that selectively bind FN1 and, optionally, with specific antibodies that selectively bind the hMena V6 isoform of the hMena protein and/or anti-total hMena antibodies, i.e. antibodies that bind to the hMena protein in all its known isoforms. Hence, specific for the hMena protein, but unable to recognize the isoforms. One example of said antibodies are antibodies targeting epitopes that are not in exon 11a and that are not included in the zone in which exon 5 is joined to exon 7, typical of the deltaV6 form.

[0076] The immunohistochemistry techniques can be performed according to any method known to the technician in the field, on bioptic samples that are fresh, frozen or stored in paraffin, also by including (embedding) therein.

[0077] For the purposes of the present invention, suitable bioptic samples can be samples obtained by standard techniques known to the technician in the field, such as biopsy, surgery or suction.

[0078] For instance, the expression can be quantified on consecutive sections of the bioptic sample by using one section for each marker, and then using standard immunohistochemistry techniques with secondary antibodies labeled with peroxidase-conjugated streptavidin, or alternatively by using polymers with the Polymer Refine Detection Bond method.

[0079] According to an alternative embodiment, the expression of the markers indicated in any one of the embodiments described herein, immunohistochemistry techniques can be used that envisage the use of secondary antibodies labeled with different fluorochromes and the detection of the different markers, also on the same section of the bioptic sample, such as, e.g., by sequential or simultaneous multiplex immunohistochemistry under fluorescence (fICH) as described in the literature. Depending on the preferred detection system, the person skilled in the field will use primary antibodies as described, specially provided secondary antibodies and suitable image acquisition devices. In any case, for quantification by immunohistochemistry the technician in the field could use any standardized immunohistochemistry technique to perform the assay of the invention.

[0080] Alternatively, the quantification as described herein can be performed by detecting the mRNA of each of said markers, e.g., by QRT-PCR or RNA-Seq techniques.

[0081] On the basis of the assay described herein, object of the invention is also a method for predicting the responsiveness or non-responsiveness, of patients with solid tumors, to treatment by immunotherapy, optionally in combination with radiotherapy or chemotherapy or target therapy, which includes assessment of the tumor microenvironment with the assay as described in any one of the embodiments indicated in the present description, wherein patients presenting an immunogenic tumor microenvironment will be responsive to immunotherapy treatment, while patients presenting an immunosuppressive tumor microenvironment will be non-responsive to treatment by immunotherapy.

[0082] What described for any embodiment of the assay of the invention applies to the method for predicting the responsiveness to treatment by immunotherapy and to the therapeutic method providing a step of classifying the patient to be treated as responsive or non-responsive to immunotherapy.

[0083] Therefore, object of the invention is also a therapeutic method for the treatment of patients with solid tumors, wherein, prior to defining the therapeutic regimen, the patient is subjected to a method of classification as responsive or non-responsive to immunotherapy as described above, and wherein, if the patient is classified as responsive, immunotherapy is administered, optionally combined with radiotherapy, chemotherapy, or target therapy, while, if the patient is classified as non-responsive, chemotherapy, or radiotherapy, or a target therapy is administered, without administration of immunotherapic drugs or active ingredients.

[0084] According to the present description, immunotherapy can be carried out with any immunotherapic drug for solid tumors commonly used in therapy.

[0085] A non-limiting example of such drugs comprises therapeutic vaccines or immune (immunological) checkpoint inhibitors.

[0086] In one embodiment, immunotherapy can be carried out with one or more immune checkpoint inhibitor drugs as defined in the glossary. A non-limiting example of immune checkpoint inhibitor drugs is represented, e.g., by Ipilibumab, Nivolumab or Pembrolizumab.

[0087] Ipilimumab (Yervoy) is a monoclonal IgG1 antibody binding CTLA-4 and preventing the reduction of cytotoxic T cells.

[0088] PD-1 signaling pathway blockage is deemed to modulate the activity of T cells that have become non-responsive after contacting PD-1 ligands in the tumor microenvironment.

[0089] Currently, there are two anti-PD-1 drugs at an advanced development stage: Nivolumab and Pembrolizumab.

a) Nivolumab (BMS-936558), a completely human immunotherapic drug, IgG4 inhibitor of immune checkpoint PD-1
b) Pembrolizumab (MK-3475) is a humanized IgG4 monoclonal antibody binding PD-1. Anti-PD-L1: anti-PD-L1 therapies also target the PD-1 signaling pathway by binding with one of its ligands, e.g., PD-L1.

[0090] As mentioned above, immunotherapy could be administered in combination (simultaneous or sequential) with radiotherapy, chemotherapy or target therapy. Examples of target therapy drugs may be given from drugs of humanized monoclonal antibody category of humanized monoclonal antibodies (trastuzumab, cetuximab), which act by interfering with the receptors present on the cell surface, as in the case of the epidermal growth factor receptor (EGFR), one of the target molecules proved more effective for target therapy. Another example are Small molecules (Erlotinib, Imatinib) which penetrate the cell membrane to interact with the target protein in the cell, interfering with the enzymatic activity of the protein.

[0091] A further object of the invention is a kit for the assessment of the microenvironment of a tumor, comprising: [0092] reagents to quantify the expression of the hMena 11a marker in a biopsy sample; [0093] reagents to quantify the expression of the FN1 marker in a biopsy sample; [0094] reagents to quantify the expression of the CD3, CD8 and CD20 markers in a biopsy sample; and [0095] reagents to quantify the expression of the PDL1 marker in a biopsy sample.

[0096] According to a further embodiment, the kit can further comprise reagents to quantify the expression of the hMena V6 marker and/or reagents to quantify the expression of the total hMena marker, i.e. of all the hMena protein isoforms in a biopsy sample.

[0097] The kit of the invention can be a kit for detection by immunohistochemistry, or for detection by nucleic acids analysis, or both.

[0098] In the case of immunohistochemical kits, the reagents for detection can comprise specific antibodies that selectively bind the 11a isoform of hMENA of the hMena protein, specific antibodies that selectively bind CD3, specific antibodies that selectively bind CD8, specific antibodies that selectively bind CD20 and specific antibodies that selectively bind FN1 and, optionally, specific antibodies that selectively bind the hMena V6 isoform, and/or anti total hMena antibodies specifically binding all Hmena protein isoforms.

[0099] The kit can further comprise secondary antibodies labeled, e.g., with peroxidase-conjugated streptavidin, or labeled with a different fluorochrome for each secondary antibody depending on the desired immunohistochemical method as described above, or the detection can comprise the Polymer Refine Detection Bond system.

[0100] Alternatively, or jointly, the kit according to any embodiment described herein can comprise as reagents oligonucleotides specific for detecting mRNA of each of said markers.

[0101] The sequences of the selected markers are known in the literature (as reported above) and primers specifically amplifying said sequences are also known. For instance, as is of common knowledge, for hMena variants primer sequences selectively amplifying isoforms are used, for hMena11a primers are used which specifically amplify regions comprising in whole or in part exon 11a, for the commonly known hMena isoform primer sequences are used which amplify one or more regions comprising the whole exon 11 or part thereof, as well as the whole exon 6 or part thereof, for the hMena V6 isoform, instead, primers comprising the zone astride exon 5 and exon 7 are used.

[0102] Hereinafter, examples of embodiments of the invention are provided. These examples are not aimed at limiting the invention but are merely to provide possible embodiments as a guide for the technician in the field.

[0103] The use of conventional techniques different from those described below is envisaged among the embodiments of the invention.

[0104] The following examples aim at illustrating ways for carrying out the invention but are not intended to limit the protection scope thereof.

[0105] All analyses performed on bioptic samples reported herein were performed on samples collected from informed patients who gave their consent to the treatment of said samples.

EXAMPLES

[0106] 1.

Immunohistochemical Analysis

[0107] Tissues fixed 18-24 h in 4% formaldehyde are included in paraffin. Immunohistochemical labeling is performed on tissues of about 5-m thickness. Sections are mounted on SuperFrost Plus slides (Menzel-Glaser, Braunschweig, Germany) and, once deparaffined and rehydrated, are treated in a thermostated bath at 96 C. for 40 minutes in a 10 mmol/L citrate solution (pH 6). Immunohistochemical labeling for hMENA11a; pan-hMENA; CD3, CD20; CD4; CD8; PD-L1 is preferentially detected by Bond Polymer Refine Detection (Leica Biosystem, Milan, Italy) with an automatic autostainer (Bond Max, Leica).

[0108] Reactivity toward hMENA isoforms by immunohistochemistry is assessed with 4 different scores: no staining, score 0; weak diffused staining in <10% of cells, score 1+; moderate cytoplasmatic staining in 10%-70% of cells, score 2+; intense staining in >70% of cells; in the presence of membrane reinforcement, score 3+.

[0109] Fibronectin expression in the stroma is assessed by using a scale from 0 to 3: score 0: no staining; score 1: weak, 2: moderate, and 3: intense). Grades 0-1 are merged and labeled as Low, whereas grades 2-3 are combined and labeled as High.

[0110] PD-L1 expression is considered positive if at least 50% of tumor cells are positive.

[0111] The expression study of CD3, CD20 CD4, CD8 markers is assessed by dividing the tissues into two areas: invasive margin and tumor central area. The expression of CD3, CD20 CD4, CD8 markers is assessed by immunohistochemistry, counting the number of positive cells per mm.sup.2 in the tumor central area and its invasive margin.

[0112] Cell count per mm.sup.2 and the relative computation can be performed by digital pathology with a dedicated software, and the result is given by the combination of the 4 markers for the two regions analyzed.

[0113] Alternatively, the expression of CD3, CD20 CD4, CD8 markers in tumor tissues can also be assessed at mRNA level by QRT-PCR, using primers or probes known to a technician in the field, or by RNA-Seq.

[0114] The presence/absence of TLS is assessed by morphological analysis of the immunohistochemical staining of CD3 and CD20, in which CD20+ cell clusters surrounded by CD3-positive areas are considered TLSs.

Experimental Results

[0115] 1. hMENA.sup.11a Modifies ECM Composition.

[0116] A proteomic analysis was performed by mass spectrometry on the secretoma of two clones of breast carcinoma cells BT549 transfected for 11a isoform of hMENA and on control BT549 cells not expressing hMENA.sup.11a but expressing the hMENADv6 isoform.

[0117] Among 1719 proteins identified in the conditioned medium (CM) of the cells, 373 proteins (10% false discovery rate (FDR), q=0.1) differently expressed in the two hMENA.sup.11a clones with respect to control cells were observed (FIG. 1a). Upstream regulator analysis, performed with Ingenuity Pathway (Qiagen, MD) had identified the downregulation of different pathways important in BT549/hMENA.sup.11a clones. TGF1 activation pathway demonstrated to be the most significant (Activation z-score=4,389, p-value=1.12E-20), along with the FN1 pathway (Activation z-score=2.472 p-value=1.79E-06) and 1 integrin (Activation z-score=2.182, p-value=0.000282) (FIG. 1c).

[0118] Said data were confirmed in another breast carcinoma cell line, DAL, which expresses undetectable levels of hMENA(t), stably transfected with hMENA.sup.11a. Secretoma analysis showed that 1324 proteins were significantly different in the CM of cells transfected for hMENA.sup.11a and in the control cells (FIG. 1b). Among differently-expressed proteins, ECM and metalloproteinase component reduction were observed in both BT549 hMENA.sup.11a and DAL hMENA.sup.11a, compared with control cells (FIG. 1a, b).

2. hMENA.sup.11a High Expression, Along with Low Stromal FN1 Prolongs Disease-Free Survival in Node-Negative NSCLC Patients.

[0119] Considering the mass spectrometry results showing how a high hMENA.sup.11a/hMENAv6 expression ratio decreases the secretion of a cluster of proteins belonging to the FN1 pathway and considering the relevance of FN1 expression in lung carcinoma, the Inventors evaluated the effect of hMENA.sup.11a and hMENAv6 transfection on FN1 expression of lung carcinoma lines. The results have shown that hMENA.sup.11a transfection reduces, whereas hMENAv6 increases FN1 level (FIG. 2a, b).

[0120] To confirm the clinical relevance of these experimental results, the Inventors analyzed FN1 the expression in 114 tumor tissues from early node-negative NSCLC patients, in parallel with Pan-hMENA and hMENA.sup.11a staining, assessed as hybMENA.sup.11a-positive (comprising hMENA.sup.11a high/hMENA(t) low cases) or -negative (comprising all other cases, and therefore hMENAv6-expressing tumors). The results show that FN1 expression in the stroma is differently distributed in hybMENA.sup.11a-positive or -negative tumors (FIG. 2c, d) (p=0.03). High FN1 was expressed more significantly in the hybMENA.sup.11a-negative group, including hMENAv6-expressing tumors. On the contrary, hybMENA.sup.11a-positive tumors fundamentally express low FN1 levels in the stroma, in accordance with proteomics data obtained in the cell lines. Of clinical relevance, Kaplan-Meier curves for DFS, in accordance with the combination of the two variables, indicate that 72.5% of patients with hybMENA.sup.11a-positive and low FN1 are disease-free at 5 years (p<0.0001), whereas, when considering only hybMENA.sup.11a expression, the Inventors found that 63.5% of patients is disease-free at 5 years (FIG. 2e). These data indicate that the combination of high hMENA.sup.11a and low stromal FN1 represents a promising prognostic factor in early node-negative NSCLC patients.

3. hMENA.sup.11a Expression Correlates with Tertiary Lymphoid Structures (TLS) Presence and Low Fibronectin Levels in Tumor Tissues of Early Node-Negative NSCLC Patients

[0121] The research group of the Authors of the present invention reported that hMENA.sup.11a expression correlates with a favorable prognosis in a group of NSCLC patients (Bria, 2014). An accurate immunohistochemical analysis for the presence of TLS in 192 cases of early (node-negative) NSCLC showed the presence of TLS in the tumor area (assessed as clusters of CD20+ cells surrounded by CD3+ cells) in 46 cases on 158 examined. 48 (25%) cases exhibited reactivity for hMENA.sup.11a. TLS presence in the tumor area is moreover significantly associated with high positivity for hMENA.sup.11a (hMENA.sup.11a positive p<0.0001) (FIG. 3f).

[0122] Since mass spectrometry data obtained by the research group of the Inventors indicate that hMENA expression pattern influences ECM composition, which in turn can influence immune cell infiltration, the Inventors assessed fibronectin expression in the stroma of the analyzed cases, finding that tumors with high FN1 in the stroma have less TLS (FIG. 3g) and that hybMENA.sup.11a-positive tumors exhibit low levels of FN1 in the stroma, and more frequently TLS in the tumor area (p=0.003) (FIG. 3h).

4. hMENA.sup.11a Expression Inversely Correlates with PD-L1 Expression in 40 Node-Positive (N+) NSCLCs and in 45 Squamous NSCLCs. hMENA.sup.11a Specific Silencing Induces Increase of PD-L1 Ligand at Membrane Protein Level, and of mRNA in Lung Carcinoma Cells.

[0123] An immunohistochemical analysis for the expression of PD-L1 (ligand of immune checkpoint PD-1) on a case record of 40 positive-node NSCLC cases in parallel with the assessment of the expression of hMENA(t) and hMENA.sup.11a demonstrated that only 20% of tumors expressing high hMENA.sup.11a and low hMENA(t) express PD-L1, which is instead expressed in 44.4% of low hMENA.sup.11a and high hMENA(t) cases (FIG. 4e). Similar results were obtained in a case record of 45 cases of squamous NSCLCs (FIG. 4f). Considering said data, the Inventors assessed, in experimental models of lung carcinoma cell lines, whether hMENA.sup.11a expression might influence PD-L1 expression. The results obtained show that hMENA.sup.11a specific silencing by interfering RNA (siRNA) induces an increase of PD-L1 levels both at a cell membrane protein level, as evaluated by cytofluorimetry (FIG. 5a), and at RNA level, as assessed by QRT-PCR (FIG. 5c). On the contrary, the silencing of all hMENA isoforms by a siRNA that complexes to an RNA sequence common to all hMENA transcripts does not influence PD-L1 expression. These data indicate that hMENA.sup.11a expression level in tumor cells can influence T lymphocyte phenotype.

5. hMENA/hMENAv6, but not hMENA.sup.11a, are Expressed in Pancreatic Carcinoma Stroma, and in Particular in CAFs where hMENADv6 Modulates their Functional Activity

[0124] The research group of the Inventors published that the expression patterns of hMENA isoforms represents a prognostic factor in pancreatic carcinoma, one of the tumors more resistant to immune checkpoint inhibitors, with hMENA.sup.11a expression associated with a favorable prognosis (Mechionna 2016). An accurate immunohistochemical analysis shows that hMENA/hMENADv6, and not hMENA.sup.11a is expressed in the stroma and in particular in tumor-associated fibroblasts (FIG. 6A). By using primary cultures of fibroblasts obtained from surgical collections of pancreatic tumors (CAFs) or of normal pancreatic islets (NFs) and of fibroblasts obtained from the tumor area (CAFs) or the distal area (DFs) of surgical collections of NSCLC lung cancers, the Inventors demonstrated that CAFs express higher hMENADv6 levels compared to NFs or DFs (FIG. 6B). Functional studies further demonstrated that hMENADv6 isoform modulates CAF functional activity. In fact, hMENADv6 silencing or overexpression respectively reduces or increases CAF ability to invade a matrigel matrix (FIG. 6C) and to produce active metalloproteinase 2 (MMP2) (FIG. 6E). Moreover, hMENA/hMENADv6 silencing reduces a characteristic CAF function, which is the ability to contract collagen gel (FIG. 6D). Said results suggest the need to introduce in the method of the invention the assessment of the expression of hMENA in the tumor stroma.