ANTI-NEUROPILIN-1 AND ANTI-PROGRAMMED CELL DEATH-1 COMBINATION THERAPY FOR TREATING CANCER

Abstract

The present invention relates to the combined use of a neuropilin-1 (Nrp-1) neutralizing agent and of a programmed cell death-1 (PD-1) neutralizing agent for killing cancer cells, typically for treating cancer, as well as to corresponding pharmaceutical compositions and kits, and to corresponding diagnostic and therapeutic methods. The invention further relates to in vitro, ex vivo and in vivo methods for detecting CD8.sup.+ TILs capable of recognizing cancer cells, for predicting the response of a subject to anti-PD-1 treatment of cancer, and for identifying a subject who responds therapeutically to a treatment of cancer with an antibody combination therapy comprising anti-Nrp-1 and anti-PD-1 antibodies.

Claims

1-18. (canceled)

19. An in vitro or ex vivo method for detecting CD8.sup.+ TILs capable of recognizing cancer cells, the method comprising a) providing a biological sample comprising TILs and b) determining whether said TILs express Nrp-1, the expression of Nrp-1 by the TILs indicating that said TILs are capable of recognizing cancer cells.

20. The in vitro or ex vivo method for detecting CD8.sup.+ TILs capable of recognizing cancer cells according to claim 19, wherein step b) comprises determining i) whether said TILs express Nrp-1, and further comprises determining ii) whether said TILs express one or several receptors selected from CTLA-4, Tim-3 and LAG-3 and/or iii) the PD-1.sup.+ status of said TILs.

21. The in vitro or ex vivo method for detecting Nrp-1.sup.+ CD8.sup.+ TILs capable of recognizing cancer cells according to claim 19, wherein the CD8.sup.+ TILs capable of recognizing cancer cells are PD-1.sup.T TILs.

22. The in vitro or ex vivo method for detecting CD8.sup.+ TILs capable of recognizing cancer cells according to claim 20, wherein the presence of Nrp-1.sup.+ PD-1.sup.+ CD8.sup.+ TILs is indicative that said TILs are dysfunctional.

23. The in vitro or ex vivo method for detecting CD8.sup.+ TILs capable of recognizing cancer cells according to claim 19, wherein the cancer is a solid tumor, a carcinoma, a sarcoma or a blastoma.

24. The in vitro or ex vivo method for detecting CD8.sup.+ TILs capable of recognizing cancer cells according to claim 23, wherein the cancer is a carcinoma, a lung cancer, non-small cell lung cancer (NSCLC) or a melanoma.

25. The in vitro or ex vivo method for detecting CD8.sup.+ TILs capable of recognizing cancer cells according to claim 19, wherein the cancer is a hematological tumor, a lymphoma, a leukemia or a multiple myeloma.

26. An in vitro or ex vivo method for predicting the response of a human subject to a treatment of lung cancer combining anti-PD-1 and anti-Nrp-1 agents, the method comprising detecting CD8.sup.+ TILs recognizing cancer according to the method of claim 19, the presence, in the CD8.sup.+ TILs of a biological sample of the subject, of about 15±5% of Nrp-1.sup.+ PD-1V CD8.sup.+ TILs being indicative that the human subject responds to the treatment of cancer.

27. A method for activating PD-1.sup.+ CD8.sup.+ tumor-infiltrating T lymphocytes (TILs) anti-cancer cell effector functions and/or for enhancing cancerous tumor infiltration by PD-1.sup.+ CD8.sup.+ TILs having active anti-cancer cells effector functions, or for treating cancer, in a subject, wherein the method comprises a step of administering, to a subject having a cancer Nrp-1.sup.+ CD8.sup.+ tumor-infiltrating T lymphocytes (TILs) exhibiting anti-cancer cell effector functions and/or enhanced cancerous tumor infiltration ability, wherein said TILs have been obtained from the subject having a cancer and have been amplified ex vivo before the administration step.

28. The method according to claim 27, wherein Nrp-1.sup.+ CD8.sup.+ TILs are PD-1.sup.T TILs.

29. The method according to claim 27, wherein the cancer is a solid tumor, a carcinoma, a sarcoma or a blastoma.

30. The method according to claim 29, wherein the cancer is a carcinoma, a lung cancer, non-small cell lung cancer (NSCLC) or a melanoma.

31. The method according to claim 27, wherein the cancer is a hematological tumor, a lymphoma, a leukemia or a multiple myeloma.

32. A method for treating cancer in a subject, wherein the method comprises a step of administering an antibody combination comprising an effective amount of i) a Nrp-1 neutralizing antibody and ii) an immunotherapeutic antibody to a subject having a cancer, said subject expressing Nrp-1.sup.+ CD8.sup.+ TILs or Nrp-1.sup.+ PD-1.sup.+ CD8.sup.+ TILs.

33. The method for treating cancer according to claim 32, wherein the immunotherapeutic antibody is selected from a PD-1 neutralizing antibody, a TIM-3 neutralizing antibody, a LAG-3 neutralizing antibody and a CTLA-4 neutralizing antibody.

34. The method for treating cancer according to claim 32, wherein the immunotherapeutic antibody is a PD-1 neutralizing antibody.

35. The method for treating cancer according to claim 32, wherein the method further comprises a step of administering Nrp-1.sup.+ CD8.sup.+ TILs to the subject and wherein said TILs have been obtained from the subject having a cancer and have been amplified ex vivo before the administration step.

36. The method for treating cancer according to claim 32, wherein the cancer is a solid tumor, a carcinoma, a sarcoma or a blastoma.

37. The method for treating cancer according to claim 36, wherein the cancer is a carcinoma, a lung cancer, non-small cell lung cancer (NSCLC) or a melanoma.

38. The method for treating cancer according to claim 32, wherein the cancer is a hematological tumor, a lymphoma, a leukemia or a multiple myeloma.

Description

LEGENDS TO THE FIGURES

[0170] FIG. 1. Expression of Nrp-1 and PD-1 on T cells infiltrating human NSCLC tumors.

[0171] a. Surface expression of Nrp-1 on CD4.sup.+ and CD8.sup.+ T cells from TIL and PBL. TIL from freshly resected NSCLC tumors were isolated and then directly analysed by flow cytometry for Nrp-1 expression on CD3.sup.+ CD4.sup.+ and CD3.sup.+ CD8.sup.+ lymphocytes. Expression of Nrp-1 on CD4.sup.+ and CD8.sup.+ T cells from HD and NSCLC patient's PBL was evaluated. Percentages of positive cells are indicated. Right: percentages of Nrp-1.sup.+ CD8.sup.+ and Nrp-1.sup.+ CD4.sup.+ T cells in TIL (n=28) and PBL from HD (n=12) and NSCLC patients (n=11). b. Expression of CD25 and PD-1 on Nrp-1.sup.+ CD8.sup.+ T cells from NSCLC TIL. Percentages of Nrp-1.sup.+ T cells among CD8.sup.+ TIL expressing or not CD25 or PD-1 (n=13-to-16). Right: dot plot showing co-expression of Nrp-1 and PD-1 on CD8.sup.+ TIL from one representative patient. c. Expression of CD25 and PD-1 on Nrp-1.sup.+ CD4.sup.+ T cells from NSCLC TIL. Percentages of Nrp-1.sup.+ T cells among CD4.sup.+ TIL expressing or not CD25 or PD-1 (n=9-to-20). Right: dot plot showing co-expression of Nrp-1 and PD-1 on CD4.sup.+ TIL from one representative patient. d. Expression of Foxp3 in Nrp-1.sup.+ CD4.sup.+ T cells. Percentages of Nrp-1.sup.+ T cells among CD4.sup.+ TIL expressing or not Foxp3. Right: dot plot for co-expression of Nrp-1 and PD-1 on CD4.sup.+ TIL expressing or not Foxp3 from one representative patient. Data presented as mean±SEM. * p<0.05; ** p<0.01 and *** p<0.001.

[0172] FIG. 2. Expression of Sema-3A and Nrp-1 in human lung tumor cell lines and CTL clone.

[0173] a. Sema-3A protein expression in human lung tumor cell lines. Total protein extracts from lung tumor cell lines were analysed by western blot using anti-Sema-3A mAb. The bronchial epithelial cell line 16HBE was included as a control. Full length and proteolytically processed proteins are indicated. Anti-F-actin was included as a loading control. b. Expression of Nrp-1 on the P62 CTL clone unstimulated and stimulated with anti-CD3 mAb. Right: Co-expression of Nrp-1 and CD25, and Nrp-1 and PD-1 on P62 T cells stimulated with immobilized anti-CD3. c. Sema-3A-Fc binds to Nrp-1 on the P62 CTL clone surface. The P62 T-cell clone was unstimulated or stimulated with anti-CD3 for 48 h, pre-incubated for 30 min with Sema-3A-Fc (12 μg/mL), and then labelled with mouse anti-human IgG Fc fragment secondary mAb. d. Sema-3A-Fc inhibits CTL clone migration toward a CXCL12 gradient. The P62 T-cell clone was stimulated with anti-CD3 for 48 h, pre-incubated for 30 min with BSA or Sema-3A-Fec, and then seeded in the upper chambers of transwell plates and exposed to a gradient of CXCL12 chemokine loaded in the lower chambers. The number of T cells that had migrated into the lower chambers was determined. Results are represented as mean chemotaxis index ±SD of triplicate samples. e. Cytotoxic activity of the CTL clone toward autologous tumor cells. The P62 T-cell clone was stimulated with plastic-coated anti-CD3 for 48 h, left in media for 24 h to detach anti-CD3 mAb and then pre-incubated in medium or with Sema-3A-Fc. Cytotoxicity toward the cognate IGR-Pub tumor cell line was determined by a conventional 4 h .sup.51Cr release assay at indicated E:T ratios. Data shown correspond to one of 3 independent experiments. * p<0.05; ** p<0.01.

[0174] FIG. 3. Expression of Sema-3B and Nrp-1 in B16F10 mouse melanoma model.

[0175] a. Expression of Sema-3B in B16F10 tumor cells. Total protein extracts from B16F10 tumor cells cultured in vitro or isolated ex vivo from tumor grafts were analysed by western blot using anti-Sema-3B mAb. Anti-F-actin was included as a loading control. b. Surface expression of Nrp-1 on CD4.sup.+ and CD8.sup.+ T cells infiltrating B16F10 melanoma engrafted in C57BL/6 mice. TIL from individual tumors were isolated at day 15 after tumor cell inoculation using CD45 beads, and then labelled with anti-Nrp-1, -CD3, -CD4 and -CD8 mAb. T lymphocytes from spleens and TdLN of tumor-bearing mice were analysed in parallel. Percentages of positive cells are included. Right: percentages of Nrp-1.sup.+ cells among CD8.sup.+ and CD4.sup.+ T cells in TIL (n=30) and splenocytes and TdLN (n=16-to-30). c. Expression of Nrp-1 and Foxp3 in CD4.sup.+ T cells. T lymphocytes from tumors, spleens and TdLN of B16F10 melanoma-bearing C57BL/6 mice were analysed at day 15 by flow cytometry. Right: Percentages of Nrp-1.sup.+ cells among Foxp3.sup.+ and Foxp3.sup.− CD4.sup.+ T lymphocytes from B16F10 (n=11-to-20). d. Expression of CD44 and CD62L on Nrp-1.sup.+ and Nrp-1.sup.− CD8.sup.+ T cells from B16F10 TIL. Right: Distribution of Nrp-1.sup.+ cells among CD8.sup.+ TIL subpopulations among naive, effector and memory T cells (n=10). Results are representative of 3-5 independent experiments. Data are presented as mean±SEM. *** p<0.001.

[0176] FIG. 4. Expression of T-cell activation/exhaustion markers on the CD8.sup.+ TIL surface.

[0177] a. Expression of Nrp-1, PD-1, LAG-3, CTLA-4 and Tim-3 on CD8.sup.+ T cells from B16F10 TIL isolated at day 15. Down: percentages of Nrp-1.sup.+ T cells among CD8.sup.+ T cells expressing or not PD-1, Tim-3, LAG-3 or CTLA-4 (n=16-to-24). Data are means±SEM. *** p<0.001. b. Expression of Nrp-1 on CD8.sup.+ T cells correlated with that of PD-1. c. Expression of Nrp-1 on CD8.sup.+ TIL is restricted to a PD-1.sup.hi T-cell population. Expression of Nrp-1 and PD-1 on CD8.sup.+ TIL; identification of three T-cell subsets: Nrp-1.sup.+ PD-1.sup.hi (31%), Nrp-1.sup.− PD-1.sup.+ (40%) and Nrp-1.sup.− PD-1.sup.− (24%). Down: Expression of exhaustion/activation markers, functional proteins and transcription factors in Nrp-1.sup.+ PD-1.sup.hi, Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− CD8.sup.+ T-cell subsets. Heat maps including percentages of cells positive for expression of PD-1, Nrp-1, Tim-3, CTLA-4, LAG-3, granzyme B (GrzB) and Ki-67, and gMFI for NFATc1, IRF-4, Helios, Blimp-1 and T-bet. d. gMFI of PD-1 expression on Nrp-1.sup.+ PD-1.sup.hi, Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− TIL subsets. Results are representative of 3 to 5 independent experiments.

[0178] FIG. 5. The Nrp-1.sup.+ PD-1.sup.hi TIL subset is enriched with activated antigen-specific CD8.sup.+ T lymphocytes.

[0179] a. Staining of CD8.sup.+ TIL with Trp2 and gp100 dextramers. C57BL/6 mice were engrafted with B16F10 melanoma cells and then vaccinated with Trp2 and gp100 antigenic peptides as described in Materials and Methods. On day 15, CD8.sup.+ T cells were isolated from TIL, labelled with anti-CD8, -Nrp-1 and -PD-1 mAb, and Trp-2 and gp100 dextramers, and then percentages of antigen-specific T cells were determined. Right: Percentages of Trp2 and gp100 dextramer-positive T cells among Nrp-1.sup.+ PD-1.sup.hi, Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− TIL. b. Cytoplasmic expression of IFNγ and TNFα in Nrp-1.sup.+ PD-1.sup.hi, Nrp-1-PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− CD8.sup.+ T cells. TIL were stimulated for 4 h with autologous tumor cells, stained with anti-CD8, -Nrp-1 and -PD-1 mAb and, after membrane permeabilization, with anti-IFNγ and -TNFα. mAb (n=18). c. Degranulation capacity of CD8.sup.+ TIL. TIL were stimulated with autologous tumor cells; then, Nrp-1.sup.+ PD-1.sup.hi, Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− T-cell subsets were analysed for surface expression of CD107a. Mean percentages of CD107a.sup.+ lymphocytes are shown (n=16). d. Cytotoxic activity of freshly isolated CD8.sup.+ TIL toward B16F10 tumor cells. CD8.sup.+ TIL were pre-incubated in medium or with anti-Nrp-1, anti-PD-1, a combination of both blocking mAb or an isotype control; then, cytotoxicity toward autologous tumor cells was determined by 4 h and 12 h .sup.51Cr release assay at 50:1 E:I ratio. e. Increase in MHC class I and PD-L1 expression on B16F10 tumor cells co-cultured with autologous CD8.sup.+ TIL. Kinetic studies of H-2K.sup.b/D.sup.b and PD-L1 expression on B16F10 cells co-cultured with CD8.sup.+ TIL for indicated time points. Expression profiles (left), percentages of positive cells (middle) and gMFI (right) of MHC-class I (upper panels) and PD-L1.sup.+ (lower panels) tumor cells are shown. f. Anti-Nrp-1 re-establishes migration of Nrp-1.sup.+ PD-1.sup.hi T cells toward B16F10 tumor cell supernatant. TIL were pre-incubated for 30 min in medium or with neutralizing anti-Nrp-1, -PD-1, a combination of both mAb or an isotype control. Cells were seeded in the upper chambers of transwell plates and then exposed to a gradient of B16F10 supernatant loaded in the lower chambers. The numbers of Nrp-1.sup.+ PD-1.sup.hi, Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− T cells that had migrated into the lower chambers were determined by flow cytometry. Results are representative of 3-to-4 independent experiments. Data presented as mean±SEM. * p<0.05; ** p<0.01 and *** p<0.001.

[0180] FIG. 6. Anti-Nrp-1 mAb combined with anti-PD-1 improve tumor progression control.

[0181] a. C57BL/6 mice were engrafted with B16F10 melanoma cells and then treated i.p. with anti-PD-1, i.t. with anti-Nrp-1, or a combination of both mAb injected in parallel at each site at days 6, 8, 10, 12 and 14 after tumor inoculation. An isotype control was included. Tumor volumes, measured every second day in mm.sup.3, are given as means (±SEM) of 5-to-7 mice/group. Mice were sacrificed at day 17, since tumor size exceeded the tolerated institutional limit. Data are means of two independent experiments out of three. * p<0.05; ** p<0.01 and *** p<0.001. b. Weight of tumors from mice untreated and treated with blocking mAb. Tumors were recovered at day 16 after engraftment and weighed. c. Absolute cell counts of CD3.sup.+ CD8.sup.+ TIL. Numbers of CD8.sup.+ T cells per milligram of tumor were determined at day 16 as described in Materials and Methods. d. Ratio of CD8.sup.+/CD4.sup.+ Treg cells in tumors from mice treated with blocking mAb or an isotype control. e. Absolute cell counts of KLRG1.sup.+, Ki-67.sup.+ and granzyme B+CD8.sup.+ T cells from B16F10 TIL. Numbers of T-cell subsets per milligram of tumor from mice treated with anti-Nrp-1, anti-PD-1, a combination of both mAb or an isotype control. Data are means from two independent experiments out of three. * p<0.05; ** p<0.01 and *** p<0.001.

[0182] FIG. 7. Expression of NRP, PLXN and SEMA3 genes in human lung tumor samples.

[0183] a. Relative expression of NRP1 and NRP2 transcripts in fresh NSCLC tumors performed by qRT-PCR analysis. b. Relative expression of PLXNA1, PLXNA2, PLXNA3, PLXNA4 and PLXND1 transcripts in NSCLC tumors performed by qRT-PCR analysis. c. Relative expression of SEMA3A, SEMA3B, SEMA3C, SEMA3D, SEMA3E, SEMA3F and SEMA3G transcripts in fresh human lung tumors. Expression was normalized to autologous healthy lung tissues (n=8).

[0184] FIG. 8. Expression of Nrp-1, Sema-3A and CXCR4 in human cells.

[0185] a. Expression of Nrp-1 on CD8.sup.+ T lymphocytes from HD PBL unstimulated (medium) and stimulated with immobilized anti-CD3 mAb. Right: Percentages of Nrp-1.sup.+ CD8.sup.+ T cells in HD PBL unstimulated and stimulated with anti-CD3 (n=10). b. Expression of Nrp-1 on CD4.sup.+ T lymphocytes from HD PBL unstimulated (medium) and stimulated with immobilized anti-CD3. Right: Percentages of Nrp-1.sup.+ CD4.sup.+ T cells in HD PBL unstimulated and stimulated with anti-CD3mAb (n=10). c. Expression of Sema-3A in human lung tumor cell lines. Intracellular immunofluorescence analysis of Sema-3A expression in NSCLC tumor cells. Full line: Anti-Sema-3A; dashed line: isotypic control mAb. d. Expression of CXCR4 chemokine receptor on the human P62 CTL clone surface. Full line: anti-CXCR4 mAb; dashed line: isotypic control.

[0186] FIG. 9. Expression of Sema-3B and Nrp-1 in the B16F10 mouse melanoma model.

[0187] a. Expression of Sema3 genes in B16F10 tumor cells cultured in vitro (left) or engrafted in C57BL/6 mice (right). Relative expression of Sema3A, Sema3B, Sema3C, Sema3D, Sema3E, Sema3F, 41 Sema3G and Sema4A transcripts in B16F10 tumor cells was determined by qRT-PCR analysis. b. Kinetic studies of Nrp-1 expression on CD8.sup.+ T cells from naive mouse splenocytes unstimulated or stimulated with immobilized anti-CD3 mAb. c. NFAT inhibitor inhibits Nrp-1 induction on CD8.sup.+ T cells. The NFAT inhibitor 11R-VIVIT inhibits expression of Nrp-1 induced by anti-CD3 activation of naive mouse splenocytes in a dose-dependent manner. d. Kinetic studies of Nrp-1, PD-1, LAG-3, CTLA-4 and Tim-3 induction on CD8.sup.+ TIL from B16F10 melanoma engrafted in C57BL/6 mice. Percentages of inhibitory receptor-expressing CD8.sup.+ T lymphocytes are shown at indicated time points. e. Absolute numbers of CD8.sup.+ TIL expressing Nrp-1, PD-1, LAG-3, CTLA-4 or Tim-3 per milligram of tumor. f. Expression of Batf; Cd244, Tigit and Egr2 genes in mouse CD8.sup.+ TIL. Relative expression of Batf; Cd244, Tigit and Egr2 transcripts in Nrp-1.sup.+ PD-1.sup.hi, Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− T-cell subsets from B16F10 TIL was determined by qRT-PCR analysis at day 15 after tumor engraftment. g. Expression of inhibitory receptors on CD4.sup.+ TIL. Percentages of Nrp-1.sup.+ T cells co-expressing or not PD-1, Tim-3 or CTLA-4 among Foxp3.sup.− CD4.sup.+ T lymphocytes from B16F10 melanoma (n=5-9). h. Expression of Ki-67 in Nrp-1.sup.+Foxp3.sup.− CD4.sup.+ T cells. Percentages of Nrp-1.sup.+ T cells co-expressing or not Ki-67 among Foxp3.sup.− CD4.sup.+ T lymphocytes infiltrating B16F10 melanoma (n=10).

[0188] FIG. 10. Tumor growth and infiltration of B16F10 melanoma.

[0189] a. C57BL/6 mice were engrafted with B16F10 cells and then vaccinated with Trp2 peptide delivered with poly(I:C) at indicated time points (arrows) or with poly(I:C) control alone (vehicle). Tumor volumes, measured every third day in mm.sup.3, are given as means (±SEM) of 5 mice/group. b. Kinetic studies of inhibitory receptor expression on CD8.sup.+ T cells infiltrating B16F10 tumors. Absolute numbers of CD8.sup.+ T cells expressing inhibitory receptors in TIL from Trp-2-vaccinated mice and control mice. Numbers of PD-1.sup.+, Nrp-1.sup.+, LAG-3.sup.+, CTLA-4.sup.+ and Tim-3.sup.+ CD8.sup.+ TIL per milligram of tumor are determined at indicated time points. c. H-2Kb/Db expression profiles of B16F10 cells cultured in vitro in medium alone or with IFNγ for 12 h, or isolated ex vivo at day 15. d. Growth of B16F10 tumors engrafted in C57BL/6 mice treated with blocking anti-Nrp-1, -PD-1, a combination of both mAb or an isotype control as described in Materials and Methods. Individual mouse tumors are shown.

[0190] FIG. 11. The Nrp-1.sup.+ PD-1.sup.hi TIL subset is enriched with activated antigen-specific CD8.sup.+ T cells.

[0191] a. Cytotoxicity of freshly isolated CD8.sup.+ TIL. CD8.sup.+ TIL were pre-incubated in a medium or with anti-Nrp-1, anti-PD-1 or a combination of both mAb; then, cytotoxicity toward autologous tumour cells was determined.

[0192] b. Expression of perforin in CD8.sup.+ T cells. TILs were stimulated with autologous tumour cells in the absence or presence of neutralizing anti-Nrp-1, anti-PD-1 or anti-Nrp-1 plus anti-PD-1, then, T-cell subsets were analyzed for expression of perforin (n=5).

[0193] FIG. 12. Trp-2 peptide vaccine increases the number of Nrp-1+PD-1hi TILs that are enriched with activated antigen-specific CD8.sup.+ T cells.

[0194] a. Kinetic studies of inhibitory receptor expression on CD8.sup.+ T cells infiltrating B16F10. Absolute numbers of CD8.sup.+ T cells expressing inhibitory receptors in TILs from Trp-2-vaccinated mice and control mice. Numbers of PD-1.sup.+, Nrp-1.sup.+, LAG-3.sup.+, CTLA-4.sup.+ and Tim-3.sup.+ CD8.sup.+ TIL per milligram of tumour are determined.

[0195] b. Left: Staining of CD8.sup.+ TIL with dextramers. C57BL/6 mice were engrafted with B16F10 and then vaccinated with Trp2 and gp100 peptides. On day 15, TILs were isolated from tumours.

[0196] Right: Percentages of Trp2 (n=8) and gp100 (n=5) dextramer-positive T cells among Nrp-1.sup.+ PD-1.sup.hi Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− CD8.sup.+ TIL.

[0197] FIG. 13. Nrp-1 is enriched on PD-1.sup.+ CD8.sup.+ TILs.

[0198] a. Expression of Nrp-1 in PD-1 CD8.sup.+ T cells subsets from NSCLC TILs. Percentages of Nrp-1 among CD8.sup.+ TILs expressing or not PD-1 (n=13).

[0199] b. Expression of Nrp-1 in PD-1 CD8.sup.+ T cells subsets from B16-F10, LLC, MC-38 and TC-1 mouse tumors.

EXAMPLES

[0200] Abbreviations: CTL: cytotoxic T lymphocyte; CTLA-4: cytotoxic T lymphocyte antigen 4; MHC-I: major histocompatibility complex class I; mAb: monoclonal antibody; NSCLC: non-small-cell lung cancer; Nrp-1: neuropilin-1, PD-1: programmed cell death-1; qRT-PCR: quantitative real-time polymerase chain reaction; r: recombinant; Sema: semaphorin; TCR: T-cell receptor; TIL: tumor-infiltrating T lymphocyte; Treg: regulatory T.

[0201] Materials & Methods

[0202] Human Lung Tumors and Freshly Isolated Lung TIL

[0203] Fresh NSCLC (non-small-cell lung cancer) tumors were obtained from the Centre chirurgical Marie Lannelongue and the Institut mutualiste Montsouris. RNA was immediately extracted with TRIzol reagent (Invitrogen), reverse-transcribed and then subjected to qRT-PCR (quantitative real-time polymerase chain reaction).

[0204] For freshly isolated TIL (tumor-infiltrating T lymphocyte), human lung tumors were dissociated mechanically and enzymatically using a tumor dissociation kit (MACS, Miltenyi Biotec). Mononuclear cells were then isolated by a Ficoll-Hypaque gradient. All human experiments were approved by the Institutional Review Board of the Gustave Roussy Institute.

[0205] Derivation and Culture of the P62 CTL Clone and IGR-Pub Autologous Tumor Cell Line

[0206] NSCLC cell line IGR-Pub was derived from the tumor specimen of patient Pub adenocarcinoma as described.sup.17. Autologous CTL (cytotoxic T lymphocyte) clone P62 was derived from TIL of the same patient.sup.49. T-cell clone P62 was stimulated every month with irradiated autologous IGR-Pub tumor cells and irradiated allogeneic Laz509 EBV-transformed B cells in RPMI-1640 medium supplemented with 10% human AB serum and rIL-2.sup.17.

[0207] The allogeneic NSCLC cell lines IGR-B2, IGR-Heu, ADC-Coco, ADC-Tor and ADC-Let were derived from tumor specimens in one of inventors' laboratories and maintained in culture as described.sup.49. A549 (ADC), SK-Mes, Ludlu (SCC), DMS53 (SCLC), H460, H1155 (LCC) and H1355 (ADC) were previously reported.sup.50.

[0208] The SV40-immortalized human bronchial epithelial cell line 16HBE14o- (16HBE), used as a control, was previously described.sup.51.

[0209] Quantitative Real-Time (RT)-PCR

[0210] Total RNA was immediately extracted from sorted cell populations using the Single Cell RNA Purification Kit (Norgen Biotek) or TRIzol reagent (Invitrogen) for human samples. cDNA was synthesised using the Maxima First Strand cDNA Synthesis Kit (ThermoFischer Scientific). qRT-PCR was performed on a Step-One Plus (Applied Biosystems) using Maxima SYBR Green Master Mix (ThermoFischer Scientific). Expression levels of transcripts were normalized to 18S expression. PCR primers and probes for human (NRP1-2, PLXNA1-4, PLXND1, SEMA3A-G, 18S) and mouse (Batf, Cd244, Tigit, Egr2, 18S) genes were designed by Sigma-Aldrich and used according to the manufacturer's recommendations.

[0211] B16F10 Melanoma Cell Line, Tumor Engraftment and Peptide Cancer Vaccine

[0212] The B16F10 melanoma cell line (H-2.sup.b) was purchased from the American Type Culture Collection (ATCC). Tumor cells were grown in DMEM/F-12 medium (ThermoFischer Scientific) supplemented with 10% FCS, 2 mM L-glutamine, 1 mM sodium pyruvate, and antibiotics (50 U/ml penicillin and 50 g/ml streptomycin).

[0213] Female C57BL/6J mice were purchased from Envigo. For each experiment, groups of 4-to-10 mice 7-9 weeks of age received 2×10.sup.5 B16F10 melanoma cells subcutaneously (s.c.) in the right flank. All animals were housed at Gustave Roussy's animal facility and treated in accordance with institutional animal guidelines.

[0214] For the cancer vaccine, C57BL/6J mice were immunized s.c. with 100 μg of gp100 (KVPRNQDWL—SEQ ID NO: 8) and/or Trp2 (SVYDFFVWL—SEQ ID NO: 9) peptides (GeneCust) plus 25 μg of poly(I:C) adjuvant (InvivoGen), at day 5 and every week at the tail base.

[0215] Murine TIL Isolation

[0216] Tumors were harvested at days 8-to-20 and digested for 40 min at 37° C. according to Tumor Dissociation Kit protocol (Miltenyi Biotec). Tumors were crushed on 100 μm cell strainers and washed twice with PBS 2% fetal calf serum (FCS). Single cell suspensions were enriched for CD45.sup.+ cells or CD8.sup.+ cells using the MultiMACS system (Miltenyi Biotec). Briefly, cells from tumor tissues were labelled with anti-CD45 or anti-CD8a microbeads (Miltenyi Biotec), and then purified using the POSSEL program on MultiMACS. The positive fraction was recovered for TIL analysis by flow cytometry or ex vivo assays.

[0217] Antibodies and Flow Cytometry

[0218] For human cell surface and intracellular staining, anti-CD3 (UCHT1), -CD4 (RPA-T4), -CD8α (RPA-T8), -CD25 (CD25-3G10), -FoxP3 (259D), -PD-1 (J105) and -Nm-1 (12C2) mAb, and rSema-3A-Fc were used. Cell surface and intracellular staining of mouse cells was performed on single-cell suspensions using antibodies specific to the following molecules: CD3 (17A2), CD4 (RM4-5), CD8α (53-6.7), CTLA-4 (UC10-4B9), PD-1 (29F.1A12), Tim-3 (RMT3-23), LAG-3 (C9B7W), Nrp-1 (3E12), T-bet (REA102), NFATc1 (7A6), Blimp-1 (5E7), Helios (22F6), IRF-4 (REA201), Ki-67 (REA183) and granzyme B (GB11). Dead cells were excluded using the Live/Dead Fixable Blue Dead Cell Stain Kit (Invitrogen).

[0219] For intracellular staining, cells were fixed/permeabilized with the Foxp3 Staining Buffer Set according to the manufacturer's instructions (eBioscience). Staining of Trp2 and gp100-specific T cells was performed using H-2Kb/SVYDFFVWL and H-2Db/KVPRNQDWL dextramers, respectively (Immudex). Flow cytometric analysis was conducted on an LSR Fortessa (BD) and analyzed using FlowJo software (Tree Star).

[0220] In Vitro Migration Assay

[0221] Freshly isolated CD8.sup.+ TIL were incubated for 30 min to 1 h either in medium or in the presence of neutralizing anti-Nrp-1 mAb (R&D system MAB59941), anti-PD-1 (Biolegend clone RMP1-14) or isotype control (Biolegend Isotype Rat IgG.sub.2a RTK2758). B16F10 tumor cells were cultured for 2 days in the lower chambers of Transwell plates (Corning) and then TIL were seeded in the upper chambers to trigger T-cell migration. After 2 h at 37° C., the number of CD8.sup.+ T cells that had migrated into the lower chambers was counted by flow cytometry and phenotyped.

[0222] For experiments with CTL clone P62, activated T cells were incubated for 30 min to 1 h either with BSA or Sema-3A-Fc (100 ng/ml, R&D system), and their ability to migrate toward the human rCXCL12 (50 nM) was evaluated. Results were expressed as chemotaxis index.

[0223] In Vitro T-Cell Stimulation, Cytokine Production and Cytotoxic Experiments

[0224] Purified CD45.sup.+ TIL were co-cultured for 4 h with B16F10 tumor cells, pulsed with Trp2 (1 μg) and gp100 (1 μg) peptides, in the presence of Brefeldin A (eBioscience), monensin (Merck) and anti-CD107a mAb (1D4B). TIL were stained with mAb specific for surface proteins prior to fixation and permeabilization. Permeabilized cells were then stained with anti-IFN-γ (XMG1.2) and anti-TNF-α (MP6-XT22) mAb.

[0225] For cytotoxicity experiments, freshly isolated CD8.sup.+ TIL were either kept in medium or pretreated with neutralizing anti-Nrp-1 (R&D system MAB59941), anti-PD-1 (clone RMP1-14) or isotype controls (Biolegend Isotype Rat IgG.sub.2a RTK2758). Cytotoxic activity toward the B16F10 cell line, pulsed with Trp2 and gp100 peptides, was evaluated using a conventional 4 h overnight chromium (.sup.51Cr) release assay.

[0226] For experiments with CTL clone P62, activated T cells were incubated for 30 min to 1 h either with BSA or Sema-3A-Fc (100 ng/ml), and their cytotoxic activity toward the autologous tumor cell was evaluated.

[0227] Western Blot Analyses

[0228] Equivalent amounts of protein extracts from tumor cell lines were separated by SDS-PAGE and transferred to a nitrocellulose membrane as described.sup.52. Blots were then probed with rat anti-Sema-3B mAb (R&D system MAB5440), mouse anti-Sema-3A (R&D system MAB1250) or anti-β-actin-peroxidase (Merck A3854), followed by secondary HRP-conjugated Ab.

[0229] In Vivo PD-1 and Nrp-1 Blockade

[0230] Mice were treated i.p. with 100 μg/mouse of anti-PD-1 (Bio-X-Cell; RMP1-14) mAb and/or i.t. with 25 μg/mouse of anti-Nrp-1 (R&D system; MAB59941) mAb. For tumor outgrowth experiments, mice were treated on days 6, 8, 10, 12 and 14 after tumor inoculation, and TIL were sorted and analysed on day 16.

[0231] Statistical Analysis

[0232] Statistical significance was determined with the one-way or two-way ANOVA test with Bonferroni correction, or with the two-tailed Student t test (GraphPad Prism, GraphPad software).

[0233] Results

[0234] Nrp-1 is Expressed on a Subset of CD8.sup.+ T Lymphocytes Infiltrating Human Lung Tumors

[0235] Nrp-1 is expressed on CD4.sup.+ Treg cells in human lymph nodes and in TIL from colorectal cancer metastases..sup.14,15 However, little is known about its expression on CD8.sup.+ TIL. Therefore, inventors first evaluated the expression of NRP1 transcripts in primary human lung tumors and autologous normal lungs. Quantitative real-time PCR (qRT-PCR) showed high expression levels of NRP1 mRNA in some lung tumor samples compared to the cognate normal lung (FIG. 7a). Moreover, freshly isolated lung TIL were found to express NRP1 mRNA (data not shown). Human NSCLC were also found to express NRP2 and certain Plexin (Plxn) A or D family member transcripts, as well as genes encoding Np ligands of the Sema 3 family (FIG. 7).

[0236] Inventors next evaluated the expression of Nrp-1 protein in freshly isolated TIL from 28 NSCLC tumors. Immunofluorescence analyses showed that Nrp-1 was expressed on a subset of human lung CD3.sup.+ TIL. In contrast, parallel analysis of peripheral blood lymphocytes (PBL) from lung cancer patients and healthy donors (HD) showed that circulating T cells did not express the receptor (FIG. 1a). Expression of Nrp-1 was observed on both CD8.sup.+ and CD4.sup.+ TIL, but with a higher frequency on CD8.sup.+ TIL (14.2±2.1% vs 8.4±0.9%). Moreover, expression of Nrp-1 correlated with the activation state of T lymphocytes, since it was more frequent on CD25.sup.+ and PD-1.sup.+ T cells from both CD8.sup.+ and CD4.sup.+ TIL subsets than on CD25.sup.− and PD-1.sup.− subsets (FIGS. 1b and c). Indeed, 70.1±8.3% of Nrp-1.sup.+ CD8.sup.+ TIL and 76.9±4.2% of Nrp-1.sup.+ CD4.sup.+ TIL also expressed high levels of PD-1 (PD-1.sup.hi). Consistently, activation of HD PBL with immobilized anti-CD3 mAb induced expression of the protein (FIGS. 8a and b). In contrast, no correlation between Nrp-1 and Foxp3 expression was observed in CD4.sup.+ TIL, since equal percentages of Nrp-1.sup.+ T cells were found in both Foxp3.sup.+ and Foxp3.sup.− subsets, which showed similar percentages of PD-1.sup.hi cells (FIG. 1d). These results show that Nrp-1 is expressed on a subset of activated human CD4.sup.+ and CD8.sup.+ TIL displaying PD-1.sup.hi status in NSCLC tumors.

[0237] Interaction of Human Nrp-1 with Sema-3A Impairs T-Cell Effector Functions In Vitro

[0238] Sema-3A, a secreted member of the Sema-3 family, is a well-known ligand of Nrp-1.sup.9,10. As human CD8.sup.+ TIL expressed substantial amounts of Nrp-1, this raises the question as to whether its interaction with Sema-3A contributes to the T-cell dysfunction often observed in the tumor microenvironment. To evaluate this hypothesis, inventors first used the IGR-Pub lung adenocarcinoma cell line and autologous CTL clone P62 established from TIL of a NSCLC patient.sup.17. Initial studies indicated that the IGR-Pub cell line, as well as several other human lung cancer cell lines, and to a lesser extent, normal bronchial epithelial cell line 16HBE, produced Sema-3A, as detected by western blot (FIG. 2a) and intracellular immunofluorescence staining (FIG. 8c). In contrast, the P62 T-cell clone did not constitutively express Nrp-1. However, stimulation with plastic-coated anti-CD3 mAb induced strong Nrp-1 expression on the surface of P62 T cells, most of which also expressed IL-2 receptor subunit CD25 and inhibitory receptor PD-1 (FIG. 2b). Therefore, inventors used the anti-CD3-stimulated P62 CTL clone to examine the consequences of Sema-3A ligation to Nrp-1 on the migratory behaviour and cytotoxic activity of these T lymphocytes in vitro.

[0239] Previous immunofluorescence experiments indicated that P62 CTL bound rSema-3A-Fc (FIG. 2c) and expressed CXCL12 receptor CXCR4 (FIG. 8d). Consequently, P62 T cells were able to migrate toward a gradient of rCXCL12 chemokine in in vitro transwell assays (FIG. 2d). Inventors found that this T-cell migratory response was inhibited in the presence of rSema-3A-Fc (FIG. 2d). This impaired T-cell migration was correlated with inhibition by rSema-3A-Fc of the cytotoxic activity of the P62 CTL clone toward autologous IGR-Pub target cells (FIG. 2e). These results indicate that Nrp-1 triggering by its soluble ligand of the Sema-3 family Sema-3A negatively regulates effector functions of Nrp-1.sup.+ cytotoxic T cells, and shows a T-cell inhibitory receptor function for Nrp-1.

[0240] Nrp-1.sup.+ CD8.sup.+ T Cells Infiltrating Murine B16F10 Melanoma Display an Exhausted PD-1.sup.hi State

[0241] To investigate the impact of Nrp-1 engagement with its ligand on CD8.sup.+ TIL functions in vivo, inventors used C57BL/6 mice engrafted with B16F10 melanoma cells, which express the Nrp-1 soluble ligand Sema-3B, as shown by RT-PCR analyses (FIG. 9a) and confirmed by western blot (FIG. 3a). At day 15, tumors were removed and TIL were isolated and analysed by flow cytometry. T cells from spleens and tumor-draining lymph nodes (TdLN) of the same mice were examined in parallel for Nrp-1 expression. Results showed that a large fraction of CD8.sup.+ TIL expressed Nrp-1 (45.5%±2.5), as opposed to spleens (1.9%±0.4) and TdLN (1.3%±0.1). The frequency of Nrp-1.sup.+ T cells was also higher in CD4.sup.+ TIL (37.7%±2.6), compared to spleens (15.7%±0.8) and TdLN (12%±0.4) (FIG. 3b). Moreover, while most Nrp-1.sup.+ CD4.sup.+ cells from spleens (80.5%±1.2) and TdLN (73%±2.7) expressed the Treg cell marker Foxp3, Nrp-1.sup.+ CD4.sup.+ TIL included both Foxp3.sup.+ (54.2%±1.7) and Foxp3.sup.− (27.2%±2.2) T cells (FIG. 3c). It should be noted that most, if not all, CD8.sup.+ TIL displayed a CD44.sup.high CD62L.sup.low phenotype, characteristic of effector/memory T (T.sub.EM) cells, especially when they co-expressed Nrp-1 (FIG. 3d). Notably, as in humans, expression of Nrp-1 was dependent on the activation state of T cells, since stimulation of CD3.sup.+ CD8.sup.+ mouse splenocytes with immobilized anti-CD3 induced Nrp-1 expression, which reached a plateau at 72 h (FIG. 9b) and was inhibited with the NFAT inhibitor 11R-VIVIT, indicating that the TCR signalling pathway was likely involved (FIG. 9c).

[0242] Inventors then focused on induction of Nrp-1 on CD8.sup.+ TIL during tumor progression. Expression of well-known inhibitory receptors PD-1, CTLA-4, Tim-3 and LAG-3 was monitored in parallel. Kinetic studies revealed that the percentage of Nrp-1.sup.+ CD8.sup.+ TIL increased during tumor growth, with similar induction of PD-1, LAG-3 and Tim-3 and, to a lesser extent, CTLA-4 (FIG. 9d). The absolute number of T cells (i.e. the number of TIL per mg of tumor) co-expressing these inhibitory receptors also increased during melanoma progression, and reached a peak at day 14 after tumor implantation (FIG. 9e). Remarkably, at day 14, all Nrp-1.sup.+ CD8.sup.+ TIL also expressed PD-1, and LAG-3, Tim-3 and CTLA-4 on most of them (FIG. 4a). This Nrp-1.sup.+ T-cell subset displayed a PD-1.sup.hi profile (FIG. 4a), with a strong correlation between expressions of the two cell surface molecules (FIG. 4b). Heat map analyses showed that this Nrp-1.sup.+ PD-1.sup.hi TIL subset included much higher percentages of T cells expressing other T-cell inhibitory receptors than the Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.− PD-1.sup.− subsets (FIG. 4c), with a much higher PD-1 geometric mean of fluorescence intensity (gMFI) than the Nrp-1.sup.− PD-1.sup.+ subset (FIG. 4d). They also indicated that the Nrp-1.sup.+ PD-1.sup.hi TIL subset included higher percentages of T cells expressing granzyme B and proliferation marker Ki67, as well as exhaustion-associated transcription factors NFATc1, IRF-4, Helios, Blimp-1 and T-bet, compared to the Nrp-1.sup.− PD-1.sup.+ subset (FIG. 4c). Nrp-1.sup.+ PD-1.sup.hi T cells also expressed high levels of Batf, Cd244 and Tigit transcripts, the products of which were associated with dysfunctional T-cell status, but not Egr2 mRNA (FIG. 9f).sup.18. It should be noted that most Nrp-1.sup.+Foxp3.sup.− CD4.sup.+ T cells were also found to express PD-1, Tim-3 and CTLA-4, as well as Ki-67 (FIG. 9g). These results indicate that Nrp-1 characterises a highly activated intratumoral CD8.sup.+ T-cell subset displaying PD-1.sup.hi status and expressing several T-cell inhibitory receptors involved in immune suppression during cancer diseases.

[0243] Nrp-1 Typifies a Highly Activated Tumor-Specific CD8.sup.+ TIL Subset with Impaired Functional Activities

[0244] Next, inventors investigated the specificity and functionality of these Nrp-1.sup.+ PD-1.sup.hi T cells. They first examined whether these lymphocytes were enriched with T cells specific to melanoma-associated antigens (MAA). To do so, C57BL/6 mice engrafted with B16F10 melanoma cells were vaccinated with Trp2 plus gp100 antigenic peptides, together with the poly(I:C) adjuvant. Antigenic specificity of intratumoral CD8.sup.+ TIL was then analysed with H-2K.sup.b-Trp2 and H-2D.sup.b-gp100 dextramers. Expression of Nrp-1 and PD-1 molecules on these TIL subpopulations was also monitored. Initial experiments showed that vaccinated mice more efficiently controlled tumor growth than unvaccinated mice (FIG. 10a). Moreover, at day 14 after melanoma cell transplantation, tumors from vaccinated mice showed increased absolute numbers of CD8.sup.+ TIL expressing PD-1, Nrp-1 and Tim-3 and, to a lesser extent, CTLA-4 and LAG-3 (FIG. 10b). More importantly, the percentage of Trp2-specific and gp100-specific CD8.sup.+ TIL in the Nrp-1.sup.+ PD-1.sup.hi T-cell subset was over two-fold higher (8.8%±1.6 and 10.4%±2.6, respectively) than in Nrp-1.sup.− PD-1.sup.+ (2.9%±0.7 and 5.6%±1.2, respectively) TIL (FIG. 5a). In contrast, the Nrp-1.sup.− PD-1.sup.− TIL subset showed very low percentages of MAA-specific T cells (0.9%±0.3 and 0.5%±0.3, respectively). Inventors' results also revealed that ex vivo stimulation of Nrp-1.sup.+ PD-1.sup.hi TIL with autologous tumor cells induced higher percentages of IFNγ-producing T cells than in Nrp-1.sup.− PD-1.sup.− and the Nrp-1.sup.− PD-1.sup.+ TIL subsets (FIG. 5b). Much stronger production of TNFα was also found with Nrp-1.sup.+ PD-1.sup.hi CD8.sup.+ TIL compared to Nrp-1.sup.− PD-1.sup.− T cells, but in this case, percentages of TNFα-producing T cells remained lower than in the Nrp-1.sup.− PD-1.sup.+ TIL population. This could be explained by a more advanced exhaustion stage of the Nrp-1.sup.+ PD-1.sup.hi CD8.sup.+ TIL subset than the Nrp-1.sup.− PD-1.sup.+ TIL subset as previously suggested during chronic viral infection.sup.19. Inventors then further assessed tumor-reactivity of Nrp-1.sup.+ PD-1.sup.hi and Nrp-1.sup.− PD-1.sup.+ TIL by measuring CD107a surface expression, a marker usually used to evaluate the degranulation capacity of CTL. Their results showed that ex vivo stimulation of TIL with the cognate B16F10 cell line induced much higher percentages of CD107a.sup.+ cells among Nrp-1.sup.+ PD-1.sup.hi and Nrp-1.sup.− PD-1.sup.+ TIL populations than did the Nrp-1.sup.− PD-1.sup.− T-cell subset (FIG. 5c).

[0245] Next, inventors analysed the cytotoxic activity of CD8.sup.+ TIL toward autologous melanoma cells in the absence or presence of neutralizing anti-Nrp-1 and/or anti-PD-1 mAb, with an isotype-matched mAb as negative control. A 4 h chromium (Cr.sup.51) release assay revealed that CD8.sup.+ TIL were poorly effective in killing B16F10 tumor cells, whether blocking mAb were present or not (FIG. 5d). In contrast, in a 12 h cytotoxicity assay, anti-Nrp-1 mAb, anti-PD-1 mAb, or a combination of both, strongly increased T-cell-mediated lysis. Notably, a combination of anti-Nrp-1 plus anti-PD-1 did not further increase target cell killing, suggesting that cytotoxicity is mainly mediated by the Nrp-1.sup.+ PD-1.sup.hi TIL subset that includes most tumor-specific effector T cells (FIG. 5d). This cytotoxicity was correlated with upregulation of MHC-I molecules on the tumor cell surface (FIG. 5e). Indeed, parallel experiments in which inventors monitored H2-K.sup.b and H2-D.sup.b expression, as well as PD-L1, on B16F10 cells co-cultured for 12 h with CD8.sup.+ TIL, showed increase of the three molecules expression, a phenomenon in which IFNγ secreted by T cells was likely involved. Consistently, inventors observed such an upregulation of H2-K.sup.b/-D.sup.b surface expression after 12 h of stimulation of tumor cells with rIFNγ and on melanoma cells collected from in vivo tumor grafts (FIG. 10c).

[0246] To further investigate the influence of Nrp-1 on CD8.sup.+ TIL functions, inventors also performed migration assays in the presence of anti-Nrp-1 and/or anti-PD-1 neutralizing mAb of CD8.sup.+ TIL seeded for 24 h in upper chambers of transwell plates, with B16F10 target cells in the lower chambers. FACS analysis of T cells having migrated to the lower chambers in control conditions showed no difference between Nrp-1.sup.− PD-1.sup.−, Nrp-1.sup.− PD-1.sup.+ and Nrp-1.sup.+ PD-1.sup.hi T-cell subsets (FIG. 5f). However, when TIL were pre-incubated with anti-Nrp-1 neutralizing mAb, the migration index of Nrp-1.sup.+ PD-1.sup.hi T cells was strongly increased. Whatever the TIL subset, no effect was observed with anti-PD-1 alone. Moreover, within the Nrp-1.sup.+ PD-1.sup.hi T-cell subset, no further increase was observed with a combination of anti-Nrp-1 and PD-1 mAb. These results support the conclusion that Nrp-1 behaves as a true CD8.sup.+ T-cell inhibitory receptor in vitro to impair the effector functions of anti-tumor CD8.sup.+ TIL.

[0247] Therapeutic Nrp-1 Blockade Potentiates Tumor Growth Control by Anti-PD-1 In Vivo

[0248] The above experiments suggested that Nrp-1 is a novel immune checkpoint expressed by tumor-reactive T cells that could impair their functional activities following interaction with its ligand Sema-3B. To directly test this hypothesis in vivo and therefore investigate the potential of neutralizing Nrp-1 via immunotherapeutic approaches, C57Bl/6 mice were engrafted with B16F10 melanoma and treated at days 6, 8, 10, 12 and 14 with anti-Nrp-1, anti-PD-1, a combination of both, or an isotype control mAb. Both tumor volume and tumor weight were followed up. Results showed that intratumoral (i.t.) administration of anti-Nrp-1 mAb, or intraperitoneal (i.p.) injection of anti-PD-1 mAb, inhibited tumor growth as compared to control treatment (FIGS. 6a and 6b). Importantly, a combination of anti-Nrp-1 plus anti-PD-1 was clearly additive, with much better control of tumor progression (FIG. 6a, FIG. 10d). A strong reduction in tumor weight measured at the experiment endpoint was also observed (FIG. 6b).

[0249] To further assess mechanisms involved in the therapeutic effect of anti-Nrp-1 plus anti-PD-1 combination, inventors analysed CD8.sup.+ TIL from tumors of the different groups of mice used above. Results showed much larger CD8.sup.+ T-cell infiltrates in tumors obtained from mice treated with the anti-Nrp-1 plus anti-PD-1 combination than from mice treated with each mAb alone or the isotype control (FIG. 6c). Increased CD8.sup.+/CD4.sup.+ Treg ratios were also observed in this group of mice compared to control mice (FIG. 6d). These CD8.sup.+ TIL expressed much higher levels of terminally differentiated effector T-cell marker KLRG1, as well as increased levels of proliferation marker Ki-67 and of serine protease granzyme B (FIG. 6e). Overall, these results demonstrate that anti-Nrp-1 plus anti-PD-1 in vivo treatment enhances tumor infiltration by Nrp-1.sup.+ PD-1.sup.hi CD8.sup.+ TIL, with increased proliferative and killing capacities, leading to strong tumor regression. They also further emphasize the therapeutic potential and benefits of the anti-Nrp-1/anti-PD-1 combination in cancer immunotherapy.

[0250] Nrp-1.sup.+ TIL for Adoptive T-Cell Transfer Immunotherapy.

[0251] Inventors' results showed that blocking Nrp-1 induces an increase in the cytotoxic potential of CD8.sup.+ TILs revealed by increased cytotoxicity toward autologous tumor cells (FIG. 11a) and perforin intracellular expression (FIG. 11b). Inventors also demonstrated that CD8.sup.+ TILs expressing Nrp-1 are enriched with tumor-antigen-specific T cells (cf. FIG. 12).

[0252] Since these Nrp-1.sup.+ CD8.sup.+ TILs are enriched with tumor-specific T cells, they can be isolated using an anti-Nrp-1 antibody, amplified ex vivo and then injected into the patient for an adoptive transfer therapy.

[0253] Combination of a Therapeutic Anti-Cancer Vaccine with Anti-Nrp-1 and Optionally with Other Immune Check-Point Inhibitor(s)

[0254] Therapeutic cancer vaccines result in increase in the number of CD8.sup.+ T cells expressing inhibitory receptors into the tumor: Nrp-1, PD-1, Tim-3, CTLA-4 and LAG-3 (cf. FIG. 12a). These TILs are also enriched with tumor antigen-specific T lymphocytes (the tumor antigen used in the vaccine). Therefore, combining the cancer vaccine with anti-Nrp-1 antibodies (in combination or not with additional immune check-point blockers (ICB)/inhibitors (ICI), such as anti-PD-1, anti-TIM-3, anti-LAG-3 and/or anti-CTLA-4) will optimize the antitumor immune response and the clinical benefit of the immunotherapy, the cancer vaccine allowing the amplification of tumor-specific T cells and the ICB reactivating the induced T cells. FIG. 12 shows that the number of Nrp-1.sup.+ CD8.sup.+ TILs is enriched among tumor-specific T cells (cf. T cells specific to the tumor antigen used in the cancer vaccine and identified with the dextramers in FIG. 12b).

[0255] Relative or Minimum Amount of CD8.sup.+Nrp1.sup.+ (PD-1.sup.+) TILs Required to Identify Relevant Populations.

[0256] Inventors' data showed that the expression of Nrp-1 on CD8.sup.+ TILs differs from one tumor to another in human subjects and according to the tumor histological type (mouse studies): [0257] in human lung cancer, the mean percentage of Nrp-1.sup.+ TILs among CD8.sup.+ TILs is around 15%±5% (FIG. 13a). Therefore, inventors believe that patients with a relative percentage of Nrp-1.sup.+ PD-1.sup.+ CD8.sup.+ TILs among TILs of the tumor sample of the subject of 15%±5%, identify the responding population. [0258] in mouse tumor models (FIG. 13b), this percentage is much higher: around 50% for melanoma (B16F10), LLC lung cancer and colon cancer (MC-38). However, in TC-1 lung cancer, it is much lower (around 10%).

DISCUSSION

[0259] Nrp-1 is essential in axonal guidance, through its capacity to bind class 3 chemorepulsive semaphorin proteins.sup.9,20, and in angiogenesis, via its interaction with VEGF and VEGF receptors (VEGFR).sup.21,22. Nrp-1 is also involved in cardiovascular and neuronal development, cell migration, immunity and cancer.sup.8,23,24. Indeed, tumor cells frequently express semaphorins and their receptors neuropilins and plexins, which can regulate malignant cell behaviour and contribute to malignant potential.sup.24. Nrp-1 is also highly expressed on tumor vasculature, functioning as a mediator of tumor initiation and progression, associated with poor clinical outcome.sup.8. For instance, high expression of Nrp-1 observed in lung cancer correlates with invasive capacity and short disease-free survival.sup.25. Moreover, cancer cells often produce secreted members of the semaphorin family, including Sema-3A and Sema-3B, also contributing to tumor escape from the immune response.sup.24,26. Inventors now report, in both human NSCLC tumors and murine B16F10 melanoma, that Nrp-1 delineates a particular subset of CD8.sup.+ TIL, enriched with tumor antigen-specific T lymphocytes, and also expressing high levels of the PD-1 inhibitory receptor. In particular, they show that Nrp-1 works as a negative regulator of antitumoral activities of this CTL subset, and that blocking of this receptor in vivo strongly improves tumor regression elicited by anti-PD-1 immunotherapy.

[0260] Multiple roles for neuropilins and semaphorins in the immune system have emerged in recent years.sup.27,28. Sema-3A has been reported to inhibit primary human T-cell proliferation and cytokine production under anti-CD3 plus anti-CD28 stimulation.sup.26. It also promotes T-cell apoptosis.sup.29 and inhibits non-specific cytotoxic activity of NK cells in mixed lymphocyte cultures.sup.26. The influence of Sema-3A on T-cell migration has also been studied. In particular, this chemorepulsive molecule has been reported to inhibit immune cell migration and response of human T cells to chemokine gradients.sup.30,31. Chemorepulsive effects of Sema-3A on human thymocytes have been reported, as well as inhibition of T-cell migratory responses triggered by chemokine CXCL1232,33. Inventors found in the context of the present invention that soluble Sema-3A binds to Nrp-1 molecules expressed on the surface of human lung tumor-specific CTL and inhibits their migratory capacity to a gradient of CXCL12 chemokine. Importantly, their results also showed that TCR-mediated killing of autologous human lung cancer cells was inhibited in the presence of Sema-3A, highlighting an inhibitory role of the Nrp-1/Sema-3A axis in CTL functions.

[0261] Nrp-1 is usually not expressed by resting T cells. However, its expression is triggered after T-cell activation, suggesting that Nrp-1 is an additional T-cell activation biomarker.sup.34. Regulation of Nrp-1 expression during adaptive immune responses is likely an essential element for understanding its physiological role. In this context, activated T cells derived from inflammatory environments were described as expressing this surface molecule.sup.16. Indeed, Nrp-1 was highly induced on CD8.sup.+ T lymphocytes engaging self-antigen, including human melanoma TIL. Inventor now show that Nrp-1 is co-expressed with CD25 and PD-1 on both CD4.sup.+ and CD8.sup.+ TIL from human NSCLC. Moreover, on CD8.sup.+ TIL, this expression is restricted to a PD-1.sup.hi T-cell subset. Various populations of CD8.sup.+ TIL have been recently described based on PD-1 expression levels: negative (PD-1.sup.N), intermediate (PD-1.sup.hi) and high (PD-1.sup.T).sup.35. Interestingly, inventors found a very good correlation between PD-1 and Nrp-1 expression (see FIG. 4b). Thus, the aforementioned PD-1.sup.T TIL subset likely corresponds to cells expressing the highest levels of PD-1 and Nrp-1 identified in the context of the present invention. This is a crucial and surprising discovery, as this subset has been recently reported to correspond to an exhausted TIL subset.sup.35. Accordingly, in the present experiments, inventors found that Nrp-1.sup.+ PD-1.sup.hi TIL express transcription factors Helios, Blimp-1, IRF-4 and NFATc1, associated with activation/exhaustion.sup.36-39. Co-expression of Nrp-1 and Helios on CD4.sup.+ T cells was also observed on TIL from human liver metastases of colorectal cancer.sup.15. Another characteristic of PD-1.sup.T TIL is that they display high tumor recognition capacity and are predictive of responses of human NSCLC patients to PD-1 blockade.sup.35. Likewise, inventors found, in B16F10 tumors, that a significant percentage of Nrp-1.sup.+ PD-1.sup.hi CD8.sup.+ TIL strongly expressed granzyme B and cell proliferation marker Ki-67, and secreted high levels of IFNγ. Moreover, this Nrp-1.sup.+ PD-1.sup.hi T-cell subpopulation was the most strongly enriched with MAA-specific CD8.sup.+ T lymphocytes. Inventors therefore assume that Nrp-1 can be used as a novel marker for identifying this important CD8.sup.+ tumor-reactive TIL population. Overall, these findings suggest that, in an anti-tumor immune response context, Nrp-1 is a very late T-cell activation marker and that its co-expression with high levels of PD-1 might be another characteristic of exhausted antigen-experienced CD8.sup.+ TIL.

[0262] Expression of CTLA-4, PD-1 and Tim-3 molecules was associated with T-cell exhaustion in chronic viral infections and tumor progression.sup.40-44. These T-cell inhibitory receptors are important for regulating immune responses in peripheral tissues and maintaining self-tolerance. Along the same lines, Nrp-1 is induced on tolerant self-reactive CD8.sup.+ T cells expressing known regulatory receptors.sup.16. In the present report, inventors show that Nrp-1 is often associated with a broad panel of inhibitory receptors, such as CTLA-4, Tim-3 and LAG-3, and characterizes a dysfunctional PD-1.sup.hi CD8.sup.+ TIL subpopulation. This particular phenotype, characteristic of T cells infiltrating an inflamed tumor microenvironment, may further explain the paradox of tumor progression, despite the presence of an ongoing T-cell response toward malignant cells. Inventors also show here that anti-Nrp-1 neutralising mAb, in particular when associated with anti-PD-1, re-established ex vivo the functionality of these T cells, such as externalization of CD107a and TCR-mediated cytotoxicity toward the cognate tumor. Remarkably, anti-Nrp-1 also restore the migratory capacity of Nrp-1.sup.+ PD-1.sup.hi T cells, likely by blocking the interaction of Nrp-1 with its ligands Sema-3A or Sema-3B. Anti-PD-1 blocking mAb have no such effect. Semaphorins have been reported to govern cell migration by regulating integrin-mediated adhesion and actin cytoskeleton.sup.45. A similar process may occur in tumors to explain how Nrp-1 signalling specifically affects T-cell migratory potential, possibly downstream plexin-A, which forms stable heterodimers with Nrp-1 and acts as a signal-transducing module for the Sema-3/Nrp-1/Plxn-A complex at the plasma membrane.sup.46.

[0263] It is becoming increasingly clear that Nrp-1 and its ligands Sema-3A/-3B play important roles during the effector phase of various immune processes, including anti-tumor T-cell responses. Inventors' work now show that Nrp-1 negatively influences CD8 T-cell immunity and responses to anti-PD-1 cancer immunotherapy. Indeed, their in vivo experiments have revealed that, like the anti-PD-1 blockade, anti-Nrp-1 immunotherapy is able to inhibit melanoma progression in C57BL/6 mice. Moreover, a combination of the two antibodies is more efficient at reducing tumor growth associated with enhanced tumor infiltration by CD8.sup.+ effector T cells, with an increase in the CD8.sup.+/CD4.sup.+ T-cell ratio. This is an important finding, since the limited success of anti-PD-1 cancer immunotherapy is often associated with weak tumor infiltration by specific CD8.sup.+ T cells.sup.47′48. Concomitant blockade of Nrp-1 and PD-1 could remedy this “cold-tumor” situation, thus providing a new immunotherapeutic strategy for further improving specific immune responses during cancer disease.

REFERENCES

[0264] 1. Mrass, P. et al. Random migration precedes stable target cell interactions of tumor-infiltrating T cells. J Exp Med 203, 2749-2761 (2006). [0265] 2. Topalian, S. L., Drake, C. G. & Pardoll, D. M. Immune checkpoint blockade: a common denominator approach to cancer therapy. Cancer Cell 27, 450-461 (2015). [0266] 3. McGranahan, N. et al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science 351, 1463-1469 (2016). [0267] 4. Phan, G. Q. et al. Cancer regression and autoimmunity induced by cytotoxic T lymphocyte-associated antigen 4 blockade in patients with metastatic melanoma. Proc Natl Acad Sci USA 100, 8372-8377 (2003). [0268] 5. Topalian, S. L. et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med 366, 2443-2454 (2012). [0269] 6. Tumeh, P. C. et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature 515, 568-571 (2014). [0270] 7. Rizvi, N. A. et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science 348, 124-128 (2015). [0271] 8. Prud'homme, G. J. & Glinka, Y. Neuropilins are multifunctional coreceptors involved in tumor initiation, growth, metastasis and immunity. Oncotarget 3, 921-939 (2012). [0272] 9. Kolodkin, A. L. et al. Neuropilin is a semaphorin III receptor. Cell 90, 753-762 (1997). [0273] 10. He, Z. & Tessier-Lavigne, M. Neuropilin is a receptor for the axonal chemorepellent Semaphorin III. Cell 90, 739-751 (1997). [0274] 11. Yadav, M. et al. Neuropilin-1 distinguishes natural and inducible regulatory T cells among regulatory T cell subsets in vivo. J Exp Med 209, 1713-1722, S1711-1719 (2012). [0275] 12. Delgoffe, G. M. et al. Stability and function of regulatory T cells is maintained by a neuropilin-1-semaphorin-4a axis. Nature 501, 252-256 (2013). [0276] 13. Hansen, W. et al. Neuropilin 1 deficiency on CD4+Foxp3+ regulatory T cells impairs mouse melanoma growth. J Exp Med 209, 2001-2016 (2012). [0277] 14. Battaglia, A. et al. Neuropilin-1 expression identifies a subset of regulatory T cells in human lymph nodes that is modulated by preoperative chemoradiation therapy in cervical cancer. Immunology 123, 129-138 (2008). [0278] 15. Chaudhary, B. & Elkord, E. Novel expression of Neuropilin 1 on human tumor-infiltrating lymphocytes in colorectal cancer liver metastases. Expert Opin Ther Targets 19, 147-161 (2015). [0279] 16. Jackson, S. R., Berrien-Elliott, M., Yuan, J., Hsueh, E. C. & Teague, R. M. Neuropilin-1 expression is induced on tolerant self-reactive CD8.sup.+ T cells but is dispensable for the tolerant phenotype. PLoS One 9, e110707 (2014). [0280] 17. Dorothee, G. et al. Tumor-infiltrating CD4.sup.+ T lymphocytes express APO2 ligand (APO2L)/TRAIL upon specific stimulation with autologous lung carcinoma cells: role of IFN-alpha on APO2L/TRAIL expression and -mediated cytotoxicity. J Immunol 169, 809-817 (2002). [0281] 18. Williams, J. B. et al. The EGR2 targets LAG-3 and 4-1BB describe and regulate dysfunctional antigen-specific CD8.sup.+ T cells in the tumor microenvironment. J Exp Med 214, 381-400 (2017). [0282] 19. Wherry, E. J., Blattman, J. N., Murali-Krishna, K., van der Most, R. & Ahmed, R. Viral persistence alters CD8 T-cell immunodominance and tissue distribution and results in distinct stages of functional impairment. J virol 77, 4911-4927 (2003). [0283] 20. Chen, H., He, Z., Bagri, A. & Tessier-Lavigne, M. Semaphorin-neuropilin interactions underlying sympathetic axon responses to class III semaphorins. Neuron 21, 1283-1290 (1998). [0284] 21. Staton, C. A., Kumar, I., Reed, M. W. & Brown, N. J. Neuropilins in physiological and pathological angiogenesis. J Pathol 212, 237-248 (2007). [0285] 22. Roskoski, R., Jr. Vascular endothelial growth factor (VEGF) signaling in tumor progression. Crit Rev Oncol Hematol 62, 179-213 (2007). [0286] 23. Fujisawa, H. Discovery of semaphorin receptors, neuropilin and plexin, and their functions in neural development. J Neurobiol 59, 24-33 (2004). [0287] 24. Muratori, C. & Tamagnone, L. Semaphorin signals tweaking the tumor microenvironment. Adv Cancer Res 114, 59-85 (2012). [0288] 25. Hong, T. M. et al. Targeting neuropilin 1 as an antitumor strategy in lung cancer. Clin Cancer Res 13, 4759-4768 (2007). [0289] 26. Catalano, A. et al. Semaphorin-3A is expressed by tumor cells and alters T-cell signal transduction and function. Blood 107, 3321-3329 (2006). [0290] 27. Suzuki, K., Kumanogoh, A. & Kikutani, H. Semaphorins and their receptors in immune cell interactions. Nat immunol 9, 17-23 (2008). [0291] 28. Kumanogoh, A. & Kikutani, H. Immunological functions of the neuropilins and plexins as receptors for semaphorins. Nat Rev Immunol 13, 802-814 (2013). [0292] 29. Moretti, S. et al. Semaphorin3A signaling controls Fas (CD95)-mediated apoptosis by promoting Fas translocation into lipid rafts. Blood 111, 2290-2299 (2008). [0293] 30. Mendes-da-Cruz, D. A., Linhares-Lacerda, L., Smaniotto, S., Dardenne, M. & Savino, W. Semaphorins and neuropilins: new players in the neuroendocrine control of the intrathymic T-cell migration in humans. Exp physiol 97, 1146-1150 (2012). [0294] 31. Mendes-da-Cruz, D. A. et al. Developing T-cell migration: role of semaphorins and ephrins. FASEB journal 26, 4390-4399 (2012). [0295] 32. Lepelletier, Y. et al. Control of human thymocyte migration by Neuropilin-1/Semaphorin-3A-mediated interactions. Proc Natl Acad Sci USA 104, 5545-5550 (2007). [0296] 33. Garcia, F. et al. Inhibitory effect of semaphorin-3A, a known axon guidance molecule, in the human thymocyte migration induced by CXCL12. J Leukoc Biol 91, 7-13 (2012). [0297] 34. Milpied, P. et al. Neuropilin-1 is not a marker of human Foxp3.sup.+ Treg. Eur J Immunol 39, 1466-1471 (2009). [0298] 35. Thommen, D. S. et al. A transcriptionally and functionally distinct PD-1(+) CD8(+) T cell pool with predictive potential in non-small-cell lung cancer treated with PD-1 blockade. Nat Med 24, 994-1004 (2018). [0299] 36. Bengsch, B. et al. Epigenomic-Guided Mass Cytometry Profiling Reveals Disease-Specific Features of Exhausted CD8 T Cells. Immunity 48, 1029-1045 e1025 (2018). [0300] 37. Shin, H. et al. A role for the transcriptional repressor Blimp-1 in CD8(+) T cell exhaustion during chronic viral infection. Immunity 31, 309-320 (2009). [0301] 38. Man, K. et al. Transcription Factor IRF4 Promotes CD8(+) T Cell Exhaustion and Limits the Development of Memory-like T Cells during Chronic Infection. Immunity 47, 1129-1141 e1125 (2017). [0302] 39. Martinez, G. J. et al. The transcription factor NFAT promotes exhaustion of activated CD8(+) T cells. Immunity 42, 265-278 (2015). [0303] 40. Trautmann, L. et al. Upregulation of PD-1 expression on HIV-specific CD8.sup.+ T cells leads to reversible immune dysfunction. Nat Med 12, 1198-1202 (2006). [0304] 41. Freeman, G. J., Wherry, E. J., Ahmed, R. & Sharpe, A. H. Reinvigorating exhausted HIV-specific T cells via PD-1.sup.− PD-1 ligand blockade. J Exp Med 203, 2223-2227 (2006). [0305] 42. Kaufmann, D. E. et al. Upregulation of CTLA-4 by HIV-specific CD4.sup.+ T cells correlates with disease progression and defines a reversible immune dysfunction. Nature immunology 8, 1246-1254 (2007). [0306] 43. Blackburn, S. D., Shin, H., Freeman, G. J. & Wherry, E. J. Selective expansion of a subset of exhausted CD8 T cells by alphaPD-L1 blockade. Proc Natl Acad Sci USA 105, 15016-15021 (2008). [0307] 44. Fourcade, J. et al. Upregulation of Tim-3 and PD-1 expression is associated with tumor antigen-specific CD8+ T cell dysfunction in melanoma patients. J Exp Med 207, 2175-2186 (2010). [0308] 45. Casazza, A., Fazzari, P. & Tamagnone, L. Semaphorin signals in cell adhesion and cell migration: functional role and molecular mechanisms. Adv Exp Med Biol 600, 90-108 (2007). [0309] 46. Takahashi, T. et al. Plexin-neuropilin-1 complexes form functional semaphorin-3A receptors. Cell 99, 59-69 (1999). [0310] 47. Herbst, R. S. et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature 515, 563-567 (2014). [0311] 48. Melero, I., Rouzaut, A., Motz, G. T. & Coukos, G. T-cell and NK-cell infiltration into solid tumors: a key limiting factor for efficacious cancer immunotherapy. Cancer Discov 4, 522-526 (2014). [0312] 49. Echchakir, H. et al. Evidence for in situ expansion of diverse antitumor-specific cytotoxic T lymphocyte clones in a human large cell carcinoma of the lung. Inter Immunol 12, 537-546 (2000). [0313] 50. Le Floc'h, A. et al. Alpha E beta 7 integrin interaction with E-cadherin promotes antitumor CTL activity by triggering lytic granule polarization and exocytosis. J Exp Med 204, 559-570 (2007). [0314] 51. Le Maux Chansac, B. et al. Potentiation of NK cell-mediated cytotoxicity in human lung adenocarcinoma: role of NKG2D-dependent pathway. Inter Immunol 20, 801-810 (2008). [0315] 52. Le Floc'h, A. et al. Minimal engagement of CD103 on cytotoxic T lymphocytes with an E-cadherin-Fc molecule triggers lytic granule polarization via a phospholipase Cgamma-dependent pathway. Cancer Res 71, 328-338 (2011).