From basic to tumour immunology for oncologists

Transcription

1 From basic to tumour immunology for oncologists Andrea Anichini Human Tumors Immunobiology Unit, Dept. of Experimental Oncology and Molecular Medicine Fondazione IRCCS Istituto Nazionale dei Tumori, Milan

2 1. Development and regulation of anti-tumor immunity: some of the main cellular subsets CD137 NK cell SLAMF7 CD16 Tumor cell KIR Antigen Gene PD-L1 PD-1 mature CD8 + CTL Plasmacell Antigen release B cell IL-12 TH2 (IL-4) CAFs:cancer-associated fibroblasts MHC Class II TCR Dendritic cell CTLA-4 PD-1 CD4 + T cell CD80 PD-L1 CD86 PD-L2 CD80 CD86 PD-L1 PD-L2 cytokines MHC Class I Co-stimulatory molecules and inhibitory receptors CTLA-4 PD-1 TH1 (IFN-γ) (-) TCR CD8 + naive T cell M2 macrophages CD163/CD204 (-) Immunosuppressive and pro-tumoral subsets Tregs (CD4 + CD25+ FOXP3+ CD39+) CTLA-4 MDSCs (HLA-DR - CD14 + CD33 + ) (HLA-DR - CD15 + CD33 + )

3 Plasmacell B cell Antigen release 2. How tumors evade the immune system. CD137 SLAMF7 PD-L2 Upregulation of Tumor cell CEACAM-1 NK cell constitutive expression ligands of CD16 TIM-3 of immunoregulatory PD-1inhibitory genes receptors KIR Antigen Gene (COX-2, β-catenin..) production of immunosuppressive factors IL-12 (IL-10, TGF-β1, VEGF-A...) MHC Class II Skewing Reduced of expression T cell TH2 polarization (IL-4) of towards ligands TH2 for profile activatory receptors of NK cells active recruitment, activation/differentiation of immunosuppressive and pro-tumoral subsets TCR Dendritic cell CTLA-4 PD-1 CD4 + T cell CD80 PD-L1 CD86 PD-L2 PD-L1 CD80 CD86 PD-L1 PD-L2 PD-1 cytokines MHC Class I Co-stimulatory molecules and inhibitory receptors TH1 (IFN-γ) mature CD8 + CTL CTLA-4 PD-1 (-) TCR CD8 + naive T cell M2 macrophages CD163/CD204 (-) PD-L1 Treg (CD4 + CD25+ FOXP3+ CD39+) Loss of TAA by dedifferentiation T cell Loss of expression BTLA of HLA molecules intra-tumor TAA heterogeneity (branched evolution) Immunosuppressive and pro-tumoral subsets CTLA-4 HVEM LAG-3 TIGIT MDSCs HLA-II CD155 CD112 (HLA-DR - CD14 + CD33 + ) (HLA-DR - CD15 + CD33 + )

4 NK activationtumor targeting Plasmacell 3. Main immunotherapy strategies CD137 SLAMF7 Vaccines Immune checkpoint -Adoptive T cell transfer, mabs Tumor cell Blockade -CAR-T cells KIR NK cell CD16 Antigen release Antigen Gene PD-L1 PD-1 mature CD8 + CTL -Promoting immunogenic cell death. -Inducing tumor cell death by oncolytic viruses. -Promoting Type I IFN response by DNMT inhibitors. -Targeting immunoregulatory genes. B cell TH2 (IL-4) IL-12 MHC Class II TCR Dendritic cell CTLA-4 PD-1 CD4 + T cell CD80 PD-L1 CD86 PD-L2 CD80 CD86 PD-L1 PD-L2 cytokines MHC Class I Co-stimulatory molecules and inhibitory receptors CTLA-4 PD-1 TH1 (IFN-γ) (-) TCR CD8 + naive T cell M2 macrophages CD163/CD204 (-) Tregs (CD4 + CD25+ FOXP3+ CD39+) Targeting suppressive cells Immunosuppressive and pro-tumoral subsets CTLA-4 MDSCs (HLA-DR - CD14 + CD33 + ) (HLA-DR - CD15 + CD33 + )

5 The relevance of the spontaneous anti-tumor response and why it is generated only in a subset of patients

6 Immune reaction to melanocytic lesions The Temptation of Saint Anthony ( ) Matthias Grünewald Issenheim altarpiece Clark WH et al. Model predicting survival in stage I melanoma based on tumor progression. J Natl Cancer Inst. 81:1893, Borroni G, Vignati G. Should Sutton nevus really be called Grünewald-Sutton nevus? Am J Dermatopathol ct;15(5): Del Vecchio M, Mortarini R, Tragni G, Di Guardo L, Bersani I, Di Tolla G, Agustoni F, Colonna V, Weber JS, Anichini A. T-cell activation and maturation at tumor site associated with objective response to Ipilimumab in metastatic melanoma. J Clin Oncol. 29:e783-8,2011. J. AM. ACAD. DERMATOL. VOLUME 67:583, 2012

7 Genomic classification of cutaneous melanoma: four genomic subtypes and three gene expression subsets. Proliferation Cell cycle progression Apoptosis Genetic stability Transcriptional control Survival Epigenetic regulation Proliferation

8

9 Four molecular subtypes of pancreatic cancer

10 "T -cell-inflamed tumors" vs "non-t-cell-inflamed tumors" in bladder cancer Sweis Cancer Immunol Res 2016 CD8A non-t-cell-inflamed tumors T -cell-inflamed tumors

11 MHC-II + melanomas express immune-related signatures ( PD-1 signalling, allograft rejection and T-cell receptor signalling ) Nature Commun.

12 MITF/AXL alternative gene programs: role in shaping the tumor microenvironment Tirosh I et al. Science , 2016 AXL program MITF program

13 Some molecular subsets of NSCLC are associated with inflammatory reaction -Simultaneous PD-L1 and JAK2 amplification in 5/94 (5.3%) NSCLC surgical specimens. 2 Cancer Res. -ADC or SCC cell lines with distinct mutations (NRAS, KRAS, BRAF, PIK3CA, EML4-ALK, RET) or FGFR1 amplification have active AKT/mTOR pathway driving constitutive PD-L1 expression -EGFR activation by EGF stimulation, exon-19 deletions, and L858R mutation induces PD-L1 in cell lines. -PD-L1 expression in ADC surgical specimens with EGFR Del-19 and L858R mutations. PD-L1 expression in cell lines and surgical specimens positive for ALK-EML4 fusion gene. 1 Clin. Cancer Res. CDKN2A/B LKB1 KC KL p53 KP - KP subset (KRAS -mut p53 -mut ) ADC show higher levels of somatic mutations, inflammatory markers, immune checkpoint effector molecules, and improved relapse-free survival. PD-1 and PD-L1, TIM-3, BTLA, CTLA-4, LAG-3, CD3+ TIL, CD80, CD86, OX40L, 4-1BB, CD127, IFN-γ-induced genes, CD4+Foxp3+ Tregs in ADC with a mesenchymal profile, not associated with mutational lad.

14 Constitutive PD-L1 expression? H. Borghaei, NEJM 2015 Figure Higher mutational load, more immunogenic tumors 2. Frequent PD-L1 expression PD-L1 expression more likely to be immuno-mediated? High frequency of patients in the KP subset in this study?

15 The arrow of tumor immunity and response to immunotherapy Neoantigens +++ T cell response +++ immune-related upregulation of PD-L1response to ICB 5 1 * * * * * *

16 Tumor Evidence Reference Melanoma Melanoma NSCLC The arrow of tumor immunity Neoantigens T cell infiltrate PD-L1 response to ICB A neoantigen landscape present in tumors with a strong response to CTLA-4 blockade. Overall mutational load, neoantigen load, and expression of cytolytic markers in the immune microenvironment were significantly associated with clinical benefit. higher nonsynonymous mutation burden in tumors associated with improved objective response. Snyder, NEJM 2014 Van Allen, Science 2015 Rizvi, Science 2015 MSI-CRC Higher response rate to anti-pd-1 in mismatch-repair deficient tumors Le, NEJM, 2015 Colorectal cancer Melanoma, Stage III Giannakis, Cell Rep Madore, Clin Cancer Res 2016

17 Hypermutant tumors : good candidates for spontaneous development of tumor immunity? Tumor Evidence Implication Reference Glioblastoma resulting from biallelic mismatch repair deficiency (bmmrd) syndrome Neoantigen loads 7-16 times higher than in immunoresponsive melanomas or lung cancers Highly responsive to anti-pd-1 Bouffet, JCO 2016 Colorectal cancer with mismatch-repair deficiency Mean of 1782 somatic mutations /tumor in mismatch repair deficient tumors, vs 73/tumor in mismatch repair proficient tumors Response to anti-pd-1: immunerelated ORR 40% in mismatch repair deficient colorectal cancers vs 0% for mismatch repair proficient tumors Le, NEJM 2015 Polymerase e Mutated and Microsatellite- Instable Endometrial Cancers Median predicted neoantigen load: POLE-tumors: 8,342 Higher MSI tumors: 541 PD-1+ T infiltration MSS tumors: 70,5. Polymerase e mutated and MSI EC tumors may be excellent candidates for PD-1 targeted immunotherapies. Howitt, JAMA Oncology 2015 BRCA1/2- mutated high grade serousovarian cancer -Higher predicted neoantigen load in BRCA1/2- mutated tumors compared to tumors without alterations in homologous recombination (HR) genes. -Better prognosis in BRCA1/2-mutated tumors with high number of TILs BRCA1/2-mutated HGSOCs may be more sensitive to PD-1/PD-L1 inhibitors compared to HR-proficient HGSOCs. Strickland, Oncotarget 2016 APOBEC gene signature in Urothelial cancer Some APOBEC family genes associated with improved OS and tumor expression of PD-L1 High mutational load in urothelial cancer explains responsiveness to ICB Mullane Sci. Rep. 2016

18 Which antigens are relevant? Antigens encoded by non mutated genes (tissue-specific TAA or cancer-testis antigens) Neoantigens resulting from nonsynonimous somatic mutations

19 Targeting tissue-specific melanocyte-lineage antigens (in the era of neoantigens) Adoptive transfer of MART1-specific CTLs generated by priming with peptide-pulsed dendritic cells in the presence of interleukin-21, followed by anti-ctla-4 therapy Chapuis et al JCO 2016

20 In addition... we should not forget the relevance of response to antigens encoded by viral genomes HPV+ HNSCC HPV-positive HNSCC Lancet Oncol. 2016

21 A high neoantigen load is not enough: relevance of the clonal architecture analysis

22 Genes and genetic alterations associated with suppression of tumor immunity Peng W. et al. Cancer Discovery 6:202 BRFA V600E PTEN -null inducible melanoma model

23 The emerging role of master immunoregulatory genes

24 COX-2 suppresses development of tumor immunity

25 Targeting melanoma dedifferentiation C-Myc NFATc2 Brn2 TNF-α MITF GPNMB RAB27a PGC-1α Gp100 Tyrosinase MART-1 Melanin production Pigmentation T cell recognition CD271 TNF-RI/II TNF-α Perotti, Oncogene, 2016

26 Pre-existing immune response associated with response to immunotherapy Responding patients Tumor Invasive margin Non responding patients Proximity of PD-1/PD-L1 expression Tumeh PC et al.

27 Relevance of immune-related signatures in pre-therapy lesions NSCLC, anti-pd-l1 Urothelial ca., anti-pd-l1 PD-L1 on immune cells

28 Multispectral immunohistochemical analsysis of neoplastic sections. A multiparametric approach to the tumor immune contexture Orange: Merkel carcinoma cells Yellow: CD8+ T cells Red: CD68+ macrophages White: PD-1 Green: PD-L1 Blue: nuclear DNA Nghiem et al NEJM 2016

29 Translational approaches for improving immunotherapy: targeting the PD-1/PD-L1 axis as an example Success vs failure in targeting the PD-1/PD-L1 axis: a working model Translational approaches 1. If: High nonsynonymous mutational load high, but clonal neoantigen load A 1. Identify hypermutant tumors. 2.Whole exome sequencing of matched tumor/normal tissue. 3.Identification and validation of candidate neoantigens. 4. Clonal architecture analysis. Increased likelihood of spontaneous activation of anti-tumor response (=T cell infiltration in pre-therapy lesions) B 1. (Multispectral) immunohistochemical analysis. 2. TH1-, Teff- and IFN-related gene signatures. 2. Then: The microenvironment will likely express PD-1 ligands on tumor and/or on immune cells C 1.Immunohistochemistry for PD-1 ligands. 2. Oncogene-dependent PD-L1 expression 3. PD-L1 genetic alterations. 3. However: Increased likelihood of response to immunotherapy targeting the PD-1/PD-L1 axis by re-activation of functionally impaired tumor-specific T cells Immunosuppressive mechanisms, fostered by gene programs in neoplastic cells, and immune escape processes, limit efficacy of spontaneous and therapy-rescued immunity D Predict response by A + B+ C On pre/post-therapy lesions 1. Transcriptional signatures of resistance/response. 2. Expression of master immunoregulatory genes. 3. Genetic changes with immunoregulatory effects. 4. Immune contexture of the lesions for Tregs, MDSCs, TAM. 5. Analysis for main immune escape mechanisms.

30

31

32 PD-L1/PD-L2: Genomic alterations Tumor Evidence Reference 22 major cancer types Copy number alterations. tumors with PD-L1 gains harbored significantly higher mutational load compared to non-amplified cases. Several tumors, including, adult T- cell leukaemia/lymphoma, diffuse large B-cell lymphoma, and stomach adenocarcinoma. Squamous Cell Carcinomas of the Cervix and Vulva. Structural variations (SVs) commonly disrupting the 3 region of the PD-L1 gene leading to a marked elevation of aberrant PD-L1 transcripts. Co-gain or co-amplification of CD274 and PDCD1LG2 in 32/48 cervical SCCs and 10/23 vulvar SCCs. Budczies, Genes, Chrom. Cancer, 2016 Kataoka, Nature 2016 Howitt, JAMA Oncol 2016