1
ST imulator of IN terferon G enes (STING) is a critical component of an innate immune pathway that activates robust anti-viral and anti-tumor responses in mouse models. Activation of the STING pathway by intratumoral (IT) injection of synthetic cyclic dinucleotides (CDNs) is being explored as a cancer therapy and has shown potent anti-tumor activity in preclinical models. Here we assessed the benefit of combining immune checkpoint blockade with ADU-S100 (MIW815), a CDN under clinical evaluation, in different syngeneic mouse tumor models. In mice bearing dual flank 4T1 mammary carcinoma tumors, adding a single dose of ADU-S100 with αPD-1 induced eradication of both injected and non-injected tumors, leading to near complete responses, demonstrating that ADU-S100 potentiates the activity of checkpoint blockade. Tumor control was CD8 + T cell-dependent and correlated with an enhanced CD8 + T cell effector profile in both the periphery and non-injected tumors. Remarkably, the resistance of CT26 Pten -/- tumor to αPD-1 was overcome by combining with ADU-S100. In addition, combining a single injection of ADU- S100 with αPD-1 elicited enhanced tumor control in the dual flank MC-38 colon carcinoma model compared to ADU-S100 or αPD-1 treatment alone. Those mice which cleared tumor by combination treatment were also protected from tumor re- challenge. Moreover, in the poorly immunogenic B16.F10 model, adding ADU- S100 to the combination therapy of αPD-1 and αCTLA4 induced tumor-specific CD8 + T cell responses and tumor control, leading to multiple complete responses and durable immunity in surviving animals. Together, these results highlight the immune correlates of STING-mediated anti-tumor efficacy and illustrate the potential of combining ADU-S100 with checkpoint inhibitors for the treatment of human cancer. Clinical trials of ADU-S100 in combination with αPD-1 or with αCTLA4 are ongoing and could further elucidate the immunological mechanism of action and therapeutic effect in humans. MC38 DF tumor bearing mice S100 1x ± αPD-1 Tumor outgrowth Vehicle + isotype (0/8) Vehicle + αPD-1 (0/8) S100 + isotype (0/8) S100 + αPD-1 (2/8) Presented at the 2018 Tumor Immunology and Immunotherapy, November 27 - 30, 2018, Miami, FL Statistical Significance: *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 by ANOVA (exceptions: 8B, log-rank (Mantel-Cox) test; 8C, Fisher’s exact test); Statistics are versus Naïve or Vehicle (V) controls unless indicated otherwise. Lines (except 8C) or bars and error represent the mean ± SEM and each symbol in bar graphs represents an individual animal. In legends, (x/x) notation indicates (number of cures/total mice in experimental group). Arrow on tumor growth curves indicates day of IT injection. Abbreviations: IT, intratumoral; CDN, cyclic dinucleotide; IFN, type I interferon; TDLN, tumor-draining lymph node; CPI, checkpoint inhibitor; V, vehicle; ns, non-significant; SF, single flank; DF, dual flank; S100, ADU-S100/MIW815; 1x, single IT injection; 2x, two IT injections; 3x three IT injections; GrzB, Granzyme B; SFC, spot-forming cells; (x/x), (cures/total n of mice) References: 1) Diamond MS et al., The Journal of Experimental Medicine. 2011. 2) Fuertes MB et al., The Journal of Experimental Medicine. 2011. 3) Ishikawa H et al., Nature. 2008. 4) Woo S-R et al., Immunity. 2014. 5) Corrales L et al., Cell Reports. 2015. 6) Francica BJ et al., Cancer Immunology Research. 2018. 7) Corrales L et al., Journal of Clinical Investigation. 2016. Ethics Approval: All animals were used according to protocols approved by Institutional Animal Use Committee of Aduro Biotech, Inc. and maintained in specific pathogen-free conditions in a barrier facility. ADU-S100 Combines with Checkpoint Blockade to Elicit an Anti-Tumor CD8 + T Cell Response to Control Non-Injected Tumors in Preclinical Models Weiwen Deng 1 , Anthony L. Desbien 1 , Kelsey Sivick Gauthier 1 , Gabrielle Reiner 1 , Leticia Corrales 1 , Tamara Schroeder 1 , Natalie H. Surh 1 , Brian Francica 1 , Justin J. Leong 1 , Ken Metchette 1 , Lianxing Zheng 2 , Charles Cho 3 , Yan Feng 2 , Jeffrey M. McKenna 2 , Steven L. Bender 3 , Chudi Ndubaku 1 , Meredith L. Leong 1 , Andrea van Elsas 1 , and Sarah M. McWhirter 1 1 Aduro Biotech, Inc., Berkeley, CA, 2 Novartis Institutes for BioMedical Research, Cambridge MA; 3 Genomics Institute of the Novartis Research Foundation, San Diego, CA A,B) Mice bearing a single flank 4T1 tumor received one IT injection of vehicle alone or 1, 10, 100, or 500 μg ADU-S100, A) ADU-S100 cleared injected tumor in a dose dependent manner. B) The frequency of replicating tumor- specific T cells, as measured by H-2L d -AH1 (AH1) tetramer staining, exhibited a bell-shaped curve in a single flank setting. C,D) Mice bearing dual flanks 4T1 tumor received one IT injection of vehicle alone or 1, 10, 100, or 500 μg ADU-S100, C) Both immunogenic and ablative doses of ADU-S100 elicited a robust tumor-specific CD8 + T cell response in the dual flank setting. D) Significant amounts of S100 were found in non-injected tumors at the 500 μg dose (134.2 ng/ml ± 29.62), reaching a concentration on the same order of magnitude as tumors injected with 10 μg (923.8 ng/ml ± 371.4). • The magnitude of tumor-specific CD8 + T cell responses is dependent on the dose of ADU-S100. • Lower immunogenic dosing regimens result in local STING activation and durable adaptive immune responses. • More aggressive ablative dosing regimens, while effective in clearing injected tumors, result in systemic drug distribution and compromised T cell immunity. • Immunogenic doses of ADU-S100 combine effectively with checkpoint inhibitors αPD-1 and/or αCTLA4 in multiple tumor models including CPIs resistant models. • Together, these results identify immune correlates of STING-mediated anti- tumor efficacy in mice and illustrate the potential of combining ADU-S100 with checkpoint inhibitors for the treatment of human cancer. CONCLUSIONS A) Treating MC38 tumor-bearing mice with an immunogenic dose of S100 (10 μg) combined with αPD-1 induced eradication of the non- injected tumors, while either agent alone was insufficient for tumor control (right: green squares vs. blue squares or red circles). B) Surviving mice demonstrated durable immunity. Figure 5. Combination of ADU-S100 and αPD-1 Enhances Non- Injected Tumor Control in a CD8 + T Cell-Dependent Manner in 4T1 Mammary Carcinoma Tumor-Bearing Mice A) Treating 4T1 tumor bearing mice with an immunogenic dose of S100 (10 μg) combined with αPD-1 induced eradication of both injected and non-injected tumors in a CD8 + T cell-dependent manner (green inverted triangles vs. yellow diamonds). B) An immunogenic dose of S100 (10 μg) activated and expanded tumor-specific effector T cells, while αPD-1 potentiated the functionality of these cells in the non-injected tumor microenvironment. BACKGROUND • T cell inflamed tumors in humans are correlated with an interferon-β (IFNβ) transcriptional signature in the tumor microenvironment. • Approaches to stimulate priming of tumor- specific CD8 + T cells for any individual and/or initiate productive immune responses in “cold” tumors have potential as cancer immunotherapies. • Different mechanisms, including the loss of function mutations, were involved in relapse to checkpoint inhibitors (CPIs). • Intratumoral (IT) injection of cyclic dinucleotide (CDN) STING agonists induces IFNβ, and activates tumor- resident dendritic cells capable of priming tumor-specific CD8 + T cells in mice. • Here, the benefit of combining IT CDN therapy optimized for CD8 + T cell induction with immune checkpoint blockade was explored using ADU-S100, a CDN under clinical evaluation, in different syngeneic mouse flank tumor models. Figure 1. STING-Activating CDNs Drive T Cell Priming o 4T1 DF tumor bearing mice S100 1x ± αPD-1, ± αCD8 Tumor outgrowth • Phosphorothioate substitution increases resistance to phosphodiesterase cleavage. • Mixed-linkage configuration facilitates broad activation of human STING alleles. • ADU-S100 has enhanced potency over natural CDN ligands in humans cells and mouse models. • ADU-S100 was selected based on balance of efficacy and tolerability in pre- clinical studies. • X-ray crystal structure (stick model) of ADU-S100 • Structure contains two adenine bases covalently bonded by a 2’-5’ and a 3’-5’ linkage (aka “mixed” linkage) and phosphorothioate substitutions in the R,R configuration • X-ray crystal structure of ADU-S100 bound to the C-terminal domain of human STING Figure 2. Development of Clinical Compound ADU-S100 (S100) Carbon (white) Phosphorous Nitrogen Oxygen Sulfur ABSTRACT B 7 days S100 1x 4T1 SF tumor bearing mice PBMC Flow cytometry 0 5 10 15 20 25 0 500 1000 1500 2000 Naive (0/4) S100 + aPD-1 (4/4) 0 200 400 600 800 1000 * ** 0 5 10 15 20 25 0 500 1000 1500 **** 30 0 5 10 15 20 25 0 500 1000 1500 **** 30 * *** 0 5 10 15 20 25 0 5 10 15 20 25 0 200 400 600 800 1000 Vehicle + isotype Vehicle + aPD-1 S100 + isotype S100 + aPD-1 Vehicle + isotype S100 + aCD8 S100 + isotype S100 + aPD-1 S100 + aPD-1 + aCD8 4T1 DF tumor bearing mice S100 1x ± αPD-1 12 days Non-injected tumor Flow cytometry 4 6 2 0 AH1 + (PD-1 MFlx10 3 ) **** 10 15 5 0 IFNγ + TNFα + (% of CD8 + ) *** 40 60 20 0 CD11c + (% of AH1 + ) **** * 10 20 5 0 AH1 + (GrzB MFlx10 3 ) 0 **** 15 V V S100 + isotype S100 + αPD-1 Vehicle + isotype Vehicle + αPD-1 Figure 7. Combination of ADU-S100 and αPD-1 Enhances Non- Injected Tumor Control and Induced Durable Immunity Figure 3. IT STING Activation Can Be Modulated to Induce Local Versus Systemic Immune Activation Day Post Implantation Vehicle S100 10 µg S100 500 µg 0 0 Injected Tumor Volume (mm 3 ) 100 200 300 400 500 5 20 25 10 15 0 5 10 15 20 25 0 *** **** Isotype Non-Injected Tumor Volume (mm 3 ) 0 0 100 200 300 400 500 5 10 15 20 25 ** Isotype 0 0 500 1000 1500 5 10 15 20 25 0 5 10 15 20 * αCD8 0 0 500 1000 1500 5 10 15 20 25 *** ns αCD8 CD8 + T cells were necessary for control of injected tumors afforded by an immunogenic dose of S100 (1x 10 μg, top: gray triangles) and for control of non- injected tumors afforded by an ablative dose of S100 (1x 500 μg, BOTTOM: black squares). In contrast, CD8 + T cells were dispensable for control of injected tumors afforded by an ablative dose of S100 (1x 500 μg, TOP: black squares). 4T1 DF tumor bearing mice S100 1x ± αCD8 Tumor outgrowth Figure 4. CD8 + T Cells Are Necessary for Anti-Tumor Immunity Elicited by Immunogenic, but Not Ablative, Doses of ADU-S100 Tumor Volume (mm 3 ) RESULTS A) CT26 tumor partially responded to αPD-1 while CT26 Pten -/- tumor was resistant to the treatment. B) Treating CT26 Pten -/- tumor bearing mice with an immunogenic dose of S100 (10 μg) combined with αPD-1 induced eradication of non-injected tumors, while S100 was insufficient for tumor control (right: green squares vs red circles). CT26 Pten -/- DF tumor bearing mice S100 1x ± αPD-1 Tumor outgrowth ± αPD-1 D7 Figure 6. The Resistance of CT26 Pten -/- Tumors to αPD-1 Was Overcome by Combining with ADU-S100 0 10 20 30 0 500 1000 1500 2000 -/- 3 * ** Vehicle, isotype (0/8) S100+isotype (0/8) S100+a PD-1 (1/8) 0 10 20 30 0 500 1000 1500 2000 -/- * * Tumor Volume (mm 3 ) Day Post Implantation 0 10 20 30 0 500 1000 1500 **** **** 0 10 20 30 0 500 1000 1500 * * Days after implantation 0 5 10 15 20 25 30 0 50 100 150 200 250 300 350 500 1000 1500 2000 Days Post Tumor Challenge Vehicle (0/8) 1 µg S100 (1/8) 10 µg S100 (5/8) 100 µg S100 (8/8) 500 µgS100 (8/8) *** **** ** **** 7 days S100 1x 4T1 SF tumor bearing mice Tumor outgrowth Injected Non-Injected B A A) In the poorly immunogenic B16.F10 model, combining an immunogenic dose of S100 (10 μg) to the αPD-1 + αCTLA4 CPIs combination therapy induced tumor- specific CD8 + T cell responses B) and tumor clearance, resulting in a significant survival benefit. C) Surviving mice demonstrated durable immunity as a significant number of animals were resistant to autologous rechallenge. B Percent of Survival 0 100 0 10 20 30 25 50 75 40 50 Day Post Implantation **** **** Vehicle + isotype (n=21) Vehicle + αPD-1 + αCTLA4*** (n=22) S100 10 µg + isotype*** (n=20) S100 10 µg + αPD-1 + αCTLA4**** (n=42) 0 10 20 30 40 50 6 0 2000 Tumor Volume (mm 3 ) 0 5 10 15 500 1000 1500 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50 Naïve (0/21) S100 10 µg + αPD-1 + αCTLA4 (9/12) *** IFNγ SFC/5x10 PBMCs 40 80 120 100 60 20 Vehicle + isotype Vehicle + αPD-1 + αCTLA4 S100 10 µg + isotype S100 10 µg + αPD-1 + αCTLA4 0 0 *** *** **** p15E: 0 - + - + - + - + Figure 8. Combination of ADU-S100, αPD-1, and αCTLA4 Enhance Anti-Tumor Immunity in B16.F10 Melanoma Tumor- Bearing Mice B16 SF tumor bearing mice S100 1x ± CPI 7 days PBMC ELISpot Tumor rechallenge Cure mice C A B B 0 10 20 30 0 1000 2000 3000 CT26 isotype CT26 aPD1 CT26 Pten -/- isotype CT26 Pten -/- aPD1 * Tumor Volume (mm 3 ) Day Post Implantation Day Post Implantation Day Post Tumor Rechallenge Day Post Tumor Rechallenge 2 AH1 + (% of CD8 + ) 4 3 1 0 µg S100: V 1 10 100500 V 1 10 100 500 0: ns ** *** * A A Tumor Volume (mm 3 ) Injected Tumor Volume (mm 3 ) Non-Injected Tumor Volume (mm 3 ) Tumor Volume (mm 3 ) Injected Non-Injected A CT26 vs. CT26 Pten -/- CT26 Pten -/- Injected CT26 Pten -/- Non-Injected C 7 days S100 1x 4T1 DF tumor bearing mice PBMC Flow cytometry days µg S100: V 1 10100 500 AH1 + (% of CD8 + ) 30 20 10 0 *** ** **** ns D days 4T1 DF tumor bearing mice S100 1x 30 min Tumor LC/MS/MS 10 10 0 ng/mL ADU-S100 10 1 10 2 10 3 10 4 10 5 µg S100: 500 10 500 0: 0 1 2 3 4 5 10 500 10 500 ** Injected Non-Injected Tumor Volume (mm 3 ) Day Post Implantation
