4T1 cells were transfected with SLR20 or OH-SLR20. At 4 hours after transfection, 4T1 cells were washed 5 times, then cultured in serum-free media for 16 hours. 4T1-cultured supernatant was collected, passed through a 0.2ï­m filter, then added neat to cultures of Raw264.7 cells for 30 minutes. Western analysis of RAW264.7 whole cell lysates was used to measure P-STAT1. Cells were transfected, and after 16 h, total RNA was assessed by RT-qPCR to measure expression of the indicated gene.. Each point represents the average of three experimental replicates, N = 3. Midlines are average {plus minus} S.D. Student's T-test.

4T1 mouse mammary tumors were grown orthotopically in athymic Balb/C (nu/nu) female mice. Tumor-bearing mice were randomized into treatment groups for intratumoral delivery of SLR20 (or OH-SLR20, or saline) on treatment days 1, 5, and 9. Tumors were measured throughout treatment (days 1-10) and for 5 days after treatment ceased (days 10-15). N = 7 per group.

Supplemental Figure S3. A. Tumors harvested on treatment d 7 were assessed by IHC for Arg-1 and FoxP3. Representative images are shown. Original magnification, 400X. N = 5. B-C. RT-qPCR gene expression analysis of Ripk1 and Ripk3 (B) and Gas6 (C) from tumors treated with vehicle or lapatinib on treatment day 1, followed by 6 daily treatments with vehicle or BMS-777607. Tumors were harvested 24 h after final treatment. Individual data points represent the average value of 5 experimental replicates from RNA harvested from a single tumor. Midlines are the average ({plus minus} S.D.) of the biological replicates (N = 3-4 per group). P values calculated using Student's T-test.

Immunohistochemistry was used to measure CD45, F4,80, CD4, and CD8 in tumors harvested at day 5. Representative Images are shown. N = 5

Supplemental Figure S6. Tumors harvested on treatment d 7 were assessed by IHC for FoxP3 and Arg-1. Representative images are shown. Original magnification = 400X. Asterisks represent areas of acellular debris. Black arrows indicate tumor infiltrating lymphocytes (TILs). Yellow arrows indicate hyper-condensed nuclei characteristic of apoptotic bodies / apoptotic debris.

A. Schematic of treatment strategy for intra-tumoral nanoparticle delivery of SLR20 (or OH-SLR20) to WT Balb/C mice harboring 4T1 mammary tumors. Saline was delivered intratumorally as a control. Tumors were measured throughout treatment . B. Tumor volume was measured beginning at treatment day 0. N = 7-8 per group.

Supplemental Figure S1. Tumors harvested on treatment d 1 and d 7 were assessed by IHC for P-Neu Y1248. Representative images are shown. Original magnification, 400X. N = 5. Histological analysis of H&E-stained tumor sections. Tumors were collected on treatment day 2 (i.e., 1 day after treatment {plus minus} lapatinib). Representative images are shown. N = 5. Original magnification was 400X. Tumors harvested on treatment d 2 were assessed by IHC for CD3 and FoxP3. Representative images are shown. Original magnification, 400X. N = 5. Tumors harvested on treatment d 2 were assessed by IHC for CD3 and FoxP3. Representative images are shown. Original magnification, 400X. N = 5. RT-qPCR analysis of tumor RNA harvested at treatment d 7 measuring relative levels of indicated gene transcripts. Values were calculated using the ddCT method. Each data point represents the average value of 5 technical replicates, N = 5 tumors. For each transcript, values were corrected for the average value measured in vehicle-treated samples. Student's T-test.

Supplemental Figure S5. Decreased tumor cellularity of tumors treated with the combination of lapatinib + BMS-777607 + epacadostat. Low power images of H&E-stained sections of tumors collected at treatment day 7. Arrows point to tumor cells, which stain darker purple, and which are evident as solid sheets in vehicle-treated samples, but only as pockets in the samples treated with lapatinib + BMS-777607 + Epacadostat. Asterisks represent areas of acellular debris.

Supplemental Figure S7. MMTV-Neu tumors were treated lapatinib on days 1 and 7, and with BMS-777607 on days 14. Average tumor volume is shown.

Supplemental Figure S2. Schematic representation of treatment groups and experimental timeline to measure the impact of PtdSer liposomes on the TME of 4T1 mouse mammary tumors IHC to detect CD3, and FoxP3 in tumors collected on treatment day 7. Quantitation of the number of positive cells per 200X field is shown. Each data point is the average of 5 random fields per tumor, midlines are the average of N = 3 samples, {plus minus}S.D. P values, Student's unpaired 2-tailed T-test.

Western analysis of whole cel lysates harvested from BT474, MCF7, and MDA-MB-361 cells using antibodies indicated at the left of each panel. Western analysis of whole cell lysates harvested from MDA-MB-361 cells tranfected with SLR20 or OH-SLR20 at 12 hours after transfection. Immunohistochemistry was used to measure RIG-I and P-STAT1 in tumors harvested at day 5. Representative Images are shown. N = 5.

Supplemental Figure S4. THP-1 cells differentiated with M-CSF were assessed for their ability to polarize to an M1 phenotype in response to LPS treatment. Cells were treated for 24 hrs. with LPS (700ng/mL). Activation was analyzed via flow cytometry comparing CD86 expression on LPS treated cells to untreated cells and quantified. Differentiated THP-1 cells were treated with either BMS777607 (final concentration of 1uM) or DMSO (1:1000) for 2hrs. then co-cultured with BT474 cells [treated with either DMSO (1:1000) or Lapatinib (1uM) for 2hrs.] for a total of 4 hours, before being collected for RNA isolation and qPCR. TUNEL analysis of tumors harvested from MMTV-Neu tumor-bearing mice at treatment day 2 and treatment day 7. Representative images are shown.

Supplemental Figure S8. MMTV-Neu tumors were measured on treatment day 1 and again on treatment day 29. measurements from each tumor is represented by a datapoint. The line connects measurements from a single tumor on day 1 to the measurement of the same tumor on day 29. Vehicle-treated tumors are shown in gray. A. Vehicle-treated tumors are compared to tumors treated with lapatinib (shown in black). B. Vehicle-treated tumors are compared to tumors treated with BMS-777607 + Epacadostat (B/E, shown in red). C. Vehicle-treated tumors are compared to tumors treated with Lapatinib + BMS-777607 + Epacadostat (L/B/E, shown in blue). D. Vehicle-treated tumors are compared to all remaining treatment groups: lapatinib-treated tumors in black, B/E-treated tumors in red, and L/B/E-treated tumors in blue.

A. METABRIC-curated clinical dataset of invasive breast cancers (N = 817, Ref. here) was assessed for samples harboring genomic loss (solid blue boxes) or mRNA down-regulation (defined as < -2 S.D. from the mean DDX58 expression among the entire dataset, and shown in blue outline). B. A lollipop graph was used to show the relative positioning of DDX58 missense mutations found in a TCGA-curated clinical dataset of invasive breast tumors (N = 817). No recurrent mutations were identified.

Cells were transfected, and after 16 h, total RNA was assessed by RT-qPCR to measure expression of the indicated gene.. Each point represents the average of three experimental replicates, N = 3. Midlines are average {plus minus} S.D. Student's T-test.

Immunohistochemistry was used to measure Ki67 and TUNEL+ cells in tumors harvested at day 5. Representative Images are shown. N = 5. Immunohistochemistry was used to measure Ki67 and TUNEL+ cells in tumors harvested at day 5. Representative Images are shown. N = 5. Western analysis of whole cell lysates collected 16 hours after transfection using antibodies indicated at the left of each panel. MCF7 Cells were transfected, and after 16 h, total RNA was assessed by RT-qPCR to measure expression of the indicated genes involved in pyroptosis. Each point represents the average of three experimental replicates, N = 3. Midlines are average {plus minus} S.D. Student's T-test.

Abstract RIG-I like receptors, RNA helicases that sense viral oligonucleotide motifs and activate innate immunity, are gaining interest in cancer therapy, given their ability to redirect immune responses within the tumor microenvironment (TME), and increase efficacy of experimental cancer vaccines. RIG-I agonists are not well studied in breast cancers, a type of cancer that is often considered immunologically “silent.” We recently reported that therapeutic delivery of RIG-I agonists increase tumor-infiltrating leukocytes (TILs) and expression of proinflammatory Th1 cytokines in the 4T1 mouse model of aggressive, metastatic breast cancer through tumor cell-intrinsic mechanisms. However, these studies do not rule out the importance of myeloid immune responders (e.g., macrophages and dendritic cells) in propagating the effects of RIG-I agonists against tumor cells in vivo, nor do they rule out the impact of RIG-I agonists on adaptive antitumor immunity, a subject that is relatively understudied. We assessed the effects of the RIG-I agonist SLR20 on the the activity of effector T-lymphocytes (TEff) and regulatory T-lymphocytes (TReg) in the TME. Interestingly, SLR20 treatment of mouse and human breast tumor cells increased expression of FAS and MHC-I on tumor cells, and caused tumor cells to express T-cell chemoattractants (e.g., CXCL10, RANTES), potentially increasing T-cells recruitment to tumors, and increasing tumor cell susceptibility to TEff recognition and killing. Using an ex vivo co-culture assay in which 4T1 mouse mammary tumor cells were co-cultured with CD8+ T-cells harvested from mice pre-inoculated with SLR20-treated 4T1 tumor cells, we measured the rate of CD8+-mediated tumor cell killing. This approach revealed that T-cells harvested from mice inoculated with SLR20-treated cells caused greater tumor cell killing than what was seen by CD8+ T-cells harvested from untreated mice. We also found that conditioned media harvested from 4T1 cells treated with SLR20 increased clonal expansion of CD3/CD28-activated T-cells above what was seen with conditioned media harvested from 4T1 cells treated with a control oligonucleotide or from untreated 4T1 cells. TGFβ-mediated differentiation of CD4+ T-cells into tolerogenic and immunosuppressive TRegs was measured in cultures of CD4+ T-cells treated with cultured media derived from SLR20-treated 4T1 cells. These studies showed that cultured media harvested from 4T1 cells treated with SLR20, but not from untreated 4T1 cells or 4T1 cells treated with a control ligand, diminished TReg differentiation, and decreased CD4+ T-cells surface expression of PD-1, CTLA4, and CCR8. Importantly, in vivo experiments assessing therapeutic treatment of 4T1 tumors with SLR20 revealed greater tumor growth inhibition when SLR20 was combined with PD-L1 targeting antibodies. Taken together, these findings indicate that therapeutic activation of RIG-I signaling operates at the interface of innate and adaptive immunity within breast tumors to redirect the TME from an immunosuppressed state to one that is immunogenic and receptive to clinically relevant checkpoint inhibitors. Citation Format: David L. Elion, Max E. Jacobson, Donna J. Hicks, Bushra Rahman, Violeta Sanchez, Paula I Gonzales-Ericsson, Olga Fedorova, Anna M. Pyle, John T. Wilson, Rebecca S. Cook. RIG-I agonists reinforce antitumor adaptive immunity and decrease Treg activity in breast cancer [abstract]. In: Proceedings of the Fourth CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; Sept 30-Oct 3, 2018; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2019;7(2 Suppl):Abstract nr A187.

Abstract Cancer immunotherapies that remove checkpoint restraints on adaptive immunity are gaining clinical momentum but have not achieved widespread success in breast cancers, a tumor type considered poorly immunogenic and which harbors a decreased presence of tumor-infiltrating lymphocytes. Approaches that activate innate immunity in breast cancer cells and the tumor microenvironment are of increasing interest, based on their ability to induce immunogenic tumor cell death, type I IFNs, and lymphocyte-recruiting chemokines. In agreement with reports in other cancers, we observe loss, downregulation, or mutation of the innate viral nucleotide sensor retinoic acid-inducible gene I (RIG-I/DDX58) in only 1% of clinical breast cancers, suggesting potentially widespread applicability for therapeutic RIG-I agonists that activate innate immunity. This was tested using an engineered RIG-I agonist in a breast cancer cell panel representing each of three major clinical breast cancer subtypes. Treatment with RIG-I agonist resulted in upregulation and mitochondrial localization of RIG-I and activation of proinflammatory transcription factors STAT1 and NF-κB. RIG-I agonist triggered the extrinsic apoptosis pathway and pyroptosis, a highly immunogenic form of cell death in breast cancer cells. RIG-I agonist also induced expression of lymphocyte-recruiting chemokines and type I IFN, confirming that cell death and cytokine modulation occur in a tumor cell–intrinsic manner. Importantly, RIG-I activation in breast tumors increased tumor lymphocytes and decreased tumor growth and metastasis. Overall, these findings demonstrate successful therapeutic delivery of a synthetic RIG-I agonist to induce tumor cell killing and to modulate the tumor microenvironment in vivo. Significance: These findings describe the first in vivo delivery of RIG-I mimetics to tumors, demonstrating a potent immunogenic and therapeutic effect in the context of otherwise poorly immunogenic breast cancers. Cancer Res; 78(21); 6183–95. ©2018 AACR.