Search

This table contains the sequences of the used neoantigen peptides, their affinity according to MHC-NET and the magnitude of the induced CD8 T cell immune responses

This figure demonstrates the spectrum of secreted cytokines in CD8 T cells induced by DC/CoAT vaccinations

Antagonistic antibodies targeting coinhibitory receptors have revolutionized the treatment of cancer by inducing durable immune responses and clinical remissions in patients. In contrast, success of agonistic costimulatory antibodies has thus far been limited because of the insufficient induction of adaptive immune responses. Here, we describe a novel vaccination method consisting of a primary dendritic cell (DC) immunization followed by a composite vaccination, including an agonistic CD40 antibody, soluble antigen, and a TLR3 agonist, referred to as CoAT. In mice, DC/CoAT prime-boost vaccinations targeting either MHC class I or II neoantigens or tumor-associated antigens rendered up to 60% of the total T-cell population specific for a single tumor epitope. DC/CoAT induced durable and complete remissions of large subcutaneous tumors without detectable side effects. Thus, booster vaccinations with agonistic costimulatory antibodies represent an ideal means to amplify DC vaccinations and induce robust T-cell immune responses while providing maximum flexibility regarding the choice of antigen. Cancer Res; 77(8); 1918–26. ©2017 AACR.

Analysis of of activation markers on CD8 T cells after neoadjuvant treatment.

Apoptosis induction, cell proliferation and viability after MAPK inhibitor treatment. (A) Apoptosis was determined in HepG2, Hep3B and Huh-7 cells after treatment with 0.2 (light gray bar), 1 (dark gray bar) or 5µM (black bar) inhibitor or DMSO (white bar) at 48h by measuring the subG1 peak using propidium iodide. (B) Cell division was determined by eFluor670 labeling of HepG2, Hep3B and Huh-7 cells after 96h treatment with the different inhibitor concentrations and detected by flow cytometry. (C) The WST-1 metabolic assay was used to detect the viability/proliferation of the three cell lines after treatment with the respective inhibitors for 72h. Each bar represents the mean {plus minus} SDs from three to six replicates.

Figure Legends for Figure S1-S5

NK cell cytotoxicity assay

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal and disseminating cancer resistant to therapy, including checkpoint immunotherapies, and early tumor resection and (neo)adjuvant chemotherapy fails to improve a poor prognosis. In a transgenic mouse model of resectable PDAC, we investigated the coordinated activation of T and natural killier (NK) cells in addition to gemcitabine chemotherapy to prevent tumor recurrence. Only neoadjuvant, but not adjuvant treatment with a PD-1 antagonist effectively supported chemotherapy and suppressed local tumor recurrence and improved survival involving both NK and T cells. Local T-cell activation was confirmed by increased tumor infiltration with CD103+CD8+ T cells and neoantigen-specific CD8 T lymphocytes against the marker neoepitope LAMA4-G1254V. To achieve effective prevention of distant metastases in a complementary approach, we blocked the NK-cell checkpoint CD96, an inhibitory NK-cell receptor that binds CD155, which was abundantly expressed in primary PDAC and metastases of human patients. In gemcitabine-treated mice, neoadjuvant PD-1 blockade followed by adjuvant inhibition of CD96 significantly prevented relapse of PDAC, allowing for long-term survival. In summary, our results show in an aggressively growing transgenic mouse model of PDAC that the coordinated activation of both innate and adaptive immunity can effectively reduce the risk of tumor recurrence after surgery, facilitating long-term remission of this lethal disease.Significance: Coordinated neoadjuvant and adjuvant immunotherapies reduce the risk of disease relapse after resection of murine PDAC, suggesting this concept for future clinical trials. Cancer Res; 78(2); 475–88. ©2017 AACR.

(C) To detect the proliferation/adhesion profile of HepG2, Hep3B and Huh-7 cells were grown 24h prior treatment (-24h) with 5 (dark gray line) or 10µM (black line) of PLX4720, U0126, AZD6244, PD0325901or 6.5 and 13µM sorafenib and DMSO as control (light gray line) for up to 96h while the cell index (CI, relative change in electrical impedance of the cells) was measured in real time every 4h and normalized to the start-point at treatment (0h, CI=1). The Asterisks indicate no measurement of 10µM PD0325901 in all three cell lines and 10µM PLX4720 in Huh-7.

Two way ANOVA statistics of phosphorylated MEK1 and ERK1/2 after MAPKi treatment. Values of phosphoplex data of mean fluorescence intensities (MFI) from 30-60 beads for each kinase at each time point were calculated against the respective DMSO value using the 2 way ANOVA statistic. Indicated p-values are defined as *p{less than or equal to}0.05, **p{less than or equal to}0.01, ***p{less than or equal to}0.001, ns= not significant.

This figure demonstrates the magnitude of DC/CoAT vaccinations induced with different agonistic antibodies (anti-CD40, anti-CD137 and anti-OX40)

This figure demonstrates the ability of CD8 T cells induced by DC/CoAT immunizations to migrate to the bronchoalveolar space

Supplementary Methods and Legend

Immunohistochemical analysis of E-Cadherin expression

Analysis of CD4+CD25+ cells in immune cell populations isolated from primary tumors after neoadjuvant treatment.

Effects of MAPK inhibitors on JNK, c-Jun, Akt and ATF2 phosphorylation. (A) Kinetics of p-JNK, p-c-Jun, p-Akt and p-ATF2 phosphorylation in HepG2 (left) and Hep3B (right) cell lines after treatment with 6.5µM sorafenib, 5µM AZD6244, U0126 (left) or 5µM PLX4720, PD0325901 (right) or DMSO at the given time points are displayed as MFI.

In this figure, the liver damage induced by heterologous, DC-based immune responses is illustrated. Mice were primed with dendritic cells and subjected to different booster immunizations consisting of variable combinations of anti-CD40, Poly I:C and peptide antigen. Liver damage was assessed via ALT measurements 24 and 72h after booste immunizations

Analysis of CD8+ CD90.2+ cells in immune cell populations isolated from primary tumors after neoadjuvant treatment.

Modulation of MEK1 and ERK1/2 phosphorylation by MAPK inhibitors. Phospho-MEK1 and p-ERK1/2 at 0.5, 6, 12, 24, 48, 72 and 96h were detected in lysates of HCC cell lines HepG2 (left), Hep3B (middle) and Huh-7 (right) cells following treatment with DMSO as control solvent, 6.5µM AZD6244, U0126, sorafenib (A), PLX4720, PD0325901(B) and with 0.2, 1 or 5µM GW5074 and sunitinib (C). Values represent MFI of 30-60 beads for each kinase and time point. One of two or three representative experiments is shown.

Apoptosis induction and Proliferation/adhesion profiles of HCC cell lines treated with MAPK inhibitors. Detection of apoptosis by measuring the sub-G1 peak using propidium iodide in HepG2 (left), Hep3B (middle) and Huh-7 (right) cells treated with DMSO, 5 (A) or 10µM (B) MAPK inhibitor or 6.5 (A) and 13µM (B) sorafenib for 48 and 96h respectively (n{greater than or equal to}3 {plus minus} SD). Mean and {plus minus}SD from n{greater than or equal to}2 experiments.

Alteration of total protein amount of MEK1, ERK1/2, c-Jun, Akt and ATF2 by MAPK inhibition. Total amount of MEK1, ERK1/2, c-Jun, Akt and ATF2 of HepG2 (A), Hep3B (B) and Huh-7 (C) cells after treatment with 0.2, 1 or 5µM sorafenib, PLX4720, U0126, AZD6244, PD0325901, GW5074 or sunitinib at different time points.

Purpose: Small molecule inhibitors of the mitogen-activated protein kinase (MAPK) pathway, such as sorafenib, represent novel treatment options for advanced hepatocellular carcinoma. The aim of our study was to identify downstream targets as biomarker candidates that are directly linked to the oncogenic MAPK pathway in hepatocellular carcinoma and correlate with inhibition of this pathway by multikinase inhibitors.Experimental Design: Hepatocellular carcinoma cell lines and fresh tumor and tumor-free liver tissues from patients with hepatocellular carcinoma were incubated with different BRaf or MEK inhibitors and analyzed for kinase phosphorylation, proliferation, induction of apoptosis, and chemokine secretion.Results: Hepatocellular carcinoma cell lines responded differentially to these inhibitors in a dose-dependent manner, even those targeting the same kinase. Sorafenib inhibited both MEK1 and ERK1/2 phosphorylation at high but increased signaling at low concentrations. Similarly, PLX4720 increased MEK/ERK signaling independently from mutations in BRaf or NRas. MEK inhibitors decreased ERK1/2 phosphorylation in a dose-dependent manner. These signaling characteristics correlated with inhibition of proliferation, induction of apoptosis, and chemokine secretion. Fresh tissues derived from patients diagnosed with primary hepatocellular carcinoma responded to these inhibitors with changes in their microenvironment following the patterns observed in hepatocellular carcinoma cells.Conclusions: Oncogenic signaling of the MAPK pathway influences hepatocellular carcinoma sensitivity to treatment with BRaf and MEK inhibitors about cell fate independently from mutations in BRaf and NRas. MAPK inhibitors have a strong impact on chemokine secretion as a consequence of interference with oncogenic signaling. Therefore, novel biomarker candidates associated with the hepatocellular carcinoma microenvironment may be developed for prediction and monitoring of treatment response to small molecule inhibitors. Clin Cancer Res; 20(9); 2410–23. ©2014 AACR.

Treatment options for pancreatic cancer are limited and often ineffective. Immunotherapeutic approaches are one possible option that needs to be evaluated in appropriate animal models. The aim of the present study was to analyze tumor-specific immune responses in a mouse model of pancreatic cancer, which mimics the human disease closely. C57BL/6 EL-TGF-α × Trp53−/− mice, which develop spontaneous ductal pancreatic carcinoma, were generated. EL-TGF-α × Trp53−/− mice developed spontaneous pancreatic tumors with pathomorphologic features close to the human disease. Tumor-specific CD8+ T-cell responses and IgG responses were analyzed in EL-TGF-α × Trp53−/− mice during tumor development and compared with mice with s.c. growing pancreatic tumors. In contrast to spontaneous pancreatic tumors, cell lines generated from these tumors were rejected after s.c. injection into wild-type mice but not in nude or RAG knockout mice. Direct comparison of spontaneous and s.c. injected tumors revealed an impaired infiltration of CD8+ T cells in spontaneous pancreatic tumors, which was also evident after adoptive transfer of tumor-specific T cells. Intratumoral cytokine secretion of tumor necrosis factor-α, IFN-γ, IL-6, and MCP-1 was lower in spontaneous tumors as well as the number of adoptively transferred tumor-specific T cells. Our data provide clear evidence for tumor-specific immune responses in a genetic mouse model for pancreatic carcinoma. Comparative analysis of s.c. injected tumors and spontaneous tumors showed significant differences in tumor-specific immune responses, which will help in improving current immune-based cancer therapies against adenocarcinoma of the pancreas. (Cancer Res 2006; 66(1): 508-16)

Supplementary Figure 1 and Table S2 from Genetically Induced Pancreatic Adenocarcinoma Is Highly Immunogenic and Causes Spontaneous Tumor-Specific Immune Responses

Treatment options for pancreatic cancer are limited and often ineffective. Immunotherapeutic approaches are one possible option that needs to be evaluated in appropriate animal models. The aim of the present study was to analyze tumor-specific immune responses in a mouse model of pancreatic cancer, which mimics the human disease closely. C57BL/6 EL-TGF-α × Trp53−/− mice, which develop spontaneous ductal pancreatic carcinoma, were generated. EL-TGF-α × Trp53−/− mice developed spontaneous pancreatic tumors with pathomorphologic features close to the human disease. Tumor-specific CD8+ T-cell responses and IgG responses were analyzed in EL-TGF-α × Trp53−/− mice during tumor development and compared with mice with s.c. growing pancreatic tumors. In contrast to spontaneous pancreatic tumors, cell lines generated from these tumors were rejected after s.c. injection into wild-type mice but not in nude or RAG knockout mice. Direct comparison of spontaneous and s.c. injected tumors revealed an impaired infiltration of CD8+ T cells in spontaneous pancreatic tumors, which was also evident after adoptive transfer of tumor-specific T cells. Intratumoral cytokine secretion of tumor necrosis factor-α, IFN-γ, IL-6, and MCP-1 was lower in spontaneous tumors as well as the number of adoptively transferred tumor-specific T cells. Our data provide clear evidence for tumor-specific immune responses in a genetic mouse model for pancreatic carcinoma. Comparative analysis of s.c. injected tumors and spontaneous tumors showed significant differences in tumor-specific immune responses, which will help in improving current immune-based cancer therapies against adenocarcinoma of the pancreas. (Cancer Res 2006; 66(1): 508-16)

Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide with a poor prognosis and one for which immunotherapy remains a viable option. Experimental tumor models have shown that regulatory T cells, a functionally unique subset of T cells, can suppress effective antitumor immune responses. This suppression might explain the poor outcome of some of the immunotherapy protocols currently being used. A better understanding of the role of regulatory T cells in HCC is important for design of future immunotherapy-based clinical protocols. We have studied regulatory T cells from 84 patients with HCC and 74 controls, including healthy donors, patients with chronic hepatitis B virus and hepatitis C virus infection and nonviral liver cirrhosis. Regulatory T cells were identified by fluorescence-activated cell sorting using a panel of antibodies and by real-time PCR analysis for Foxp3 expression. Functional studies were done to analyze their inhibitory role. Finally, regulatory T cells were analyzed in tumors and ascites from patients with HCC. Patients with HCC have increased numbers of CD4+CD25+ regulatory T cells in their peripheral blood, which express high levels of HLA-DR, GITR, and low or no CD45RA. These cells were anergic toward T-cell receptor stimulation and, when cocultured with activated CD4+CD25− cells, potently suppressed their proliferation and cytokine secretion. There were also high numbers of regulatory T cells in tumor-infiltrating lymphocytes of HCC patients comparable with the increase in their peripheral blood. Our data suggest that the increase in frequency of regulatory T cells might play a role in modulation of the immune response against HCC and could be important in design of immunotherapeutic approaches.