Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.1304
Emmanuel Prestat, Elsa Schalck, Antoine Bonnefoy, Antoine Sabourin, Cyrielle Gasc, Carole Schwintner, Nathalie Corvaia
Background
Studies for the last 8 years have connected gut microbiome composition to ICI efficacy in cancer therapy, including pilot studies1 2 demonstrating that FMT from responders to non-responders can improve response rates. However, interstudy inconsistencies were observed in microbiome signature findings3 also confirmed when reprocessing internally raw data from multiple studies. It remains critical to tackle this heterogeneity to learn from stable patterns and develop a robust and reliable drug candidate screening algorithm. Hence, MaaT Pharma developed an AI framework to train models from microbiota Whole Metagenome Sequencing (WMS) datasets that predict the responder status to ICIs. We focused on performance and robustness, which was achieved by monitoring the AUC as a standard approach, and precision to control for false positive rate and to emphasize the positive classification criticality in a drug candidate selection approach.
Methods
We collected baseline WMS datasets from 10 cohorts in 3 ICI indications: melanoma, non-small cell lung cancer and renal cell carcinoma, along with clinical evaluation of ICI treatment. Those datasets were processed by gutPrint® MgRunner software, before being included in the AI framework. About 70 experiments were conducted within a Leave-One-Dataset-Out cross-validation scheme. Various factors such as taxonomic or functional inputs, dataset bias correction, data augmentation approaches, ML algorithms and data representation methods, were tested to select the top ones. Finally, a model was refit with the best performing parameters on the entire dataset, and applied to score MaaT Pharma mono-donor and healthy-pooled-donors-derived drug substances (DS).
Results
The best performing experiment provided models based on the XGBoost algorithm with AUCs ranging from 0.52 to 0.73 depending on the left-out cohort (average AUC = 0.65), and a precision that ranges between 0.55 and 0.81 (average precision = 0.65). Those results outperform melanoma-centered study with a comparable method.4 Despite the diverse data sources and indications, the multi-indication approach surpassed the mono-indication (melanoma) training approach for predictions related to melanoma patients. Considering the scoring of DS derived from healthy donors, 73% of mono-donors and 91% of healthy-pooled-donors-derived DS were classified as ‘Responder-like’.
Conclusions
Present study highlights the significance of dataset size in ICI microbiota models and presents a methodology to enhance the performances of a multi-cohort-based ML approach. Conditioned to the performances we obtained, the healthy-pooled-donors-derived DS harbor a considerable ratio (91%) of ‘ICI Responder-like’, significantly higher than the mono-donor stools (73%) suggesting that pooled ecosystems from healthy donors could better convert ICI-non responders into responders.
{"title":"1304 Robust machine learning (ML) approach for screening microbiome ecosystem therapies (MET) drug candidates in combination with immune checkpoint inhibitors","authors":"Emmanuel Prestat, Elsa Schalck, Antoine Bonnefoy, Antoine Sabourin, Cyrielle Gasc, Carole Schwintner, Nathalie Corvaia","doi":"10.1136/jitc-2023-sitc2023.1304","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.1304","url":null,"abstract":"<h3>Background</h3> Studies for the last 8 years have connected gut microbiome composition to ICI efficacy in cancer therapy, including pilot studies<sup>1 2</sup> demonstrating that FMT from responders to non-responders can improve response rates. However, interstudy inconsistencies were observed in microbiome signature findings<sup>3</sup> also confirmed when reprocessing internally raw data from multiple studies. It remains critical to tackle this heterogeneity to learn from stable patterns and develop a robust and reliable drug candidate screening algorithm. Hence, MaaT Pharma developed an AI framework to train models from microbiota Whole Metagenome Sequencing (WMS) datasets that predict the responder status to ICIs. We focused on performance and robustness, which was achieved by monitoring the <i>AUC</i> as a standard approach, and <i>precision</i> to control for false positive rate and to emphasize the positive classification criticality in a drug candidate selection approach. <h3>Methods</h3> We collected baseline WMS datasets from 10 cohorts in 3 ICI indications: melanoma, non-small cell lung cancer and renal cell carcinoma, along with clinical evaluation of ICI treatment. Those datasets were processed by gutPrint® <i>MgRunner</i> software, before being included in the AI framework. About 70 experiments were conducted within a Leave-One-Dataset-Out cross-validation scheme. Various factors such as taxonomic or functional inputs, dataset bias correction, data augmentation approaches, ML algorithms and data representation methods, were tested to select the top ones. Finally, a model was refit with the best performing parameters on the entire dataset, and applied to score MaaT Pharma mono-donor and healthy-pooled-donors-derived drug substances (DS). <h3>Results</h3> The best performing experiment provided models based on the XGBoost algorithm with AUCs ranging from 0.52 to 0.73 depending on the left-out cohort (average AUC = 0.65), and a precision that ranges between 0.55 and 0.81 (average precision = 0.65). Those results outperform melanoma-centered study with a comparable method.<sup>4</sup> Despite the diverse data sources and indications, the multi-indication approach surpassed the mono-indication (melanoma) training approach for predictions related to melanoma patients. Considering the scoring of DS derived from healthy donors, 73% of mono-donors and 91% of healthy-pooled-donors-derived DS were classified as ‘Responder-like’. <h3>Conclusions</h3> Present study highlights the significance of dataset size in ICI microbiota models and presents a methodology to enhance the performances of a multi-cohort-based ML approach. Conditioned to the performances we obtained, the healthy-pooled-donors-derived DS harbor a considerable ratio (91%) of ‘ICI Responder-like’, significantly higher than the mono-donor stools (73%) suggesting that pooled ecosystems from healthy donors could better convert ICI-non responders into responders. <h3>References</h3","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.0644
Naifei Chen, Chengfei Pu, Lingling Zhao, Ning Li, Chang Wang, Yusheng Huang, Suxia Luo, Xun Li, Zhenzhou Yang, Jun Bie, Ruihong Zhu, Xi Huang, Haiyang Tang, Tingting Liang, Yizhuo Wang, Beibei Jia, Dongqi Chen, Victor Lu, Zhao Wu, Yongping Song, Lei Xiao, Jiuwei Cui
Background
GCC19CART, the first clinical candidate from the CoupledCAR® solid tumor platform, targets guanylate cyclase-C (GCC) which is expressed in colorectal cancers. CoupledCAR utilizes multiple vectors to make both solid tumor targeting CAR-T and CD19 CAR-T in a single manufacturing step. An investigator-initiated dose escalation trial in China for patients with relapsed or refractory metastatic colorectal cancer (R/R mCRC) is reported here.
Methods
Subjects are screened for GCC expression by immunohistochemistry. Eligible subjects undergo leukapheresis, a single dose of lymphodepleting chemotherapy (fludarabine 30mg/m2 and cyclophosphamide 300mg/m2) 3 days prior to infusion, and then administration of a single infusion of GCC19CART at one of two preassigned doses: 1x106 or 2x106 CAR T-cells/kg. Endpoints are safety and preliminary evidence of efficacy as determined by CT or PET/CT per RECIST 1.1 or PERCIST 1.0. All responses were confirmed by an independent third-party imaging contract research organization (CRO).
Results
13 subjects have been enrolled to dose level 1 (1x106 cells/kg) and 8 subjects have been enrolled to dose level 2 (2x106 cells/kg). The most common adverse events were cytokine release syndrome (CRS) in 21/21 subjects (Grade 1 19/21 (90.48%) or Grade 2 2/21 (9.52%)) and diarrhea in 21/21 subjects (Grade 1 6/21 (28.57%) Grade 2 5/21 (23.81%) Grade 3 9/21 (42.86%) or Grade 4 1/21 (4.76%)). All patients with grade 3 and higher side effects were well managed. Immune effector cell-associated neurotoxicity syndrome(ICANS) was observed in 2/21 (9.52%) subjects at Grade 3 or 4 and resolved with corticosteroids. The combined overall response rate (ORR) for both dose levels was 28.6% (6/21). For dose level 1, the overall response rate (ORR) per RECIST 1.1 was 15.4% (2/13). Two subjects demonstrated a partial response (PR) while 3 additional subjects had partial metabolic response (PMR) on PET/CT with stable disease (SD) or progressive disease (PD) per RECIST 1.1. For dose level 2, The ORR per RECIST 1.1 was 50% (4/8). 4 subjects demonstrated a PR (3 at month 1, 1 at month 3 after being SD at month 1) and 2 additional subjects had PMR on PET/CT with SD per RECIST 1.1.
Conclusions
Preliminary results demonstrate that GCC19CART has meaningful dose-dependent clinical activity and an acceptable safety profile in relapsed or refractory metastatic colorectal cancer. This trial is ongoing and updated data will be presented. A Phase 1 trial of GCC19CART in the US has opened for accrual and is expected to enroll patients in mid-2022.
{"title":"644 A phase 1 dose escalation study of GCC19CART – A novel CoupledCAR® therapy for subjects with metastatic colorectal cancer","authors":"Naifei Chen, Chengfei Pu, Lingling Zhao, Ning Li, Chang Wang, Yusheng Huang, Suxia Luo, Xun Li, Zhenzhou Yang, Jun Bie, Ruihong Zhu, Xi Huang, Haiyang Tang, Tingting Liang, Yizhuo Wang, Beibei Jia, Dongqi Chen, Victor Lu, Zhao Wu, Yongping Song, Lei Xiao, Jiuwei Cui","doi":"10.1136/jitc-2023-sitc2023.0644","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.0644","url":null,"abstract":"<h3>Background</h3> GCC19CART, the first clinical candidate from the CoupledCAR® solid tumor platform, targets guanylate cyclase-C (GCC) which is expressed in colorectal cancers. CoupledCAR utilizes multiple vectors to make both solid tumor targeting CAR-T and CD19 CAR-T in a single manufacturing step. An investigator-initiated dose escalation trial in China for patients with relapsed or refractory metastatic colorectal cancer (R/R mCRC) is reported here. <h3>Methods</h3> Subjects are screened for GCC expression by immunohistochemistry. Eligible subjects undergo leukapheresis, a single dose of lymphodepleting chemotherapy (fludarabine 30mg/m<sup>2</sup> and cyclophosphamide 300mg/m<sup>2</sup>) 3 days prior to infusion, and then administration of a single infusion of GCC19CART at one of two preassigned doses: 1x10<sup>6</sup> or 2x10<sup>6</sup> CAR T-cells/kg. Endpoints are safety and preliminary evidence of efficacy as determined by CT or PET/CT per RECIST 1.1 or PERCIST 1.0. All responses were confirmed by an independent third-party imaging contract research organization (CRO). <h3>Results</h3> 13 subjects have been enrolled to dose level 1 (1x10<sup>6</sup> cells/kg) and 8 subjects have been enrolled to dose level 2 (2x10<sup>6</sup> cells/kg). The most common adverse events were cytokine release syndrome (CRS) in 21/21 subjects (Grade 1 19/21 (90.48%) or Grade 2 2/21 (9.52%)) and diarrhea in 21/21 subjects (Grade 1 6/21 (28.57%) Grade 2 5/21 (23.81%) Grade 3 9/21 (42.86%) or Grade 4 1/21 (4.76%)). All patients with grade 3 and higher side effects were well managed. Immune effector cell-associated neurotoxicity syndrome(ICANS) was observed in 2/21 (9.52%) subjects at Grade 3 or 4 and resolved with corticosteroids. The combined overall response rate (ORR) for both dose levels was 28.6% (6/21). For dose level 1, the overall response rate (ORR) per RECIST 1.1 was 15.4% (2/13). Two subjects demonstrated a partial response (PR) while 3 additional subjects had partial metabolic response (PMR) on PET/CT with stable disease (SD) or progressive disease (PD) per RECIST 1.1. For dose level 2, The ORR per RECIST 1.1 was 50% (4/8). 4 subjects demonstrated a PR (3 at month 1, 1 at month 3 after being SD at month 1) and 2 additional subjects had PMR on PET/CT with SD per RECIST 1.1. <h3>Conclusions</h3> Preliminary results demonstrate that GCC19CART has meaningful dose-dependent clinical activity and an acceptable safety profile in relapsed or refractory metastatic colorectal cancer. This trial is ongoing and updated data will be presented. A Phase 1 trial of GCC19CART in the US has opened for accrual and is expected to enroll patients in mid-2022.","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161136","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.0074
Julia Cuende, Stephanie Ma, Iain Welsby, Nicolas Rosewick, Noemie Wald, Anais Vezzu, Marjorie Mercier, Paola Tieppo, Bart Claes, Yvonne McGrath, Olivier De Henau, Gregory Driessens, Maurizio Ceppi, Marion Libouban
Background
T cell immunoglobulin and ITIM domain (TIGIT) is an immune checkpoint inhibitor expressed mainly on NK and T cell populations. Antagonist a-TIGIT mAbs in combination with a-PD(L)-1 demonstrated clinical proof of concept in 1L NSCLC.1 2 Belrestotug (previously known as EOS-448 or GSK4428859A) is an antagonistic anti-TIGIT human immunoglobulin G1 (hIgG1) antibody, designed to prevent ligand binding and to engage Fc gamma receptors (FcγR), resulting in a multifaceted mode of action: (i) activate effector T cells and NK cells (ii) modulate antigen-presenting cells, and (iii) deplete suppressive regulatory T cells (Tregs) as well as terminally exhausted CD8 T cells that express the highest levels of TIGIT.3–5 In a first-in-human trial, belrestotug demonstrated a good tolerability profile with early signs of efficacy6 and we reported for the first time intratumoral TIGIT target engagement in the patient tumors.5 Belrestotug is currently being tested in combination with anti-PD1 in solid tumors.
Results
Pharmacodynamic assessment made by flow cytometry in the blood of patients treated with belrestotug in monotherapy and in combination with anti-PD1 (pembrolizumab or dostarlimab) showed (i) increased proportion of proliferating memory CD8+ T and NK cells during the first treatment cycle, (ii) sustained depletion of immunosuppressive Tregs, and (iii) decreased proportion of TIGIT high CD8+ T cells. We demonstrated the terminally exhausted phenotype of the TIGIT high CD8+ T cells targeted by belrestotug by isolating peripheral blood mononuclear cells (PBMCs) from treatment naïve cancer patients. Tregs in the tumor microenvironment (TME) hinder effective tumor immune response and are mainly localized in the stromal area. Stromal Tregs co-expressing several immune checkpoint inhibitors, including TIGIT, have a high immunosuppressive profile and its density is associated with poor clinical outcome.7 We investigated the functional effect of belrestotug combined with anti-PD1 on the TME using a multiplex immunofluorescence (mIF) panel. We observed a decrease of immunosuppressive TIGIT+ Tregs and PD1+ Tregs in the stroma, as well as a spatial reorganization.
Conclusions
Overall, our data suggests a mechanism whereby belrestotug in combination with anti-PD1 induces a more immunocompetent TME driven by immunosuppressive Treg depletion. These data support the clinical development of the doublet therapy to enhance an anti-tumor immune response.
References
Cho BC, et al. Abstract #LBA2 ESMO-IO; 2021 Jonhson ARC-7 ASCO Dec 2022 Yu X, et al, Nature, 2008 Preillon J, et al. Mol Cancer Ther, 2021. Cuende, et al. AACR 2022 Van den Mooter TF, et al. Abstract #CT118 AACR; 2021 Devi-Marulkar, et al, comms Bio, 2022
{"title":"74 a-TIGIT mAb belrestotug in combination with anti-PD1 induces an immunocompetent tumor microenvironment (TME)","authors":"Julia Cuende, Stephanie Ma, Iain Welsby, Nicolas Rosewick, Noemie Wald, Anais Vezzu, Marjorie Mercier, Paola Tieppo, Bart Claes, Yvonne McGrath, Olivier De Henau, Gregory Driessens, Maurizio Ceppi, Marion Libouban","doi":"10.1136/jitc-2023-sitc2023.0074","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.0074","url":null,"abstract":"<h3>Background</h3> T cell immunoglobulin and ITIM domain (TIGIT) is an immune checkpoint inhibitor expressed mainly on NK and T cell populations. Antagonist a-TIGIT mAbs in combination with a-PD(L)-1 demonstrated clinical proof of concept in 1L NSCLC.<sup>1 2</sup> Belrestotug (previously known as EOS-448 or GSK4428859A) is an antagonistic anti-TIGIT human immunoglobulin G1 (hIgG1) antibody, designed to prevent ligand binding and to engage Fc gamma receptors (FcγR), resulting in a multifaceted mode of action: (i) activate effector T cells and NK cells (ii) modulate antigen-presenting cells, and (iii) deplete suppressive regulatory T cells (Tregs) as well as terminally exhausted CD8 T cells that express the highest levels of TIGIT.<sup>3–5</sup> In a first-in-human trial, belrestotug demonstrated a good tolerability profile with early signs of efficacy<sup>6</sup> and we reported for the first time intratumoral TIGIT target engagement in the patient tumors.<sup>5</sup> Belrestotug is currently being tested in combination with anti-PD1 in solid tumors. <h3>Results</h3> Pharmacodynamic assessment made by flow cytometry in the blood of patients treated with belrestotug in monotherapy and in combination with anti-PD1 (pembrolizumab or dostarlimab) showed (i) increased proportion of proliferating memory CD8+ T and NK cells during the first treatment cycle, (ii) sustained depletion of immunosuppressive Tregs, and (iii) decreased proportion of TIGIT high CD8+ T cells. We demonstrated the terminally exhausted phenotype of the TIGIT high CD8+ T cells targeted by belrestotug by isolating peripheral blood mononuclear cells (PBMCs) from treatment naïve cancer patients. Tregs in the tumor microenvironment (TME) hinder effective tumor immune response and are mainly localized in the stromal area. Stromal Tregs co-expressing several immune checkpoint inhibitors, including TIGIT, have a high immunosuppressive profile and its density is associated with poor clinical outcome.<sup>7</sup> We investigated the functional effect of belrestotug combined with anti-PD1 on the TME using a multiplex immunofluorescence (mIF) panel. We observed a decrease of immunosuppressive TIGIT+ Tregs and PD1+ Tregs in the stroma, as well as a spatial reorganization. <h3>Conclusions</h3> Overall, our data suggests a mechanism whereby belrestotug in combination with anti-PD1 induces a more immunocompetent TME driven by immunosuppressive Treg depletion. These data support the clinical development of the doublet therapy to enhance an anti-tumor immune response. <h3>References</h3> Cho BC, <i>et al.</i> Abstract #LBA2 ESMO-IO; 2021 Jonhson ARC-7 ASCO Dec 2022 Yu X, <i>et al</i>, <i>Nature</i>, 2008 Preillon J, <i>et al. Mol Cancer Ther</i>, 2021. Cuende, <i>et al. AACR</i> 2022 Van den Mooter TF, <i>et al.</i> Abstract #CT118 AACR; 2021 Devi-Marulkar, <i>et al</i>, <i>comms Bio</i>, 2022","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161142","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.1422
JinHo Kang, Hyo-Hyun Park, Jin Kyeong Choi, Eunkyo Joung, Hun Jung
Background
Th-Vac® discovery platform, consists of module 1 (Immunoinformatics-based in-silico) and module 2 (in-vitro and vaccine efficacy evaluation) under the comprehensive immunologic algorithms, is aimed to identify antigen-specific MHC class II epitopes for CD4+ T cell with optimal binding affinity and promiscuity across multiple alleles. In a therapeutic cancer area, Th1-specific epitopes prediction followed by cancer vaccine discovery is fully applicable.1–3 Therefore, an immunological mode of action of cancer vaccine powered by Th-Vac® discovery platform is to generate the inflammatory T-cell immunity (type 1 immunity) against tumor antigens with enough potency to overcome either the absence of a T-cell immune response to the tumor or a preexisting immune tolerant response which tends to be the immune-suppressing (type 2 immunity) phenomenon. TROP2 protein is a transmembrane glycoprotein encoded by the Tacstd2 gene. It is known to be overexpressed on the surface of various epithelial cancer cells, including prostate cancer, colorectal cancer, pancreatic cancer, and ovarian cancer, although the role of TROP2 protein in cancer cell growth and proliferation is not well understood.4 In this study, Th1-specific TROP2 epitopes were precisely identified via module 1 and 2 of Th-Vac® discovery platform.
Methods
MHC class II binding epitope predictions were conducted thru engaging common human MHC class II alleles in module 1 of Th-Vac® discovery platform, and MHC class II-specific peptide sequences (up to 15 mers) were initially selected based on the rank order of the predicted binding affinity which are representing potential immunogenic hot spots to MHC class II. Each peptides were synthesized to be stepping into in-vitro immunologic assessment (module 2 of Th-Vac® discovery platform). Th1-specific immunologic response was evaluated with ELISpot and/or FACS analysis.
Results
In a module 1 of Th-Vac® discovery platform, 10 sequences (consists of 15 mers) were finally predicted as potential epitopes that have a high-affinity to MHC class II. In module 2 (2a for in-vitro and 2b for in-vivo immunologic evaluation), four epitopes out of 10 epitopes showed only strong Th1 immune responses without type 2 immunity in both ELISpot and FACS analysis.
Conclusions
TROP2 epitopes that were identified in the Th-Vac® discovery platform has been developed as a promising ‘off-the-shelf’ type Th1-specific therapeutic cancer vaccine (AST-07X). Additionally, the Th-Vac® platform were fully validated in terms of performance and its application would be expanded beyond a cancer vaccine.
References
Watt WC, Cecil DL, Disis ML. Selection of epitopes from self-antigens for eliciting Th2 or Th1 activity in the treatment of autoimmune disease or cancer. Semin Immunopath
{"title":"1422 Th1-specific TROP2 vaccine (AST-07X) elicits strong antigen-specific T cell responses<i>in vitro</i>","authors":"JinHo Kang, Hyo-Hyun Park, Jin Kyeong Choi, Eunkyo Joung, Hun Jung","doi":"10.1136/jitc-2023-sitc2023.1422","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.1422","url":null,"abstract":"<h3>Background</h3> Th-Vac<sup>®</sup> discovery platform, consists of module 1 (Immunoinformatics-based <i>in-silico</i>) and module 2 (<i>in-vitro</i> and vaccine efficacy evaluation) under the comprehensive immunologic algorithms, is aimed to identify antigen-specific MHC class II epitopes for CD4<sup>+</sup> T cell with optimal binding affinity and promiscuity across multiple alleles. In a therapeutic cancer area, Th1-specific epitopes prediction followed by cancer vaccine discovery is fully applicable.<sup>1–3</sup> Therefore, an immunological mode of action of cancer vaccine powered by Th-Vac<sup>®</sup> discovery platform is to generate the inflammatory T-cell immunity (type 1 immunity) against tumor antigens with enough potency to overcome either the absence of a T-cell immune response to the tumor or a preexisting immune tolerant response which tends to be the immune-suppressing (type 2 immunity) phenomenon. TROP2 protein is a transmembrane glycoprotein encoded by the Tacstd2 gene. It is known to be overexpressed on the surface of various epithelial cancer cells, including prostate cancer, colorectal cancer, pancreatic cancer, and ovarian cancer, although the role of TROP2 protein in cancer cell growth and proliferation is not well understood.<sup>4</sup> In this study, Th1-specific TROP2 epitopes were precisely identified via module 1 and 2 of Th-Vac<sup>®</sup> discovery platform. <h3>Methods</h3> MHC class II binding epitope predictions were conducted thru engaging common human MHC class II alleles in module 1 of Th-Vac<sup>®</sup> discovery platform, and MHC class II-specific peptide sequences (up to 15 mers) were initially selected based on the rank order of the predicted binding affinity which are representing potential immunogenic hot spots to MHC class II. Each peptides were synthesized to be stepping into <i>in-vitro</i> immunologic assessment (module 2 of Th-Vac<sup>®</sup> discovery platform). Th1-specific immunologic response was evaluated with ELISpot and/or FACS analysis. <h3>Results</h3> In a module 1 of Th-Vac<sup>®</sup> discovery platform, 10 sequences (consists of 15 mers) were finally predicted as potential epitopes that have a high-affinity to MHC class II. In module 2 (2a for <i>in-vitro</i> and 2b for <i>in-vivo</i> immunologic evaluation), four epitopes out of 10 epitopes showed only strong Th1 immune responses without type 2 immunity in both ELISpot and FACS analysis. <h3>Conclusions</h3> TROP2 epitopes that were identified in the Th-Vac<sup>®</sup> discovery platform has been developed as a promising ‘off-the-shelf’ type Th1-specific therapeutic cancer vaccine (AST-07X). Additionally, the Th-Vac<sup>®</sup> platform were fully validated in terms of performance and its application would be expanded beyond a cancer vaccine. <h3>References</h3> Watt WC, Cecil DL, Disis ML. Selection of epitopes from self-antigens for eliciting Th2 or Th1 activity in the treatment of autoimmune disease or cancer. <i>Semin Immunopath","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.1411
Robert O Dillman, Gabriel I Nistor, Hans S Keirstead
Background
Personal therapeutic cancer vaccines consisting of autologous dendritic cells (DC) loaded ex vivo with autologous tumor antigens (ATA) derived from cells that are self-renewing in culture (cancer-initiating cells, stem cells) have been clinically tested for more than 20 years. The current study addressed feasibility, safety, and efficacy of the DC-ATA approach.
Methods
ATA, as irradiated tumor cells (ITC) or ITC lysates, were derived from short-term tumor cell lines established from resected cancer tissue. Peripheral blood monocytes (MC) were collected by leukapheresis; MC were differentiated into DC by culturing with interleukin-4 and granulocyte-macrophage colony-stimulating-factor (GM-CSF). Cryopreserved doses were thawed, suspended in 500 mcg GM-CSF, and injected at weeks 1, 2, 3, 8, 12, 16, 20, and 24. Data was derived from clinical trials conducted during 2000–2023 in patients with metastatic renal, hepatocellular, and ovary cancers, melanoma, and glioblastoma (GBM). Key end-points were DC-ATA manufacturing success rates, treatment emergent adverse events (TEAE), objective response rates, progression-free survival (PFS) and overall survival (OS).
Results
The success rate for establishing tumor cell lines in proprietary stem cell media was 173/178 (97.2%) including 80/82 GBM, 73/75 ovary, 17/18 liver, 3/3 other. During 2002–2023, leukapheresis procedures yielded sufficient monocytes in 218/223 patients (97.8%) including 74/74 melanoma, 72/74 GBM, 50/53 ovary, 11/11 renal cell, 8/8 hepatoma, and 3/3 other. Of 187 treated patients, no one discontinued DC-ATA because of TEAE; there were no grade 4 TEAE. Most common TEAE were mild to moderate, self-limited local injection site reactions and flu-like symptoms. There was no difference in TEAE frequency or grade in a blinded, randomized trial of DC-ATA vs MC in ovary patients. Immune responses were more favorable for DC-ATA vs ITC in metastatic melanoma, and for DC-ATA vs MC in advanced ovary cancer. During treatment there were no objective responses among 30 patients with measurable metastatic disease, but 3 (10%) had delayed, durable, complete remissions (2 renal, 1 melanoma). In GBM PFS of 10.4 months was 50% longer than in standard treatment arms of six randomized trials. In metastatic melanoma DC-ATA was associated with better OS compared to historical treatment with ITC vaccine (median 60 vs 20.4 months, p<0.001), and better OS compared to ITC in a randomized trial (43.4 vs 20.5 months with a 70% reduction in death p=0.0053)
Conclusions
This DC-ATA approach is feasible and reproducible across tumor types, treatment is well-tolerated, and there is efficacy in some patients. Additional investigation is warranted.
All clinical studies obtained approvals from local ethics committees or institutional review bo
个人治疗性癌症疫苗由自体树突状细胞(DC)在体外装载来自培养中自我更新的细胞(癌起始细胞、干细胞)的自体肿瘤抗原(ATA)组成，已经进行了20多年的临床试验。目前的研究探讨了DC-ATA方法的可行性、安全性和有效性。方法ATA作为辐照肿瘤细胞(ITC)或ITC裂解物，来源于切除肿瘤组织建立的短期肿瘤细胞系。白细胞分离法采集外周血单核细胞(MC);通过白细胞介素-4和粒细胞-巨噬细胞集落刺激因子(GM-CSF)培养将MC分化为DC。冷冻保存的剂量解冻，悬浮在500 mcg GM-CSF中，并在第1、2、3、8、12、16、20和24周注射。数据来源于2000-2023年期间在转移性肾癌、肝细胞癌和卵巢癌、黑色素瘤和胶质母细胞瘤(GBM)患者中进行的临床试验。主要终点为DC-ATA制造成功率、治疗紧急不良事件(TEAE)、客观缓解率、无进展生存期(PFS)和总生存期(OS)。结果在专有干细胞培养基中建立肿瘤细胞系的成功率为173/178(97.2%)，其中GBM 80/82，卵巢73/75，肝脏17/18，其他3/3。2002-2023年期间，218/223例患者(97.8%)的白细胞分离术获得了足够的单核细胞，包括74/74例黑色素瘤、72/74例GBM、50/53例卵巢、11/11例肾细胞、8/8例肝癌和3/3其他。在187名接受治疗的患者中，没有人因为TEAE而停用DC-ATA;无4级TEAE。最常见的TEAE是轻至中度、自限性局部注射部位反应和流感样症状。在一项针对卵巢患者的DC-ATA和MC的盲法随机试验中，TEAE的频率和级别没有差异。在转移性黑色素瘤中DC-ATA和ITC的免疫应答更有利，在晚期卵巢癌中DC-ATA和MC的免疫应答更有利。在治疗期间，30例可测量的转移性疾病患者没有客观反应，但3例(10%)有延迟，持久，完全缓解(2例肾脏，1例黑色素瘤)。在6个随机试验中，GBM患者的PFS为10.4个月，比标准治疗组长50%。在转移性黑色素瘤中，与ITC疫苗的历史治疗相比，DC-ATA具有更好的OS(中位60个月vs 20.4个月，p<0.001)，在一项随机试验中，与ITC相比，DC-ATA具有更好的OS(43.4个月vs 20.5个月，死亡率降低70% p=0.0053)。结论DC-ATA方法在不同肿瘤类型中是可行的，可重复的，治疗耐受性良好，并且在一些患者中有效。有必要进行进一步调查。临床试验注册Clinicaltrials.gov: NCT00014131、NCT00012064、NCT00436930、NCT00331526、NCT03400917伦理批准所有临床研究均获得当地伦理委员会或机构审查委员会的批准，所有参与者均给予书面知情同意。
{"title":"1411 Feasibility, safety, and efficacy of personal vaccines consisting of autologous dendritic cells loaded<i>ex vivo</i>with autologous tumor antigens from self-renewing cancer cells","authors":"Robert O Dillman, Gabriel I Nistor, Hans S Keirstead","doi":"10.1136/jitc-2023-sitc2023.1411","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.1411","url":null,"abstract":"<h3>Background</h3> Personal therapeutic cancer vaccines consisting of autologous dendritic cells (DC) loaded <i>ex vivo</i> with autologous tumor antigens (ATA) derived from cells that are self-renewing in culture (cancer-initiating cells, stem cells) have been clinically tested for more than 20 years. The current study addressed feasibility, safety, and efficacy of the DC-ATA approach. <h3>Methods</h3> ATA, as irradiated tumor cells (ITC) or ITC lysates, were derived from short-term tumor cell lines established from resected cancer tissue. Peripheral blood monocytes (MC) were collected by leukapheresis; MC were differentiated into DC by culturing with interleukin-4 and granulocyte-macrophage colony-stimulating-factor (GM-CSF). Cryopreserved doses were thawed, suspended in 500 mcg GM-CSF, and injected at weeks 1, 2, 3, 8, 12, 16, 20, and 24. Data was derived from clinical trials conducted during 2000–2023 in patients with metastatic renal, hepatocellular, and ovary cancers, melanoma, and glioblastoma (GBM). Key end-points were DC-ATA manufacturing success rates, treatment emergent adverse events (TEAE), objective response rates, progression-free survival (PFS) and overall survival (OS). <h3>Results</h3> The success rate for establishing tumor cell lines in proprietary stem cell media was 173/178 (97.2%) including 80/82 GBM, 73/75 ovary, 17/18 liver, 3/3 other. During 2002–2023, leukapheresis procedures yielded sufficient monocytes in 218/223 patients (97.8%) including 74/74 melanoma, 72/74 GBM, 50/53 ovary, 11/11 renal cell, 8/8 hepatoma, and 3/3 other. Of 187 treated patients, no one discontinued DC-ATA because of TEAE; there were no grade 4 TEAE. Most common TEAE were mild to moderate, self-limited local injection site reactions and flu-like symptoms. There was no difference in TEAE frequency or grade in a blinded, randomized trial of DC-ATA vs MC in ovary patients. Immune responses were more favorable for DC-ATA vs ITC in metastatic melanoma, and for DC-ATA vs MC in advanced ovary cancer. During treatment there were no objective responses among 30 patients with measurable metastatic disease, but 3 (10%) had delayed, durable, complete remissions (2 renal, 1 melanoma). In GBM PFS of 10.4 months was 50% longer than in standard treatment arms of six randomized trials. In metastatic melanoma DC-ATA was associated with better OS compared to historical treatment with ITC vaccine (median 60 vs 20.4 months, p<0.001), and better OS compared to ITC in a randomized trial (43.4 vs 20.5 months with a 70% reduction in death p=0.0053) <h3>Conclusions</h3> This DC-ATA approach is feasible and reproducible across tumor types, treatment is well-tolerated, and there is efficacy in some patients. Additional investigation is warranted. <h3>Trial Registration</h3> Clinicaltrials.gov: NCT00014131, NCT00012064, NCT00436930, NCT00331526, NCT03400917 <h3>Ethics Approval</h3> All clinical studies obtained approvals from local ethics committees or institutional review bo","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.0418
Yanjie Li, Lei Ding, Jixue Li, Mariska Ter Haak, Kate Rochlin, Lawrence Lamb
Background
Gamma-delta (γδ) T cells are depleted during cancer progression resulting in the progressive loss of anti-cancer activity. Elevated numbers of γδ T cells are associated with greater survival outcomes in both hematopoietic and solid malignancies. Induced pluripotent stem cell (iPSC) derived γδ T cells could address the therapeutic challenges of multiple allogeneic γδ T cell infusions as iPSCs possess nearly unlimited self-renewal and multi-lineage differentiation potential. These can be genetically modified, selected, and propagated to provide a source of potentially ‘off-the-shelf’ immune cells.
Methods
Precursor cells obtained from healthy volunteer donors were reprogrammed into iPSCs using non-integrating Yamanaka factors. A feeder-free multi-step strategy was used to differentiate iPSCs, leading to the generation of Vδ1+ γδ T cells. Characterization of the Vδ1+ T cell product included multiplex genomic PCR assays and Sanger sequencing to examine the rearrangement of the TCRγ and TCRδ gene loci, and G-band karyotype analysis. Pluripotent markers (Tra-1–60, OCT3/4 & SSEA4), HPC markers (CD34, CD43), γδ T cell surface markers (CD3, γδ TCR, CD4, CD8, CD16, CD56), effector memory markers (CD45RA, CD27), natural cytotoxicity receptors (NKG2D) were identified using multiparameter flow cytometry. T cell function was determined by flow cytometric cytotoxicity assays against K562, OLM13, U87MG, OVSAHO, OVCAR-3, KURAMUCHI targets at increasing Effector to Target (E:T) ratios. Th1/2/17 cytokine release was determined following PMA/ionomycin stimulation and LEGENDplex™ bead-based immunoassays.
Results
We generated Vδ1T-iPSC lines (iVδ1T) identified as Vγ5-to-Jγ1/2 and Vδ1-to-Jδ1 recombination. One iPSC line showed normal karyotype with 99% cells expressing OCT3/4 & SSEA4. The differentiation process generated 70+ million iVδ1T cells from 3 million iPSCs expressing γδ T cell markers CD45, CD3, Vδ1-TCR, CD16, CD56, NKG2D, CD45RA, and CD27. Cytokine release following PMA/ionomycin stimulation showed increases of at least 50x for Granzyme A, 300x for IFN-γ, 1400x for TNF-α, ~10 to 20x for Granzyme B, ~5 to 10x for Perforin, ~6x for Granulysin. IL-6 was not detected either before or after stimulation, and IL17A was at low concentration. At a 16:1 E:T ratio, preliminary data shows that Vδ1+ γδ T cells killed K562 (CML) 95.7%; MOLM13 (AML) 60.3%; U87MG (glioblastoma) 70.3%; and ovarian cancer lines OVSAHO 57.1%, OVCAR-3 69.6%, and KURAMUCHI 55.1%.
Conclusions
We generated Vδ1+ iPSC derived γδ T cells with effector cytokine phenotype and low risk for cytokine release syndrome. Robust cytotoxic activity was seen across a variety of cancer cell lines, potentially providing an off-the-shelf platform for allogeneic cell therapy.
{"title":"418 The development of ‘off-the-shelf’ manufacturing strategies of iPSC-based gamma-delta T cells","authors":"Yanjie Li, Lei Ding, Jixue Li, Mariska Ter Haak, Kate Rochlin, Lawrence Lamb","doi":"10.1136/jitc-2023-sitc2023.0418","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.0418","url":null,"abstract":"<h3>Background</h3> Gamma-delta (γδ) T cells are depleted during cancer progression resulting in the progressive loss of anti-cancer activity. Elevated numbers of γδ T cells are associated with greater survival outcomes in both hematopoietic and solid malignancies. Induced pluripotent stem cell (iPSC) derived γδ T cells could address the therapeutic challenges of multiple allogeneic γδ T cell infusions as iPSCs possess nearly unlimited self-renewal and multi-lineage differentiation potential. These can be genetically modified, selected, and propagated to provide a source of potentially ‘off-the-shelf’ immune cells. <h3>Methods</h3> Precursor cells obtained from healthy volunteer donors were reprogrammed into iPSCs using non-integrating Yamanaka factors. A feeder-free multi-step strategy was used to differentiate iPSCs, leading to the generation of Vδ1+ γδ T cells. Characterization of the Vδ1+ T cell product included multiplex genomic PCR assays and Sanger sequencing to examine the rearrangement of the TCRγ and TCRδ gene loci, and G-band karyotype analysis. Pluripotent markers (Tra-1–60, OCT3/4 & SSEA4), HPC markers (CD34, CD43), γδ T cell surface markers (CD3, γδ TCR, CD4, CD8, CD16, CD56), effector memory markers (CD45RA, CD27), natural cytotoxicity receptors (NKG2D) were identified using multiparameter flow cytometry. T cell function was determined by flow cytometric cytotoxicity assays against K562, OLM13, U87MG, OVSAHO, OVCAR-3, KURAMUCHI targets at increasing Effector to Target (E:T) ratios. Th1/2/17 cytokine release was determined following PMA/ionomycin stimulation and LEGENDplex™ bead-based immunoassays. <h3>Results</h3> We generated Vδ1T-iPSC lines (iVδ1T) identified as Vγ5-to-Jγ1/2 and Vδ1-to-Jδ1 recombination. One iPSC line showed normal karyotype with 99% cells expressing OCT3/4 & SSEA4. The differentiation process generated 70+ million iVδ1T cells from 3 million iPSCs expressing γδ T cell markers CD45, CD3, Vδ1-TCR, CD16, CD56, NKG2D, CD45RA, and CD27. Cytokine release following PMA/ionomycin stimulation showed increases of at least 50x for Granzyme A, 300x for IFN-γ, 1400x for TNF-α, ~10 to 20x for Granzyme B, ~5 to 10x for Perforin, ~6x for Granulysin. IL-6 was not detected either before or after stimulation, and IL17A was at low concentration. At a 16:1 E:T ratio, preliminary data shows that Vδ1+ γδ T cells killed K562 (CML) 95.7%; MOLM13 (AML) 60.3%; U87MG (glioblastoma) 70.3%; and ovarian cancer lines OVSAHO 57.1%, OVCAR-3 69.6%, and KURAMUCHI 55.1%. <h3>Conclusions</h3> We generated Vδ1+ iPSC derived γδ T cells with effector cytokine phenotype and low risk for cytokine release syndrome. Robust cytotoxic activity was seen across a variety of cancer cell lines, potentially providing an off-the-shelf platform for allogeneic cell therapy.","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.0069
Laetitia Chatelain, Anji Bei, Nancy George, Ge-Ah Kim, Sonali Deshpande, Steve Zhou, Li-Chong Wang, Maithreyan Srinivasan
Background
Spatial biology methods are increasingly used for the characterization of complex tissue microenvironments, the understanding of which can shed light on fundamental biological mechanisms and better inform development of targeted therapeutics. RNAscope™ in situ hybridization (ISH) technology, capable of highly sensitive single-molecule RNA detection, can be combined with immunohistochemistry (IHC) or immunofluorescence (IF) for the co-detection of clinically relevant biomarkers on the same slide with morphological context. This application is especially important in immuno-oncology research to profile immune cell populations using protein markers and characterize their activation states by detecting cytokine and chemokine expression with RNA. However, RNAscope necessitates the use of proteases to digest RNA-associated proteins and facilitate probe access to RNA targets, which can negatively impact epitopes targeted by some antibodies. Previously, we developed the Integrated Co-detection Workflow (ICW) to partially solve this problem, with the fixation of the primary antibody-target complex prior to the protease application. While ICW rescues signal for many previously incompatible antibodies, proteases can still adversely impact some of the primary antibody-target complex. To address this deficiency, we have developed a novel RNA-protein co-detection workflow that eliminates the need for protease, resulting in high detection sensitivities for both protein and RNA markers.
Methods
To maintain the same RNA detection sensitivity without the use of proteases, we formulated a new protease-free pretreatment buffer to replace the existing protease step within the current workflow which allows adequate accessibility of RNAscope™ probes to the target RNAs. Following this protease-free pretreatment, tissue specimens were assayed with RNAscope Multiplex to detect RNA species in the tissue, followed by standard IF staining to co-detect protein biomarkers. Antibodies which previously exhibited degraded protein signal in both sequential ISH-IF and ICW were tested in the new protocol.
Results
Here, we present results from FFPE human tissues to co-detect several protease-sensitive antibodies, including degranulating cytotoxic lymphocyte marker CD107a along with human house-keeping genes TBP, POLR2A and PPIB. The protease-free RNAscope co-detection workflow restored the protein staining pattern at nominal antibody concentrations used for IHC while maintaining mRNA dot counts for TBP, POLR2A and PPIB, indicating minimal impact of protease-free pretreatment buffer on both RNA and protein signal.
Conclusions
The protease-free RNAscope co-detection workflow will serve as a powerful multi-omics staining technique for a wider range of antibodies by enabling visualization of RNA-protein co-detection for the comprehensive profiling of tissue microenvironments, facilitating fast
{"title":"69 A novel protease-free method for the co-detection of RNA and protein biomarkers using the RNAscope™ technology","authors":"Laetitia Chatelain, Anji Bei, Nancy George, Ge-Ah Kim, Sonali Deshpande, Steve Zhou, Li-Chong Wang, Maithreyan Srinivasan","doi":"10.1136/jitc-2023-sitc2023.0069","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.0069","url":null,"abstract":"<h3>Background</h3> Spatial biology methods are increasingly used for the characterization of complex tissue microenvironments, the understanding of which can shed light on fundamental biological mechanisms and better inform development of targeted therapeutics. RNAscope™ <i>in situ</i> hybridization (ISH) technology, capable of highly sensitive single-molecule RNA detection, can be combined with immunohistochemistry (IHC) or immunofluorescence (IF) for the co-detection of clinically relevant biomarkers on the same slide with morphological context. This application is especially important in immuno-oncology research to profile immune cell populations using protein markers and characterize their activation states by detecting cytokine and chemokine expression with RNA. However, RNAscope necessitates the use of proteases to digest RNA-associated proteins and facilitate probe access to RNA targets, which can negatively impact epitopes targeted by some antibodies. Previously, we developed the Integrated Co-detection Workflow (ICW) to partially solve this problem, with the fixation of the primary antibody-target complex prior to the protease application. While ICW rescues signal for many previously incompatible antibodies, proteases can still adversely impact some of the primary antibody-target complex. To address this deficiency, we have developed a novel RNA-protein co-detection workflow that eliminates the need for protease, resulting in high detection sensitivities for both protein and RNA markers. <h3>Methods</h3> To maintain the same RNA detection sensitivity without the use of proteases, we formulated a new protease-free pretreatment buffer to replace the existing protease step within the current workflow which allows adequate accessibility of RNAscope™ probes to the target RNAs. Following this protease-free pretreatment, tissue specimens were assayed with RNAscope Multiplex to detect RNA species in the tissue, followed by standard IF staining to co-detect protein biomarkers. Antibodies which previously exhibited degraded protein signal in both sequential ISH-IF and ICW were tested in the new protocol. <h3>Results</h3> Here, we present results from FFPE human tissues to co-detect several protease-sensitive antibodies, including degranulating cytotoxic lymphocyte marker CD107a along with human house-keeping genes <i>TBP</i>, <i>POLR2A</i> and <i>PPIB</i>. The protease-free RNAscope co-detection workflow restored the protein staining pattern at nominal antibody concentrations used for IHC while maintaining mRNA dot counts for <i>TBP</i>, <i>POLR2A</i> and <i>PPIB</i>, indicating minimal impact of protease-free pretreatment buffer on both RNA and protein signal. <h3>Conclusions</h3> The protease-free RNAscope co-detection workflow will serve as a powerful multi-omics staining technique for a wider range of antibodies by enabling visualization of RNA-protein co-detection for the comprehensive profiling of tissue microenvironments, facilitating fast","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.0047
Ze Zhang, Kartik Sehgal, Keisuke Shirai, Rondi Butler, John Wiencke, Devin Koestler, Geat Ramush, Min Kyung Lee, Annette Molinaro, Hannah Stolrow, Lucas A Salas, Robert Haddad, Karl Kelsey, Brock Christensen
Background
Immune checkpoint inhibitors (ICIs) are approved to treat patients with recurrent/metastatic head and neck squamous cell carcinoma (HNSCC). However, currently approved biomarkers are limited due to the heterogeneity and availability of tumor samples. The development of peripheral biomarkers offers an alternative noninvasive approach to assess immunotherapy response. DNA methylation-based immune cell deconvolution provides opportunities for developing blood-based biomarkers to predict immunotherapy response outcomes in HNSCC.
Methods
Our study is an ongoing prospective multi-center study aimed at identifying blood DNA methylation biomarkers of therapy response in patients with HNSCC undergoing standard-of-care, FDA-approved ICIs. Blood was drawn prior to immunotherapy initiation. DNA isolated from these samples underwent methylation profiling using the Illumina EPIC microarray. Peripheral blood immune profiles were generated using cellular deconvolution.1 69 HNSCC patients with anti-PD-1 monotherapy were included (figure 1). 47 patients’ tumor samples were sequenced to evaluate tumor mutational burden (TMB) (figure 2). We investigated 48 immune variables and TMB for relation with progression-free survival (PFS) and overall survival (OS) using Cox proportional-hazard models adjusted for age, sex, and a marker for corticosteroid exposure.2 12 primary immune cell proportions and TMB were investigated for interaction. A p-value < 0.05 was used as the cut-off for statistical significance.
Results
In 69 patients with HNSCC who received anti-PD-1 therapy, neutrophil proportion, monocyte count, and total B cell count were associated with worse PFS outcomes while CD4T memory cell count and total T cell count were associated with better PFS outcomes (figure 3). Regarding OS, NLR, neutrophil proportion, total naïve lymphocyte proportion, monocyte count, CD4T naïve percentage, CD4T naïve to memory ratio, and T regulatory cell percentage were found to be associated with poorer survival while total CD4 T cell count, total CD4 T cell proportion, CD4 T memory cell proportion, total T cell count, lymphocyte to monocyte ratio, and total lymphocyte proportion were found to be associated with better outcomes (figure 4). Higher TMB was found to be associated with better survival outcomes. TMB exhibits significant interaction with peripheral monocyte proportion. TMB is a better predictor of survival in individuals with a lower level of monocyte proportion (figure 5).
Conclusions
DNA methylation-based immune profiling in peripheral blood at baseline identifies clinically relevant biomarkers of benefit from ICIs. Our results demonstrate the potential of new blood DNA methylation-based biomarkers to predict immunotherapy response prior to the initial treatment, and connected peripheral immune profile with TMB through their interactive impact on survival outcomes.
{"title":"47 Baseline blood DNA methylation-based immune profiles and tumor mutational burden predict survival outcomes in anti-PD-1 treated head and neck cancer patient","authors":"Ze Zhang, Kartik Sehgal, Keisuke Shirai, Rondi Butler, John Wiencke, Devin Koestler, Geat Ramush, Min Kyung Lee, Annette Molinaro, Hannah Stolrow, Lucas A Salas, Robert Haddad, Karl Kelsey, Brock Christensen","doi":"10.1136/jitc-2023-sitc2023.0047","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.0047","url":null,"abstract":"<h3>Background</h3> Immune checkpoint inhibitors (ICIs) are approved to treat patients with recurrent/metastatic head and neck squamous cell carcinoma (HNSCC). However, currently approved biomarkers are limited due to the heterogeneity and availability of tumor samples. The development of peripheral biomarkers offers an alternative noninvasive approach to assess immunotherapy response. DNA methylation-based immune cell deconvolution provides opportunities for developing blood-based biomarkers to predict immunotherapy response outcomes in HNSCC. <h3>Methods</h3> Our study is an ongoing prospective multi-center study aimed at identifying blood DNA methylation biomarkers of therapy response in patients with HNSCC undergoing standard-of-care, FDA-approved ICIs. Blood was drawn prior to immunotherapy initiation. DNA isolated from these samples underwent methylation profiling using the Illumina EPIC microarray. Peripheral blood immune profiles were generated using cellular deconvolution.<sup>1</sup> 69 HNSCC patients with anti-PD-1 monotherapy were included (figure 1). 47 patients’ tumor samples were sequenced to evaluate tumor mutational burden (TMB) (figure 2). We investigated 48 immune variables and TMB for relation with progression-free survival (PFS) and overall survival (OS) using Cox proportional-hazard models adjusted for age, sex, and a marker for corticosteroid exposure.<sup>2</sup> 12 primary immune cell proportions and TMB were investigated for interaction. A p-value < 0.05 was used as the cut-off for statistical significance. <h3>Results</h3> In 69 patients with HNSCC who received anti-PD-1 therapy, neutrophil proportion, monocyte count, and total B cell count were associated with worse PFS outcomes while CD4T memory cell count and total T cell count were associated with better PFS outcomes (figure 3). Regarding OS, NLR, neutrophil proportion, total naïve lymphocyte proportion, monocyte count, CD4T naïve percentage, CD4T naïve to memory ratio, and T regulatory cell percentage were found to be associated with poorer survival while total CD4 T cell count, total CD4 T cell proportion, CD4 T memory cell proportion, total T cell count, lymphocyte to monocyte ratio, and total lymphocyte proportion were found to be associated with better outcomes (figure 4). Higher TMB was found to be associated with better survival outcomes. TMB exhibits significant interaction with peripheral monocyte proportion. TMB is a better predictor of survival in individuals with a lower level of monocyte proportion (figure 5). <h3>Conclusions</h3> DNA methylation-based immune profiling in peripheral blood at baseline identifies clinically relevant biomarkers of benefit from ICIs. Our results demonstrate the potential of new blood DNA methylation-based biomarkers to predict immunotherapy response prior to the initial treatment, and connected peripheral immune profile with TMB through their interactive impact on survival outcomes. <h3>References</h3> Salas LA, Zhang Z, Ko","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.1466
Camila A Bach, Ada Blidner, Ramiro Perrotta, Joaquin P Merlo, Mariana Salatino, Gabriel Rabinovich
Background
Extracellular vesicles (EVs) suppress effector cells and activate immunosuppressive cells in the ‘escape phase’ of tumor immunoediting. Tumor derived EVs contribute to the differentiation and expansion of immunosuppressive cell subsets. Bone marrow-derived cells (BMDCs) can uptake EVs released by 4T1 -a triple-negative breast cancer cell line- and induce myeloid-derived suppressor cells (MDSCs).1 Immunosuppressive myeloid cells expansion can also be promoted by their own EV in an autocrine manner. Moreover, EVs secreted by immunosuppressive myeloid cells, may inherit their parental functions.2 Nevertheless, the molecular circuits and mediators of tumor microenvironment EV release remain uncertain.
Methods
We generated immunosuppressive myeloid cells by culturing mouse BMDCs with GM-CSF for three days in the presence of Gal1. Additionally, we cultured 4T1 WT and Lgals1 knocked out cells. We purified Small EVs by size exclusion chromatography from conditioned medium of immunosuppressive myeloid cells (control and Gal1-treated) or 4T1 cells (WT or Gal1 KO). We further purified 4T1 EVs with CD63+ beads, a tumoral EV marker. ELISA, flow cytometry, miRNA sequencing, proteomic and metabolomic analyses verified EV identity. We performed functional assays by co-culturing tumoral or myeloid EVs with BMDC, activated T or B cells.
Results
When exposed to EVs from 4T1 cells in vitro, BMDCs switched their differentiation pathway to a M-MDSCs phenotype (CD11b+Ly6G-Ly6Chi) and showed higher immune checkpoint molecules expression, including PD-L1 (p=0.005) and VISTA (p=0.003). However, when exposed to tumoral EVs lacking Gal1, BMDCs reinforced a PMN-MDSCs phenotype (CD11b+Ly6G+Ly6Clo). In contrast with Gal1+ EVs, tumoral EVs lacking Gal1 failed to inhibit CD4+ and CD8+ T and B cell proliferation and activation. Gal1 is both in the cargo and corona of 4T1 EVs as revealed by flow cytometry and ELISA (p=0.0032). Moreover, EV production inhibition with GW4869 in 4T1 cells decreased Gal1 levels secreted (p=0.037). To analyze EV-derived Gal1 contribution as an MDSC autocrine signal, we added EVs from Gal1-treated immunosuppressive myeloid cells to new BMDCs and observed CD11b+ cells expansion and higher VISTA expression on cells with M-MDSCs phenotype. Moreover, these EVs showed greater T cell-suppressive capacity and has different protein, metabolic, and miRNA cargo compared to control EVs.
Conclusions
Here, we propose that Gal1+ EVs released by tumoral cells interact with myeloid cells and potentiate their immunosuppressive properties, including the release of EV with autocrine and paracrine functions. Targeting Gal1 as a molecular mechanism of EV-mediated tumor development may help overcome therapy resistance.
{"title":"1466 Galectin-1 orchestrates a hierarchical tumor to stromal extracellular vesicles system that fosters immune suppression","authors":"Camila A Bach, Ada Blidner, Ramiro Perrotta, Joaquin P Merlo, Mariana Salatino, Gabriel Rabinovich","doi":"10.1136/jitc-2023-sitc2023.1466","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.1466","url":null,"abstract":"<h3>Background</h3> Extracellular vesicles (EVs) suppress effector cells and activate immunosuppressive cells in the ‘escape phase’ of tumor immunoediting. Tumor derived EVs contribute to the differentiation and expansion of immunosuppressive cell subsets. Bone marrow-derived cells (BMDCs) can uptake EVs released by 4T1 -a triple-negative breast cancer cell line- and induce myeloid-derived suppressor cells (MDSCs).<sup>1</sup> Immunosuppressive myeloid cells expansion can also be promoted by their own EV in an autocrine manner. Moreover, EVs secreted by immunosuppressive myeloid cells, may inherit their parental functions.<sup>2</sup> Nevertheless, the molecular circuits and mediators of tumor microenvironment EV release remain uncertain. <h3>Methods</h3> We generated immunosuppressive myeloid cells by culturing mouse BMDCs with GM-CSF for three days in the presence of Gal1. Additionally, we cultured 4T1 WT and <i>Lgals1</i> knocked out cells. We purified Small EVs by size exclusion chromatography from conditioned medium of immunosuppressive myeloid cells (control and Gal1-treated) or 4T1 cells (WT or Gal1 KO). We further purified 4T1 EVs with CD63+ beads, a tumoral EV marker. ELISA, flow cytometry, miRNA sequencing, proteomic and metabolomic analyses verified EV identity. We performed functional assays by co-culturing tumoral or myeloid EVs with BMDC, activated T or B cells. <h3>Results</h3> When exposed to EVs from 4T1 cells <i>in vitro</i>, BMDCs switched their differentiation pathway to a M-MDSCs phenotype (CD11b<sup>+</sup>Ly6G<sup>-</sup>Ly6C<sup>hi</sup>) and showed higher immune checkpoint molecules expression, including PD-L1 (p=0.005) and VISTA (p=0.003). However, when exposed to tumoral EVs lacking Gal1, BMDCs reinforced a PMN-MDSCs phenotype (CD11b<sup>+</sup>Ly6G<sup>+</sup>Ly6C<sup>lo</sup>). In contrast with Gal1<sup>+</sup> EVs, tumoral EVs lacking Gal1 failed to inhibit CD4<sup>+</sup> and CD8<sup>+</sup> T and B cell proliferation and activation. Gal1 is both in the cargo and corona of 4T1 EVs as revealed by flow cytometry and ELISA (p=0.0032). Moreover, EV production inhibition with GW4869 in 4T1 cells decreased Gal1 levels secreted (p=0.037). To analyze EV-derived Gal1 contribution as an MDSC autocrine signal, we added EVs from Gal1-treated immunosuppressive myeloid cells to new BMDCs and observed CD11b<sup>+</sup> cells expansion and higher VISTA expression on cells with M-MDSCs phenotype. Moreover, these EVs showed greater T cell-suppressive capacity and has different protein, metabolic, and miRNA cargo compared to control EVs. <h3>Conclusions</h3> Here, we propose that Gal1<sup>+</sup> EVs released by tumoral cells interact with myeloid cells and potentiate their immunosuppressive properties, including the release of EV with autocrine and paracrine functions. Targeting Gal1 as a molecular mechanism of EV-mediated tumor development may help overcome therapy resistance. <h3>References</h3> Xiang X, Poliakov A, Liu C, Liu Y, D","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1136/jitc-2023-sitc2023.1397
Melissa Vrohlings, Stephanie Jungmichel, Ivana Tosevski, Alessio Vantellini, Philip Knobel, Nadia Sanchez, Elizabeth Ross, Marian Van Kerckhoven, Giorgia Giacomazzi, Maria Liivrand, Reija Hieta, Nicholas Dupuis, Dieter Rondas, Pamela Swatkowski, Daniel Lenherr-Frey, Swethajit Biswas, Gilberto Lopes, Leonardo Borras
Background
Squamous non-small cell lung cancer (SQ-NSCLC) is the 2nd most common type of lung cancer. Given the paucity of actionable oncogene drivers, and lack of efficacy from multiple therapies in the Lung-MAP trial, there is a high unmet need in SQ-NSCLC to develop effective 2nd-line immunotherapies for patients with disease progression after immune checkpoint inhibitors (ICI). The melanoma antigen gene A4 (MAGE-A4) is exclusively expressed in cancer and absent in somatic tissues. MAGE-A4-derived peptides presented on HLA molecules at the cell surface recently emerged as a novel therapeutic opportunity. Thus, the two key objectives of this study were to: 1). Evaluate MAGE-A4 expression in human SQ-NSCLC; 2). Demonstrate the anti-cancer activity of CDR404, an antibody-based bispecific and bivalent T-cell engager targeted against MAGE-A4230–239 peptide in vitro and in vivo xenograft models of SQ-NSCLC.
Methods
MAGE-A4 mRNA prevalence and expression in SQ-NSCLC was analyzed using the TCGA database (https://www.cancer.gov/tcga). Protein expression of MAGE-A4 was confirmed using immunohistochemistry (IHC) in fifty FFPE human SQ-NSCLC samples (clone E7O1U). CDR404 target cell killing in the presence of human PBMCs was assessed using the human SQ-NSCLC cell line NCI-H1703. HLA-A*02:01+MAGE-A4neg cancer cells were used as controls. To exclude reactivity of CDR404 in healthy tissues, HLA-A*02:01+ primary cells presenting peptides with high MAGE-A4 similarity were co-cultured with human PBMCs. In vivo activity of CDR404 in SQ-NSCLC was evaluated with an NCI-H1703 xenograft model in NSG mice.
Results
SQ-NSCLC had the highest MAGE-A4 mRNA expression levels among solid cancers in the TCGA database. IHC showed positive MAGE-A4 staining in 28/50 (56%) of SQ-NSCLC samples. In vitro, CDR404 showed efficient target cell lysis across all effector-to-target ratios tested. Similarly, simultaneous target engagement and resulting synapse formation induced T cell activation and secretion of cytolytic molecules in an effector-to-target ratio-dependent fashion. No reactivity was observed using co-cultured HLA-A*02:01+MAGE-A4neg cancer cells. Lack of T cell activation/cytolytic molecule release in the presence of HLA-A*02:01+ primary cells confirmed the specificity profile of CDR404. In vivo, treatment with four different doses of CDR404 induced complete tumor regression in the SQ-NSCLC NCI-H1703 xenograft model.
Conclusions
The high MAGE-A4 expression levels and the highly specific anti-cancer cell activity of CDR404 make it a highly attractive immunotherapy for development post-progression on ICI for patients with HLA-A*02:01+ SQ-NSCLC. A multi-tumor phase 1 trial of CDR404, including SQ-NSCLC, is expected to begin in 2024 with prospective patient selection for both HLA-A*02:01 and tumor MAGE-A4.
{"title":"1397 CDR404, an antibody-based bispecific & bivalent T-cell engager targeted against MAGE-A4, for Squamous Non-Small Cell Lung Cancer (SQ-NSCLC)","authors":"Melissa Vrohlings, Stephanie Jungmichel, Ivana Tosevski, Alessio Vantellini, Philip Knobel, Nadia Sanchez, Elizabeth Ross, Marian Van Kerckhoven, Giorgia Giacomazzi, Maria Liivrand, Reija Hieta, Nicholas Dupuis, Dieter Rondas, Pamela Swatkowski, Daniel Lenherr-Frey, Swethajit Biswas, Gilberto Lopes, Leonardo Borras","doi":"10.1136/jitc-2023-sitc2023.1397","DOIUrl":"https://doi.org/10.1136/jitc-2023-sitc2023.1397","url":null,"abstract":"<h3>Background</h3> Squamous non-small cell lung cancer (SQ-NSCLC) is the 2<sup>nd</sup> most common type of lung cancer. Given the paucity of actionable oncogene drivers, and lack of efficacy from multiple therapies in the Lung-MAP trial, there is a high unmet need in SQ-NSCLC to develop effective 2<sup>nd</sup>-line immunotherapies for patients with disease progression after immune checkpoint inhibitors (ICI). The melanoma antigen gene A4 (MAGE-A4) is exclusively expressed in cancer and absent in somatic tissues. MAGE-A4-derived peptides presented on HLA molecules at the cell surface recently emerged as a novel therapeutic opportunity. Thus, the two key objectives of this study were to: 1). Evaluate MAGE-A4 expression in human SQ-NSCLC; 2). Demonstrate the anti-cancer activity of CDR404, an antibody-based bispecific and bivalent T-cell engager targeted against MAGE-A4<sub>230–239</sub> peptide <i>in vitro</i> and <i>in vivo</i> xenograft models of SQ-NSCLC. <h3>Methods</h3> MAGE-A4 mRNA prevalence and expression in SQ-NSCLC was analyzed using the TCGA database (https://www.cancer.gov/tcga). Protein expression of MAGE-A4 was confirmed using immunohistochemistry (IHC) in fifty FFPE human SQ-NSCLC samples (clone E7O1U). CDR404 target cell killing in the presence of human PBMCs was assessed using the human SQ-NSCLC cell line NCI-H1703. HLA-A*02:01<sup>+</sup>MAGE-A4<sup>neg</sup> cancer cells were used as controls. To exclude reactivity of CDR404 in healthy tissues, HLA-A*02:01<sup>+</sup> primary cells presenting peptides with high MAGE-A4 similarity were co-cultured with human PBMCs. <i>In vivo</i> activity of CDR404 in SQ-NSCLC was evaluated with an NCI-H1703 xenograft model in NSG mice. <h3>Results</h3> SQ-NSCLC had the highest MAGE-A4 mRNA expression levels among solid cancers in the TCGA database. IHC showed positive MAGE-A4 staining in 28/50 (56%) of SQ-NSCLC samples. <i>In vitro</i>, CDR404 showed efficient target cell lysis across all effector-to-target ratios tested. Similarly, simultaneous target engagement and resulting synapse formation induced T cell activation and secretion of cytolytic molecules in an effector-to-target ratio-dependent fashion. No reactivity was observed using co-cultured HLA-A*02:01<sup>+</sup>MAGE-A4<sup>neg</sup> cancer cells. Lack of T cell activation/cytolytic molecule release in the presence of HLA-A*02:01<sup>+</sup> primary cells confirmed the specificity profile of CDR404. <i>In vivo</i>, treatment with four different doses of CDR404 induced complete tumor regression in the SQ-NSCLC NCI-H1703 xenograft model. <h3>Conclusions</h3> The high MAGE-A4 expression levels and the highly specific anti-cancer cell activity of CDR404 make it a highly attractive immunotherapy for development post-progression on ICI for patients with HLA-A*02:01+ SQ-NSCLC. A multi-tumor phase 1 trial of CDR404, including SQ-NSCLC, is expected to begin in 2024 with prospective patient selection for both HLA-A*02:01 and tumor MAGE-A4. <h3","PeriodicalId":500964,"journal":{"name":"Regular and Young Investigator Award Abstracts","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135161403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：