Subtype Classification Research Articles

Abstract 4433: Secondary electrospray ionization mass spectrometry for volatile analysis of lung cancer

Abstract Lung cancer remains a leading cause of cancer-related mortality globally, necessitating innovative approaches for early detection and improved understanding of its metabolic landscape and consequences. Secondary Electrospray Ionization Mass Spectrometry (SESI-MS), an innovative analytical technique, has emerged as a powerful tool for the analysis of volatile organic compounds (VOCs) associated with various diseases, from metabolic disorders to bacterial infections. In this proof-of-concept study, we first utilized SESI-MS analyses of four representative lung cancer cells of both non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) to demonstrate its capability to profile VOCs released by lung cancer cells. The technique allows for real-time, non-invasive analysis of the headspace above cultured cells, enabling the detection of unique VOC signatures associated with different lung cancer subtypes. Meanwhile, SESI-MS also facilitated the discrimination specific cells based on distinct VOC patterns. Furthermore, we also performed cisplatin treatment experiments and observed the metabolic signature shift post drug treatment. These results collectively offered potential applications in early diagnosis and personalized treatment strategies. Next, we continued with in vivo investigations to further extend the utility of SESI-MS in lung cancer research. Animal models, mimicking human lung cancer pathophysiology, have been subjected to SESI-MS analysis to identify volatile biomarkers indicative of tumor development and progression. Several representative cancer drug treatments were performed to the lung cancer mice model, and the drug-induced responses were also monitored via SESI-MS breath analyses. The in vivo SESI-MS approach detected many interesting breath VOC features that could provide valuable insights into the dynamic changes in VOC profiles during various stages of lung cancer, and therefore has the potential to offer a deeper understanding of the disease's metabolic alterations and potential avenues for targeted therapeutic interventions. In conclusion, SESI-MS-based volatile analysis presents a robust and versatile approach for investigating lung cancer, and our ongoing work has the potential to bridge the gap between in vitro and in vivo models for better translational research outcomes. The detection of unique VOC signatures associated with lung cancer cells and tumors holds significant promise for early detection, subtype classification, and monitoring of treatment responses. As technology continues to advance, SESI-MS stands poised to play a pivotal role in transforming the landscape of lung cancer diagnostics and research. Citation Format: Jiangjiang (Chris) Zhu, Fouad Choueiry. Secondary electrospray ionization mass spectrometry for volatile analysis of lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4433.

Abstract 7406: Visualization and analysis of cancer genomics data using UCSC Xena

Abstract UCSC Xena (http://xena.ucsc.edu/) is a web-based visual integration and exploration tool for multi-omic data and associated clinical and phenotypic annotations. Researchers can easily view and explore public data, their own private data, or both using the Xena Browser. Private data are kept on the researcher's computer and are never uploaded to our public servers. We support Mac, Windows, and Linux. Questions Xena can help you answer:* Is overexpression of this gene associated with lower/higher survival?* What genes are differentially expressed between these two groups of samples?* What is the relationship between mutation, copy number, expression, etc for this gene? Xena showcases seminal cancer genomics datasets from TCGA, the Pan-Cancer Atlas, GDC, PCAWG, ICGC, and more; a total of more than 1500 datasets across 50 cancer types. We support virtually any type of functional genomics data: SNPs, INDELs, copy number variation, gene expression, ATAC-seq, DNA methylation, exon-, transcript-, miRNA-, lncRNA-expression and structural variants. We also support clinical data such as phenotype information, subtype classifications and biomarkers. All of our data is available for download via python or R APIs, or using our URL links. Our signature Visual Spreadsheet view shows multiple data types side-by-side enabling discovery of correlations across and within genes and genomic regions. We also have dynamic Kaplan-Meier survival analysis, powerful filtering and subgrouping, differential gene expression analysis, GSEA, charts, statistical analyses, genomic signatures, and the ability to generate URLs to live views. New features include: * Genome-wide differential gene expression analysis * GSEA analysis * Select samples directly from the screen for filtering and creating subgroups * Loading of Microsoft Excel files Our beta prototype site for visualizing single-cell data delivers million-cell-scale multi-omics data for interactive visualization in a web browser. Contact us for access to our beta prototype site. If you use us please cite our publication in Nature Biotechnology: https://www.nature.com/articles/s41587-020-0546-8 Citation Format: Mary Goldman, Brian Craft, Jingchun Zhu, David Haussler. Visualization and analysis of cancer genomics data using UCSC Xena [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7406.

Abstract 4905: Multimodal transformer model improves survival prediction in lung cancer compared to unimodal approaches

Abstract Integrating multimodal lung data including clinical notes, medical images, and molecular data is critical for predictive modeling tasks like survival prediction, yet effectively aligning these disparate data types remains challenging. We present a novel method to integrate heterogeneous lung modalities by first thoroughly analyzing various domain-specific models and selecting the optimal model for embedding feature extraction per data type based on performance on representative pretrained tasks. For clinical notes, the GatorTron models showed the lowest regression loss on an initial evaluation set, with the large GatorTron-medium model achieving 12.9 loss. After selecting the top performers, we extracted robust embeddings on the full lung dataset built using the Multimodal Integration of Oncology Data System (MINDS) framework. MINDS provides an end-to-end platform for aggregating and normalizing multimodal patient data. We aligned the multimodal embeddings to a central pre-trained language model using contrastive representation learning based on a cosine similarity loss function. To adapt the language model to the new modalities, we employed a parameter-efficient tuning method called adapter tuning, which introduces small trainable adapter layers that leave the base model weights frozen. This avoids catastrophic forgetting of the pretrained weights. We evaluated our multimodal model on prognostic prediction tasks including survival regression and subtype classification using both public and internal lung cancer datasets spanning multiple histologic subtypes and stages. Our aligned multimodal model demonstrated improved performance over models utilizing only single modalities, highlighting the benefits of integrating complementary information across diverse lung data types. This work illustrates the potential of flexible multimodal modeling for critical lung cancer prediction problems using heterogeneous real-world patient data. Our model provides a strong foundation for incorporating emerging data types, modalities, and predictive tasks in the future. Citation Format: Aakash Tripathi, Asim Waqas, Yasin Yilmaz, Ghulam Rasool. Multimodal transformer model improves survival prediction in lung cancer compared to unimodal approaches [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4905.

Abstract 2079: Tumor-derived fibrinogen promotes pancreatic ductal adenocarcinoma progression

Abstract Background: Hypercoagulability is a common clinical manifestation in patients with cancers, especially in pancreatic ductal adenocarcinoma (PDAC). Over 20% of PDAC patients with advanced-stage and worse prognosis undergo a venous thromboembolism (VTE) incidence. Fibrinogen (FG) may contribute to VTE due to its high expression in advanced PDAC tumors. Here, we set to determine the biological roles of tumor-derived FG in PDAC. In addition, our previous study showed that Von Willebrand Factor (VWF) A2 domain can effectively bind to FG and reduce platelet clot formation. Therefore, the functional roles of A2 on FG in PDAC were also studied. Methods: The correlation between survival rate and FG expression in PDAC was analyzed using RNAseq profiles of 149 PDAC patient samples from the TCGA database. PDAC subtyping of Baylor College of Medicine (BCM) PDAC patient-derived xenograft (PDX) cohort was performed using RNAseq data according to Moffitt’s criteria. The FG (three chains FGA, FGB, FGG) overexpression (OE) and knock-out (KO) PDAC cell lines were established. The gene expression and biological functions of FG were performed in PDX and cell lines using IHC and immunoblot assays. The FG effect on PDAC progression in vivo was studied in mouse models. Results: TCGA analysis showed that elevated tumor-derived FG gene expression is significantly correlated with worse overall survival and disease-free survival rates in PDAC patients. PDX lines RNAseq analysis clearly separated BCM PDAC PDXs into two subtypes: basal-like and classical PDACs. Basal-like PDXs showed significantly stronger FG staining than that in classical ones. Compared to the classical subtype, the basal-like subtype showed significantly upregulated genes that are associated with the epithelial-mesenchymal transition (EMT), IL6/JAK/STAT3, and TNFα signaling pathways. Accordingly, basal-like subtype PDAC exposed to IL6 or TNFα significantly increased FG and FG receptor expression with activated STAT3 and AKT pathways and elevated EMT markers, whereas classical subtype cells have minimal changes. In addition, we found that the FG-OE cells exhibit enhanced cell proliferation and migration properties whereas FG-KO cells reduced the effects. Consistent with in vitro data, FG-OE PDAC showed significantly higher tumor burdens than that of control PDAC in both orthotopic and subcutaneous PDAC mouse models. Furthermore, A2 protein treatment can effectively block the FG function and dramatically inactivate the FG-induced EMT process and AKT signaling pathway. Conclusions: Tumor-derived fibrinogen is upregulated in a basal-like subtype of PDAC cells and may contribute to tumor growth and metastasis by activating aberrant signaling pathways and promoting EMT. Our findings imply that elevated fibrinogen in PDAC could contribute to tumor progression and the A2 protein which target to fibrinogen may be a promising therapy for basal-like subtype PDAC. Citation Format: Dongliang Liu, Lisa Brubaker, Yichi Niu, Emily LaPlante, George Van Buren, Aleksandar Milosavljevic, Changyi Chen, Chenghang Zong, Miguel Cruz, Qizhi Yao. Tumor-derived fibrinogen promotes pancreatic ductal adenocarcinoma progression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2079.

Abstract 3939: Ror2, a novel key regulator driving cell fate decisions throughout pancreatic tumor progression

Abstract Introduction: Reprogramming of pancreas cell fate drives development of pancreatic ductal adenocarcinoma (PDAC). Acinar cells, the most probable origin of pancreatic cancer, undergo a rapid cell identity switch towards a duct-like phenotype upon KrasG12D expression and when combined with pancreatitis or the loss of acinar differentiation factors. Metaplastic and dysplastic duct-like cells are heterogeneous with a proportion acquiring features reminiscent of gastric lineages. While some gastric signatures are maintained in the classical PDAC subtype, they are eroded in the more aggressive, basal-like PDAC. Since subtype identity has a major impact on prognosis and therapeutic targetability, druggable targets that regulate cellular reprogramming in pancreatic cancer can be exploited to increase sensitivity to therapy. Methods: To elucidate mechanisms responsible for early reprogramming, pancreatic tissue of mice with conditional KrasG12D expression and loss of Pdx1 was analyzed by single-nucleus ATAC-seq. Expression of identified target genes was studied in precancerous lesions and PDAC by using multiplex RNAscope and IHC staining. Computational analyses of publicly available sequencing data were used to establish correlation to PDAC subtype identity. Genes were functionally studied in PDAC cell lines. Results: By performing snATAC-seq and RNA-seq of early transformed pancreatic tissue, we discovered that acinar cells with the combined expression of KrasG12D and loss of Pdx1 activate expression of a gastric metaplastic gene signature accompanied by elevated levels of the receptor kinase Ror2. Ror2 is also highly expressed in distinct subpopulations of metaplastic and dysplastic cells, associated with a gastric neck cell phenotype and enhanced proliferative capacity. In contrast, Ror2Low lesions are characterized by a gastric pit cell-like and a senescent phenotype. Genetic ablation of Ror2 resulted in a shift in lesion identity, enriching those with a pit cell and senescent identity. In PDAC, we found that Ror2 anti-correlates with a similar gastric pit cell phenotype that is maintained in the classical subtype PDAC but is strongly associated with the more aggressive basal-like subtype. Overexpression of ROR2 in human PDAC cell lines with a classical differentiation induced loss of the classical gene signature as well as epithelial-to-mesenchymal transition. Moreover, ROR2 enforces a strong dependency on AKT signaling, causing increased vulnerability of ROR2-expressing cells to AKT inhibition, but increased resistance to the KRAS inhibitor MRTX1133. Conclusions: We discovered Ror2 as a critical determinant of precancerous lesion as well as PDAC subtype identity. Its role in driving an aggressive PDAC phenotype that is inherently resistant to Kras inhibition suggests that inhibiting this receptor tyrosine kinase will enhance sensitivity to the new generation of targeted therapies. Citation Format: Simone Benitz, Alexander Steep, Malak Nasser, Jonathan Preall, Ujjwal Mahajan, Ian Loveless, Holly McQuithey, Erick Davis, Hui-Ju Wen, Daniel Long, Thomas Metzler, Samuel Zwernik, Daniel Salas-Escabillas, Ling Huang, Nina Steele, Ivonne Regel, Filip Bednar, Howard Crawford. Ror2, a novel key regulator driving cell fate decisions throughout pancreatic tumor progression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3939.

Abstract 7644: Prognostication of genomic characteristics of residual breast cancer after neoadjuvant chemotherapy

Abstract Residual cancer burden (RCB) based on pathologic characteristics is the most powerful prognostic factor in breast cancer (BC) which received neoadjuvant chemotherapy (NAC) followed by surgery. According to RCB classification, class III predicted short survival duration regardless of BC subtypes according to hormone receptor (HR) and human epidermal growth factor receptor-2 (HER-2) state. However, up to 20-40% of BCs with RCB class III did not experience BC recurrence and had long term survival without disease recurrence. In our study, we evaluated genomic characteristics of BC which had high RCB class. We collected 96 fresh frozen BC specimens which had undergone surgery after NAC between 2010 and 2018 from Samsung Medical Center Biobank. We performed whole genome or exome sequencing, RNA sequencing and high-plex single-cell spatial transcriptomics. Of 96 BCs, 65 BCs had experienced BC recurrence and 31 BCs have not been BC recurrence. Of 65 BCs with dismal prognosis, 12 were HR+HER2-, four in HR+HER2+, 39 in triple negative breast cancer (TNBC) and 10 in HR-HER2+ BC. Of 31 BCs without recurrence, 11 were HR+HER2-, two in HR+HER2+, 14 in TNBC and four in HR-HER2+ BC. In HR+HER2- BC, ten of 12 BCs with dismal prognosis were categorized into basal like intrinsic subtype and two were into luminal B subtype compared that six of luminal A and five of luminal B subtype categorization in good prognosis BC group. Age associated mutation signature were enriched in HR+HER2- BC with good prognosis (p=0.027). Tumor microenvironment (TME) analysis using Kassandra suggested that high monocytes were associated with BC with dismal prognosis (p=0.00031). In single-cell spatial transcriptome analysis, T-cell abundance in the tumoral niche is associated with dismal prognosis in HR+HER2- BC (p=0.039). In TNBC, almost BCs (88.7%) were categorized into basal-like intrinsic subtype regardless of their prognosis. TME analysis presented that immune cells including macrophages and lymphocytes with CD4 and CD8 T cells were significantly enriched in TNBC with good prognosis (ps <0.05, respectively). In HER2+ BC, intrinsic subtypes were heterogeneous in both groups. Focal amplification in HER2+BC patient harbored DNA segments from different chromosomes more frequently than in those with good prognosis (p=0.00091), which was prominent in oncogenic extrachromosomal DNA amplifications compared to other oncogenic amplifications. In TME analysis, there were no specific cell enrichments according to BC prognosis but sing-cell spatial transcriptome analysis revealed that higher T-cell abundance in the tumoral niche was associated with HR+HER2+ BC with good prognosis (p=0.047). In conclusion, each BC subtype had their own genomic characteristics to determine their prognosis. Our study also suggested that TME characteristics was common prognostic factor for survival outcome of BC with residual tumor after NAC. Citation Format: Ji-Yeon Kim, Sabin Park, Sepil Ahn, Eun Seop Seo, Soyeon Kim, Mark Gregory, Emily Killingbeck, Eun Yoon Cho, Jeong Eon Lee, Jin Seok Ahn, Yeon Hee Park, Woong-Yang Park, Hoon Kim, Semin Lee, Young-Hyuck Im. Prognostication of genomic characteristics of residual breast cancer after neoadjuvant chemotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7644.

Abstract 2339: GPNotebook: A toolkit for identifying cancer-related glycosylation alterations in glycoproteomic analysis

Abstract This study is dedicated to the development and application of 'GPNotebook', an innovative Python-based data analysis tool, specifically designed to enhance glycopeptide research in cancer studies by enabling the detailed investigation of glycosylation patterns and their correlation with various clinical cancer phenotypes. In our study, we recognize the critical role of protein glycosylation in biological processes and its relevance in cancer. To address the lack of specialized tools for glycoproteomic analysis, we developed GPNotebook, a comprehensive Python package designed for detailed investigation of intact glycopeptides (IGPs). This tool integrates multiple functionalities, including statistical profiling, differential expression analysis, categorization of glycosylation subtypes, and exploration of the interactions between glycosylation and phosphorylation. It also encompasses survival analysis and assessments of glycosylation enzymes. We applied GPNotebook in a research project focused on 10 cancer types collected from Clinical Proteomic Tumor Analysis Consortium (CPTAC) projects to validate its efficacy and demonstrate its wide-ranging capabilities. Through this application, we were able to analyze and interpret complex glycoproteomic data, revealing potential IGPs as biomarkers for cancer detection and differentiation of cancer subtypes. We identified hundreds of cancer-specific IGPs, with several showing promise as early-stage cancer biomarkers. In addition, our analysis revealed unique glycosylation patterns linked to specific molecular subtypes and immune cell distributions in tumor microenvironments. We also clarified the roles of certain glycosylation enzymes across different cancer types, highlighting their potential significance in cancer progression. GPNotebook has proven to be a critical tool in glycopeptide research. Our findings not only enhance the understanding of glycosylation in cancer but also demonstrate GPNotebook's potential in identifying novel biomarkers and providing insights into cancer subtype differentiation. This study positions GPNotebook as a valuable resource for exploring the complex relationships between protein glycosylation and cancer phenotypes. Citation Format: Hui Zhang, Trung Hoang, Yingwei Hu. GPNotebook: A toolkit for identifying cancer-related glycosylation alterations in glycoproteomic analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2339.

Abstract 1587: Multimodal spatial transcriptomics uncover distinct tumor microenvironment states and cell-cell communication networks in molecular pancreatic cancer subtypes

Abstract The tumor microenvironment (TME) presents a complex ecosystem comprising a diversity of immune and stromal cell populations that influence tumor progression, drug delivery and therapy outcome. High levels of inter- and intra-tumor heterogeneity in TME state are driven by molecular tumor phenotypes. Pancreatic ductal adenocarcinoma (PDAC), one of the most lethal cancer types, displays a high genetic heterogeneity along with an immunosuppressive TME. However, it remains largely elusive how distinct TME states are mechanistically influenced, and which molecular processes dictate the emergence of different modes of immunosuppression within the TME of molecular PDAC subtypes.Here, we performed a systematic analysis of functional associations between the major molecular PDAC subtypes, the classical and mesenchymal subtype, and their TME states. We describe the TME composition and complex cell-cell communication networks within PDAC subtypes using a multimodal integration of spatial transcriptomics (ST) with complementary approaches, such as scRNA-seq, MS-based Secretomics and multiplexed histocytometry. We generated ST data sets (Visium) of a tumor cohort derived from a Kras-driven PDAC mouse model. To analyze TME composition as well as spatial niches and cell type communities, we computationally enhanced the spot-resolution of ST datasets, followed by cell type deconvolution and cell-cell communication analysis to identify subtype-specific TME communities and communication networks.To this end, we generated a large resource of PDAC mouse models which represent molecular subtypes of the disease and mimic the heterogeneity of TME states found in human PDAC cohorts. This analysis revealed that a set of subtype-specific secreted factors shape the immunosuppressive PDAC TME via direct and indirect communication networks with immunosuppressive myeloid and T cells in molecular PDAC subtypes. Multimodal ST analysis delineated spatial subtype-specific TME communities as well as spatial communication patterns. We functionally investigate the tumor cell intrinsic regulation of the identified subtype-specific secreted factors, providing potential therapeutic vulnerabilities for more effective combinatorial subtype-specific therapies including immunotherapeutic approaches. Citation Format: Stefanie Baerthel, Chiara Falcomatà, Vanessa Goelling, Daniele Lucarelli, Rushin Gindra, Constantin Schmitt, Jonathan Swietlik, Felix Meissner, Marc Schmidt-Supprian, Dieter Saur. Multimodal spatial transcriptomics uncover distinct tumor microenvironment states and cell-cell communication networks in molecular pancreatic cancer subtypes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1587.

Abstract 1770: Right oncogene, wrong tumor - CBL mutations in pediatric CNS and solid tumors

Abstract The molecular analysis of individual patient tumors by personalized medicine programs like the Zero Childhood Cancer Program (ZERO) enables the detection of potentially targetable lesions, unlocking new therapeutic opportunities for patients. Somatic mutations in the E3-ubiquitin ligase, CBL, are known to activate receptor tyrosine kinases (RTKs) in cancer and have been exclusively characterized in hematological malignancies. This includes acute myeloid leukemia, where CBL mutations lead to FLT3 activation. Using molecular data from ZERO we have identified known and novel CBL variants in novel tumor contexts. Our findings raise the possibility that CBL mutation may be a marker of RTK activation in a range of paediatric cancer types and may represent a group of patients who could benefit from tyrosine kinase inhibitor (TKI) therapy. We analyzed whole genome and RNA sequencing data from the ZERO cohort and identified 26 somatic CBL variants in 22 individual patients, the majority of which were in CNS tumors (14 patients). In 8 patients with CNS tumors, we identified missense and splicing variants, which were either novel or not previously seen in this tumor type, that are predicted to have a functional impact on CBL E3 ligase activity due to their location in the linker region or RING finger domain of CBL. Interestingly, most of these patients did not fall into a pre-defined DNA methylation-based subtype classification of CNS tumors and had no significant upregulation of RTK genes. In addition to the novel variants, we identified an established oncogenic CBL deletion variant, CBL exon 8/9 deletion (CBL ex8/9Δ), in novel tumor types - neuroblastoma and a germ cell tumor. Functionally, we show that overexpression of CBL ex8/9Δ enhances cell proliferation and maintains EGFR signaling by blocking CBL mediated degradation of phosphorylated EGFR in neuroblastoma cells. In this study, we demonstrate that CBL is mutated in a range of pediatric cancer types beyond hematological malignancies and may represent a marker of RTK activation and a new target for high-risk patients with limited therapeutic options. Citation Format: Lauren M. Brown, Chelsea Mayoh, Pablo Acera Mateos, Antoine De Weck, Robert Salomon, Teresa Sadras, Neevika Manoharan, Marie Wong, Mark J. Cowley, Paul G. Ekert. Right oncogene, wrong tumor - CBL mutations in pediatric CNS and solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1770.

Abstract 4232: Exploring chemotherapy resistance in pancreatic cancer: Insights from a 3D organoid model

Abstract Pancreatic cancer is marked by a dismal prognosis, with a 5-year survival rate at approximately 10%. A key contributor to its lethality is the development of chemotherapy resistance, followed by early metastasis. However, studies elucidating the mechanisms behind chemotherapy resistance in pancreatic cancer are limited. Previous research has relied on 2D cell line models, which present limitations, such as uniform cell types that fail to capture tumor heterogeneity and an inaccurate representation of clinical realities, wherein resistance often develops gradually rather than through continuous exposure to high drug concentrations. In this study, we developed a more sophisticated organoid model of chemotherapy resistance. Six organoids from three patients were matched to study chemotherapy resistance in a personalized context. Comprehensive whole-exome sequencing and transcriptomic analyses were performed, along with an examination of chemotherapy resistance protein expression and EMT-related markers. Notably, differential expression levels of CK19 and Vimentin were observed between samples. Our analysis also focused on the expression of the ATP-Binding Cassette Subfamily G Member 2 (ABCG2) gene in chemoresistant organoids. Western blot analysis indicated a significant increase in ABCG2 expression in these organoids compared to controls. According to the Moffitt RNA signature, the primary and corresponding chemoresistant organoids retained their molecular subtype classifications both before and after developing chemoresistance. Gene set enrichment analysis uncovered significant upregulation of pathways related to DNA repair and cell cycle regulation, particularly the G2M checkpoint, in chemoresistant organoids. Furthermore, gene ontology functional analysis suggested a notable overrepresentation of activities related to amino acid transmembrane transport and extracellular matrix constituents, underscoring their potential role in the chemoresistance phenotype. Citation Format: Hee Seung Lee. Exploring chemotherapy resistance in pancreatic cancer: Insights from a 3D organoid model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4232.

Abstract 5059: Robust single sample consensus molecular subtype classification for primary and metastatic colorectal cancer

Abstract Introduction Consensus Molecular Subtypes (CMS) represent a well-established molecular stratification framework for colorectal cancer (CRC). Existing methods for CMS classification rely on a representative input cohort as a preprocessing step (CMScaller) or have difficulty generalizing to metastatic samples (CMSclassifier). We developed Tempus CMS to overcome these limitations. Our method normalizes gene expression data from input samples to a static reference cohort to enable single sample classification of both primary and metastatic tumors. We evaluated the performance of this classifier on a large, de-identified CRC cohort comprising 8,489 samples from primary and metastatic sites. Methods To normalize input data, our method shifts and scales each gene expression value based on the mean and standard deviation of the expression of that gene in the reference cohort (n=2,787 primary and metastatic CRC). CMS calls are then generated via nearest template prediction similar to CMScaller (Eide et al., 2017). The analysis cohort was selected from clinical biopsies within 30 days of primary or metastatic diagnosis excluding reference cohort samples. CMS calls were assessed by comparing subtype prevalence to reported rates and by testing for known enrichments of pathways and genomic markers. Results Subtype prevalences for lower gastrointestinal (GI) biopsies and resections (n=5,090) were comparable to reported rates for primary CRC tumors: CMS1 - 12%, CMS2 - 28%, CMS3 - 19%, and CMS4 - 33% with 7.6% no calls. Biopsies from metastatic tissue showed site-specific changes in prevalence: CMS2 increased to 43% in liver samples (n=1,715) and CMS4 was identified in 49% of samples from metastatic sites excluding the liver (n=1,684). In addition, we found that purity had an effect on prevalence: in lower GI tumors with ≥ 50% tumor purity rates of CMS2 and CMS4 were 34% and 26%, respectively, while rates in tumors with purity between 20% and 50% were 20% CMS2 and 41% CMS4. Across all tissue groups, pathway enrichment showed significant associations, including TGF-ꞵ signaling in CMS4 (adj P ≤ 4.4e-15) and cell differentiation in CMS3 (adj P ≤ 3.0e-27). Expected enrichments of CMS markers were also observed: BRAF mutations and microsatellite instability in CMS1 and KRAS mutations in CMS3. In liver tissue, higher KRAS mutation rates were noted in CMS1 (57%) and CMS3 (83%) compared to GI tissue (35% and 67%, respectively). Conclusions In this study, we developed an enhanced, single sample CMS-calling algorithm optimized for primary and metastatic sites. Application of this algorithm in a large cohort recapitulated known biology, which suggests that the tool can be used to support clinical studies requiring robust molecular stratification. Citation Format: Andrew J. Sedgewick, Tushar Chandra, Timothy Taxter, Justin Guinney. Robust single sample consensus molecular subtype classification for primary and metastatic colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5059.

Abstract 3526: GAIN-BRCA: A graphical explainable AI-net framework for breast cancer subtype classification using multiomics data

Abstract Higher standards for precise distinction and classification among the four breast cancer subtypes, i.e., Luminal A. Luminal B, HER-2-enriched, and Basal, are essential to delivering customized treatment in precision oncology. The effective integration of readily accessible multi-omics datasets from the same patient may improve the accuracy of subtype prediction algorithms and help decipher reliable molecular features for the prediction and prognosis of breast cancer. Previous models for predicting breast cancer subtypes relied mostly on concatenation and autoencoder methods but rarely utilized graph-based integration to leverage the biological associations among omics datatypes. To overcome these limitations, we developed a graph-based multi-omics integration method using features from mRNA, DNA methylation, and miRNA data and synthesized features from their interactions. GAIN-BRCA computes weightage scores from miRNA-mRNA and DNA methylation-mRNA interactions to derive a new transformed feature vector without compromising biological information. The neural network architecture was designed and trained on the transformed features to classify breast cancer subtypes. Neural network prediction accuracies were compared using concatenated, autoencoder, and GAIN-BRCA-based integrated datasets. GAIN-BRCA outperforms the existing similar models MOGONET (Acc: 0.72) and moBRCA-net (Acc: 0.86) with an accuracy of 0.91. These transformed features were then processed using SHAP, and the top 357 features were selected for functional characterization and pathway analysis. We identified specific pathways and genes related to each breast cancer subtype separately. The most enriched pathways found in each subtype are ribonucleotide reductase signaling and glucocorticoid signaling pathways in luminal A, PFKB4 signaling pathway in luminal B, adipogenesis and DNA methylation and transcriptional signaling in HER2+, and the role of OCT4 in mammalian embryonic stem cell pluripotency in basal. Similarly, the most discriminated gene features are BRAF, ESR1, KRAS, MAPK1, SMARCE1, PPP3CA, and TGFBR2 in luminal A, TGFB3 and NCOA3 in luminal B, HNF4A, SIN3A, and SOX9 in HER2+, and POU5F1 and FOXA2 in basal subtypes. The GAIN-BRCA framework, combined with SHAP, weighs independent omics as cross-disciplinary rather than multidisciplinary and presents a distinctive picture for each subtype. As a result, the framework presents the importance of features about the corresponding subtype, which might lead to a greater understanding of the molecular basis of breast cancer subtypes. Citation Format: Jai Chand Patel, Sushil Kumar Shakyawar, Sahil Sethi, Chittibabu Guda. GAIN-BRCA: A graphical explainable AI-net framework for breast cancer subtype classification using multiomics data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3526.

TOAST stroke subtype classification in clinical practice: implications for the Get With The Guidelines-Stroke nationwide registry.

The TOAST (Trial of ORG 10172 in Acute Stroke Treatment) is the most commonly used ischemic stroke subtype classification system worldwide and a required field in the US National Get With The Guidelines-Stroke (GWTG-Stroke) registry. However, stroke diagnostics have advanced substantially since the TOAST classification was designed 30 years ago, potentially making it difficult to apply reliably. In this prospective diagnostic accuracy study, we analyzed consecutive ischemic stroke patients admitted to a Comprehensive Stroke Center between July-October 2021. Clinical practice TOAST classification diagnoses rendered by the stroke team in the electronic medical record (EMR) at discharge were retrieved from GWTG-Stroke registry and compared to a reference ("gold") standard diagnosis derived from agreement between two expert raters after review of the EMR and patient imaging. Among 49 patients; age was 72.3 years (±12.1), 53% female, and presenting NIHSS median 3 (IQR 1-11). Work-up included: brain imaging in 100%; cardiac rhythm assessment in 100%; cervical/cerebral vessel imaging in 98%; TTE ± TEE in 92%; and TCD emboli evaluation in 51%. Reference standard diagnoses were: LAA-6%, SVD-14%, CE-39%, OTH-10%, UND-M (more than one cause)-20%, and UND-C (cryptogenic)-10%. GWTG-Stroke TOAST diagnoses agreed with reference standard diagnoses in 30/49 (61%). Among the 6 subtype diagnoses, specificity was generally high (84.8%-97.7%), but sensitivity suboptimal for LAA (33%), OTH (60%), UND-M (10%), and UND-C (20%). Positive predictive value was suboptimal for 5 of the 6 subtypes: LAA (13%), SVD (58%), OTH (75%), UND-M (50%), and UND-C (50%). Clinical practice TOAST classification subtype diagnoses entered into the GWTG-Stroke registry were accurate in only 61% of patients, a performance rate that, if similarly present at other centers, would hamper the ability of the national registry to provide dependable insights into subtype-related care. Development of an updated ischemic stroke subtype classification system, with algorithmic logic embedded in electronic medical records, is desirable.

Abstract 6548: Leveraging compact feature sets for TCGA-based molecular subtype classification on new samples

Abstract The NCI's The Cancer Genome Atlas (TCGA) project profiled over 10,000 tumor samples over the course of 10 years. As different tissue-specific working groups reviewed all of the available data, these patient samples were separated into distinct molecular subtypes, and these clusters were reported in various marker papers. While these assignments provided invaluable information about the common patterns of molecular characteristics in different types of cancer there was no consistent methodology for assigning new samples to these defined molecular subtypes.The NCI's Tumor Molecular Pathology group was formulated to create machine learning-based models that could be applied to non-TCGA samples and determine their TCGA mapped subtypes. Five modeling systems, JADBio, SKGrid by the Oregon Health and Science University, CloudForest by the Institute of Systems Biology, AKLIMATE by University of California Santa Cruz and subSCOPE by BC Cancer’s Genome Sciences Centre, were trained to recognize TCGA subtypes using multi-omic measurements from gene expression, DNA methylation, miRNA expression, copy number, and somatic mutation calls. While the TCGA samples were profiled using multi-omic technologies, single platform and/or compact feature set models also were assessed for their ability to assign these classifications. Each machine learning system created predictive models for 106 subtypes from 26 cancer types using as few features as possible, with a maximum of 100 features allowed for scored models. A set of 411,706 models was developed, composed of results of each of the learning methods across the various omic platforms. Top models, both multi-omic and single platform, were selected for each cancer type. On average, models were able to achieve an overall weighted F1 score of 0.895 with 42 features. While the top models for each cancer type had an overall weighted F1 mean performance of 0.936 with a mean of 29 features, in 20 of the 26 cancer types models using only gene expression provided the best performance. Analysis of features selected by the models showed some known onco-drivers were selected by many models, but many times different models would utilize features of different genes with similar levels of performance. Network-level analysis revealed that many genes of these selected features operated within the same pathways.Transferability of these models to external datasets was tested, taking TCGA breast cancer trained models and applying them to AURORA and METABRIC datasets. Interestingly, despite the data platform difference between TCGA (RNAseq) and METABRIC (microarray), model performance saw only minimal degradation of F1 values in transfer. This set of models and the training dataset will provide new opportunities for researchers and translational scientists to connect new tumors to the subtypes seen in the TCGA cohorts. Citation Format: Kyle Ellrott, Chris K. Wong, Christina Yau, Mauro A. Castro, Jordan Lee, Brian Karlberg, Jasleen K. Grewal, Vincenzo Lagani, Bahar Tercan, Verena Friedl, Toshinori Hinoue, Vladislav Uzunangelov, Lindsay Westlake, Xavier Loinaz, Ina Felau, Peggy Wang, Anab Kemal, Samantha J. Caesar-Johnson, Ilya Shmulevich, Alexander J. Lazar, Ioannis Tsamardinos, Katherine A. Hoadley, The Cancer Genome Atlas Analysis Network, Gordon A. Robertson, Theo A. Knijnenburg, Christopher C. Benz, Joshua M. Stuart, Jean C. Zenklusen, Andrew D. Cherniack, Peter W. Laird. Leveraging compact feature sets for TCGA-based molecular subtype classification on new samples [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6548.

Abstract 3491: Tumor subtype classification of HPV-associated head and neck cancers is central to key clinically relevant variables

Abstract Cancer types are typically categorized according to the cell of origin, but within these groupings there exists vast heterogeneity. Thus, subtypes based on molecular features are often defined that have clinical utility as prognostic biomarkers, aid in selection of therapeutic strategies, or are associated with treatment response. Head and neck cancer (HNC) is the seventh most common cancer globally and projected to have 54,540 new cases in the U.S. in 2023. Within the US, HPV(+) HNC is now more prevalent than HPV(+) cervical cancer, and it is expected to continue to rise. Although there is wide morphologic and epigenetic diversity within HPV(+) HNC, tumor subtyping is not yet widely used for this cancer population, largely because of HPV(+) HNC’s unique tumorigenesis process. Unlike other cancers that are driven by specific genetic signatures, HNC is driven by viral gene mechanisms. Given the large degree of heterogeneity among HPV(+) HNC cases and the lack of key mutations to define subtypes, the classification of HPV(+) HNSC subtypes requires a more sophisticated approach. In this study, we introduce a robust, ensemble machine learning (ML) classifier for subtyping HPV(+) HNC that was trained on multiple cohorts and gene feature sets from RNA-seq data, and rigorously tested to ensure high reproducibility. The results classify HPV(+) HNC into the two main recognized subtypes: IMU (immune strong) and KRT(highly keratinized). Using a cohort of 227 patients, we show that the IMU/KRT classification is highly correlated with key clinically relevant variables in HNC. Of 41 molecular, clinical, and epidemiologic variables tested, 24 significantly associated with subtype. The IMU subtype was significantly associated with CD8 T-cells (p-value = 4.21 × 10−9), Dendritic cells (p-value = 1.65 × 10−8), B-cells (p-value = 2.20E−8), and epithelial to mesenchymal transition (p-value = 0.00013146). The KRT subtype is significantly associated with keratinization (p-value = 7.11 × 10−9), HPV integration (p-value = 3.14 × 10−6), radiation resistance (p-value = 0.00307), female sex (p-value = 0.00641), and high T stage (p-value = 0.0324). Genetic, epigenetic, and HPV gene-related variables were also among those significantly associated with subtype. HPV(+) HNC subtypes have been shown to be associated with survival, with KRT-like patients having worse clinical outcomes than IMU. Given the different carcinogenic processes underlying IMU and KRT tumors, our ensemble ML subtype classifier web tool will help inform future studies of HPV(+) HNC. Future work to assess how HPV(+) subtypes can be incorporated into precision treatment strategies is well-motivated by our findings. Citation Format: Bailey F. Garb, Shiting Li, Tingting Qin, Elizabeth Lopez, Sarah Soppe, Snehal Patil, Laura Rozek, Nisha D'Silva, Maureen Sartor. Tumor subtype classification of HPV-associated head and neck cancers is central to key clinically relevant variables [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3491.

Abstract 7567: In-depth mass cytometry analysis reveals different immune cell composition of unique molecular subtypes in extensive stage small-cell lung cancer

Abstract Background: A novel molecular subtype classification was fashionable for small cell lung cancer (SCLC). However, the distribution of SCLC molecular subtype classification in Chinese patients were unknown, especially extensive stage (ES)-SCLC. It should validate and define their associated microenvironments across subtypes in ES-SCLC. Method: We used immunohistochemical (IHC) to evaluate SCLC markers of the subtypes including ASCL1, NEUROD1, POU2F3, SCLC-A/N/P in 135 ES-SCLC tumors. Peripheral blood of 24 ES-SCLC patients was test by mass cytometry using time-of-flight (CyTOF) to evaluate the differences in immune subsets regard to the subtype tumor microenvironment (TME). Validation was performed using flow cytometry and multiplex IF. Results: ASCL1, NEUROD1, POU2F3, and Inflamed were account for 60.7%, 28.9%, 7.4%, and 18.5% of SCLC tumors, respectively. Significant intratumoral heterogeneity was observed with 15.6% of the tumors being both positive for SCLC-A and -N subtype markers, respectively. The SCLC-I type possessed more CD8+ T-cells infiltration and had higher frequency an ‘inflamed’ immunophenotyped. CyTOF data showed greater CD161+CD127+CD8+T cells proportion in SCLC-I than SCLC-N subtypes before and after anti-immunotherapy. Furthermore, we validated that lung tissue of SCLC-I were infiltrated more CD161+CD127+CD8+T cells compared to SCLC-N subpopulation(P=0.0281). Flow cytometry further showed the stronger immune activation of CD161+CD127+CD8+T cells in SCLC-I types induced higher expression of GZMB and GNLY and lower-expression of T-cell exhaustion markers PD-1,TIGIT and LAG3 in comparison with SCLC-N types. We further confirmed that the high level of CD161+CD127+CD8+T associated with a favorable treatment response and prolonged overall survival. Conclusion: The present study validated the molecular subtypes in ES-SCLC patients. Furthermore, SCLC-I subtype has longer PFS and OS than SCLC-A, -N and -P patients. Our research not only provides an insight into the immune landscape of ES-SCLC but also sheds light on a special subset of CD161+CD127+CD8+T cells with prognostic and therapeutic significance. Citation Format: Yuekang Li, Jingjing Qu, Binggen Wu, Qian Shen, Wenjia Sun, Lijun Chen, Bo Wang, Lixiong Ying, Jianya Zhou, Jianying Zhou. In-depth mass cytometry analysis reveals different immune cell composition of unique molecular subtypes in extensive stage small-cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7567.

The prognostic values and immune characteristics of polo-like kinases (PLKs) family: A pan-cancer multi-omics analysis

BackgroundIn the realm of tumor-targeted therapeutics, Polo-like kinases (PLKs) are a significant group of protein kinases that were found recently as being related to tumors. However, the significance of PLKs in pan-cancer remains systematically studied. Methods and materialsWe integrated multi-omics data to comprehensively investigate the expression patterns of the PLK family across various cancer types. Subsequently, study examined the associations between tumor mutation burden (TMB), microsatellite instability (MSI), immune subtype classification, immune infiltration, tumor microenvironment scores, immune checkpoint gene expression, and the PLKs expression profiles within various tumor types. Furthermore, using our mRNA sequencing data (TRUCE01) and four bladder cancer (BLCA) cohorts (GSE111636, GSE176307, and IMvigor210), We examined the correlation between the expression level of PLK and immunotherapy effectiveness. Next, Gene set enrichment analysis (GSEA) was evaluated to find that potentially enriched PLK signaling pathways. Utilizing TIMER 2.0, we conducted an immune infiltration analysis underlying transcriptome expression, copy number variations (CNV), or somatic mutations of PLKs in BLCA. Finally, mRNA expression validation of PLK1/3/4 by real-time PCR within 10 paired BLCA tissues, protein expression verification through the Human Protein Atlas (HPA), and PLK4 in vitro cytological studies have been employed in BLCA. ResultsThe expression of most of the PLK family members exhibits variation between cancerous tissues and adjacent normal tissues across various cancer species. Furthermore, the expression of PLKs demonstrates a significant association with immunotyping, infiltration of immune cell, tumor mutational burden (TMB), microsatellite instability (MSI), immunological checkpoint gene activity and therapeutic effectiveness in pan-tumor tissues. Additional investigation into the correlation between the PLK family and BLCA has revealed that the expression of the PLK genes holds substantial significance in the biological processes of BLCA. Furthermore, a notable association has been observed between the copy number variation, variant status, and the degree of certain immunological cell infiltration. Of note, the expression validation and in vitro phenotypic experiments have demonstrated that PLK4 has a significant function in promoting the BLCA cell proliferation, migration, and invasion. ConclusionCollectively, based on various databases, our results highlight the involvement of PLK gene family in the formation of different types of tumors and identify PLK-related genes that may be used for therapy.

Prediction uncertainty estimates elucidate the limitation of current NSCLC subtype classification in representing mutational heterogeneity

The heterogeneous pathogenesis and treatment response of non-small cell lung cancer (NSCLC) has led clinical treatment decisions to be guided by NSCLC subtypes, with lung adenocarcinoma and lung squamous cell carcinoma being the most common subtypes. While histology-based subtyping remains challenging, NSCLC subtypes were found to be distinct at the transcriptomic level. However, unlike genomic alterations, gene expression is generally not assessed in clinical routine. Since subtyping of NSCLC has remained elusive using mutational data, we aimed at developing a neural network model that simultaneously learns from adenocarcinoma and squamous cell carcinoma samples of other tissue types and is regularized using a neural network model trained from gene expression data. While substructures of the expression-based manifold were captured in the mutation-based manifold, NSCLC classification accuracy did not significantly improve. However, performance was increased when rejecting inconclusive samples using an ensemble-based approach capturing prediction uncertainty. Importantly, SHAP analysis of misclassified samples identified co-occurring mutations indicative of both NSCLC subtypes, questioning the current NSCLC subtype classification to adequately represent inherent mutational heterogeneity. Since our model captures mutational patterns linked to clinical heterogeneity, we anticipate it to be suited as foundational model of genomic data for clinically relevant prognostic or predictive downstream tasks.

A pocket companion to cell-free DNA (cfDNA) preanalytics.

The cumulative pool of cell-free DNA (cfDNA) molecules within bodily fluids represents a highly dense and multidimensional information repository. This "biological mirror" provides real-time insights into the composition, function, and dynamics of the diverse genomes within the body, enabling significant advancements in personalized molecular medicine. However, effective use of this information necessitates meticulous classification of distinct cfDNA subtypes with exceptional precision. While cfDNA molecules originating from different sources exhibit numerous genetic, epigenetic, and physico-chemical variations, they also share common features that complicate analyses. Considerable progress has been achieved in mapping the landscape of cfDNA features, their clinical correlations, and optimizing extraction procedures, analytical approaches, bioinformatics pipelines, and machine learning algorithms. Nevertheless, preanalytical workflows, despite their profound impact on cfDNA measurements, have not progressed at a corresponding pace. In this perspective article, we emphasize the pivotal role of robust preanalytical procedures in the development and clinical integration of cfDNA assays, highlighting persistent obstacles and emerging challenges.

New Strategies for Macrophage Re-Education in Cancer: An Update.

The association between cancer and inflammation is well established. Chronic inflammation represents a fundamental step in the development and progression of some types of cancer. Tumors are composed of a heterogeneous population of infiltrating cells including macrophages, fibroblasts, lymphocytes, granulocytes, and mast cells, which respond to signals from the microenvironment and, in turn, produce cytokines, chemokines, transcription factors, receptors, and miRNAs. Recent data demonstrate that, in addition to classical (M1) and alternative (M2) macrophage subtypes, there are many intermediate subtypes that potentially play different roles in response to environmental stimuli. Tumors are infiltrated by macrophages called TAMs that mainly display an M2-like phenotype and tumor growth-permissive activities. There is a bidirectional interaction between tumor cells and tumor-infiltrating cells that determines macrophage polarization and ultimately tumor progression or regression. These complex interactions are still unclear but understanding them is fundamental for the development of new therapeutic strategies. Re-educating tumor-permissive macrophages into anti-tumor macrophages is a new focus of research. This review aims to analyze the most recent articles investigating the interplay between tumors, tumor-infiltrating cells, and TAMs, and the strategies for re-educating tumor-permissive macrophages.