An Integrated Simulation And Analysis Of Glioblastoma Heterogeneity And Drug Sensitivity Based On A Novel Tumor Organoid Model

Glioblastoma(GBM)is the most malignant type of primary brain tumor in adults,and patients have a poor prognosis,with a median survival of only 12-15 months and a 5-year survival rate of less than 5%.Earlier studies,particularly high-throughput sequencing analysis of tumor tissue,suggest a high degree of genetic,metabolic and epigenetic heterogeneity potentially associated with drug resistance and high mortality in GBM.The classification of GBM into several different subtypes at the transcriptome level has greatly facilitated the resolution of individual tumor heterogeneity in GBM patients,but clinical studies and applications related to subtyping have been slow and the clinical implications are highly controversial.Further studies suggest that a key cause of GBM drug resistance and treatment failure may be the complex and dynamic intratumorally heterogeneity,which is influenced by a number of factors,particularly the complex and variable tumor microenvironment(TME).TME is critical for the development,maintenance and drug response of GBM stem cell subtypes,and the dynamic evolution of TME during GBM progression and treatment leads to dynamic transitions between GBM stem cell subtypes,posing a significant challenge to current therapeutic strategies that do not account for this complexity and dynamic therapeutic strategies that do not take this complexity and dynamism into account.A key challenge in the study of intra-tumor heterogeneity and drug resistance is the lack of suitable preclinical research models and paradigms to resolve the structural and dynamic evolution of GSC subpopulations and clone-like tissues.Single cell analyses,such as single cell transcriptomes,are changing the understanding of intra-tumor heterogeneity,enabling precise and detailed resolution of GSC subpopulations,clonal organization and genealogical composition and dynamic evolution.However,single cell transcriptome sequencing is very limited in depth and inevitably leads to cell damage and destruction during experiments,making it difficult to capture rare cells.Therefore,it is still difficult to use the single-cell transcriptome for in-depth,precise analysis and prediction of function and mechanism at a systematic and comprehensive,histological level,which greatly hinders its application in drug development and personalized precision medicine based on intra-tumor heterogeneity and drug resistance.The development of new in vivo and ex vivo research models and paradigms to systematically resolve intra-tumor heterogeneity and dynamic genealogical and clonal evolution issues to guide personalized therapeutic strategies remains a huge basic and translational clinical research challenge.In the last decade,patient tissue-derived organoids are rapidly emerging as versatile in vitro models for tumor research and personalized medicine.However,the high degree of intratumorally heterogeneity and the unique extracellular matrix and immune microenvironment of the brain make it exceptionally challenging to develop organoid models that can better mimic the internal environment of GBM and reproduce its in vivo pathological processes and drug sensitivity dynamics.In 2016,Rich’s group reported for the first time the Glioblastoma Organoid Culture System(GBO),which uses patient-derived glioblastoma CSCs to successfully recapitulate the hypoxic gradients and stem cell heterogeneity of tumors in vivo.Subsequent GBM organoid studies have mostly followed a similar strategy of using Matrigel as a substrate to obtain three-dimensional structures.However,Matrigel is rich in basement membrane components,which are mainly used to support epithelial tumor growth,and is very different from the brain and GBM microenvironment,which is rich in proteoglycans such as hyaluronic acid,and suffers from batch-to-batch variability and mechanical instability.in 2020,the Song group succeeded in establishing a patient-derived GBO model and biobank by culturing sections of surgical tumor tissue,preserving a more complete TME,reproducing inter-and intra-tumor heterogeneity.In addition,Marra’s group confirmed the ability to better preserve inter-and intra-tumor heterogeneity in GBM patient-derived explants(PDEs)and glioma spheroids using single-cell RNA sequencing.For dynamic drug selection and efficacy prediction and long-term management of cancer patients,ideal GBM organoid models and experimental paradigms need to be able to reproduce the unique intracerebral and GBM microenvironment,intra-tumor heterogeneity and drug sensitivity,as well as their clonal organization and dynamic evolution,while also enabling in-depth,comprehensive studies at the histological level.To achieve these goals,in this study a novel patient-derived GBM heterogeneous organoid(PHO)model was developed by mixing Matrigel and HA hydrogels to develop a three-dimensional scaffold capable of mimicking the brain’s intracellular matrix,which has comparable cell growth and genetic properties to the Matrigel-based PDO model,but the PHO model can be maintained in long-term suspension without The PHO model has comparable growth and genetic properties to the Matrigel-based PDO model,but the PHO model can be maintained in long-term suspension culture without passaging,and exhibits more complex oxygen gradients,proliferative capacity and stemness properties,as well as enhanced tumorigenicity.More importantly,the PHO model is able to support clonal growth of GSC,retaining more complex phenotypic heterogeneity and GBM-like clonal organization,enabling rapid characterization of the killing effects of chemotherapeutic agents,potentially offering greater advantages over the Matrigel-based PDO model in the personalized treatment of GBM.Using this well-defined clonal growth characteristics and organization in the PHO model,the thesis has developed a comprehensive translational research protocol system based on experiments and bioinformatics that enables the isolation of individual clones and in-depth profiling of their cellular and molecular biology,with different clones exhibiting different cellular and molecular properties and drug sensitivity;developed monoclonal RNA-seq(scl RNA-seq)and Transcriptomic data were obtained for monoclonal clones from different patients and the analysis confirmed that the different clones were able to characterize inter-and intra-tumor heterogeneity and plasticity.Potentially clinically significant,scl RNA-Seq sequencing is much deeper than sc RNA-Seq,allowing unbiased whole-transcriptome level analysis of clones from individual GSCs,thus enabling genome-wide resolution of intratumorally heterogeneity in specific patients.scl RNA-Seq analysis not only confirms the presence of individual intratumorally heterogeneity,but also the presence of PN and MES clones.The results of scl RNA-Seq analysis not only confirm the existence of individual intra-tumoral heterogeneity,such as the coexistence of PN and MES clones in the same PHO,but also resolve,for the first time at single-cell resolution,whole-transcriptomic level differences between GBM subtypes(in the form of single-cell-derived clones),revealing a strong ribosome-associated metabolic profile in the PN subtype and a strong immune profile in the MES subtype,and are further independently validated in single-cell RNA-seq and published datasets.Finally,based on the above transcriptomic predictions,the paper uses PHO and monoclonal culture models for validation studies of combined treatment with ribosomal inhibitors and immune molecules,strongly suggesting that PHO models and scl RNA-Seq can provide a powerful system of research platforms and analytical protocols for GBM modelling,heterogeneity studies,drug testing and personalized treatment.