TMC5 is Highly Expressed in Human Cancers and Corelates to Prognosis and Immune Cell Infiltration: A Comprehensive Bioinformatics Analysis

Background: The oncological role of TMC5 in human cancers has only been revealed partially. We performed integrated bioinformatics analysis to provide a thorough and detailed insight of associations between TMC5 and tumorigenesis, cancer progression, and prognosis.Methods: With reference to the accessible online databases, the TMC5 expressions in tumor tissues and corresponding normal tissues, different pathological stages, and various cancer cells were analyzed, while the protein levels of TMC5 in different cancers were also inspected. Meanwhile, the prognostic value of TMC5 expression in multiple cancers as well as in advanced-stage patients was investigated. Furthermore, the mutational data of TMC5 and its correlation with cancer prognosis were assessed. Moreover, the association between the TMC5 level and immune cell infiltration was evaluated. Next, TMC5-related pathway alterations and drug responses were summarized. Finally, the TMC5 based protein network was generated, and relevant enrichment was performed.Results: In our study, the expression level of TMC5 was significantly higher in the tumor tissue than that of the normal tissues in most cancer types. Fluctuations of TMC5 levels were also observed among different pathological stages. In the meantime, the protein level elevated in the tumor tissue in the cancers enrolled. Moreover, the expression of TMC5 was not only prognostic for overall survival (OS) or recurrence free survival (RFS) in various types of cancers but also correlated to OS in patients with more advanced cancers. Additionally, the mutational status of TMC5 is also associated with prognosis in cancer patients. It is worth noting that the TMC5 level was closely related to immune cell infiltrations, especially in ESCA, TGCT, and USC. The TMC5 expression was also identified as an activator for pathways including PI3K/AKT, RAS/MAPK, and TSC/mTOR, proved to be associated with multiple drug responses and assessed to be interactive with the TMEM family.Conclusion: TMC5 might function as a potential marker for cancer survival and immune responses.

TMOD-29. STANDARDIZED GENERATION OF TUMOR-ORGANOIDS AS NOVEL DRUG SCREENING MODEL IN MENINGIOMA

Tumor-organoids (TOs) are novel, complex three-dimensional ex vivo tissue cultures that under optimal conditions accurately reflect genotype and phenotype of the original tissue with preserved cellular heterogeneity and morphology. They may serve as a new and exciting model for studying cancer biology and directing personalized therapies. The aim of our study was to establish TOs from meningioma (MGM) and to test their usability for large-scale drug screenings. We were capable of forming several hundred TO equal in size by controlled reaggregation of freshly prepared single cell suspension of MGM tissue samples. In total, standardized TOs from 60 patients were formed, including eight grade II and three grade III MGMs. TOs reaggregated within 3 days resulting in a reducted diameter by 50%. Thereafter, TO size remained stable throughout a 14 days observation period. TOs consisted of largely viable cells, whereas dead cells were predominantly found outside of the organoid. H&E stainings confirmed the successful establishment of dense tissue-like structures. Next, we assessed the suitability and reliability of TOs for a robust large-scale drug testing by employing nine highly potent compounds, derived from a drug screening performed on several MGM cell lines. First, we tested if drug responses depend on TO size. Interestingly, drug responses to these drugs remained identical independent of their sizes. Based on a sufficient representation of low abundance cell types such as T-cells and macrophages an overall number of 25.000 cells/TO was selected for further experiments revealing FDA-approved HDAC inhibitors as highly effective drugs in most of the TOs with a mean z-AUC score of -1.33. Taken together, we developed a protocol to generate standardized TO from MGM containing low abundant cell types of the tumor microenvironment in a representative manner. Robust and reliable drug responses suggest patient-derived TOs as a novel drug testing model in meningioma research.

DDRE-42. TOWARDS PRECISION MEDICINE: AUTOMATED DRUG SCREENING PLATFORM UTILIZING TUMOR-ORGANOIDS TO IDENTIFY PATIENT-SPECIFIC DRUG-RESPONSES IN MENINGIOMA

Tumor-organoids (TO) are mini-tumors generated from tumor tissue preserving its genotype and phenotype by maintaining the cellular heterogeneity and important components of the tumor microenvironment. We recently developed a protocol to reliably establish TOs from meningioma (MGM) in large quantities. The use of TOs in combination with lab automation holds great promise for drug discovery and screening of comprehensive drug libraries. This might help to tailor patient-specific therapy in the future. The aim of our study was to establish an automated drug screening platform utilizing TOs. For this purpose, we established TOs by controlled reaggregation of freshly prepared single cell suspension of MGM tissue samples in the high-throughput format of 384-well plates. The drug screening was performed fully automated by utilizing the robotic liquid handler Hamilton Microlab STAR and a drug library containing 166 FDA-approved oncology agents. Viability was assessed with CellTiterGlo3D. In total, we performed the drug screening with 166 drugs on TOs from 11 patients suffering from MGM (n=8 WHO°I, n=2 WHO°II, n=1 WHO°III). The top five most effective drugs resulted in a decrease of TO viability ranging from 84.6–63.3%. K-means clustering analysis resulted in groupings of drugs with similar modes of action. One cluster consisted of epigenetic drugs while another cluster consisted of several proteasome inhibitors. However, when looking at a patient-individual level, in 11 patients 44 of 166 drugs, were among the top 10 most effective drugs, providing strong evidence for heterogeneous drug-responses in MGM patients. Taken together, we successfully developed an automated drug screening platform pipeline utilizing TOs from MGM to identify patient-specific drug-responses. The observed intra-individual differences of drug responses mandate for a personalized testing of comprehensive drug libraries in TOs to tailor more effective therapies in MGM patients.

Preclinical Application of Conditional Reprogramming Culture System for Laryngeal and Hypopharyngeal Carcinoma

Management of laryngeal and hypopharyngeal squamous cell carcinoma (LHSCC) remains highly challenging due to highly variable therapeutic responses. By establishing an in vitro model for LHSCC based on conditional reprogramming (CR), a cell-culture technique, we aim to investigate its potential value on personalized cancer therapies. Herein, a panel of 28 human LHSCC CR cells were established from 50 tumor tissues using the CR method. They retained tumorigenic potential upon xenotransplantation and recapitulated molecular characteristics of LHSCC. Differential responses to anticancer drugs and radiotherapy were detected in vitro. CR cells could be transformed to xenograft and organoid, and they shared comparable drug responses. The clinical drug responses were consistent with in vitro drug responses. Collectively, the patient-derived CR cell model could promisingly be utilized in clinical decision-making and assisted in the selection of personalized therapies for LHSCC.

An overview of machine learning methods for monotherapy drug response prediction

For an increasing number of preclinical samples, both detailed molecular profiles and their responses to various drugs are becoming available. Efforts to understand, and predict, drug responses in a data-driven manner have led to a proliferation of machine learning (ML) methods, with the longer term ambition of predicting clinical drug responses. Here, we provide a uniquely wide and deep systematic review of the rapidly evolving literature on monotherapy drug response prediction, with a systematic characterization and classification that comprises more than 70 ML methods in 13 subclasses, their input and output data types, modes of evaluation, and code and software availability. ML experts are provided with a fundamental understanding of the biological problem, and how ML methods are configured for it. Biologists and biomedical researchers are introduced to the basic principles of applicable ML methods, and their application to the problem of drug response prediction. We also provide systematic overviews of commonly used data sources used for training and evaluation methods.

Network Modeling Predicts Personalized Gene Expression and Drug Responses in Valve Myofibroblasts Cultured with Patient Sera

Aortic valve stenosis (AVS) patients experience pathogenic valve leaflet stiffening due to excessive extracellular matrix (ECM) remodeling. Numerous microenvironmental cues influence pathogenic expression of ECM remodeling genes in tissue-resident valvular myofibroblasts, and the regulation of complex myofibroblast signaling networks depends on patient-specific extracellular factors. Here, we combined a manually curated myofibroblast signaling network with a data-driven transcription factor network to predict patient-specific myofibroblast gene expression signatures and drug responses. Using transcriptomic data from myofibroblasts cultured with AVS patient sera, we produced a large-scale, logic-gated differential equation model in which 11 biochemical and biomechanical signals are transduced via a network of 334 signaling and transcription reactions to accurately predict the expression of 27 fibrosis-related genes. Correlations were found between personalized model-predicted gene expression and AVS patient echocardiography data, suggesting links between fibrosis-related signaling and patient-specific AVS severity. Further, global network perturbation analyses revealed signaling molecules with the most influence over network-wide activity including endothelin 1 (ET1), interleukin 6 (IL6), and transforming growth factor β (TGFβ) along with downstream mediators c-Jun N-terminal kinase (JNK), signal transducer and activator of transcription (STAT), and reactive oxygen species (ROS). Lastly, we performed virtual drug screening to identify patient-specific drug responses, which were experimentally validated via fibrotic gene expression measurements in VICs cultured with AVS patient sera and treated with or without bosentan - a clinically approved ET1 receptor inhibitor. In sum, our work advances the ability of computational approaches to provide a mechanistic basis for clinical decisions including patient stratification and personalized drug screening.