Human thymoma-associated mutation of the GTF2I transcription factor impairs thymic epithelial progenitor differentiation in mice

Machine learning for predicting the prognosis of patients with thymoma and thymic carcinoma

Background

Thymoma and thymic carcinoma are the most common tumors of the anterior mediastinum. However, there are little research on applying machine learning (ML) approaches to the prognostic prediction of thymoma and thymic carcinoma. The study aims to develop predictive models utilizing ML techniques to accurately forecast the 5-year survival of patients with thymoma and thymic carcinoma.

Methods

Patients with malignant thymic neoplasms were identified in the Surveillance, Epidemiology, and End Results (SEER) 17 database, and their demographic and clinicopathological characteristics were collected. ML classifiers, including elastic net regularized logistic regression, random forest (RF), non-linear support vector machine (SVM), extreme gradient boosting (XGBoost) machine, and categorical boosting (CatBoost) were trained. The hyper-parameter of the algorithms was optimized by a grid search with five repeats of 10-fold cross-validation. Ensemble models were built based on the three algorithms with the highest area under the receiver operator characteristic (ROC) curve (AUC) in the validation set. The best model among the single models and ensemble model was selected as the final model. Calibration curve and decision curve were adopted to evaluate the calibration performance and clinical utility. For comparison, we constructed a baseline model consisting of age and Masaoka stages using logistic regression.

Results

After data cleaning, 1,363 patients and 841 patients were included in the overall survival (OS) dataset and disease-specific survival (DSS) dataset, respectively. CatBoost [AUC: 0.755; 95% confidence interval (CI): 0.698-0.811] had the best performance in the OS prediction for the original dataset. The ensemble model achieved the highest prognostic efficiency for the original dataset, with an AUC of 0.833 (95% CI: 0.765-0.901). Calibration showed favorable goodness of fit and was further verified with the Hosmer-Lemeshow test (CatBoost: χ2=12.63, P=0.13; ensemble model: χ2=7.61, P=0.47). The decision curve showed that the final model provided a high net benefit. The model could significantly distinguish the prognosis of patients (all P values <0.001). Finally, World Health Organization (WHO) histological classification, Masaoka stage, and age were the variables that significantly contributed to the models' prediction of OS and DSS.

Conclusions

We trained ML-based predictive models that could accurately predict the 5-year OS and DSS of patients with thymoma and thymic carcinoma.

GTF2I acts as a novel tumor suppressor transcription factor and shows Favorable prognosis in renal cancer

The role of GTF2I (General Transcription Factor2I) alteration has already been reported in thymic cancer as a valuable biomarker. However, the association of GTF2I mutation with renal cancer for prognosis of immunotherapy is not yet examined. The biologic and oncologic significance of GTF2I in renal cancer was examined at multiomics level such as mutation, copy number alteration, structural variants. The Cancer Genome Atlas (TCGA), Human Protein Atlas (HPA) were used to retrieve the omics data. The expression of GTF2I mRNA was quite significant in case of renal caner. Correlation among the GTF2I mRNA, mutation, CNA and structural variants was also studied. Interactome of GTF2I was also constructed using STRING database. Gain, amplification, and missense mutation exhibited a positive correlation between GTF2I mRNA expression and non-structural variants. Similarly, GTF2I mRNA expression and copy number alterations from GISTIC were positively correlated. High expression of GTF2I was associated with better overall survival indicating the less aggressive clinical features. Insight Box Investigating GTF2I's complex function as a tumor suppressor transcription factor in renal carcinoma provides fresh insights into its biologic and oncologic importance, especially when considering the prognosis of immunotherapy. Little is known about its possible use as a biomarker for renal cancer. Using a multiomics approach and utilizing information from the Human Protein Atlas (HPA) and The Cancer Genome Atlas (TCGA), our study clarifies the intricate relationship between mRNA expression, GTF2I changes, and clinical outcomes in renal cancer. Our results indicate that GTF2I expression may be used as a prognostic indicator because it is positively correlated with favorable survival outcomes. Furthermore, the molecular interactions behind GTF2I's functional significance in renal cancer are revealed by interactome analysis utilizing the STRING database, providing important information for further study and treatment approaches.

Single-cell sequencing has revealed a more complex array of thymic epithelial cells

Thymic epithelial cells participate in the maturation and selection of T lymphocytes. This review explores recent insights from single-cell sequencing regarding classifying thymic epithelial cells in both normal and neoplastic thymus. Cortical thymic epithelial cells facilitate thymocyte differentiation and contribute to positive selection. Medullary epithelial cells are distinguished by their expression of AIRE. Cells progress from a pre-AIRE state, containing precursors with cortical and medullary characteristics, termed junctional cells. Mature medullary epithelial cells exhibit promiscuous gene expression and after that downregulate AIRE mRNA. Post-AIRE cells can adopt a Hassall corpuscle-like phenotype or exhibit distinctive differentiation characteristics including tuft cells, ionocytes, neuroendocrine cells, and myoid cells.

CIBRA identifies genomic alterations with a system-wide impact on tumor biology

MOTIVATION

Genomic instability is a hallmark of cancer, leading to many somatic alterations. Identifying which alterations have a system-wide impact is a challenging task. Nevertheless, this is an essential first step for prioritizing potential biomarkers. We developed CIBRA (Computational Identification of Biologically Relevant Alterations), a method that determines the system-wide impact of genomic alterations on tumor biology by integrating two distinct omics data types: one indicating genomic alterations (e.g. genomics), and another defining a system-wide expression response (e.g. transcriptomics). CIBRA was evaluated with genome-wide screens in 33 cancer types using primary and metastatic cancer data from the Cancer Genome Atlas and Hartwig Medical Foundation.

RESULTS

We demonstrate the capability of CIBRA by successfully confirming the impact of point mutations in experimentally validated oncogenes and tumor suppressor genes (0.79 AUC). Surprisingly, many genes affected by structural variants were identified to have a strong system-wide impact (30.3%), suggesting that their role in cancer development has thus far been largely under-reported. Additionally, CIBRA can identify impact with only 10 cases and controls, providing a novel way to prioritize genomic alterations with a prominent role in cancer biology. Our findings demonstrate that CIBRA can identify cancer drivers by combining genomics and transcriptomics data. Moreover, our work shows an unexpected substantial system-wide impact of structural variants in cancer. Hence, CIBRA has the potential to preselect and refine current definitions of genomic alterations to derive more nuanced biomarkers for diagnostics, disease progression, and treatment response.

AVAILABILITY AND IMPLEMENTATION

The R package CIBRA is available at https://github.com/AIT4LIFE-UU/CIBRA.

Cell Culture Models for Translational Research on Thymomas and Thymic Carcinomas: Current Status and Future Perspectives

Cell culture model systems are fundamental tools for studying cancer biology and identifying therapeutic vulnerabilities in a controlled environment. TET cells are notoriously difficult to culture, with only a few permanent cell lines available. The optimal conditions and requirements for the ex vivo establishment and permanent expansion of TET cells have not been systematically studied, and it is currently unknown whether different TET subtypes require different culture conditions or specific supplements. The few permanent cell lines available represent only type AB thymomas and thymic carcinomas, while attempts to propagate tumor cells derived from type B thymomas so far have been frustrated. It is conceivable that epithelial cells in type B thymomas are critically dependent on their interaction with immature T cells or their three-dimensional scaffold. Extensive studies leading to validated cell culture protocols would be highly desirable and a major advance in the field. Alternative methods such as tumor cell organoid models, patient-derived xenografts, or tissue slices have been sporadically used in TETs, but their specific contributions and advantages remain to be shown.

Single-Cell Sequencing Illuminates Thymic Development: An Updated Framework for Understanding Thymic Epithelial Tumors

Thymic epithelial tumors (TETs) are rare tumors for which treatment options are limited. The ongoing need for improved systemic therapies reflects a limited understanding of tumor biology as well as the normal thymus. The essential role of the thymus in adaptive immunity is largely effected by its epithelial compartment, which directs thymocyte (T-cell) differentiation and immunologic self-tolerance. With aging, the thymus undergoes involution whereby epithelial tissue is replaced by adipose and other connective tissue, decreasing immature T-cell production. Against this natural drive toward involution, a fraction of thymuses will instead undergo oncologic transformation, leading to the formation of TETs, including thymoma and thymic carcinoma. The rarity of these tumors restricts investigation of the mechanisms of tumorigenesis and development of rational treatment options. To this end, the development of technologies which allow deep molecular profiling of individual tumor cells permits a new window through which to view normal thymic development and contrast the malignant changes that result in oncogenic transformation. In this review, we describe the findings of recent illuminating studies on the diversity of cell types within the epithelial compartment through thymic differentiation and aging. We contextualize these findings around important unanswered questions regarding the spectrum of known somatic tumor alterations, cell of origin, and tumor heterogeneity. The perspectives informed by single-cell molecular profiling offer new approaches to clinical and basic investigation of thymic epithelial tumors, with the potential to accelerate development of improved therapeutic strategies to address ongoing unmet needs in these rare tumors.

The Way Ahead: Lessons Learned from Decades of Cancer Research on Thymomas and Thymic Carcinomas

The history of thymoma (TH) research begins in the early 20th century, when Bell first recognized the epithelial nature of these tumors and their association with myasthenia gravis (MG) [...].

Non-Mutational Key Features in the Biology of Thymomas

Thymomas (THs) are a unique group of heterogeneous tumors of the thymic epithelium. In particular, the subtypes B2 and B3 tend to be aggressive and metastatic. Radical tumor resection remains the only curative option for localized tumors, while more advanced THs require multimodal treatment. Deep sequencing analyses have failed to identify known oncogenic driver mutations in TH, with the notable exception of the GTF2I mutation, which occurs predominantly in type A and AB THs. However, there are multiple alternative non-mutational mechanisms (e.g., perturbed thymic developmental programs, metabolism, non-coding RNA networks) that control cellular behavior and tumorigenesis through the deregulation of critical molecular pathways. Here, we attempted to show how the results of studies investigating such alternative mechanisms could be integrated into a current model of TH biology. This model could be used to focus ongoing research and therapeutic strategies.

Molecular and Functional Key Features and Oncogenic Drivers in Thymic Carcinomas

Thymic epithelial tumors, comprising thymic carcinomas and thymomas, are rare neoplasms. They differ in histology, prognosis, and association with autoimmune diseases such as myasthenia gravis. Thymomas, but not thymic carcinomas, often harbor GTF2I mutations. Mutations of CDKN2A, TP53, and CDKN2B are the most common thymic carcinomas. The acquisition of mutations in genes that control chromatin modifications and epigenetic regulation occurs in the advanced stages of thymic carcinomas. Anti-angiogenic drugs and immune checkpoint inhibitors targeting the PD-1/PD-L1 axis have shown promising results for the treatment of unresectable tumors. Since thymic carcinomas are frankly aggressive tumors, this report presents insights into their oncogenic drivers, categorized under the established hallmarks of cancer.

Somatic mutations of thymic epithelial tumors with myasthenia gravis

Background

Thymic epithelial tumors are rare malignant neoplasms that are frequently associated with paraneoplastic syndromes, especially myasthenia gravis. GTF2I is an oncogene mutated in a subgroup of thymomas that is reputed to drive their growth. However, for GTF2I wild-type tumors, the relevant mutations remain to be identified.

Methods

We performed a meta-analysis and identified 4,208 mutations in 339 patients. We defined a panel of 63 genes frequently mutated in thymic epithelial tumors, which we used to design a custom assay for next-generation sequencing. We sequenced tumor DNA from 67 thymomas of patients with myasthenia gravis who underwent resection in our institution.

Results

Among the 67 thymomas, there were 238 mutations, 83 of which were in coding sequences. There were 14 GTF2I mutations in 6 A, 5 AB, 2 B2 thymomas, and one in a thymoma with unspecified histology. No other oncogenes showed recurrent mutations, while sixteen tumor suppressor genes were predicted to be inactivated. Even with a dedicated assay for the identification of specific somatic mutations in thymic epithelial tumors, only GTF2I mutations were found to be significantly recurrent.

Conclusion

Our evaluation provides insights into the mutational landscape of thymic epithelial tumors, identifies recurrent mutations in different histotypes, and describes the design and implementation of a custom panel for targeted resequencing. These findings contribute to a better understanding of the genetic basis of thymic epithelial tumors and may have implications for future research and treatment strategies.

Histogenetic and disease-relevant phenotypes in thymic epithelial tumors (TETs): The potential significance for future TET classification

Thymic epithelial tumors (TETs) encompass morphologically various subtypes. Thus, it would be meaningful to explore the expression phenotypes that delineate each TET subtype or overarching multiple subtypes. If these profiles are related to thymic physiology, they will improve our biological understanding of TETs and may contribute to the establishment of a more rational TET classification. Against this background, pathologists have attempted to identify histogenetic features in TETs for a long time. As part of this work, our group has reported several TET expression profiles that are histotype-dependent and related to the nature of thymic epithelial cells (TECs). For example, we found that beta5t, a constituent of thymoproteasome unique to cortical TECs, is expressed mainly in type B thymomas, for which the nomenclature of cortical thymoma was once considered. Another example is the discovery that most thymic carcinomas, especially thymic squamous cell carcinomas, exhibit expression profiles similar to tuft cells, a recently discovered special type of medullary TEC. This review outlines the currently reported histogenetic phenotypes of TETs, including those related to thymoma-associated myasthenia gravis, summarizes their genetic signatures, and provides a perspective for the future direction of TET classification.