Computational Analysis of Prognostic Factors, Immune Microenvironment, and T-Cell Receptor Repertoires in Lymphoma
Doctoral thesis
Date of Examination:2024-10-25
Date of issue:2024-11-22
Advisor:Prof. Dr. Michael Altenbuchinger
Referee:Prof. Dr. Michael Altenbuchinger
Referee:Dr. Johannes Söding
Files in this item
Name:Dissertation_NicoleSeifert.pdf
Size:21.3Mb
Format:PDF
This file will be freely accessible after 2025-10-23.
Abstract
English
The immune system plays an important role in the development and progression of cancer. While solid cancers like carcinomas primarily affect specific organs and tissues, hematologic malignancies are not always so localized, as they originate from cells in the blood. A hematologic malignancy that specifically affects lymphocytes like T and B cells, which are a crucial part of the immune system, is called lymphoma. Lymphomas are diverse in their pathophysiology and are mainly categorized into Non-Hodgkin lymphoma (NHL) and Hodgkin lymphoma (HL). NHL and HL present unique therapeutic challenges. While some NHL subtypes exhibit aggressive behaviors and have a poor prognosis, HL has generally better outcomes but affects younger populations who suffer from the long-term side effects of traditional therapies. Addressing these challenges requires a comprehensive approach, involving interdisciplinary collaboration of physicians, and laboratories to gather and generate relevant data from patients and computational scientists to analyze and interpret these data. This thesis focuses on the computational analysis of clinical data, genomic and histochemical data, and T-cell receptor (TCR) sequences as part of this collaboration. The main objectives of this thesis were to (i) identify prognostic factors in NHL with central nervous system (CNS) involvement, (ii) investigate the tumor microenvironment (TME) in HL with a focus on human leukocyte antigen class I (HLA-I) downregulation as an immune escape mechanism and T-cell activity following immune checkpoint blockade (ICB), and (iii) develop a user-friendly and efficient tool for the analysis of TCR sequences. To address the first objective, clinical data from 124 patients with SCNSL (secondary CNS lymphoma) treated at five hematological centers were analyzed computationally. The analyses revealed that the initial lymphoma subtype leading to SCNSL significantly influenced patient outcomes. Additionally, patients who received high-dose chemotherapy and autologous stem cell transplantation (HDT-ASCT), and those who reached complete or partial remission early after initial treatments, showed significantly better survival rates. On the other hand, differences in initial treatment strategies showed no significant correlation with patient outcomes. These findings underscore the importance to perform HDT-ASCT and to align future SCNSL treatment strategies on histopathological subtypes and on achieving early treatment responses. For the second objective, gene expression profiles and the cellular composition of the TME in HL patients were analyzed. The analyses revealed that HLA-I expression on Hodgkin Reed-Sternberg cells (HRSCs) correlated with a distinct gene expression profile and specific cellular composition in the TME. HLA-I positive HL cases showed fewer HRSCs and reduced expression of key cytokines, such as CCL17/TARC, compared to HLA-I negative cases. HLA-I positive HL also showed more CD8+ cytotoxic T cells, though with increased expression of immune checkpoint inhibitors like LAG3. Additionally, TCR sequences in tissue biopsies and blood samples of HL patients before and during anti-PD1 ICB were computationally analyzed. The analyses showed that anti-PD1 ICB did not result in significantly increased T-cell activation and clonal expansion, neither in the TME nor in the blood of HL patients. However, CD8+ T cells in the peripheral blood of HL patients showed an overall increase in clonality compared to the TME. The findings strongly suggest that HLA-I expression on HRSCs and the exclusion of clonally expanded T cells from the TME should be considered in future research, especially concerning the development of immunotherapeutic strategies. To address the third objective, TCRanalyzer was developed as a tool to facilitate comprehensive TCR sequence analysis. TCRanalyzer integrates several computational tools and provides a robust and fast solution for analyzing TCR repertoires. TCRanalyzer can quantify TCR diversity metrics, track dynamic changes, and identify clonally expanded T-cell populations. TCRanalyzer was validated through a case study, providing clear insights into T-cell activity and antigen specificity following immunotherapy. The key findings of this thesis highlight the importance of tailored treatment strategies for SCNSL patients, the impact of HLA-I expression on the HL TME, and the surprisingly low T-cell expansion in the HL TME, even after anti-PD1 ICB. The development of TCRanalyzer addresses limitations in traditional TCR analysis tools, offering an efficient and user-friendly solution for analyzing and monitoring T-cell responses. These findings could collectively contribute to the development of more effective and personalized treatment strategies for lymphomas, with the ultimate goal to improve patient outcomes and their quality of life.
Keywords: Lymphoma; Hodgkin; SCNSL; Microenvironment; Immunology; TCR