Preview

PDF, English Download (3MB) | Terms of use

Abstract

This thesis combines the use of novel radiation qualities with in-vivo functional genetic screens in prototypic head and neck squamous cell carcinoma (HNSCC, FaDu) and non-small cell lung cancer (NSCLC, A549) preclinical xenograft models. Using whole-genome FaDu and A549 CRISPR/Cas9 libraries, in-vivo tumor response to conventional photon radiotherapy as well as a series of particles with gradually enhanced ionization density (LET), i.e., proton, helium, carbon- and oxygen ion irradiation were investigated. Two genome wide - GeCKO A and Brunello – gRNA libraries were integrated in Cas9 expressing FaDu and A549 cells. Transcriptome and guide representation studies using next-generation sequencing (NGS) and quantitative analysis provided insights into the tumor evolutionary landscape and genes/pathways contributing to tumor fitness under different in-vivo and therapeutic selection pressures. A pronounced reduction of library complexity was detected in-vivo vs in-vitro, regardless of tumor model and library. Additional reduction was observed in irradiated samples. Clonal tumor evolution could be recapitulated, as outgrown tumors showed an expansion of few clones constituting the majority of the tumors. Tumor growth rate was negatively linked to library complexity – a potential surrogate for intratumoral heterogeneity – strengthening the previously described association between high intratumoral complexity and increased tumor aggressiveness. High-LET irradiation showed improved capacity for cell killing, with either tumor regression (in FaDu model) or enhanced growth delay (in A549 model). Irradiation type specific transcriptome fingerprints revealed both favorable- and non-favorable changes, comprising modulation of the tumor microenvironment (TGF-β), inflammation and senescence, pro- and anti-angiogenic alterations, DNA damage repair, hypoxia and antigen presentation. Main factors influencing transcriptome profiles differences within particles were LET (high vs. low LET groups). On guide level, radio-resistance seems to be mediated by loss of TGF-β signaling components (BMP8A), chromatin remodeling (KDM5C) and metabolic pathways (MDH2). LET dependency analyses also identified TGF-β signaling as key component, cell-death mediations (BAD, BNIP3) and tumor suppressors (ARID2). Common pattern of upregulation across both tumors and libraries was a gradual LET dependent increase of genes associated immune signaling processes (HLAs, interferon response, MX1, etc.) which correlated well with enrichment of pathways found to be related to immune response. Together, this work provides novel insights into the molecular fitness landscape of tumors under conventional photon irradiation and ion beam therapy.