Sir Peter MacCallum Department of Oncology - Theses
Permanent URI for this collection
Search Results
Now showing
1 - 1 of 1
-
ItemNo Preview AvailableDevelopment of new methods for accurate estimation of tumour heterogeneityHollizeck, Sebastian ( 2022)It is now understood that intra-tumor heterogeneity is one of the leading determinants of therapeutic resistance and treatment failure and one of the main reasons for poor overall survival in cancer patients. However, the possibility to study this phenomenon is so far underexplored as the acquisition of multi region data sets from the different tumour sites can be ethically challenging. With circulating tumour DNA (ctDNA) used as a proxy for tumour biopsies, it is possible to analyse a snapshot of the unified heterogeneity in each patient, but there is still an unmet need for new methods to optimize the analysis of these large-scale, high-dimensional data to derive new treatment targets. The contributions of this work include the development of multiple new methods, which show that the analysis of bulk sequencing from tumour tissue and ctDNA has unrealised potential for both diagnostic and research questions. This thesis presents three distinct but related projects, which explore the analysis of tumour heterogeneity at different levels and depths, focusing on method development. First, we developed a workflow to improve the detection of somatic variants present at very low allele frequencies. When multiple samples, separated in time or space, from the same patient were available, we were able to substantially improve the detection threshold of variants. These low abundance variants are invaluable in a clinical setting, where they can indicate an arising resistance mechanism or relapse of disease. With the improved sensitivity of our method, the treatment of patients can be adjusted earlier and more accurately. We then used our new analysis workflows to explore evolutionary trajectories and resistance pathways of five lung cancer patients enrolled in the CASCADE autopsy program. In addition to analysis of somatic variants, we used copy number analysis and structural variants to contrast and compare each sample within a patient to generate phylogenies to visualise the evolutionary distances and a pseudo time scale to assess the timing of mutations. Clear genomic determinants of treatment resistance were identified for three of the five cases with non-small cell lung cancer and the diversity of these genomic mechanisms profoundly highlighted the true extent of inter-patient heterogeneity. This work included the identification of a novel genomic resistance mechanism to the drug selpercatinib, a small molecule inhibitor of REK kinase. Among the remaining two patients, treatment resistance was mediated by transformation of their disease from non-small cell lung cancer to a small cell lung cancer histological phenotype. These two cases showed distinct evolutionary trajectories compared to the other non-small cell lung cancer cases, with similarity in their nuclear and mitochondrial phylogenies, but no clear genetic determinant for the small cell transformation, highlighting the additional importance of non-genomic mechanisms which can drive resistance in this disease. Finally, we developed a method, called MisMatchFinder, to monitor tumour heterogeneity and evolution over time through ctDNA. We tailored the method to be fully tumour agnostic and enabled it to be easily applied in the clinical setting by using low-coverage whole genome sequencing. The method uses highly specialised filtering steps to enrich the tumour signal and eliminate the background noise from normal cell-free DNA and sequencing errors in these data. We showed that the method could accurately detect specific cancer-related signatures at low tumour purity and tumour burden in simulated and patient data for melanoma and breast cancer. In summary, with this work we contributed multiple new methods to study, measure and understand genetic tumour heterogeneity. This understanding is crucial for the continuous optimisation of cancer management and the development of new and effective treatments for patients.