Due to copyright issues, an electronic copy of the thesis must be ordered from the faculty. For the faculty to have time to process the order, the order must be received by the faculty at the latest 2 days before the public defence. Orders received later than 2 days before the defence will not be processed. After the public defence, please address any inquiries regarding the thesis to the candidate.
Trial Lecture – time and place
See Trial Lecture.
Adjudication committee
- First opponent: Assistant Professor Javier Antonio Alfaro, University of Gdansk, Poland
- Second opponent: Associate Professor Marcela Dávila, University of Gothenburg, Sweden
- Third member and chair of the evaluation committee: Professor II Vessela Nedelcheva Kristensen, University of Oslo
Chair of the Defence
Associate Professor Victor Greiff, University of Oslo
Principal Supervisor
Scientific Officer Trevor Clancy, NEC OncoImmunity AS
Summary
The accumulation of somatic mutations in tumor cells leads to the display of immunogenic neoantigens bound to HLA molecules on their surface, enabling T-cells to selectively eradicate them. However, the very existence of cancer means that this immune response can, at times, fail. Advanced bioinformatics and machine learning (ML) pipelines leveraging next generation sequencing (NGS) data have facilitated antitumor neoantigen-based personalized immunotherapy, where tumor’s immunogenicity is restored by therapeutically guiding T-cells against HLA-neoantigen bearing tumor cells. Moreover, neoantigens' value as biomarkers is evident from their robust association with immune checkpoint inhibitor (ICI) therapy response. While clinical trials are encouraging, the computational prediction of neoantigens from NGS data remains challenging, requiring precise detection of numerous patient-specific somatic mutations and HLA molecules. Yet, current bioinformatics tools are hindered by the complexity of cancer sequencing, particularly tumor heterogeneity and HLA polymorphism.
This project presents novel bioinformatics pipelines leveraging whole exome and RNA sequencing data to address some of the current computational shortfalls in somatic variant calling and HLA typing, ultimately reinforcing the prediction of patient-specific neoantigens for their use as immunotherapeutic agents and/or biomarkers.
- We developed a somatic variant calling computational framework integrating multiple variant callers, sequencing features, and supervised ML algorithms. Its superior capacity to detect low-frequency variants could aid the prediction of neoantigens targeting all tumor subclonal populations, potentially reducing the risk of therapy resistance and recurrence.
- HLA alleles encode for HLA molecules which might bind neoantigens with varying affinities. Misidentifying HLA alleles or targeting neoantigens not displayed on the tumor cell surface due to HLA downregulation could render neoantigen-based immunotherapy futile. We developed a pioneering HLA typing computational pipeline to infer patient-specific novel HLA allele sequences and profile tumor HLA mutations and expression by integrating HLA-allele-specific germline and somatic variants. The identification of the officially named HLA-B*44:02:01:52 allele showcased the pipeline's ability to uncover novel alleles.
- We assessed the clinical relevance of the introduced bioinformatics pipelines by studying the influence of neoantigens and immune cell infiltration in metastatic sarcoma patients undergoing ICI therapy. The results revealed a positive synergistic impact on survival between neoantigens and T-cells, along with patterns of immune evasion linked to HLA downregulation.
Additional information
Contact the research support staff.