Kaja Christine Graue Berg defended her PhD on the 10th Dec 2019
She has performed her work in the Lothe lab., and has been a PhD student at the Institute for Clinical Medicine, Faculty of Medicine, U of Oslo.
The thesis “Integrated genomics of colorectal cancer” included three papers of which she is the 1st author (Molecular Cancer 2017, Oncogene 2019, submitted manuscript).
The candidate held an excellent trial lecture “The gut microbiome as an emerging target for cancer therapy”, and she had a good discussion with her two opponents professors Karin Jirstrøm and Eirik Frengen (photo, candidate to the right).
Brief summary of the work:
In this work, Kaja C. G. Berg and colleagues have performed integrated multi-level molecular analyses of primary and metastatic CRCs, as well as pre-clinical CRC models, to improve the molecular stratification of the disease. In cell lines they have shown that the primary transcriptomic program, which is related to cancer cell differentiation, is consistent on the gene-, microRNA-, and protein expression levels. Together with delineation of the effects of mutations on the expression level, such comprehensive molecular profiles of commonly used pre-clinical cancer model systems provide an important resource to the research community.
By integrated DNA copy number and gene expression analyses of primary tumors, the researchers identified novel high-level amplifications, beyond the targetable ERBB2 amplification, with a strong impact on target gene expression. These amplifications were largely mutually exclusive and associated with a poor patient outcome. However, the correlation observed between DNA copy numbers and gene expression varied among the gene expression-based consensus molecular subtypes (CMS). The epithelial-like CMS2 tumors were driven by DNA copy number gains to a much larger extent than remaining CMS groups, suggesting a novel genetic basis for this transcriptomic class.
Surgery is a potentially curative treatment for patients with CRC liver metastases, but there are currently no good patient selection criteria, and only one third of the patients have a long-term survival benefit. Berg and colleagues showed that combined mutational analyses in a tumor heterogeneity setting may improve the prognostic stratification of the patients. Specifically, co-occurring mutations in RAS and TP53 were associated with a poor survival. Analyses of intra-patient inter-metastatic heterogeneity on the DNA copy number level further improved this prognostic stratification, and identified a subgroup of patients with little benefit from invasive surgical treatment.