Structural variation in pediatric cancer: characterizing tumor genomes to improve patient care

Abstract

Cancer is characterized by uncontrolled cell division and widespread disruption of biological processes due to changes in the DNA (mutations). Cancer usually occurs later in life when cells have acquired sufficient 'driver' mutations that contribute to rapid growth and division. Childhood cancer forms an exception, as there are usually few mutations that lead to a stop or 'wrong turn' during tissue development. Certain rearrangements in the structure of the DNA (structural variants, SVs) are characteristic driver mutations for specific types of cancer. However, the effect of the majority of SVs on (cancer) cells is unknown and accurate detection of SVs is challenging. In this thesis, we addressed these issues so that analysis of SVs can be more effectively incorporated in clinical practice. Our work shows that combining DNA and RNA sequencing data enables more accurate detection of SVs that are important for diagnostics. It also provides new insights into the effect of SVs on tumor cells. This improves the usefulness of SVs as biomarkers to divide patients into risk groups (stratification). In addition, we investigated the role of complex rearrangements in childhood cancer. These have been discovered in aggressive forms of adult cancer before. This work shows that complex rearrangements commonly occur across multiple types of childhood cancers and can result in the acquisition of driver mutations. Especially for bone cancer, which has many complex rearrangements, quantifying the instability and complexity of tumor DNA is promising for stratification. In summary, characterizing structural variation in childhood cancer can provide new insights in tumor biology and ultimately improve patient care.