The aim of this thesis is to find a novel liquid biomarker for the detection of cancer and to optimize treatment. The first chapter gives an introduction to the oncology biomarker field and focuses on platelets and their role in cancer. In part 1, we evaluate extracellular vesicles (EVs). EVs are small vesicles released by all types of cells, including tumor cells, into the circulation. They carry protein kinases and can be isolated from plasma. We demonstrate that AKT and ERK kinase protein levels in EVs reflect the cellular expression levels and treatment with kinase inhibitors alters their concentration, depending on the clinical response to the drug. Therefore, EVs may provide a promising biomarker biosource for monitoring of treatment responses. Part 2 starts with reviews describing the function and role of platelets in greater depth. Chapter 3 focusses on thrombocytogenesis and several biological processes in which platelets play a role. Furthermore, the RNA processing machineries harboured by platelets are discussed. Both chapter 3 and 4 evaluate the change platelets undergo after being exposed to tumor and its environment. The exchange of biomolecules with tumor cells results in educated platelets, so-called tumor educated platelets (TEPs). TEPs play a role in several hallmarks of cancer and have the ability to respond to systemic alterations making them an interesting biomarker. In chapter 5 the diagnostic potential of platelets is first discussed. We determine their potential by sequencing the RNA of 283 platelet samples, of which 228 are patients with cancer, and 55 are healthy controls. We reach an accuracy of 96%. Furthermore, we are able to pinpoint the location of the primary tumor with an accuracy of 71%. In part 3, our developed thromboSeq platform is taken to the next level. Several potential confounding factors are taken into account such as age and comorbidity. We show that particle-swarm optimization (PSO)-enhanced algorithms enable efficient selection of RNA biomarker panels. In a validation cohort we apply these algorithms to non-small-cell lung cancer and reach an accuracy of 88% in late stage (n=518) and early-stage 81% accuracy. Finally, in chapter 7 we describe our wet- and dry-lab protocols in detail. This includes platelet RNA isolation, mRNA amplification, and preparation for next-generation sequencing. The dry-lab protocol describes the automated FASTQ file pre-processing to quantified gene counts, quality controls, data normalization and correction, and swarm intelligence-enhanced support vector machine (SVM) algorithm development. Part 4 focuses on central nervous system (CNS) malignancies especially on glioblastoma. Chapter 8 gives an overview of the different liquid biomarkers for diffuse glioma, the most common primary CNS malignancy. In chapter 9 we assess the specificity of the platelet education due to glioblastoma by comparing the RNA profile of TEPs from glioblastoma patients with a neuroinflammatory disease and brain metastasis patients. This results in a detection accuracy of 80%. Secondly, analysis of patients with glioblastoma versus healthy controls in an independent validation series provide a detection accuracy of 95%. Furthermore, we describe the potential value of platelets as a monitoring biomarker for patients with glioma, distinguishing pseudoprogression from real tumor progression. In part 5 thromboSeq is applied to breast cancer diagnostics both as a screening tool in the general population and in a high risk population, BRCA mutated women. In chapter 11 we first apply our technique to an inflammatory condition, multiple sclerosis (MS). Platelet RNA is used as input for the development of a diagnostic MS classifier capable of detecting MS with 80% accuracy in the independent validation series. In the final part we conclude this thesis with a general discussion of the main findings and suggestions for future research.
|Qualification||Doctor of Philosophy|
|Award date||21 Sep 2021|
|Publication status||Published - 22 Sep 2021|