The growing value of real-world evidence: From exploring the many potential uses of real-world data to improving its efficient use within the field of oncology

Abstract

With the transition from paper medical records to electronic medical records, data on the complete patient hospital journey became more accessible, reliable and detailed. These developments made data collected in routine clinical practice better accessible for research. This led to a growing interest in real-world evidence, which can play an essential role in improving treatments of future patients. Therefore, this thesis aimed to explore different uses of real-world data (RWD) and to facilitate its efficient use in the field of oncology. Part I focused on the effectiveness of treatments in clinical practice. In Chapter 1, the representativeness of a phase III trial for osimertinib in pretreated advanced EGFR-mutated non-small-cell lung cancer (NSCLC) patients was investigated. In Chapter 2, the treatment sequences and survival outcomes in advanced HR+ HER2- breast cancer patients were investigated. Part II focused on the toxicity of treatments in clinical practice. In Chapter 3, the exposure-toxicity relationship of kinase inhibitors was studied to determine whether high plasma trough concentrations of kinase inhibitors used in NSCLC were related to an increased toxicity risk. Chapter 4 described the incidence of immune-related adverse events (irAEs) of immune checkpoint inhibitors leading to admission to the intensive care unit. Part III focused on the treatment of obese patients, in which the effectiveness and toxicity of treatments were compared to lean patients. Chapter 5 investigated the influence of obesity on the tolerability and effectiveness of full-weight-based paclitaxel chemotherapy in women with early-stage breast cancer. Chapter 6 described the influence of obesity on the plasma trough concentrations of oral targeted oncolytics exhibiting an exposure-response relationship. Part IV focused on methodological approaches to improve the use of RWD. In Chapter 7, the use of joint models for exposure-response analyses of oral targeted anticancer agents was explored and compared to conventional exposure-response analysis methods with the application of alectinib in NSCLC patients as proof-of-concept. Chapter 8 described a systematic review investigating whether RWD-derived control arms can act as appropriate and reliable control arms for single-arm clinical trials. In Chapter 9, an exploratory study was performed in which the use of clinical text mining was investigated to facilitate data collection of RWD from electronic medical records, as manual data collection is highly labor intensive. In summary, this thesis explored a variety of different oncology related research questions that could be answered using RWD. In addition, this thesis investigated methodological approaches to improve the use of RWD. There are still several barriers preventing the optimal use of the immense amount of valuable information present in electronic medical records. Therefore, the full potential of RWE has yet to be achieved. Future research should continue to exploit the endless possibilities of RWD and use it to guide the best treatment option for individual patients.