PhD defence M.S.C. (Marscha) Verhulst-Holleman

Dissertation in short:

Increasing healthcare expenditures induces great pressure on countries’ public spending. Since the financial resources are limited, healthcare should be kept affordable. Therefore, healthcare decision makers should decide what healthcare treatments should be reimbursed. Decision makers should be informed with evidence on the (cost-)effectiveness of healthcare treatments. This thesis aimed to provide evidence on new treatments in lung and prostate cancer to inform healthcare decision making. A network meta-analysis (NMA) of five epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) for EGFR mutation-positive non-small cell lung cancer (NSCLC) considered osimertinib as most favourable first-line EGFR-TKIs. Cost-effectiveness analysis of four first-line EGFR-TKIs for EGFR mutation-positive NSCLC showed an incremental cost-effectiveness ratio of €128,343/QALY for osimertinib versus afatinib. Only when the type of risk-sharing arrangements (RSAs) matches the setting and type of uncertainty, RSAs may be useful to mitigate uncertainty around the cost-effectiveness of treatments. Real-world costs of castration-resistant prostate cancer (CRPC) treatments in the Netherlands were €67,174 per patient. 
A patient-level simulation model based on real-world data (RWD) lacked internal validity, which might be induced by unobserved differences between treatments. Further research should explore the use of a combination of randomised controlled trials (RCT) and RWD. This thesis showed that NMAs are useful to assess the (cost-)effectiveness of multiple treatments. RSAs could be used to alleviate the uncertainty around the cost-effectiveness of treatments. RWD could provide information on the costs and effects of patients in clinical practice. However, due to the limitations of RWD, RWD combined with RCT should be explored.