Physicians' behavior influences the health and economic impact of applying circulating tumor cells as response marker in metastatic castration-resistant prostate cancer

Objectives: Treatment decisions in metastatic castration-resistant prostate cancer (mCRPC) vary between physicians, even though expert opinion guidelines exist to guide the interpretation of information from multiple time-dependent imaging modalities and markers. We aimed to investigate whether physicians' behavior, regarding the use and interpretation of imaging modalities and markers, influences the health and economic impact of novel biomarkers, such as circulating tumor cells (CTC). Methods: A previous developed model on the use of response markers for informing treatment switches from ineffective docetaxel treatment to cabazitaxel treatment in mCRPC was used. This model compares two strategies for monitoring patients' response to treatment: the use of PSA and bone scans (control strategy) and the use of CTC (experimental strategy). The likelihood that a physician would follow guideline-suggested treatment switches was varied (60%- 100%) to assess its impact on several outcome measures, including the amount of overtreatment and the incremental cost-utility ratio. Results: The results show that the physicians' behavior influences both the health and economic impact of applying CTC as response marker in mCRPC treatment. Average docetaxel overtreatment varied between 19.7 and 21.5 weeks and between 10.3 and 12.5 weeks for the control and experimental strategy, respectively. Average cabazitaxel overtreatment varied between 10.6 and 11.2 weeks and between 6.5 and 7.6 weeks for the control and experimental strategy, respectively. The cost-utility of CTC as response marker varied between € -14,421.- and € 56,753.- per QALY gained. Conclusions: Physicians' behavior is expected to have a large impact on the performance of CTC as response markers in mCRPC treatment with regard to both health and economic outcomes. This indicates that also the extent to which patients benefit from health care innovations depends on physicians' preferences regarding the use of those interventions. Therefore, physicians and modelers should be aware of the impact that physicians' behavior might have in practice.