Cost-utility analysis of 177Lu-PSMA-617 radioligand therapy in second-line and third-line treatment for metastatic castration-resistant prostate cancer (mCRPC) in Germany
Carolin Brinkmann, Richard P. Baum, Tom Stargardt
Abstract
Abstract Purpose To evaluate the cost-effectiveness of 177 Lu-PSMA-617 radioligand therapy (PRLT) in metastatic castration-resistant prostate cancer (mCRPC) in Germany by comparing (I) PRLT plus standard-of-care (SoC) versus SoC alone as third-line treatment and (II) PRLT versus second-line cabazitaxel chemotherapy. Methods Cohort state-transition models were developed with (I) four health states (treatment, stable after treatment, progression, death) and (II) six health states (treatment, stable after treatment, third-line treatment after progression, stable after third-line treatment, next progression, death). Transition probabilities were derived from the VISION and TheraP trials, and quality-of-life data from the VISION and CARD trials. Costs were derived from statutory health insurance claims data (2019–2022). Models simulated a five-year horizon with one-month cycles, applying within-cycle correction and a 3% discount rate. Sensitivity analyses addressed uncertainty. Results For Model I, PRLT plus SoC compared to SoC incurred incremental costs of €27,200 per patient, with a gain of 0.39 quality-adjusted life years (QALYs) and incremental cost-effectiveness ratio (ICER) of €69,418 per QALY gained. In Model II, PRLT plus SoC compared to cabazitaxel achieved incremental savings of €1,460 per patient and a QALY gain of 0.11, making it the dominant option. Conclusion Although Germany lacks an explicit willingness-to-pay threshold for interpreting the ICER of Model I, it falls within the range of other reimbursed cancer therapies. This suggests PRLT is cost-effective as second- or third-line treatment for mCRPC. Between 2019 and 2022, hospitals mainly used self-produced 177 Lu-PSMA-617, which was less costly than the product now commercially available, limiting the generalizability of our findings.