Lifetime effectiveness of mifamurtide addition to chemotherapy in nonmetastatic and metastatic osteosarcoma: a Markov process model analysis.

Abstract

The mortality and progression rates in osteosarcoma differ depending on the presence of metastasis. A decision model would be useful for estimating long-term effectiveness of treatment with limited clinical trial data. The aim of this study was to explore the lifetime effectiveness of the addition of mifamurtide to chemotherapy for patients with metastatic and nonmetastatic osteosarcoma. The target population was osteosarcoma patients with or without metastasis. A Markov process model was used, whose time horizon was lifetime with a starting age of 13 years. There were five health states: disease-free (DF), recurrence, post-recurrence disease-free, post-recurrence disease-progression, and death. Transition probabilities of the starting state, DF, were calculated from the INT-0133 clinical trials for chemotherapy with and without mifamurtide. Quality-adjusted life-years (QALY) increased upon addition of mifamurtide to chemotherapy by 10.5 % (10.13 and 9.17 QALY with and without mifamurtide, respectively) and 45.2 % (7.23 and 4.98 QALY with and without mifamurtide, respectively) relative to the lifetime effectiveness of chemotherapy in nonmetastatic and metastatic osteosarcoma, respectively. Life-years gained (LYG) increased by 10.1 % (13.10 LYG with mifamurtide and 11.90 LYG without mifamurtide) in nonmetastatic patients and 42.2 % (9.43 LYG with mifamurtide and 6.63 LYG without mifamurtide) in metastatic osteosarcoma patients. The Markov model analysis showed that chemotherapy with mifamurtide improved the lifetime effectiveness compared to chemotherapy alone in both nonmetastatic and metastatic osteosarcoma. Relative effectiveness of the therapy was higher in metastatic than nonmetastatic osteosarcoma over lifetime. However, absolute lifetime effectiveness was higher in nonmetastatic than metastatic osteosarcoma.