Cost-Effectiveness of Proton Versus Photon Therapy for Medulloblastoma Using Updated Clinical Outcomes

The benefit of proton beam therapy (PBT) is evident in pediatric cancer, as survivors have to face long-term radiation-related side effects and lifetime costs. However, PBT's higher costs cause a challenge especially in middle-income countries. This study analyzed cost-effectiveness of PBT versus photon beam therapy (XRT) for medulloblastoma based on updated clinical outcomes within middle-income country context.

Markov and Monte Carlo models were constructed for simulation analysis. We used annual mortality rates based on actual medulloblastoma patients receiving craniospinal irradiation over the first 5 years, combined with general population mortality rates thereafter. Rates and utilities for side effects, including hormone deficiencies, hearing loss, brain tumors, thyroid cancer, and IQ decline, were derived from previous literature, while the cardiac mortality rate was estimated using a linear model. Medical and nonmedical costs, along with other economic parameters and mortality rates during the first 5 years, were derived within the context of a middle-income country, using Thailand as a representative setting where PBT is operational.

For the base-case model, the total costs for PBT and XRT were 69 349 USD (2 357 874 THB) and 80 217 USD (2 727 366 THB), respectively, with total utilities of 36.12 and 35.23, respectively. The ICER was -12 211 USD (-415 160 THB)/QALY, and the net monetary benefit for PBT and XRT were 100 627 USD (3 421 326 THB) and 85 572 USD (2 909 434 THB), respectively. Including microsimulation and probabilistic sensitivity analysis for side effects rates, mortality rates, inflation, cost reduction, and utility reduction, PBT was still cost-effective. However, XRT was found to be more cost-effective when excluding opportunity loss from IQ decline.

Despite its higher cost, PBT was more cost-effective than XRT for treating medulloblastoma in a middle-income country, based on updated clinical outcomes, particularly when incorporating IQ decline into the analysis.