A cost-effectiveness analysis of surveillance, radiotherapy, and carboplatin in the management of stage I testicular seminoma

Abstract

6108 Background: Alternative post-orchiectomy management strategies for patients with Stage I Testicular Seminoma (TS) have different costs, recurrence risks, and toxicities. This study compares the cost-effectiveness (CE) of surveillance (SVL), dogleg radiotherapy (DLRT), para-aortic radiotherapy (PART), or one cycle of Carboplatin (CARB) from a societal perspective. Methods: Using TreeAge software, a decision tree with embedded Markov process models was constructed to simulate the natural history of Stage I TS from orchiectomy until death for hypothetical cohorts of 35 year-old men. Recurrence risks, established follow-up guidelines, utilities, direct and indirect costs, and excess mortality and costs attributable to late second cancers were modeled. CE and incremental CE (ICER) ratios were calculated per life-year (LY) and quality-adjusted life year (QALY) using a 3% discount rate; one- and two- way sensitivity analyses were performed. Results: The base case results are shown in the Table . SVL is CE when both recurrence risk is decreased 50% and follow-up costs are reduced 65%. PART is CE when follow-up and indirect costs are reduced 29% while PART second cancer relative risk (RR) is reduced to 1.8. CARB is CE when follow-up and indirect costs are reduced 41%, or the PART relapse risk function is increased 41% while DLRT costs are increased 87%. The model was insensitive to changes in utilities. Conclusions: Despite an assumed increased second cancer risk, DLRT is CE. This is due to the routine use of intensive follow-up for the alternative strategies and the high cost of salvage treatment. PART may become CE if a decrease in second cancer RR is observed without increases in recurrences. SVL should be reserved for those with favorable prognostic factors who can be followed less intensely. At a reasonable willingness-to-pay ceiling CARB is optimal, however, the small differences in effectiveness between the strategies may support decision-making based on costs. [Table: see text] No significant financial relationships to disclose.