OBJECTIVES: Most economic evaluations of chemotherapies for ovarian cancer patients have used hypothetical cohorts or randomized control trials, but evidence integrating real-world survival, cost, and utility data is limited.
METHODS: A propensity score-matched cohort of 6856 elderly (greater than or = 65 years) ovarian cancer patients diagnosed from 1991 to 2005 from the Surveillance, Epidemiology, and End Results-Medicare data cohort were included. Treatment regimens (i.e., no chemotherapy, platinum-based only, platinum plus taxane, and other nonplatinum) were identified in the 6 months postdiagnosis. Patients were followed until death or end of study (December 2006). Effectiveness was measured in quality-adjusted life-years (QALYs), and total health care costs were measured by using a payer's perspective (2009 US dollars). Methodological and statistical uncertainties were accounted by including alternative scenarios (for utility values) and net monetary benefit approach. Incremental cost-effectiveness ratios (ICERs) were calculated, and stratified analyses were performed by tumor stages and age groups.
RESULTS: On comparing the platinum-based group versus no chemotherapy, we found that the ICER was $30,073/QALY and $58,151/QALY for early- and late-stage disease, respectively, while other nonplatinum and platinum plus taxane groups were dominated (less effective and more costly). Similar results were found across alternative scenarios and age groups. For patients 85 years or older, platinum plus taxane, however, was not dominated by the platinum-based group, with an ICER of $133,892/QALY.
CONCLUSIONS: Following elderly ovarian cancer patients over a lifetime using real-world longitudinal data and adjusting for quality of life, we found that treatment with platinum-based regimen was the most cost-effective treatment alternative.
