OBJECTIVES: Approximately 8.33 years of randomized clinical trial data is now available from the Arimidex, Tamoxifen Alone or in Combination (ATAC) trial, confirming that the advantage of anastrozole in recurrence free survival compared with tamoxifen in hormone receptor-positive (HR+) postmenopausal (PM) early breast cancer (EBC) patients is preserved even after stopping treatment at 5 years. Using the long-term outcome data, this study estimated the cost-utility of anastrozole versus tamoxifen from a German health service perspective. METHODS: A semi-Markov model was adapted to include the 100-month clinical trial efficacy data. Kaplan-Meier curves on time to recurrence (TTR) from the ATAC trial were fitted by survival regression including a treatment coefficient. Treatment specific progression rates were extrapolated to 10 years, and were assumed equivalent based on pooled treatment arms thereafter. Survival probabilities after recurrence were identical for all treatments, based on German national mortality data.  Published literature and physician survey provided data on resource use and unit costs and utilities. Probabilistic analysis was carried out to obtain incremental cost-effectiveness ratios and cost-effectiveness acceptability curves. Sensitivity analyses were conducted. RESULTS: Over 25 years anastrozole patients gained 0.32 QALYs at a cost of €6,819 per patient, resulting in an ICER of €21,069 (95%CI: €12,567-€46,604) per QALY gained, vs tamoxifen. With a >90% probability the ICER was below €30,000/QALY. The results were robust to plausible changes in parameters tested in sensitivity analyses, including the hazard ratio for TTR. CONCLUSIONS: This is the first cost-utility analysis of an aromatase inhibitor based on extended follow-up data. Anastrozole’s carryover treatment effect beyond therapy completion at 5 years translates into favourable cost-effectiveness results against tamoxifen for PM HR+ EBC women from a German national health insurance perspective.