This evaluation adapts a previous Canadian analysis of upfront and sequential adjuvant AI strategies in postmenopausal women with breast cancer to a Belgian perspective and includes an extended aromatase inhibitor (AI) strategy. A Markov model calculated monthly costs and outcomes in a hypothetical cohort of postmenopausal women with early-stage breast cancer. Baseline event rates and hazard ratios were derived from the Arimidex, Tamoxifen Alone or in Combination trial, International Exemestane Study and MA.17 trials. The analysis took a Belgian healthcare payer perspective with a 20-year time horizon. Costs and outcomes were discounted by 3%. Costs are in 2005 Euros. The cost-utility of all three strategies was favourable (<30,000 euros per QALY gained). Based on indirect comparisons using tamoxifen (TAM) alone as a common comparator, sequential TAM-AI was less costly and more effective than upfront or extended strategies. All three AI strategies were cost-effective alternatives to TAM alone, but sequential TAM-AI appears to be the economically preferred strategy.