OBJECTIVES: To evaluate the cost-effectiveness of primary and secondary prophylaxis (PP and SP) with pegfilgrastim, filgrastim (6-day and 11-day), or no prophylaxis for decreasing the incidence of febrile neutropenia (FN) in recurrent ovarian cancer patients receiving 3 cycles of myelosuppressive chemotherapy (docetaxel every 21 days 100 mg/m2, FN risk ≥20%) over a lifetime horizon from a U.S. payer perspective. METHODS: A Markov cycle tree model tracks FN events in chemotherapy cycles 1-3 (3-week cycles) and long-term survival (1-year cycles).  Long-term survival is modeled according to received relative dose intensity (RDI), which correlates with FN history. Model inputs, including efficacy (relative risk of FN compared to no prophylaxis) of each strategy, effects of FN on RDI, mortality, costs, and utilities, were estimated from public sources, research databases, and peer-reviewed publications.  Expected lifetime costs and quality-adjusted life-years (QALYs) (discounted 3% per year) were estimated for each strategy. Probabilistic sensitivity analyses (PSA) were conducted. RESULTS: The incremental cost-effectiveness ratio (ICER) for PP vs. SP with pegfilgrastim was $15,946 per QALY gained. Total lifetime costs for PP and SP with pegfilgrastim were $13,034 and $12,704, respectively, while lifetime QALYs were 0.916 and 0.895.  PP with pegfilgrastim dominated all other comparators.  Although ICER results of PP vs. SP with pegfilgrastim may be affected by changes in baseline FN risk, pegfilgrastim efficacy, pegfilgrastim cost, and cost of FN-related hospitalization, PP was cost-effective compared to SP in 95% and 100% of PSA simulations at thresholds of $50,000 and $100,000 per QALY, respectively.  CONCLUSIONS: From a U.S. payer perspective, PP with pegfilgrastim is cost-effective compared to other prophylaxis strategies in recurrent ovarian cancer patients receiving docetaxel.