OBJECTIVE: CPCRA-042/CTN-02, was a binational, randomized, open label trial in patients with advanced HIV receiving either Nelfinivir (NLF) or Ritonavir (RTV). A pharmacoeconomic (PE) sub-study of consenting Canadian participants prospectively captured health resource utilization (HRU) and Quality of Life (QOL) data. Our objective was to assess costs, effects and incremental cost-effectiveness of NLF compared to RTV as Protease Inhibitor therapy. METHOD: The PE sub-study recruited from 13 Canadian sites. Data collected included HRU and QOL as measured by a VAS and SF12 every 4 months. Costs were estimated using the St. Paul's Hospital (SPH) formulary, SPH Cost Model, and BCMA fee schedule. Using intent to treat analysis, the annual incremental cost effectiveness ratio (ICER) was calculated. RESULTS: In the main study, there was no difference in the time to clinical progression, immunologic or virologic responses between the two study arms. In PE sub-study, 137 patients were randomized: NLF (n=71) or RTV (n=66). The median (Q1-Q3) baseline patient age was 38 years (33-44), with median CD4+ count of 36/mm3 (12-70). Total follow-up time was >=3 years. Preliminary results show mean (SD) first year annual total cost for NFV patients to be $26,099 (14,800) and $20,475 (7591) for RTV patients; p<0.001. QOL scores showed no significant difference among groups at one year. The number of patients switching initially assigned study drug due to toxicity was lower for the NLF group 19 (27%) vs. 26 (39%) at one year and 8 (12%) vs. 24 (36%) in the first 8 months. The annual ICER per switch avoided equaled $24,071 per patient. CONCLUSION: The overall total cost and tolerability with assigned therapy were both higher for NLF. Given equal efficacy and immunologic response, the choice for one drug over the other as initial therapy depends on the importance placed on tolerability of the start-up regimen and potential for the emergence of drug resistance.