OBJECTIVES: First-line antiretroviral therapy selection provides the greatest opportunity to fully suppress HIV replication and prevent the emergence of drug-resistant strains that lead to treatment failure and compromise future drug treatment strategies. Currently, two alternative protease inhibitors PIs, lopinavir (LPV) and atazanavir (ATV), are included in the Colombian national formulary for the treatment of HIV patients. Darunavir (DRV) boosted with low-dose ritonavir has demonstrated to be an efficacious PI alternative for treatment of experienced and naïve patients. The objective of this study is to compare costs and benefits among different PI sequential algorithms of DRV, ATV and LPV from the Colombian Health Care System. METHODS: A decision tree was designed using the insurer perspective to estimate costs and benefits throughout a three years time horizon. Clinical data were obtained from the pivotal trials in order to model second line and rescue progression and benefits were measured as probability of virological response (VR). Only direct costs were considered, such as medications (first line, second line and rescue medications), laboratory tests and rescue related inpatient care costs using national tariffs and prices from Ministry of Health medication database. Four PI schemes were assessed: 1) DRV followed by ATV, 2) ATV followed by DRV, 3) LPV followed by ATZ and 4) LPV followed by DRV. Discount rate 3% and exchange rate (1 USD = 1,794 COP). RESULTS: Direct costs (USD):  DRV-ATV (29,747), ATV-DRV (30,191), LPV-ATV (24,507) and LPV-DRV (24,157). VR: DRV-ATV (51%), ATV-DRV (49%), LPV-ATV (43%) and LPV-DRV (43%). Cost per VR (USD / VR): DRV-ATV (58,327), ATV-DRV (61,614), LPV-ATV (56,993) and LPV-DRV (56,179). CONCLUSIONS: Using DRV after LPV as a first line agent is the sequencing strategy which demonstrates the better cost per virological response. DRV in first line position save costs and improves VR versus ATV in first line position.