Focusing on the HIV-1 subtype B epidemic, the most frequent HIV variant in Brazil, we extend our work on HIV modeling and propose a practical decision-making framework for the design of individualized HIV-1 drug chemotherapy strategies relying on patient-specific genotyping test data. The methodology is applied to 9 real-world problem instances regarding a set of 15 antiretroviral drugs and a set of 62 amino acid mutations along 45 positions of the HIV-1 pol gene. Optimized strategies are compared with current guidelines from the Brazilian standard HIV/AIDS treatment protocol. Computational results show that distinct optimal treatment strategies are produced for different patients and we conclude that relevant opportunities are to be promptly exploited on an individual basis through a better management of HIV-1 mutations in the long-term, which can be achieved by frequently changing the HAART scheme over time in response to and in anticipation of the emergence of drug-resistant strains. (C) 2013 Elsevier Ltd. All rights reserved.