Investigation of the clinical prognostic value for therapy failure of HIV-1 drug resistance results and its interaction with other factors, using new data

1 January, 2004 to 31 December, 2007


  • Hendrik Blockeel
  • Maurice Bruynooghe
  • Jan Struyf

In absence of antiviral therapy, the human immune deficiency virus (HIV) causes an acquired immune deficiency syndrome (AIDS) followed by death within 2 to 10 years after infection. Current approved drugs are as monotherapy not potent enough to eradicate the virus, or to completely block virus replication. Combination therapy is therefore the standard of care, generally refered to as highly active antiretroviral therapy (HAART). Any virus replication in presence of drugs ultimately leads to virus drug resistance and therapy failure. With the currently approved drugs, only a limited choice of combinations exists. Even with HAART, therapy failure is frequent, especially in those pretreated with suboptimal therapy. The first line HAART is in fact a crucial choice, since resistance and cross-resistance compromises any next combination. At initiation of therapy and at each therapy failure, many factors can influence the chances of success of the next combination. These factors are often influencing each other, such that the therapy choice is not always straightforward. It is our aim to try to find the determinants of therapy failure/success, using a relational data mining approach, and to develop algorithms that take into account each patient's individual variables, in formulating an advice for therapy choice. We specifically want to focus on HIV-1 drug resistance, drug adherence and therapeutic drug level monitoring.