Statistical and Mathematical Modeling in Biological Applications
On the use of mechanistic models for HIV clinical trials
Abstract: Mechanistic models are mathematical models defined according to biological processes and most often based on a set of differential equations. These models have first been applied in HIV infection in mid-nineties revealing the very high turnover of the virus and of the CD4+ T cells. The improvement of statistical methods to estimate parameters has allowed to use this type of models with more sparse data. Here we show how these models can be useful to analyse data from clinical trials evaluating antiretrovirals or immunotherapies. In addition to quantify the effect of an intervention, they provide very good individual predictions. A next step is the use of these models to design new clinical trials or for treatment monitoring. We show applications with real clinical trials involving antiretrovirals and/or exogeneous cytokines.