Treatment of HIV patients must balance the need to suppress viral
replication against the harmful side effects and significant cost to
the patient of antiretroviral therapy. This tradeoff has led to the
development of various drug-sparing HIV-1 treatment strategies, which
often results in the emergence of resistant viruses and overall
treatment failure. This has prompted an interest in
induction–maintenance (IM) treatment strategies, in which brief
intensive therapy is used to reduce host viral levels, which is then
followed by a simplified and more easily tolerated maintenance regimen.
IM
approaches remain an unproven concept in HIV therapy. In a study
publishing July 13, 2007 in PLoS Computational Biology, clinical
responses to antiretroviral drug therapy are simulated for the first
time, and the model is then applied to IM therapy. Marcel Curlin,
Shyamala Iyer, and John Mittler, from the University of Washington,
find that IM is expected to be successful beyond three years and that
six to ten months of induction therapy should achieve durable
suppression of HIV and maximize the possibility of eradicating viruses
resistant to the maintenance regime. They also find the
counter-intuitive result that for induction regimens of limited
duration the optimal time to initiate induction therapy may be several
days or weeks after the start of regular (maintenance) therapy.
These results are important not simply because they show how this
particular, albeit important, therapy strategy may be optimized, but
because they illustrate the more general potential for mathematical
models to influence therapy decisions. "Our experience has been that
clinicians and policy makers are often hesitant to consider, sometimes
even hostile towards, mathematical modeling approaches. Instead, they
rely on intuition or await the results of expensive, long-term clinical
trials", says Mittler. By presenting a detailed model that makes
concrete quantitative predictions and gives some interesting,
counter-intuitive qualitative results, this paper may help to change
attitudes concerning the value of dynamical modeling approaches.
Source: Public Library of Science. July 2007.
