Improving the Management of HIV Disease Vol. 6, no. 5 [World Conference on AIDS (12th: 1998: Geneva)]

QUESTIONS TO THE ANTIRETROVIRAL THERAPY AND RESISTANCE TESTING PANELS

identifying which drugs not to use. Such
combination regimens can include a
combination of nRTIs, NNRTIs, and protease inhibitors. Preliminary data suggest
that regimens that include dual protease
inhibitors may be particularly beneficial.

hot, do
useful in

DR BRUN-VZINET: According to the
data presented at the HIV Drug
Resistance Workshop in June 1998, the
correlation between genotype and phenotype is better for HIV-resistant isolates
than for sensitive strains. Several genotype-phenotype databases are currently
in development with the aim of generating software that can predict phenotype
from genotype results. Retrospective
studies reported that baseline genotype
and/or phenotype may predict the viral
load response in patients who had previously experienced several therapeutic
failures. For example, Lanier et al showed
the predictive strength of baseline genotype and phenotype in abacavir-treated
patients. Zolopa and colleagues demonstrated that genotype at baseline is a
strong predictor of virologic response in
patients receiving ritonavir/saquinavir
after a previous protease inhibitor-containing regimen failed. In this study the
genotype data had a better predictive
value than clinical and drug history. It is
not known at the present time whether
genotype or phenotype is the better predictor of drug failure. Finally, the utility
of phenotypic and genotypic testing in
HIV-infected patients must be evaluated
and validated through prospective studies.

DR LOVE DAY: One important conclusion
from these two recent conferences was
that there is an urgent need for the generation of a database(s) that documents
genotype, phenotype, and clinical outcome for thousands of patients so that
relationships may be analyzed to assist in
understanding the use of these measures
in clinical care. However, based on
results of clinical studies at these meetings and unpublished data I have seen
since, our center in the United Kingdom
has determined to include real-time
genotyping as part of our management
for the best care of our patients. I now
feel that failure to test may, in some
cases, result in ineffective therapies and a
wasteful use of drugs. We will not be able
to answer all the questions at this time,
but if we can prevent patients who are
drug-naive or undergoing their first
change in therapy from starting an
inappropriate combination-one to
which they are doomed to become unresponsive-we are making enormous clinical strides.
The economics of this philosophy
are simple: it costs the price of approximately 2 months' supply of 1 drug to test
for resistance to all drugs in 1 patient,
and this simple expedient could save
thousands in wasted therapy over the following year.
25

DR YE N I: There is no clear demonstration that drug-level monitoring is useful
in the clinic. However, on an individual
basis, it may be useful to confirm that
plasma concentrations are in the therapeutic range, especially when drug pharmacokinetic interactions are expected to
occur. The approximation in the pharmacokinetic characteristics, when assessed
by 1 blood sample only, must be recognized and complicate the interpretation
of the result, even if the exact time of the
last drug dosing in the patient is recorded. Drug-level monitoring is not adequate
for assessment of adherence, because of
the risks of overinterpretation of a result
from a single determination.
DR KURITZKES: At present, therapeutic
drug-level monitoring is not recommended for monitoring antiretroviral drugs. In
the case of nRTIs, the half-lives of the
drugs are too short to use trough levels
as a meaningful marker of adherence to
treatment. Moreover, it is the level of
intracellular dideoxynucleoside triphosphate that really matters.
In the case of the NNRTIs and protease inhibitors, drug-level monitoring
may be useful in very specific cases where
some unusual drug-drug interaction is
suspected. In such cases an in-patient
pharmacokinetic study could be performed in which serial measurements are
obtained after an observed dose of drug
is given.

VOLUME 6, OCTOBER 1998