 |
|
|
| |
Abstract 694. CLINICAL IMPLICATIONS OF A NEW TRI-PHASIC MODEL FOR HEPATITIS C
VIRAL KINETICS DURING IFN-a THERAPY
Reported by Jules Levin
|
| |
| |
The authors of this study suggest that induction IFN therapy may be crucial
for genotype 1 and high HCV viral load. Future studies should consider
exploring high induction dosing with daily interferon followed by pegylated
interferon, or perhaps higher dosing of pegylated interferon.
Claudia C Bergmann, Bar-Ilan Univ, Ramat-Gan Israel; Jennifer E Layden, Univ
of Illinois, Chicago, IL; Rachel S Levy-Drummer, Bar-Ilan Univ, Ramat Gan
Israel; Thomas J Layden, Univ of Illinois, Chicago, IL; Bart L Haagmans,
Erasmus Univ Rotterdam, Rotterdam Netherlands; Avidan U Neumann, Bar-Ilan
Univ, Ramat-Gan Israel
Background: A bi-phasic model currently describes HCV dynamics during IFN-a
therapy and its 2nd phase slope is the best predictor for sustained viral
response (SVR). According to this model the 2nd slope is mainly determined by
the strength of the immune response in killing infected cells, and to a
lesser degree by the dose of IFN. However, viral decline in a large fraction
(30-40%) of patients appears to be tri-phasic, with a first rapid decline
followed by a flat intermediate plateau region (shoulder phase) that turns
into a slow decline later. Moreover, new results show that 1st phase
parameters can predict the 2nd phase slope and consecutively SVR, although
this cannot be explained by the bi-phasic model.
Methods: We re-analyzed previous studies of HCV kinetics in patients (N=200) with frequent sampling during the period of 3-28 days. We use a new
tri-phasic model that allows the loss rate of infected cells to be
accelerated during therapy as function of decreasing viral load. We assume an
initial hypo-responsiveness of cellular immune response in presence of high
virus titers that is restored when viral load decreases below a suppression
threshold.
Results: Our model predicts that if viral load does not decline during the
1st phase below the suppression threshold then the 2nd phase slope is flat,
while if it is considerably below the threshold then a rapid 2nd slope is
achieved. A tri-phasic decline will occur only if the viral load at the end
of the 1st phase is just about the suppression threshold level. An
intermediate shoulder phase with a slow decline is then observed until the
immune response "kick-starts" and the 2nd phase slope accelerates. Indeed,
the viral decline patterns in all studies agree with these predictions. The
estimated viral load suppression threshold is 100,000 IU/ml (± 60,000) and
the length of the plateau-phase (3-21 days) is significantly correlated with
the viral load at the end of the 1st phase. The model predicts that following
a dose reduction, which gives rise to lower blocking of virion production,
there will occur a rebound in viral load and the 2nd phase slope will be
slowed. However, if the dose reduction occurs after viral load declined
considerably below the suppression threshold these effects will be diminished
and the 2nd slope will still be rapid. We found that SVR rate among patients
with a rapid slope (larger than 0.1/day) after the shoulder phase is
equivalent to that of patients with a rapid slope immediately after the first
phase.
Conclusions: The new tri-phasic model, which better describes the HCV
kinetics during IFN therapy, predicts that it is necessary to considerably
lower viral load already during the 1st phase in order to allow the patient
to have a rapid viral response in the 2nd phase and consecutively SVR.
Therefore in patients with HCV genotype 1, who have low IFN effectiveness in
blocking production, and high baseline viral load it will be important to
treat with high dose of IFN. High dose induction should continue until the
immune response overcomes its hyporesponsiveness (i.e., at the end of the
shoulder phase) and then switched to lower dose given daily or in pegylated
form. In patients with a shoulder phase it is important to correctly define
the 2nd phase slope after the end of the shoulder in order to evaluate the
probability to achieve SVR. Nevertheless, this model allows to construct a
framework to predict SVR already from the 1st phase parameters, 24 hours
after beginning of treatment.
|
|
| |
|
|
|
|
|
