Baseline
HCV RNA Level of 400,000 Is Optimal Cut-off Point to Differentiate Low and High
Probability of SVR
Among
individuals with chronic genotype 1 hepatitis C virus infection, baseline HCV
RNA is an important predictor of sustained virological response (SVR), and could
be useful in selecting the most appropriate treatment strategy, especially in
those with high HCV viral load.
In the past, researchers classified patients
as "high" or "low" viral load using a cut-off of 2 x 106 copies/mL.
This value was based on data generated using conventional interferon-based regimens
or pegylated interferon monotherapy. Furthermore, when HCV RNA assays were standardized,
2 x 106copies/mL translated (depending on the assay used) to either 600 or 800
x 103 IU/mL.
In the present study, an experimental analysis was performed
to identify a baseline HCV RNA cut-off that most effectively differentiates between
high and low probability of SVR in patients receiving current standard-of-care
treatment (pegylated interferon plus ribavirin).
For this analysis, 568
study participants were drawn from 2 Phase II trials of treatment-naive, HCV genotype
1 patients receiving Pegasys 180 mcg/week plus ribavirin 1000-1200 mg/day. The
likelihood of achieving an SVR (undetectable HCV RNA < 50 IU/mL at 72 weeks,
or 24 weeks after the end of treatment), was estimated as a function of baseline
viral load strata using a generalized additive multiple logistic regression model
taking into account discrete (gender, race, cirrhosis status) and continuous (age,
weight, baseline viral load, pretreatment ALT quotient) variables.
Results
In the logistic
regression model, which included all predictors listed above, baseline HCV RNA
was a strong independent predictor of SVR (P < 0.001).
However, this effect was found to be non-linear, indicating that above a cut-off
of 5.6 log10 IU/mL, the contribution of viral load in predicting SVR is minimal.
Based
on this model and ROC analyses, the baseline HCV RNA level that most effectively
differentiated between a high and low probability of SVR was 5.6 log10 (~ 400
x 103) IU/mL.
Using this cut-off, the SVR rate in patients with viral load ? 400 x 103 IU/mL
was 70%, compared to 43% in patients with baseline viral load > 400 x 103 IU/mL
(see table).
SVR rates in patients with high viral load were similar across all 3 cut-off definitions,
reflecting that patients in the 400-800 x 103 IU/mL viral load category had SVR
rates similar (43%) to those of patients with viral load > 800 x 103 IU/mL.
Table:
SVR in pts with HVL and LVL using different cut-offs
Baseline HCV RNA
≤ 400 vs > 400 x 103 IU/mL
≤ 600 vs > 600 x 103 IU/mL
≤ 800 vs > 800 x 103 IU/mL
n/N
90/129
189/439
112/177
167/391
156/206
156/362
SVR
70%
43%
63%
43%
60%
43%
Conclusion
Based
on these findings, the authors concluded, "This analysis shows that a baseline
HCV RNA level of approximately 400 x 103 IU/mL is optimal for use as a cut-off
point to best discriminate between low and high viral load, based on the probability
to achieve an SVR in genotype 1 patients when treated with [Pegasys] plus ribavirin
for 48 weeks."
They added, "Use of this cut-off point will allow
treatment optimization in genotype 1 patients." Department of Internal
Medicine, Saarland University Hospital, Homburg/Saar, Germany; University of North
Carolina, Chapel Hill, NC; Hospital of the University of Pennsylvania, Philadelphia,
PA; Hôpital Beaujon, Clichy, France; Hospital General, Valencia, Spain;
GI & Liver Unit, University of Palermo, Palermo, Italy; Scripps Clinic, La
Jolla, CA; Gastroenterology Division, University of Turin, Turin, Italy; North
Shore University Hospital, Manhasset, NY; Virginia Commonwealth University Medical
Center, Richmond, VA; Roche, Nutley, NJ; Henry Dunant Hospital, Athens, Greece.
11/07/06
Reference S
Zeuzem, M W Fried, K Reddy, and others. Improving the clinical relevance of pre-treatment
viral load as a predictor of sustained virological response (SVR) in patients
infected with hepatitis C genotype 1 treated with peginterferon alfa-2a (40KD)
(Pegasys) plus ribavirin (Copegus). 57th AASLD. October 27-31, 2006. Boston, MA.
Abstract 209.
In the logistic
regression model, which included all predictors listed above, baseline HCV RNA
was a strong independent predictor of SVR (P < 0.001). 
