400,000 IU/ml was the best cut-off level between Low VL and High VL
Reported by Jules Levin
AASLD, Oct 27-31, 2006, Boston, MA
As I've been reporting, several study reports at AASLD & the recent ICAAC find 400,000 IU/mL is a better lower viral load cutoff in predicting SVR compared to the old cutoff of 800,000 IU/mL or 2 million copies/ml. Here is another study by Zehnter. Another study by Berg at AASLD found similar results and in addition found relapse was better predicted by using the 400,000 cutoff. That is, relapse rates were higher if baseline HCV RNA was >400,000 Iu/ml, see below. Maribel Rodriguez reported the same at ICAAC in HIV/HCV coinfection in analysis of the APRICOT Study.
"Better prediction of SVR in patients with HCV genotype 1 (G1) with peginterferon alfa-2a (PEGASYS) plus ribavirin: Improving differentiation between low (LVL) and high baseline viral load (HVL)"
"....The best baseline predictors for chance of cure with antiviral therapy in
HCV-patients are age <40 years, normal GGT, normal serum ferritin,
normal platelets (>150,000) and a VL <400,000 IU/ml HCV-RNA...."
E. Zehnter1, S. Mauss2, C. John3, R. Heyne4, B. Möller4, T. Lutz5, B. Bokemeyer6, R. Kihn7, G. Moog8, U. Alshuth9, D. Huppe10
1Center of Gastroenterology, Dortmund; 2Center of Gastroenterology and Hepatology, Dusseldorf; 3Center of Gastroenterology, Berlin; 4Livercenter, Berlin; 5Center of Infectiology, Frankfurt; 6Center of Gastroenterology, Minden;
7Center of Gastroenterology, Frankfurt; 8Center of Gastroenterology, Kassel; 9Hepatitis/HIV/Infectiology, Roche Pharma AG, Grenzach-Wyhlen; 10Center of Gastroenterology, Herne; all Germany
Baseline viral load has recently become a very important predictive factor in the development of a treatment algorithm in patients with genotype 1 (G1), but it is not clear whether the cut-off of 600,000 or 800,000 IU/ml is ideal. Both were derived from former 2 million cp/ml due to different factors of the used PCR assay.
The "Association of German Independent Gastroenterologists" (bng, Berufsverband Niedergelassener Gastroenterologen Deutschlands e.V.) in cooperation with Roche, Germany, is conducting a nationwide observational study including screening and treatment phases to determine the quality of treatment for chronic hepatitis C (CHC) in routine clinical practice.
In preceding analyses of our data on predictive factors (DDW 2006, # 219003), the categorized baseline viral load (800,000 IU/ml) was not significant neither in uni- nor in multivariate analyses.
The aim of this evaluation was to find an optimal cut-off between low (LVL) and high viral load (HVL), which gave better prediction of SVR. This analysis was based on an ongoing observational study of patients with chronic hepatitis C and genotype 1 being treated with peginterferon alfa-2a (40KD) (PEGASYS) plus ribavirin in everyday practice.
The well accepted formerly used cut-off of 2 million cp/ml translated with different factors into IU/ml was statistically optimized for treatment with standard interferon.
In the era of pegylated interferon this cut-off is not the best way to differentiate between LVL and HVL with regard to SVR. To use VL as a reliable predictor of successful treatment outcome in hepatitis C the optimized cut-off of 400,000 IU/ml should be adopted in the future.
The determining of this new cut-off for low and high VL revealed that even under real life conditions VL is an important predictive factor of treatment outcome.
This evaluation is part of a large ongoing German multicentre, open-label observational study including anti-HCV-positive adults with detectable HCV RNA. The nature of this study allowed dosing and duration of both peginterferon alfa-2a (40KD) and RBV to be at the discretion of the physician.
This data set includes only naive genotype 1 patients who initiated treatment with peginterferon alfa-2a (40KD) plus ribavirin. The data collection was performed online via the internet.
The screening data include age, sex, weight, height, duration and source of infection, prior antiviral treatment, clinical symptoms, histology, genotype, viral load, concomitant diseases and social status.
The documented data should reflect the clinical routine as intended by the doctors in charge. Therefore, the statistical analysis remains descriptive.
Due to the ongoing character of the study, the status of data was frozen on May 31st, 2006, including queries solved.
Multivariate logistic regression (MLR) analyses: To approximate the optimal cut-off for viral load, multivariate logistic regression analyses were performed in steps of 100,000 and subsequently 10,000 IU/ml.
Statistical significance was evaluated by two-sided Fisher's exact test. Receiver operating characteristic (ROC) curve: The influence of log-arithmized viral load on SVR was estimated as a continuous variable in univariate logistic regression (ULR) analysis and by analyzing the ROC curve. The optimized cut-off was then tested against both existing cutoffs in a MLR analysis.
A total of 10326 treatment naive patient screenings (see Figure 1) have been completed and 4377 of these patients (42.4%) have been treated with peginterferon alfa-2a (40KD), in almost all cases plus ribavirin.
Genotype 1 (G1) was diagnosed in 2522/4377 patients.
In 1239/2522 G1 patients treatment documentation was finished. The mean age was 44.1 years. 58.1% of the patients were male. The mean BMI was 24.9 kg/m2 and the mean duration of infection was 12.2 years.
Sources of infection were (multiple answers possible): iv drug abuse 35.9%, transfusion 20.8% or medical measures 11.1%. The source of infection was unknown in 26.6% of the patients.
In 118/1239 patients treatment was discontinued due to reasons not related to virological nonresponse or tolerability: lost-to-follow-up (n=49), personal reasons (n=30), lack of compliance (n=24), concomitant disease (n=5) or other reason (n=10).
In 158/1239 patients the data cleaning process is still ongoing.
The remaining 963 G1 patients formed the database for the following evaluation. These patients were treated according to the current consensus recommendations (i.e. treatment according to guidelines or discontinuation due to insufficient efficacy or tolerability). In 727/963 patients (75.5%) treatment was completed as planned. 236/963 patients (24.5%) discontinued treatment either due to insufficient efficacy (n=159, 16.5%) and/or to insufficient tolerability (n=89, 9.2%).
Cut-off data set: In 642 patients viral load was documented in IU/ml.
Early Virological Response: 575 of 737 G1 patients (78.0%) achieved an Early Virological Response at week 12 (EVR; ≥2-log10 drop in HCV RNA or HCV RNA undetectable).
EOT-Responses were achieved by 672 of 963 G1 patients (69.8%).
Sustained Virological Response (SVR) was achieved by 507 of 963 G1 patients (52.6%).
MLR analyses of viral load
For the 642 patients of the cut-off data set, the optimal cut-off to discriminate between high and low viral load was 460,000 IU/ml (p<.0001).
Figure 3 demonstrates the descriptive SVR rates for different cut-offs according to the cut-off categories of the first approximation of MLR analysis.
ROC analysis of viral load
In the ULR analysis continuous viral load was a strong predictor of SVR (p<.0001, OR=0.79, CI:0.69-0.89), but the effect of viral load was nonlinear. The ROC-plot revealed a cut-off level of viral load of 5.6 log10 IU/ml (~400,000 IU/ml).
According to this result, an additional discrete cut-off level of 400,000 IU/ml in addition to the old ones (600,000 and 800,000 IU/ml) was compared by MLR analysis. 400,000 IU/ml was the best cut-off level (<.0001, OR=0.48, CI:0.37-0.63) between LVL and HVL.
Using this cut-off of 400,000 IU/ml, G1 patients with LVL reached SVR rates of 62.0% and with HVL 43.7%. SVR-rates of other viral load cutoffs are shown in Figure 4. While SVR in patients with LVL increased with decreasing cut-off levels, SVR in patients using different HVL cut-offs remained at 43%. This result indicated that patients with higher viral load than 400,000 IU/ml had similar low SVR rates as patients with viral load of more than 800,000/ml and belong to the same category.
Other predictive factors
In order to identify other reliable predictive factors for the response to treatment, the following factors were estimated: age <40 years, BMI <25 kg/m2, sex (m/f), GPT male >50, female >35 U/l, GGT male >66, female >39 U/l, platelets >150,000 /_l, serum ferritin <200 _g/l (p>0.001) and VL(<400,000 IU/ml HCV RNA).
In univariate analyses, all above predictions could be established
significantly except for gender.
In a multivariate analysis, age was the strongest independent significant predictive factor (p<.0001, OR=0.32, CI:0.19-0.52) controlled for the effect of GGT (p<.002, OR=0.48, CI:0.22-0.86), serum ferritin (p<.006, OR=0.49, CI:0.30-0.81), platelets (p<017, OR=0.43, CI:0.22-0.86) and VL (p<.022, OR=0.58, CI:0.37-0.93). GPT (p<.243), BMI (p<.592) and gender (p<.611) were non-significant.
The best baseline predictors for chance of cure with antiviral therapy in HCV-patients are age <40 years, normal GGT, normal serum ferritin, normal platelets (>150,000) and a VL <400,000 IU/ml HCV-RNA.
"Definition of a pre-treatment viral load cut-off for an optimized prediction of treatment outcome in patients with genotype 1 infection receiving peginterferon alfa-2a (40KD) plus ribavirin"
Universitaetsklinikum Charite, Berlin, Germany
New HCV RNA Cut-off Level (II) - Study Design and Methods
Population: 455 naive G1 CHC patients
Treatment: PEGASYS 180 mg/week plus COPEGUS 800 mg/d for 48 or 72 weeks
Methods: ROC (Receiver operating characteristic) analyses
New HCV RNA Cut-off Level (II) -Conclusion
In patients treated for 72 weeks, only the 400,000 IU/mL viral load cut-off was predictive for SVR
The 400,000 IU/mL cut-off also proved superior in predicting relapses
Use of a 400,000 IU/mL cut-off point will allow treatment optimization in G1 patients (prediction of SVR and relapse)
A baseline HCV RNA cut-off of 400,000 IU/mL best discriminates between LVL and HVL in HCV genotype 1 patients based on the probability of achieving an SVR
A 400,000 IU/mL HCV RNA cut-off resulted in the greatest difference in SVR rate between patients with a LVL and a HVL
While HVL remains a poor prognostic factor, it must be remembered that HCV genotype 1 patients with a LVL have a high chance of SVR, which increases as HCV RNA decreases