Highlights from
Digestive Disease Week

May 20-23, 2001
Atlanta, Georgia

Predicting Response to HCV Therapy After One Day of Inteferon

Patients in this study were genotype 1 and received Infergen (interferon) monotherapy. In this study of HCV viral kinetics Layden and colleagues find that all patients with a 1.7 log drop within the first 24 hours after starting therapy and who had less than 5.38 log viral load (240,000 copies/ml) had a sustained virologic response. And no patient cleared virus without accomplishing these two points. But 5 out of 7 who met this criteria had a SVR, so you could have a 1.7 log drop and have <5.38 log viral load and not attain an SVR. All patients who did not meet these two criteria were not SVRs. The authors said the positive predictive value for using these criteria was 71% (% pts observed to have SVR out of the patients predicted to have SVR by this criteria), and the negative predictive value was 100% (% of patients observed to be non-responders [NR} out of the pts not meeting criteria for prediction of SVR). The authors said that using both IFN effectiveness (IFN effectiveness at inhibiting viral production and intrinsic viral clearance, which occurs before the end of day 1 after starting therapy) and viral load at day one (<5.4 log copies/ml) the SVR/NR can be accurately predicted 95% of the time. The results of this study suggest potentially important implications if you could reliably predict outcome after one day of therapy. Since pegylated interferon is time released and blood levels build up slowly taking 4-6 weeks to reach steady state this approach would need exploration using Peg IFN. Such a mechanism to predict outcome needs to be tested in HCV/HIV coinfection. And in certain circumstances particularly in some situations in coinfection this approach may not be usable. For example, in coinfection or in HCV, it may be crucial given an individual’s situation to slow or stop liver disease progression despite not achieving a virologic response. This may have particular importance in coinfection since HIV accelerates HCV progression and because most HCV/HIV coinfected have genotype 1 (which has less of a response rate). As well, coinfected African-Americans may not respond virologically as well since African-Americans have not responded as well as Caucasians in IFN/RBV studies. And so, in certain circumstances it may not be appropriate to defer HCV therapy even if response in the first few days or weeks is not as good as we would like.

The study authors also concluded that the viral load decline affects the immune response. As viral load declines, infected cell death rate increases (due to heightened immune response). This suggests to me that, which I think applies in HIV as well, that the faster viral load declines the more effective therapy should be over the long term in keeping viral load well suppressed.

     Jennifer E. Layden, Univ of Illinois, Chicago, IL; Avidan U. Neumann, Bar-Ilan Univ, Ramar-Gan Israel; Thomas J. Layden, Univ of Illinois, Chicago, IL; Rachel Levi-Drummer, Bar-Ilan Univ, Ramat Gan Israel; Rajender K. Reddy, Univ of Miami, Miami, FL; Thelma E. Wiley, Univ of Illinois, Chicago, IL; Blaine Hollinger, Baylor Coll of Medicine, Houston, TX; Jennifer Poulakas, Amgen Corporation, CA

Patients(pts) with geno 1 HCV are resistant to IFN as only 12% develop SVR despite 12 months of costly therapy that is fraught with side effects. Studies have shown that HCV RNA decline with IFN is biphasic; within 24 hr (1stphase) there is a 0.5 to 2.0 log decline which reflects the effectiveness(e) of IFN in blocking viral production (Neumann et al, Science 1998;282,103). Then, viral decline slows reflecting the death of infected liver cells (2ndphase).

We have shown that the rate of 2nd phase slope over the 1st month of therapy is an excellent predictor of sustained virologic response (Neumann Hep:32,A788, 2000) with a positive predictive value (PPV) of 55% and a negative predictive value (NPV) of 100%.

In the present study we sought to determine if 1st phase viral kinetic parameters(e and viral load at day 1(V1)) correlated with 2nd phase slope and could predict SVR and Non Response(NR) in IFN treated pts within 24 hr. 38 HCV geno 1 infected patients were treated with Infergen for 1 yr. Blood was drawn every 3 hr for RNA measurement (Superquant, NGI) during the 1st day to calculate e and V1, at wks 1,2,and 4 to calculate 2nd slope, and at months 12 and 18 to determine SVR. e was calculated using described mathematical models. V1 and e were analyzed to determine if in combination or alone they could predict SVR and NR; and the correlation between these 2 parameters and 2nd slope was assessed. 5 of 38 pts had SVR.

There was a highly significant correlation(p=<.001; r=.7) between V1 and 2nd slope, and e and 2nd slope(p=.001, r=.6). Interestingly, all pt who achieved SVR had an e greater than 98% (1.7 log drop in 24hr) and a V1 less than 5.34 log. No pt cleared virus without meeting both criteria. 5 out of the 7 pts meeting these 2 criteria had SVR(PPV of 71%).

More importantly, all pts who did not meet both of these criteria were NR(NPV of 100%). By using these 2 criteria, 31 of the 38 pts(82%) were predicted to be NR with the actual rate of NR in the study being 87%. Thus, the accuracy of using these 2 criteria, e > 98% and V1 < 5.34 log, in predicting SVR and NR was 95%. When each of these criteria were used alone they were not statistically as good in predicting SVR.

In summary, this study indicates that 1st phase viral parameters(eand V1) significantly effect the rate of 2nd phase viral decline and that after 1 day of IFN therapy in geno 1 infected patients, NR can be accurately predicted. These results suggest that after one day, in pts receiving IFN monotherapy, decisions regarding therapy continuation can be made. Similar studies should be performed in pts receiving combination treatment and PEG-IFN.


to top

All DDW Articles >