icon-folder.gif   Conference Reports for NATAP  
  AASLD (American Association for the Study of Liver Diseases)
Nov 2-5, 2002, Boston, MA
Back grey_arrow_rt.gif
HCV Viral Kinetics and Early Response Predicting Sustained Response
Reported by Jules Levin
  The purpose of this study is to explore viral load responses during the first week of therapy and to see if these responses can predict achieving a sustained viral response. Patients were randomized to receive two different dosing regimens of pegylated interferon a-2b (PegIntron), in combination with ribavirin.
The results of this study are somewhat at odds with the data presented at the Schering symposium saying that response at 12 weeks is the best cutoff for predicting an SVR. Neumann finds you can predict SVR by week 4 based on early viral load response and Davis finds many individuals without a response by week 4 can still respond. Therefore Davis' findings are that the 12-week cutoff concept is much better, and evaluating viral response at week 12 is preferable to evaluating viral response before week 12 is more reliable in predicting sustained viral response. I agree that using the 12-week viral response in predicting sustained response is preferable and this is further discussed in this article below. Here is the link to NATAP's detailed report on the Davis data:
Avidan Neumann, a noted viral kineticist, reported this study (abstract 778) on 55 treatment-naive patients who had chronic HCV, genotype 1, and elevated ALT. They were randomized to receive a decreasing dose of of PegIntron, 3.0 ug/kg the first week, 1.5 for 3 weeks, and 1.0 for 44 weeks plus ribavirin 800 mg/ day (27 patients), or 0.5 ug/kg weekly plus ribavirin 800 mg/day for 48 weeks (28 patients). HCV RNA was measured at baseline, daily during the first week, weekly until week 8, and weeks 12, 18, 24, and 48 of treatment. SVR status was defined for 48 patients by undetectable (<100 IU/ml) virus at 24 weeks after the end of treatment.
HCV declined during the first days maximally (Vmax) by -1.4 log (range -2.8 to -0.1) for the low dose and -2.3 for the high dose (range -4.3 log to -0.9 log); viral load reduction was dose dependent, meaning decline was increased by increased dose. Reduction in viral load peaked at 1-4 days (mean 1.9 days independent of dose), however after that all the patients had an increase in viral load until what is called a maximal viral load (Vmax) was reached at days 5-7 (mean -0.28 log; range-1.7 to + 1.3 log for the low dose; mean -0.9 log, range -3.0 to + 0.7 for the high dose). In other words, HCV viral load reduction peaked at 1-4 days, and the peak HCV viral load decrease was -1.4 log for the low dose and -2.3 log for the high dose. But, after the first 1-4 days viral load increased and during the 5-7 day period during the week following drug administration the mean log reduction in total for the week was -0.28 for the low dose and -0.9 log for the high dose.
There was no significant correlation of the viral kinetics during the first week with baseline viral load, baseline ALT, age, weight, or gender. Both Vmin and Vmax were significantly correlated with SVR. This means that the peak viral load reduction during days 1-4 predicted SVR, the more viral load was reduced duriing days 1-4 the better the chance of achieving SVR. And the greater the rebound is of viral load during days 5-7the lesser the chance of achieving an SVR.
Neuman said that it is possible to obtain a 100% or 90% positive predictive value from the first week of viral kinetcs (see below). However, no combination of the viral kinetics parameters during the first week gave a negative predictive value higher than 75% because some of the patients with SVR had very bad response during the first week. This means that if a patient has a good viral load response by reaching a certain threshold defined by Neuman in this study of viral load level during the first week they were 100% assured of achieving an SVR. If however patients did not achieve this threshold level in the first week there is still a 25% chance in this study that they could achieve a SVR.
The second slope of decline during weeks 1-4 was strongly correlated with SVR independently of dose, and a criteria of slope faster than 0.3 log IU/ml/week gave 100% positive predictive value and 100% negative predictive value. Meaning all patients who achieved viral load decline of 0.3 log per week for weeks 1-4 had SVR and patients who did not achieve 0.3 log decline did not achieve SVR.
After the first week, viral load had a slow second phase decline with large variation between patients (mean 0.57 log IU/ml/week, range 0.03-1.20). The decline was not dose independent and not correlated with baseline viral load, baseline ALT, age, weight or gender. The second slope decline during weeks 1-4 was correlated with the viral loads at days 1-7.
The results from this study suggest that a good viral load decline during the first few days during therapy should yield a good viral load response in the following weeks and predict the ability of achieving an SVR, but not necessarily. Neumann concluded a rapid viral load decline during the first 3 days of treatment and the lack of a strong increase at the end of the first week are correlated with SVR. However, by using only first week data it is possible that 25% of the patients predicted not to respond will in fact have SVR. In Neumenn's words, the finding that a large increase in viremia at the end of the first week reduces the probability of an SVR indicates that strategies to optimize the early viral decline may be important.
Neumann concludes, based on his data from this study, that the second phase viral decline slope gives a perfect prediction of SVR already after 4 weeks of treatment with pegylated Interferon a-2b and ribavirin. And again he finds that all patients in this study with a second slope decline of >0.3 log/week achieve SVR. My feeling is to stick with the clinical data showing 12 weeks as a cutoff and further I'm concerned that 24 weeks may be better for some patients who may be later responders.
Neumann found in this study that a viral load reduction of 1.2 or greater by the end of week 1 is a high indication of SVR. But the negative predictive vale is 54-75%, meaning that many patients still have a good chance of responding.
For partial viral responders or coinfected patients, my feeling is that the doctor and patient should discuss these issues and ultimately if the patient wants to continue past 12 weeks to 24 weeks the patient judgement should be respected.
In coinfection preliminary data suggests that for some patients viral load response can be delayed. In monoinfection as well a certain percentage of patients may not respond until after 12 and up to 24 weeks. In coinfection there is a little data suggesting that response at week 12 will predict SVR outcome for most patients, but this has not been well confirmed and it appears that discontinuation of therapy before week 24 is too quick.