We have read with interest the letter by Pearlman et al. in response to our article entitled “Randomized Trial of Peginterferon Alfa-2b and Ribavirin for 48 or 72 Weeks in Patients With Hepatitis C Virus Genotype 1 and Slow Virologic Response.”1 They strongly disagree with our statement that “extending treatment in slowly responding patients requires reevaluation.” According to the results of our study, in which a slight increase in sustained virologic response (SVR) rates (5%) was observed in slow responders when the treatment was prolonged to 72 weeks (P = 0.644), it is clear that not all patients will benefit from extended therapy. In addition, more therapy discontinuations and an important increase in cost were observed with 72 weeks of treatment; these observations should prompt a reevaluation of therapy prolongation.1
As for the specific points, it was estimated that 120 slow responders would be required to detect a 25% difference in SVR rates between patients treated for 48 weeks and patients treated for 72 weeks (i.e., 45% and 70%, respectively, with a two-sided alpha value of 0.05) with at least 80% power and to avoid a type II error. For this reason, 1400 patients were included, and the large number of centers allowed for rapid enrollment. Overall, 73 centers enrolled the 120 slow responders. The proportion of slow responders in different studies has ranged from 11% to 20%.1-4 Only Pearlman et al.'s study5 had a higher proportion of slow responders (31%), and the authors concluded that this was due to a higher proportion of patients with poor baseline prognostic factors (34% of the patients had a high body mass index >30 kg/m2, 18% had a fasting glucose level >100 mg/dL, 48% were African American, and 26% had advanced fibrosis). However, in their article, the baseline characteristics of the patients were not correlated with the eventual response (e.g., rapid virologic response, slow response, or nonresponse), and this prevented any opportunity for assessing whether the poor baseline prognostic factors were associated with slow responders. However, these poor baseline factors have been associated with nonresponse, but their association with slow response is still unknown.
The population of slow responders has not been well characterized, likely because the majority of studies include relatively few slow responders. The percentage of patients with advanced fibrosis or cirrhosis has varied widely in different trials (8%-37%).1-5 In addition, baseline predictors of the virologic response, such as the percentage of patients with advanced liver fibrosis or cirrhosis, pretreatment viral loads, body mass index, and γ-glutamyltransferase levels, have varied significantly between the different trials. Only one study of slow responders reported these variables, and it demonstrated that the proportions of patients with a high body weight and advanced fibrosis were similar for those achieving rapid virologic response and slow responders.4
Because the majority of the studies did not include African American patients and did not report SVR results according to body weight, baseline glucose levels, insulin resistance, and advanced fibrosis, it is not possible to determine whether patients with these poor prognostic factors were more often slow responders and, more importantly, whether they could have benefited from extended therapy. In other words, can extending the duration of therapy to 72 weeks increase SVR rates among patients with advanced age, a higher body weight, advanced fibrosis, high levels of fasting glucose, and a slow virologic response?
As for the exclusion of patients weighing more than 125 kg in our study, the reason was our desire to evaluate the effect of an adequate dosage of ribavirin on SVR in a way similar to that used in the Individualized Dosing Efficacy Versus Flat Dosing to Assess Optimal Pegylated Interferon Therapy study.6 In addition, our study was conducted in countries outside the United States; these countries have far fewer obese patients (>125 kg) and no African American patients. Hence, patients with these characteristics were not included in our study.
Finally, two different groups have examined the use of other time points for hepatitis C virus RNA levels as thresholds for selecting patients for treatment extension to 72 weeks, but the advantage is unclear and seems limited.7, 8 For this reason, we discussed only the most relevant aspects.
The level of evidence supporting an extended duration of standard-of-care therapy is now much weaker. There exists a need to characterize the baseline prognostic factors of slow responders to determine if patients with poor prognostic factors benefit from an extended treatment duration in comparison with those with more favorable prognostic factors.