| |
Treatment of chronic hepatitis C virus genotype 1 with peginterferon, ribavirin, and epoetin alpha
|
| |
| |
Hepatology July 2007
Mitchell L. Shiffman *, Jennifer Salvatore, Sarah Hubbard, Angie Price, Richard K. Sterling, R. Todd Stravitz, Velimir A. Luketic, Arun J. Sanyal
Hepatology Section, Virginia Commonwealth University Medical Center, Richmond, VA
Funded by:
Ortho-Biotech and Schering-Plough
Abstract
Successful treatment of chronic HCV with peginterferon (PEGIFN) and ribavirin (RVN) is often limited by anemia. We performed the present study to determine if utilizing epoetin alpha (EPO) with or without a higher dose of RVN could enhance sustained virologic response (SVR).
We randomized 150 treatment-naive patients with chronic HCV genotype 1 into 3 treatment groups: (1) PEGIFN alpha-2b (1.5 ug/kg/week) + weight-based RVN (WBR) 13.3 mg/kg/day (800 to 1400 mg/day); (2) PEGIFN alpha-2b + WBRVN + EPO (40,000 U/week); or (3) PEGIFN alpha-2b + higher dose WBR 15.2 mg/kg/day (1000 to 1600 mg/day) + EPO.
We initiated EPO at the onset of therapy to maintain the hemoglobin between 12 and 15 g/dL. When required, we reduced RVN by 200-mg steps. African Americans compose 36% of the population.
A significantly smaller percentage of group 2 patients had a decline in hemoglobin to less than 10 g/dL (9% versus 34%; P < 0.05) and required that the RVN dose be reduced (10% versus 40%; P < 0.05) compared to group 1 patients. Despite this, SVR was similar in these groups (19% to 29%). SVR was significantly greater (P < 0.05) in group 3 patients (49%). This resulted from a significant decline (P < 0.05) in relapse rate; only 8% versus 38% for groups 1 and 2.
Conclusion: We conclude that using EPO in all subjects at the initiation of PEGIFN and RVN treatment will not enhance SVR given the same starting dose of RVN. In contrast, a higher starting dose of RVN was associated with a lower relapse rate and higher rate of SVR.
The results of the present study do not imply that the use of hematologic growth factors should be completely abandoned in HCV patients receiving PEGIFN and RVN. Approximately 10% of such patients experience a profound decline in serum hemoglobin, by 4 g or greater, within several weeks of initiating treatment and must interrupt or stop therapy.[30] Severe hemolytic anemia secondary to PEGIFN and RVN therapy is also much more common in patients with cirrhosis awaiting liver transplantation,[31] in patients with recurrent HCV following liver transplantation,[32] in patients with hemoglobinopathies,[33] and in those with chronic renal insufficiency.[34] It is unlikely that such patients could be successfully treated with PEGIFN and RVN without the use of a hematologic growth factor. Randomized controlled trials in these specific subpopulations are likely to demonstrate that hematologic growth factors will yield an SVR advantage.
Several studies have now demonstrated that higher doses of RVN are more effective at reducing relapse and enhancing SVR.... A significant reduction in relapse and increase in SVR were also observed in those patients who received a higher starting dose of RVN along with EPO in the current study. This does not imply that a hematologic growth factor is absolutely necessary to treat patients with higher doses of RVN or is required to enhance SVR with this treatment regimen. These questions were not addressed by the treatment protocol employed by the current study. However, it appears that the starting dose of RVN may be the most important factor associated with reducing relapse and that dose reduction, at least by small amounts, may not alter this relationship. This possibility deserves further investigation.
In summary, the routine use of EPO did not enhance SVR given the same starting dose of PEGIFN and RVN. This suggests that reducing the dose of RVN by 200-mg decrements should be the first response in an HCV patient who develops anemia secondary to PEGIFN and RVN therapy. In contrast, that subset of patients who develops anemia that is so rapid in its onset and severe that it could not be overcome by dose reduction alone could potentially benefit from the use of a hematologic growth factor initiated at the onset of therapy. A randomized controlled trial evaluating this possibility in this specific patient population is clearly needed.
Discussion
The hypothesis that hematologic growth factors could potentially improve SVR if started at the initiation of HCV therapy is based upon several observations. Over 20% of patients who receive PEGIFN and RVN must either alter the dose, temporarily interrupt, or prematurely discontinue either 1 or both of these medications.[1-3] Previous observations suggested that reducing the dose of PEGIFN or RVN to less than 80% of the cumulative maximal dose that would have been received during the first 12 weeks of treatment was associated with a significant decline in both EVR and SVR.[4][5] The use of a hematologic growth factor after patients developed anemia improved serum hemoglobin and allowed patients to remain on either full-dose RVN or to increase the dose in those patients who already required a dose reduction.[11-13] In a prospective, double-blind, placebo controlled trial a significantly greater percentage of patients treated with EPO were able to receive 80% or more of their target RVN dose, had a significantly greater cumulative RVN exposure, and reported a significantly better quality of life.[12][14] However, none of the previous trials evaluated the impact of the hematologic growth factor on SVR. Despite this, it was widely assumed that the positive effects of utilizing a hematologic growth factor would translate into an SVR advantage; and the routine use of these products during HCV treatment began to proliferate without any data to suggest that this approach would improve outcomes. Although generally safe, the use of hematologic growth factors has been associated with adverse events, which include hypertension, headache, arthralgias, paresthesias, injection site erythema, thrombosis, thromboembolic events, and antibody-mediated pure red cell aplasia.[16-19]. A recent report has also documented pure red cell aplasia in an HCV patient receiving EPO to treat PEGIFN-induced and RVN-induced hemolytic anemia.[20]
The use of a hematologic growth factor nearly doubles the medication costs associated with the treatment of HCV.[21] Despite this, a recent analysis suggested that the cost of utilizing a hematologic growth factor during HCV treatment was well within the range of other well-accepted medical therapies associated with an increase in quality-adjusted life-expectancy.[21] However, such an analysis is predicated upon the assumption that the use of a hematologic growth factor actually enhances SVR; which it has never been demonstrated to do.
The present study was performed to test the hypothesis that the routine use of EPO during HCV treatment would reduce the incidence of PEGIFN-associated and RVN-associated hemolytic anemia, reduce the need to lower the dose of RVN, and enhance SVR. Indeed, initiating EPO at the start of HCV treatment along with PEGIFN and RVN did significantly reduce the incidence of anemia and the need to reduce the dose of RVN. Unfortunately, this approach did not affect either RVR, EVR, or end-of-treatment VR and failed to enhance SVR in patients who received the same starting dose of RVN. In contrast, a significant increase in SVR was observed in those patients who received a higher starting dose of RVN.
The overall rate of EVR, end-of-treatment VR, and SVR for patients treated with standard WBR (800 to 1400 mg/day) in the present study were 65%, 43%, and 24%, respectively. This is somewhat lower than that reported for patients with HCV genotype 1 in several large multicenter, randomized controlled trials.[1][2] The reason for this lower rate of VR likely reflects the high percentage of African Americans enrolled in the current study. Overall, 39% of patients enrolled in this study were African Americans, compared to less than 10% in the studies of Manns et al.[1] and Fried et al.[2] It is now well established that the VR of African Americans with HCV genotype 1 is significantly lower than observed in patients of other races at all times evaluated.[22-24] Despite the high percentage of African Americans enrolled in the present study, the SVR rate calculated according to the per-protocol analysis for the control group treated with PEGIFN and WBR in this trial was very similar to that recently reported in a large community-based trial utilizing the same doses of PEGIFN and RVN.[25]
Achieving an SVR in patients with chronic HCV is dependent upon 2 steps: (1) the patient must first respond and become HCV RNA undetectable; and (2) the patient must not relapse. The former is primarily dependent upon the antiviral and immunologic affects of interferon. Adding RVN to either interferon or PEGIFN monotherapy enhances VR by about 15%.[1][2][26][27] However, the primary role of RVN is to reduce relapse, and in large controlled trials the use of RVN reduced the relapse rate from approximately 50% to under 20%.[1][2][26][27] Furthermore, the higher the initial starting dose of RVN, the lower the relapse rate, even though higher doses of RVN were associated with more frequent dose reduction.[28][29] A recent study has suggested that dosing RVN by body weight enhances SVR over that observed with an 800 mg/day fixed dosage.[25] This is likely the result of providing a higher dose of ribavirin to those patients who weigh more than 65 kg. The results of the present study are in agreement with these observations. Patients who were randomized to receive the higher dose of RVN (approximately 15.2 mg/kg/day) had a significantly lower relapse rate and higher SVR rate, even though nearly one-third of these patients required dose reduction.
Although previous studies suggested that reducing the dose of RVN negatively impacted SVR, more recent studies have demonstrated that dose reduction to even 60% or less of the total cumulative RVN dose does not appear to affect SVR as long as RVN dosing is not temporarily interrupted or prematurely discontinued.[7-9] This appears to be especially true in those patients who have already become HCV RNA undetectable prior to dose reduction and in those patients who achieved an RVR and were HCV RNA undetectable within 4 weeks of initiating HCV treatment.[6-9] These observations provide an explanation as to why the routine use of EPO did not enhance SVR in the current study. In addition, the dose reduction strategy employed in the current study, reducing RVN by 200 mg increments, limited the number of patients who either interrupted or prematurely discontinued this medication.
The results of the present study do not imply that the use of hematologic growth factors should be completely abandoned in HCV patients receiving PEGIFN and RVN. Approximately 10% of such patients experience a profound decline in serum hemoglobin, by 4 g or greater, within several weeks of initiating treatment and must interrupt or stop therapy.[30] Severe hemolytic anemia secondary to PEGIFN and RVN therapy is also much more common in patients with cirrhosis awaiting liver transplantation,[31] in patients with recurrent HCV following liver transplantation,[32] in patients with hemoglobinopathies,[33] and in those with chronic renal insufficiency.[34] It is unlikely that such patients could be successfully treated with PEGIFN and RVN without the use of a hematologic growth factor. Randomized controlled trials in these specific subpopulations are likely to demonstrate that hematologic growth factors will yield an SVR advantage.
Several studies have now demonstrated that higher doses of RVN are more effective at reducing relapse and enhancing SVR.[25][28][29] A significant reduction in relapse and increase in SVR were also observed in those patients who received a higher starting dose of RVN along with EPO in the current study. This does not imply that a hematologic growth factor is absolutely necessary to treat patients with higher doses of RVN or is required to enhance SVR with this treatment regimen. These questions were not addressed by the treatment protocol employed by the current study. However, it appears that the starting dose of RVN may be the most important factor associated with reducing relapse and that dose reduction, at least by small amounts, may not alter this relationship. This possibility deserves further investigation.
In summary, the routine use of EPO did not enhance SVR given the same starting dose of PEGIFN and RVN. This suggests that reducing the dose of RVN by 200-mg decrements should be the first response in an HCV patient who develops anemia secondary to PEGIFN and RVN therapy. In contrast, that subset of patients who develops anemia that is so rapid in its onset and severe that it could not be overcome by dose reduction alone could potentially benefit from the use of a hematologic growth factor initiated at the onset of therapy. A randomized controlled trial evaluating this possibility in this specific patient population is clearly needed.
Results
Comparison of Study Groups
Of the 150 patients enrolled into the study, 146 received at least 1 dose of study drug. The remaining 4 patients decided not to start treatment in the protocol after learning of the treatment they had been randomized to receive. Table 1 compares various demographic, biochemical, virologic, and histologic characteristics of the 3 groups at baseline. The mean age of the population was 47 years, 61% were male, 39% were African American, mean body weight was 82 kg, mean serum ALT was 101 IU/l, mean log serum HCV RNA was 5.5 IU/ml, and 5% of patients had cirrhosis. Although no significant differences existed in any of these parameters among the 3 groups, the group treated with standard WBR and EPO (PEGIFN+WBR+EPO) had the highest percentage of African Americans and the highest percentage of patients with cirrhosis.
Virologic Response
VR according to the intention-to-treat analysis at various times during the study are illustrated in Fig. 1A. At week 4, 9%, 8%, and 11% of patients treated with PEGIFN+WBR, PEGIFN+WBR+EPO, and PEGIFN+HDR+EPO, respectively, were HCV RNA undetectable and had achieved an RVR. At week 12, EVR had been achieved in 68%, 65%, and 63% of patients in these 3 groups, respectively. At the completion of 48 weeks of treatment, VR was achieved in 46%, 31%, and 53% of patients in the 3 groups, respectively. The somewhat lower VR rate observed in the group that received standard WBR and EPO (PEGIFN+WBR+EPO) was likely secondary to the higher percentage of African Americans and patients with cirrhosis in this group (Table 1). No significant differences in either RVR, EVR, or VR existed between the 3 treatment groups. The SVR rates observed for the 2 groups who received standard WBR (PEGIFN+WBR and PEGIFN+WBR+EPO) were not significantly different; 29% and 19% for patients in the PEGIFN+WBR and the PEGIFN+WBR+EPO groups, respectively. The group treated with HDR and EPO (PEGIFN+HDR+EPO) had an SVR of 49%; and this was significantly (P < 0.05) greater when compared to the 2 groups treated with standard dose RVN.
Because the difference between the rates of VR were not significantly different among the 3 treatment groups, any differences in SVR must have been secondary to relapse (Fig. 1B). The relapse rates in the 2 groups treated with standard WBR (PEGIFN+WBR and PEGIFN+WBR+EPO) were not significantly different; 36% and 40%, respectively. In contrast, the relapse rate of the group that received HDR (PEGIFN+HDR+EPO) was only 8%. This 4-fold to 5-fold decline in relapse rate was significantly lower (P < 0.05) than observed in the 2 groups treated with standard doses of RVN.
End-of-treatment VR and SVR rates according to the per-protocol analysis are illustrated in Fig. 2. Overall, we observed slightly higher rates for VR and SVR compared to the intention-to-treat analysis (Fig. 1A). However, we observed a similar relationship with regard to the 3 treatment groups. We observed no significant differences in end-of-treatment VR between the 3 groups, which ranged from 37% to 54%. SVR rates for the 2 groups who received standard WBR (PEGIFN+WBR and PEGIFN+WBR+EPO) were not significantly different (34% and 22%). In contrast, the group who received HDR (PEGIFN+HDR+EPO) had an SVR rate of 49% and this was significantly greater (P < 0.05) than that observed for the 2 groups that received the standard dose of RVN.
Reduction in PEGIFN Dose
For those patients who completed 48 weeks of therapy, PEGIFN dose reductions were similar across the 3 treatment groups. The percentage of patients with any PEGIFN dose reduction was 14%, 11%, and 19%, respectively. The mean decrease in PEGIFN dose across the 3 groups was 2%, 6%, and 6%, respectively. Only 5%, 11%, and 12% of the patients in the 3 groups received less than 80% of the anticipated total maximal PEGIFN dose.
Impact of EPO on Hemoglobin
The impact of EPO on serum hemoglobin is summarized in Table 2. The mean hemoglobin prior to the initiation of treatment was 15.3 g/dl. This was not significantly different between the 3 treatment groups. The routine use of EPO did not prevent RVN-induced anemia. However, the maximal decline in hemoglobin and the percentage of patients in whom the hemoglobin declined to less than 10 g/dl were significantly lower (P < 0.05) in those patients who received standard WBR along with EPO (PEGIFN+WBR+EPO), compared to patients who did not receive EPO (PEGIFN+WBR).
Impact of EPO on RVN Dose
The impact of EPO on the dose of RVN received is summarized in Table 3. The routine use of EPO in patients treated with standard WBR significantly reduced (P < 0.05) but did not eliminate the percentage of patients who required RVN dose reduction. Although the mean dose of RVN received by patients in the PEGIFN+WBR+EPO group was greater than in patients not treated with EPO (PEGIFN+WBR), this difference was not significant. Patients treated with the higher dose of RVN received on average a mean RVN dose that was significantly (P < 0.05) greater than the 2 groups treated with standard dose of RVN. However, despite the use of EPO, 31% of patients in this group still required dose reduction by a mean of 102 mg/day. RVN dose reduction did not appear to adversely impact SVR.
Impact of RVN Dose on VR
The impact of RVN dose on SVR is illustrated in Fig. 3. Because both the SVR and the mean RVN dose received by the 2 patient groups treated with standard WBR were similar, these 2 groups were combined and compared to the group treated with the higher dose of RVN. Figure 3A illustrates the impact of RVN dose on SVR in Caucasians and African Americans. Both Caucasians and African Americans had a higher SVR when treated with the higher dose of RVN. In Caucasians, this increased from 27% to 56% and in African Americans this increased from 27% to 31%. In both groups, the increase in SVR was secondary to a reduction in the rate of relapse. Figure 3B illustrates the impact of RVN dose on body weight. Regardless of body weight, patients treated with the higher dose of RVN had a higher SVR.
Reason for Discontinuation of Treatment
The reasons for discontinuing treatment are listed in Table 4. Nonresponse and adverse systemic reactions accounted for over one-half of all patients who stopped treatment in all 3 treatment groups. Given the 200-mg stepwise dose reduction scheme utilized in this study, only 4% of patients in the group treated with standard WBR without EPO (PEGIFN+WBR) discontinued treatment secondary to anemia. Interestingly, the group treated with standard WBR and EPO had the highest rate of premature dose reductions secondary to adverse events. Across all groups, 6% to 15% of patients discontinued treatment secondary to adverse events after they had become HCV RNA undetectable.
Patients and Methods
Patient Population
Patients were eligible to enter this trial if they tested positive for HCV RNA with genotype 1, had a liver biopsy consistent with chronic HCV, and had no prior treatment for chronic HCV. Patients were excluded from entering this study if they had any other form of chronic liver disease detected by appropriate serologic testing and histologic examination of liver tissue. All patients were negative for hepatitis B virus surface antigen, had either negative or insignificant elevations in serum titers of antinuclear antibody and anti-smooth muscle antibody and normal values for serum alpha-1-antitrypsin and ceruloplasmin. We tested patients with elevations in serum iron saturation or ferritin for genetic hemochromatosis. We excluded patients with a positive genetic test or those with greater than 2+ stainable iron within the histologic specimen. We also excluded patients from participation if they were infected with any non-1 HCV genotype, had previously received any type of interferon or RVN, were known to be actively utilizing intravenous drugs, consuming excessive alcohol on a regular basis, were HIV positive, pregnant, had a serum creatinine above the upper limits of normal, or had received an organ transplant. We also excluded patients with a history of documented ascites, hepatic encephalopathy, variceal hemorrhage, a Child-Pugh score greater than 6, platelet count less than 80,000/mL, total white blood cell count less than 3,000/mL, and hemoglobin less than 12 g/dl.
All patients underwent liver biopsy within 2 years of study enrollment. Liver biopsies were scored according to the histologic activity index of Knodell et al.[15] by 1 of 2 dedicated liver pathologists at our center. The liver biopsy slides from all patients who underwent this procedure outside our center were obtained and reviewed by these same 2 pathologists.
This study protocol was approved by the Committee on the Conduct of Human Investigation at the Virginia Commonwealth University Medical Center. We obtained informed consent to participate in this study from all patients prior to enrollment and treatment.
Study Design and Adjustment of Medication Dosing
We randomly assigned all patients to 1 of 3 treatment groups as follows: (1) PEGIFN alpha-2b 1.5 g/kg/week plus standard weight-based RVN (WBR) (PEGIFN+WBR), approximately 13.3 mg/kg/day; (2) PEGIFN alpha-2b 1.5 ug/kg/week, plus standard WBR, approximately 13.3 mg/kg/day, plus EPO 40,000 units/week (PEGIFN+WBR+EPO); or (3) PEGIFN alpha-2b 1.5 g/kg/week plus high-dose RVN (HDR), approximately 15.2 mg/kg/day, plus EPO 40,000 units/week (PEGIFN+HDR+EPO). We carried out randomization in blocks of 9. EPO (Procrit) was supplied to all patients by Ortho-Biotech Products LP (Bridgewater, NJ). Additional RVN capsules were supplied to those patients randomized to the high-dose RVN group by Schering-Plough (Kenilworth, NJ).
The starting dose of PEGIFN alpha-2b was 1.5 ug/kg/week. We reduced this by 0.5 ug/kg steps in patients who developed adverse events attributable to interferon. We discontinued treatment in patients who developed severe adverse events, which included the following: an absolute neutrophil count less than 500/ml, a total white cell count of less than 800/ml, a platelet count less than 30,000/ml, severe eflu-likef symptoms that could not be tolerated, and severe depression. We stopped both RVN and EPO in those patients who could not remain on PEGIFN.
We adjusted the starting dose of RVN by body weight to approximately 13.3 mg/kg/day as follows: <65 kg, 800 mg/day; 66 to 85 kg, 1000 mg/day; 86 to 105 kg, 1200 mg/day; and >105 kg, 1400 mg/day. In those patients randomized to receive HDR, approximately 15.2 mg/kg/day, the starting doses were 200 mg/day greater at each weight grouping as follows: 1000, 1200, 1400, and 1600 mg/day, respectively. We reduced the dose of RVN by 200-mg steps in patients whose hemoglobin declined to less than 10 g/dl or in those patients who developed other adverse events attributable to this drug. The dose of RVN could be increased if the hemoglobin increased above 10 g/dl or if the adverse event resolved. In general, we adjusted and maximized RVN doses according to tolerability. We permanently discontinued RVN in patients whose hemoglobin declined to less than 8.5 g/dl or in those who developed severe adverse events not responding to numerous dose reductions. Such patients could have continued treatment with PEGIFN with or without EPO.
The starting dose of EPO was 40,000 IU/week. This was initiated at the start of PEGIFN and RVN treatment in all patients randomized to groups B and C as long as the hemoglobin was less than 15 g/dl. In those patients with a hemoglobin greater than or equal to 15 g/dl we monitored the hemoglobin weekly and we initiated EPO as soon as the hemoglobin declined to less than 15 g/dl. We increased the EPO dose to 60,000 IU/ml if the hemoglobin declined by more than 2 g or if the hemoglobin did not rise by at least 1 g. If the hemoglobin rose to 15 g/dl or greater, we did not administer the weekly EPO dose and we reduced the dose to 20,000 IU/week. Thereafter, we adjusted the dose of EPO between 20,000 and 60,000 IU/weekly or every other week to maintain the hemoglobin within the range of 12 to 15 g/dl.
We monitored the serum hemoglobin at weekly intervals during the first month, at 2-week intervals during the second month, and monthly intervals thereafter as long as the hemoglobin was stable. For those patients in which the dose of EPO or RVN was being adjusted or the hemoglobin was not stable; we measured this more frequently. We did not use either granulocyte colony stimulating factor or interleukin-11 for treatment of neutropenia or thrombocytopenia, respectively.
Monitoring Virologic Response and Testing for HCV RNA
We assessed HCV RNA at baseline just prior to the initiation of treatment and then at monthly intervals until the patient was either HCV RNA undetectable or had failed to achieve an EVR. We defined rapid virologic response (RVR) as HCV RNA undetectable at treatment week 4. We defined EVR as having either a 2 log decline in HCV RNA from the pretreatment baseline or as being HCV RNA undetectable at treatment week 12. We stopped treatment in all patients who failed to achieve EVR. We also stopped treatment in all patients who had detectable HCV RNA at treatment week 24. Thus, only those patients who were HCV RNA undetectable prior to or at treatment week 24 remained on treatment. In those patients who continued treatment past week 24, we assessed HCV RNA at treatment weeks 36 and 48 (end-of-treatment) and at weeks 60 and 72 (12 and 24 weeks after treatment had been stopped). Patients who were dropped from the protocol because of adverse events after week 24 had repeat HCV RNA testing at their next scheduled monthly visit. Those patients who remained HCV RNA undetectable continued to be followed and were retained in the protocol.
At the start of the study, we measured HCV RNA by the Cobas Amplicor PCR assay (Roche Molecular Systems, Branchburg, NJ) in our center's Molecular Diagnostics laboratory. The lower limit for detection of HCV RNA with this assay was 100 IU/ml. About halfway through this study, our center acquired a TaqMan real-time PCR unit (Roche Molecular Systems). The lower limit for detection of HCV RNA with this assay was 25 IU/ml. We measured all subsequent samples with this assay. In all patients, we measured the 72-week sample utilized to define SVR with the TaqMan assay even if we had measured previous samples by Amplicor PCR. Because of cost limitations, we did not reassay by TaqMan the samples that had previously been assayed by the Amplicor assay.
Statistical Analysis
The primary endpoint of this pilot trial was to determine if SVR could be enhanced by utilizing EPO at the initiation of HCV therapy with either a standard weight-based dose or a higher dose of RVN. We did not anticipate that SVR would be affected by EPO in the group treated with the same starting dose of RVN as the control group. However, we hypothesized that a higher starting dose of RVN would reduce relapse and therefore enhance SVR. Unfortunately, no preliminary data was available from other studies upon which to estimate sample size from these anticipated changes. Our data analysis plan stated that if no significant SVR difference was present between the 2 groups treated with the standard dose of RVN, these 2 groups could be combined and SVR compared to the group treated with the higher dose of RVN. Secondary endpoints were the percentage of patients who required RVN dose reduction, the mean RVN dose received, and the percent of patients with a hemoglobin below 10 mg/dl. We reported values for various demographic, biochemical, and virologic characteristics of the study population as mean + SD. We compared differences between the means of specific variables with a paired Student t test or the Mann-Whitney rank sum test as appropriate. We performed 2 types of analyses. In the intention-to-treat analysis, we included all patients who received at least 1 dose of study drug, and we counted those who dropped out of the study as nonresponders regardless of their virologic response (VR). In addition, we also performed a per-protocol analysis. In this analysis, we excluded from the analysis patients who dropped out of the study before their VR could be characterized. Thus, we counted as nonresponders in the per-protocol analysis only those patients who were treated for 12 weeks and failed to achieve EVR, who were treated for 24 weeks and still had detectable HCV RNA, or who developed breakthrough between weeks 24 and 48. We followed patients who stopped both drugs after they were HCV RNA negative, but before week 48, through week 72 and we included them in the analysis as either an SVR or relapse. We compared the difference in VR rates between various groups by analysis of variance. A P value of 0.05 was considered significant.
|
|
| |
| |
|
|
|
