Efficacy and Tolerance of Interferon β Plus Ribavirin Treatment for Chronic Hepatitis C Patients with Depression or Thrombocytopenia Comparison with Pegylated Interferon α Plus Ribavirin Treatment

Hepatitis C virus (HCV) infection is a major cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma (HCC) and thus represents a significant public health problem [1,2]. The primary objective of treatment for chronic hepatitis C is to eradicate HCV, achieve Sustained Virological Response (SVR), and to prevent progression to cirrhosis or HCC [3]. We previously reported that the eradication of HCV infection decreases the occurrence of hepatocellular carcinoma [4].


Introduction
Hepatitis C virus (HCV) infection is a major cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma (HCC) and thus represents a significant public health problem [1,2]. The primary objective of treatment for chronic hepatitis C is to eradicate HCV, achieve Sustained Virological Response (SVR), and to prevent progression to cirrhosis or HCC [3]. We previously reported that the eradication of HCV infection decreases the occurrence of hepatocellular carcinoma [4].
Pegylated interferon α (PEG-IFNα) plus Ribavirin (RBV) combination treatment has improved the SVR rate for patients with chronic hepatitis C [5]. Previous studies reported that 25-40% of such patients discontinued the combination treatment due to adverse effects, such as psychological problems or cytopenia [5][6][7][8]. Patients with depression are not considered suitable for this combination treatment. In Japan, natural human interferon β (nIFNβ) and RBV combination treatment has been approved and recommended for depressed patients with chronic hepatitis C, and some studies have shown that nIFNβ plus RBV treatment has equivalent efficacy and mild adverse effects [9][10][11].
This case-control study was done to compare the efficacy and safety of nIFNβ plus RBV combination treatment for Japanese chronic hepatitis C patients with that of PEG-IFNα plus RBV combination treatment.
The dose of nIFNβ or PEG-IFNα2b was reduced if a patient had an adverse psychological effect, the white blood cell count fell below 1500 × 10 6 /L, or the platelet count fell below 50 × 10 9 /L. Likewise, the dose of RBV was reduced if the hemoglobin level decreased to under 100 g/L. Treatment was discontinued if the hemoglobin level, white blood cell count, or platelet count fell below 85 g/L, 1000 × 10 6 /L, or 25 × 10 9 /L, respectively.

HCV RNA detection and clinical evaluation
Blood samples were collected from the patients just before treatment, at weeks 4,8,12,24, and 48 of the treatment, at the end of treatment (EOT), and at 24 weeks after EOT. The serum HCV RNA level at each point was determined by quantitative real time Polymerase Chain Reaction (PCR) assay (COBAS TaqMan HCV assay; Roche Diagnostics) with a linear dynamic range of 1.2 to 7.8 log 10 IU/mL [13]. Biochemical and hematological tests were performed once each month during treatment. All were measured by standard laboratory techniques in our hospital laboratory. Liver biopsy was performed for 103 (85.8%) of the 120 patients by experienced hepatologists and the stage of fibrosis (F0-4) and the grade of activity (A0-3) were established according to the METAVIR score [14].

Interleukin 28B and inosine triphosphate pyrophosphatase
Human genomic DNA was extracted from peripheral blood. Genotyping by the Single-Nucleotide Polymorphism (SNP) of the interleukin 28B (IL28B) (rs8099917) and inosine triphosphate pyrophosphatase (ITPA) (rs1127354) genes was done using the TaqMan Allelic Discrimination Demonstration Kit (7500 Real-Time PCR System; Applied Biosystems, Foster City, CA). Patients were genotyped as IL28B TT, TG, or GG and as ITPA CC, CA, or AA at the polymorphic site [15].

Virological response
SVR was defined as undetectable HCV RNA at 24 weeks after the end of treatment. Early virological response during treatment was categorized as follows: Rapid Virological Response (RVR), undetectable HCV RNA at week 4. Complete early viological response (cEVR), detectable HCV RNA at week 4 but undetectable at week 12. Relapse was defined as undetectable HCV RNA at the end of treatment, but non-SVR. Non-response was defined as detectable HCV RNA through treatment. Non-SVR was defined as not achieving SVR, including relapse and non-response.

Assessment of depression
A questionnaire survey was conducted when physicians perceived the necessity or patients complained depressive symptom, using the Beck Depression Inventory II (BDI-II) and the Pittsburgh Sleep Quality Index (PSQI) [16,17]. Patients with a BDI-II score of 14 or more were considered to have the onset of depression symptoms. Patients with a PSQI score of 11 or more were identified as having sleep disorder.

Statistical analysis
All statistical analyses were performed using JMP ® ver. 9 (SAS Institute Inc., Cary, NC). Data are reported as mean ± standard error (SE), median [first quartile, third quartile], or percentage for each category. Spearman correlation coefficient analysis was used to analyze patient characteristics. The student's t test and the Mann-Whitney U test were used to compare between-group differences. A P value of <0.05 was considered statistically significant. Table 1 shows the pretreatment clinical characteristics of the patients of groups B and A. No significant differences in baseline serum HCV RNA levels, histology of fibrosis, or distribution of IL28B or ITPA genotype were found between the B1 and A1 or B2 and A2 groups. Group B1 had a significantly higher rate of history of IFN therapy, higher levels of aspartate aminotransferase and α-fetoprotein, and a significantly lower platelet count than group A1. Group B2 had a significantly higher rate of depression and a lower rate of IL28B TT genotype than group A2. The discontinuation rates of group B1 and B2 during prior IFN treatment were higher than those of group A1 and A2, but there was no significant difference. Tables 2 and 3 show the virological response to both treatments. There was no significant difference in the SVR rates of groups B1 and A1 (21.4% vs. 33.3%, P=0.328) or groups B2 and A2 (72.2% vs. 88.9%, P=0.402). Patients with RVR achieved SVR significantly more often than patients without RVR in group B1 and B2 (P = 0.001 and 0.029, respectively), but no significance was found for groupA1 and A2 (P=0.106 and 0.065, respectively). Patients with cEVR achieved SVR significantly more often than patients without cEVR in group B1, B2 and A1 (P<0.001 and P=0.008 and 0.020, respectively), but no significance was found for group A2 (P=0.111). In B1, IL28B TT patients showed significantly higher SVR rates than IL28B non-TT patients (33.3%, 8 of 24 patients and 0.0%, 0 of 14, respectively, P=0.017). In group B2 and A1, IL28B TT patients showed higher SVR rates than IL28B non-TT patients (B2; 75.0%, 9 of 12 vs. 60.0%, 3 of 5, and A1; 34.6%, 9 of 26 vs. 25.0%, 2 of 8, respectively) but the difference was not significant (P=0.600 and P>0.999, respectively). In group A2, all patients determined IL28B genotype had IL28B TT genotype.

Safety and tolerance of nIFNβ plus RBV treatment
There was no significant difference in the discontinuation rates between group B1 and A1 (P=0.164) or group B2 and A2 (P=0.486). Of the 42 group B1 patients, 11 (26.2%) discontinued the treatment because of malaise (3 patients at weeks 14, 20, 24), hyperthyroidism (1 patient at week 9), poor response (4 patients at weeks 15,16,19,38), new occurrence of HCC (2 patients at weeks 3, 25), and for personal reasons (1 patient at week 24). Of the 42 group A1 patients, 5 (11.9%) discontinued treatment, two because of malaise at week 38    Of the 60 group B patients, none had exacerbated or newly developed depression during treatment, but 7 (11.7%) of the 60 group A patients developed depression during treatment (P=0.002) and all of 7 required to reduce the dose of PEG-IFNα2b.
Of the 60 group B, none required a reduction of the dose of nIFNβ.
However, in the 42 group A1 patients, the adherence rate to PEG-IFNα2b was 85.2 [72.2, 96.9]% (median [first quartile, third quartile]), and in the 18 group A2 patients, the rate was 96.0 [87.8, 108.5]%. There was a significant difference in the adherence rates to IFN between groups B1 and A1 (P=0.005), but not between groups B2 and A2 (P=0.230). In 60 group A patients, 14 required a reduction of the dose of PEG-IFNα2b, 7 required because they developed depression, and other 7 required because their platelet counts fell below 50 × 10 9 /L. The SVR rate of these 14 group A, PEG-IFNα2b reduction patients was significantly lower (2 patients, 14.3%) than that of 46 group A patients who did not required a reduction (28 patients, 60.9%) (P=0.002). None required discontinuation the therapy because of adverse effect.

Influence of nIFNβ to blood count
The platelet count of group B increased to higher than baseline after week 8, but the platelet count of group A decreased throughout the treatment. Significant differences were found between groups B and A throughout the treatment (all P<0.001). Figure 1 shows the differences in the on-treatment changes of platelet count by ITPA genotype. The platelet counts of ITPA non-CC patients were lower than those of ITPA CC patients (significantly at weeks 4, 12 and 24, P = 0.013, 0.102, 0.039 and 0.008 at weeks 4, 8, 12 and 24, respectively) for group B (Figure 1a). The ITPA non-CC patients had a significantly higher decrease than the ITPA CC patients at week 4 (P=0.001), but there were no significant differences after week 8 (P = 0.141, 0.329 and 0.281 at weeks 8, 12 and 24, respectively) for group A (Figure 1b).
There was no significant difference in the decrease of hemoglobin level at weeks 4, 8 or 12 between groups B and A. At week 24, the decrease of the hemoglobin level of group A became significantly higher than that of group B (P=0.048), even though the adherence rate to RBV of group A was significantly lower than group B. Figure  2 shows on-treatment differences in the decrease of hemoglobin levels, by ITPA genotype. The hemoglobin levels of ITPA CC patients decreased significantly more than those of ITPA non-CC patients (P<0.001 at week 4 and P=0.002, 0.005, and 0.022 at weeks 8, 12 and 24, respectively) in group B (Figure 2a). The hemoglobin of ITPA CC patients decreased significantly more than those of ITPA non-CC patients at week 4 (P<0.001), but there was no significant difference between the genotypes after week 8 (P=0.252, 0.621 and 0.787 at weeks 8, 12 and 24, respectively) in group A (Figure 2b).

Discussion
The four major findings of the present study are as follows. First, the efficacy of the nIFNβ plus RBV treatment was equivalent to that of PEG-IFNα2b plus RBV treatment. Second, none of the patients treated with nIFNβ plus RBV had exacerbated or newly developed depression. Third, the platelet count of patients treated with nIFNβ plus RBV increased to higher than baseline after week 8 whereas they fell with PEG-IFN2b plus RBV. Finally, significantly fewer ITPA non-CC patients than ITPA CC patients treated with nIFNβ plus RBV had a decrease in the hemoglobin level.
Although the overall discontinuation rates for nIFNβ plus RBV treatment were slightly higher than for PEG-IFNα2b and RBV treatment, the number of patients discontinued because of interferoninduced side effects, such as malaise or thyroid disease, was equivalent. And the platelet count of patients taking nIFNβ increased to higher than baseline. The SVR rates for nIFNβ plus RBV treatment and PEG-IFNα2b plus RBV treatment were equivalent for chronic hepatitis C patients, however, because the mean age of the patients with HCV genotype 1 was over 65 years, the SVR rate of these patients was low as the previous studies [8,18]. These results indicate that nIFNβ plus RBV combination treatment is effective and safe for patients with chronic hepatitis C, in particular for those with depression or thrombocytopenia.  TPA CC ITPA nonCC PEG-IFNα is known to often cause and exacerbate psychological problems (the range was reported to be from 30 to 80%) [19,20]. The incidence of nIFNβ-induced psychological problems has been reported to be from 0 to 10% [9][10][11]21]. In this study, 38.3 % (23 0f 60) of the patients who received nIFNβ plus RBV treatment suffered depression before treatment, but all of them completed treatment without increasing antidepressants or reduction of nIFNβ. Of the patients who received PEG-IFNα2b and RBV treatment, 11.6% (7 of 60) developed depression during treatment. The difference in the frequency of nIFNβinduced and PEG-IFNα-induced psychological problems may be related to the higher elevation of serum level of interleukin-1 receptor antagonist (IL-1Ra) with anti-inflammatory effects by nIFNβ than by PEG-IFNα. The ratio of interleukin-1β to IL-1Ra is maintained within normal range because IL-1Ra does not decrease when nIFNβ is used [22,23].
PEG-IFNα often causes cytopenia. This side effects often lead to decrease the treatment dosage. In this study, patients who underwent nIFNβ plus RBV treatment had an increased platelet count after week 8 of treatment. Moreover, the platelet count of patients with ITPA CC was higher than at baseline after week 4 of treatment. In contrast, the platelet count of patients who underwent treatment with PEG-IFNα2b and RBV decreased throughout treatment, regardless of ITPA genotype. Fewer patients who underwent nIFNβ plus RBV treatment had a decrease in hemoglobin level than did patients who underwent PEG-IFNα2b and RBV, even though the adherence rate to RBV was significantly higher for the patients treated with nIFNβ plus RBV than for those treated with PEG-IFNα2b and RBV. Although the mechanism of the platelet count increase of the patients who receive nIFNβ plus RBV treatment remains to be clarified, the physiological linkage of hematopoietic factors may explain this phenomonon. Thrombosis has been reported in patients with iron deficiency anemia and Bilic and Bilic reported that the amino acid sequence homology of thrombopoietin and erythropoietin might explain it [24][25][26]. From our results, the ITPA CC might be associated with the increase of platelet count during nIFNβ plus RBV treatment. Further studies will be necessary to clarify the relation between nIFNβ plus RBV treatment and changes of platelet count.
The histories of prior IFN treatment of patients of both groups were different. However, patients of groups B were considered to be less tolerant to IFN treatment than those of groups A, because the discontinuation rates of prior IFN treatment among groups B were higher than those among groups A. Thus, we did not consider that that difference might weaken our result that nIFNβ plus ribavirin treatment had enough tolerance.
Because patients must go to a hospital for treatment three times a week, nIFNβ is somewhat more inconvenient than PEG-IFNα. However, this inconvenience is offset by the milder and fewer nIFNβrelated side effects and the increased platelet count during treatment.
This study has some limitations. One is the small number of patients. Although the SVR rates were not statistically different between groups B and A, there is a probability of difference when the number of patients were more. Further study with more patients will be necessary. Another is that the IL28B genotype was not determined for some patients, so careful interpretation must be made of the evaluation of the influence of IL28B on the efficacy of both treatments. Further study to determine the IL28B of all patients is desirable. The final limitation is that the study was not of a randomized design. Unfortunately, for ethical reasons it is very difficult to conduct a randomized study in Japan.
In conclusion, nIFNβ plus RBV combination treatment is an optional treatment for chronic hepatitis C, especially for patients with depression or thrombopenia.