Several groups [9-14] have investigated factors that might predict a sustained response to interferon therapy or a subsequent relapse after therapy in patients with chronic hepatitis C. However, such factors have not yet been clearly identified. Moreover, it is not well documented whether HCV can be completely eradicated in patients who had a long-term response to interferon therapy. To identify predictors of a long-term response and to determine whether HCV infection can be eradicated in patients with a long-term response, we investigated the clinical, virologic, and histologic effects of interferon- therapy during and 1 year after therapy in patients who had either a long-term or a short-term response to therapy.

Methods

Top Methods Results Discussion Author & Article Info References

Ninety patients with chronic hepatitis C were treated with human natural lymphoblastoid interferon- (Sumiferon, Sumitomo Pharmaceutical, Osaka, Japan) given intramuscularly at a dose of 6 MU three times a week for 6 months during 1991 and 1992. Written informed consent was obtained before the initiation of therapy. All patients had had increased serum aminotransferase levels (at least 1.5 times the upper limit of the normal range) for at least 6 months and had histologically proven chronic hepatitis. All patients had antibody to HCV (II) and detectable HCV RNA in serum, and all were negative for hepatitis B surface antigen. Patients with evidence of other forms of liver disease and with decompensated liver disease were excluded.

Each month, all study patients were seen and tested for liver biochemical variables during and after therapy. During and at the end of interferon- therapy, serum alanine aminotransferase levels decreased into the normal range in 47 patients (52%, defined as responders); however, in the remaining 43 patients (48%, defined as nonresponders), the levels did not return to normal. The 47 responders were followed for more than 1 year after the cessation of therapy and were assigned to one of two groups according to the outcome after therapy: long-term responders and short-term responders. Long-term responders were defined as patients whose levels of serum alanine aminotransferase became normal during and at the end of therapy and who had sustained normal levels for more than 1 year after cessation of therapy. Short-term responders were defined as patients whose serum aminotransferase levels became normal during and at the end of therapy but whose levels increased again within 1 year after the cessation of therapy. Of the 47 responders, 22 were long-term responders and 25 were short-term responders. All patients gave written informed consent before they were enrolled in our study.

Serum and liver biopsy samples (paired and 1-year follow-up) were tested for HCV RNA by reverse transcriptase-PCR using primers from the 5'-noncoding region of the HCV genome. Portions of liver tissue weighing approximately 20 mg or more were immediately frozen in liquid nitrogen and were stored at –80°C until used. Histologic diagnosis was made using both hematoxylin and eosin and reticulum-stained liver sections according to international standards, and these specimens were also graded using the numerical scoring system of Knodell and colleagues [15] by a pathologist who was blinded to the status of the patients. Nonspecific reactive hepatitis was defined as minimal inflammatory infiltration in portal tracts.

Detection of Hepatitis C Virus RNA

Hepatitis C virus RNA in the serum and the liver was assayed by the reverse transcriptase-double PCR technique using primers from the 5'-noncoding region of the HCV genome, as previously described [5, 16, 17]. In brief, total RNA was extracted from frozen liver tissue by using guanidinium-phenol-chloroform methods [18]. The amount of nucleic acid extracted from the liver was determined by spectrophotometry, and 1.5 µg of total cellular nucleic acid was subjected to reverse transcription and amplification with "nested" primers in a double-PCR technique. Nucleic acid extracted from 50 µL of serum was dissolved in 50 µL of RNase-free water, 5 µL of which was used for reverse transcriptase-PCR. The oligonucleotide primers used were synthesized from the highly conserved 5'-noncoding region of the HCV genome [19]: For the first cycle of the PCR assay, 5'-GAGGAACTACTGTCTTCACGCAG-3' (sense; nucleotide numbers 37 to 59) and 5'-CATGGTGCACGGTCTACGAGACC-3' (antisense; nucleotide numbers 310 to 332) primers were used. For the second cycle of the PCR assay, 5'-GAGGAACTACTGTCTTCACGCAG-3' (sense; nucleotide numbers 37 to 59, the same sense primer used for the first cycle) and 5'-GCACTCGCAAGCACCCTATCAGG-3' (antisense; nucleotide numbers 280 to 302) primers were used. The products of the PCR reaction were analyzed by electrophoresis in a 1% agarose gel containing ethidium bromide, 0.5 µg/mL, and were visualized under ultraviolet light. The size of the expected HCV DNA product was 265 base pairs.

The relative sensitivity of the PCR assay was determined by testing the synthetic HCV RNA; the assay detected 50 molecules of the synthetic HCV RNA. The end-point titer of HCV RNA was estimated in a semi-quantitative fashion by testing serial 10-fold dilutions of the nucleic acids extracted from 50 µL of serum. Serial samples from individual patients were tested in the same run, and samples with known titers of HCV RNA were also tested as internal controls. Throughout our study, carryover PCR contamination was avoided by using techniques described by Kwok and Higuchi [20]. For each assay, known positive and negative serum samples were analyzed as controls in parallel with test samples.

A PCR assay was also done using all the components except the reverse transcriptase enzyme, as well as the known negative and positive controls, to exclude the possibility of contamination by PCR products. In all the liver biopsy specimens, the integrity of the extracted RNA was shown by amplification of ß-actin messenger RNA as an internal control.

Hepatitis C Virus Genotyping

The HCV genotype was determined by reverse transcriptase-double PCR with type-specific primers in the core domain according to the methods of Okamoto and colleagues [21]. In brief, extracted RNA from 50 µL of serum was reverse transcribed to complementary DNA with 10 pmol of the external antisense primer and with 100 units of Moloney murine leukemia virus-reverse transcriptase in a final volume of 10 µL. Forty µL of the first PCR solution containing 10 pmol of the external sense primer and 1 unit of Taq polymerase was added to the reverse transcriptase reaction solution in a final volume of 50 µL. This mixture was amplified with 35 cycles of PCR (94 °C for 1 minute, 55 °C for 1 minute, and 72 °C for 2 minutes). A 1-µL aliquot of the amplified first PCR solution was added to the nested PCR solution with 6 pmol of the universal sense primer and with 3 pmol of each of the type-specific antisense primers, and 30 cycles of the nested PCR was done in the same program as the first PCR. A 20-µL aliquot of the amplified nested PCR solution was analyzed by electrophoresis on 2.5% agarose gels containing ethidium bromide (0.5 µg/mL) or on 4% agarose gels (NuSieve 3:1 Agarose, FMC Bioproducts, Rockland, Maine), followed by ethidium bromide staining. According to this method, type I specific DNA fragments were detected as a 49-base pair band; type II, type III, and type IV specific fragments were detected as 144-, 174-, and 123-base pair bands, respectively.

Antibodies to Hepatitis C Virus

Antibodies to HCV were assayed by HCV enzyme immunoassay (II) and by HCV passive hemagglutination second-generation kits (Dinabott, Tokyo, Japan) according to the manufacturer's instructions.

Statistical Analysis

The chi-square test or the Fisher exact test was used for the statistical analysis of comparisons of two group frequencies. Where appropriate, laboratory features, titers of HCV RNA, and histologic activity index scores were compared by the Student t-test and by the paired t-test. Multivariate assessment by logistic regression was done to determine which variables were independently associated with a subsequent relapse after therapy. A multiple logistic regression model was built using 11 variables: 8 variables before therapy (age, sex, history of blood transfusion, HCV genotype, alanine aminotransferase value, serum HCV RNA levels, hepatic HCV RNA levels, and histologic activity index scores) and 3 variables at the end of therapy (the presence of serum HCV RNA, the presence of hepatic HCV RNA, and histologic activity index scores). Each variable was transformed into categorical data consisting of two or three simple ordinal numbers for univariate and multivariate analyses.

All factors that were at least marginally associated with a future relapse (P < 0.05) were tested by multivariate logistic regression analysis. A P value of less than 0.05 was considered to be significant. The Statistical Package for the Social Sciences (SPS), version 4.0 (SPS, Chicago, Illinois), was used for all statistical computations.

Results

Top Methods Results Discussion Author & Article Info References

The characteristics of long-term and short-term responders are shown in Table 1. No significant differences were noted between the two groups in mean age, incidence of history of blood transfusion, or mean initial alanine aminotransferase levels. However, more women were in the long-term group than in the short-term group (P = 0.0003), and more patients with HCV genotype II were in the short-term group than in the long-term group (P = 0.029) (the genotype for each patient is shown in Tables 2 and 3). Mean initial scores for histologic activity index did not differ significantly between the two groups.

View this table: [in this window] [in a new window]   Table 2. Hepatitis C Virus in Long-Term Responders to Interferon-*

View this table: [in this window] [in a new window]   Table 3. Hepatitis C Virus in Short-Term Responders to Interferon-*

Changes in Serum Hepatitis C Virus RNA during and after Therapy

Changes in serum HCV RNA with therapy are shown in Tables 2 and 3. The mean (±SD) pretreatment log titer of serum HCV RNA in the 22 long-term responders (2.1 ±1.3; range, 1 to 5) did not significantly differ from that in the 25 short-term responders (2.8 ±1.2; range, 1 to 6). At the end of therapy, serum HCV RNA could no longer be detected in 21 (95%) of the 22 long-term responders; 1 year after the cessation of therapy, it remained undetectable in 20 (91%) responders. One year after therapy, the two long-term responders (patients 4 and 5) had detectable HCV RNA in their serum. One of these patients (patient 5) had detectable HCV RNA in the serum and the liver both during and 1 year after the cessation of therapy, and the other (patient 4) became positive again 1 year after therapy.

Serum HCV RNA was undetectable in 16 (64%) of the 25 short-term responders at the end of therapy and remained detectable in the remaining 9 patients. One year after the cessation of therapy, all 25 patients again had detectable levels of HCV RNA in their sera. Overall, the sera of 37 (79%) of the 47 responders no longer showed detectable levels of HCV RNA; 21 (57%) of these 37 patients had a long-term response, whereas the remaining 16 patients (43%) had a relapse after the cessation of therapy.

Changes in Hepatic HCV RNA during and after Therapy

Tables 2 and 3 show changes in hepatic HCV RNA. The pretreatment mean (±SD) log titer of hepatic HCV RNA in the 22 long-term responders (3.6 ±1.2; range, 2 to 6) was greater than that in the 25 short-term responders (4.8 ±1.6; range, 2 to 8) (P = 0.009). At the end of therapy, HCV RNA was no longer detectable in liver and serum samples from 21 (95%) of the 22 long-term responders and remained undetectable in 19 (86%) responders 1 year after the cessation of therapy. Three long-term responders (patients 4, 5, and 12) had detectable HCV RNA in their livers 1 year after the cessation of therapy; 1 patient (patient 5) remained positive for HCV RNA in the liver and the serum throughout therapy, and the other 2 patients (patients 4 and 12) became positive again for hepatic HCV RNA 1 year after the cessation of therapy.

Hepatitis C virus RNA was detectable in the liver at the end of therapy in 19 (76%) of the 25 short-term responders, but titers had decreased significantly; hepatic HCV RNA was not detectable in the remaining 6 patients (24%). All 6 short-term responders whose liver specimens were negative for HCV RNA also had sera negative for HCV RNA at the end of therapy; however, they had a relapse within 1 year after the cessation of therapy. Overall, 27 (57%) of the 47 responders did not have detectable hepatic HCV RNA; 21 (78%) of these 27 patients had a long-term response, whereas the remaining 6 (22%) patients had a relapse after the cessation of therapy.

Changes in Histologic Activity Index Score and Liver Histologic Results

Figure 1 shows changes in histologic activity index scores. The mean (±SD) initial histologic activity index scores did not differ significantly between the two groups. In long-term responders, the mean histologic activity index score decreased from an initial score 9.3 ±3.0 to a score of 4.2 ±2.4 at the end of therapy; 1 year after the cessation of therapy, the score further decreased to 2.2 ±1.7 (P = 0.0001 for each) (Figure 1). In the 22 long-term responders, liver histologic results at the end of therapy showed chronic active hepatitis in 2 patients, chronic persistent hepatitis in 11, and nonspecific reactive hepatitis in 9. One year after the cessation of therapy, liver histologic results showed chronic persistent hepatitis in 9 long-term responders, nonspecific reactive hepatitis in 9, and normal liver in 4 (Table 2). Mild-to-minimal portal cellular infiltration still remained in 18 (82%) of the 22 long-term responders, even 1 year after the cessation of therapy. In one long-term responder (patient 5) whose serum and liver samples remained positive for HCV RNA both during and 1 year after therapy, histologic activity index scores and liver histologic results did not change significantly during the follow-up period. In the two long-term responders (patients 4 and 12) whose liver had HCV RNA that was again detectable 1 year after the cessation of therapy, liver histologic results showed chronic persistent hepatitis at the end of therapy and 1 year after therapy.

View larger version (23K): [in this window] [in a new window]   Figure 1. Changes in the mean histologic activity index score after interferon- therapy. Left. Index scores in the 22 long-term responders before, immediately after, and 1 year after therapy with interferon-. Right. Index scores in the 25 short-term responders before and immediately after therapy. The bars represent SDs.

The mean (±SD) histologic activity index score in the short-term responders also decreased at the end of therapy (initial score, 9.2 ±3.8; score at the end of therapy, 5.7 ±3.7; P = 0.0001). Liver histologic results also improved in the short-term responders at the end of therapy (Table 3). However, in the seven short-term responders from whom liver biopsy specimens were obtained 1 year after the cessation of therapy for the second course of interferon therapy, the mean histologic activity index score had increased again to near the initial score (data not shown). The mean histologic activity index scores at the end of therapy did not differ significantly between the two groups.

Univariate and Multivariate Analyses

Univariate analysis was done with 11 variables for the short-term and long-term responders. The eight variables for before therapy were age, sex, history of blood transfusion, serum alanine aminotransferase level, HCV genotype, serum HCV RNA titers, hepatic HCV RNA titers, and histologic activity index score. The three variables for the end of therapy were serum HCV RNA titers, hepatic HCV RNA, and histologic activity index score. In univariate analyses, the following four variables were associated with a subsequent relapse after therapy; sex (P = 0.0003), HCV genotype (P = 0.029), pretreatment titer of hepatic HCV RNA (P = 0.009), and the presence of hepatic HCV RNA at the end of therapy (P < 0.0001). Multivariate logistic analysis was then done with these four variables (Table 4). Among them, the persistent presence of hepatic HCV RNA at the end of therapy was associated (P = 0.004) with an increased risk for relapse after therapy.

Discussion

Top Methods Results Discussion Author & Article Info References

Our study showed that HCV infection appeared to be eradicated in most of the long-term responders to interferon- because HCV RNA could no longer be detected in their serum and liver samples and because a significant improvement occurred in their liver histologic results. However, the histologic improvement seemed to be gradual even after HCV RNA could no longer be detected in the liver because approximately 80% of the long-term responders still showed minimal or mild portal inflammation 1 year after the cessation of therapy. According to the initial severity of liver injury, more time may be needed for liver histologic results to return to normal. One year after the cessation of therapy, 10% of the long-term responders had detectable HCV RNA in their liver regardless of whether their aminotransferase levels continued to be normal. A longer-term follow-up study is needed to determine whether these patients have a relapse later or whether they become asymptomatic HCV RNA carriers with normal aminotransferase levels.

Hepatic HCV RNA was detectable in most of the short-term responders at the end of therapy—even though HCV RNA could not be detected in the serum—indicating that the liver may have acted as a reservoir for the return of the virus in these patients. In some of the short-term responders, HCV RNA could no longer be detected in liver and serum samples at the end of therapy; however, these patients had a relapse after the cessation of therapy with a return of detectable HCV RNA in the serum. It is possible that the virus, previously undetectable at the end of therapy, could have achieved, through replication, detectable levels 1 year after the cessation of therapy. Alternatively, this return of the virus may have been due to sampling errors at the end of therapy because a liver biopsy specimen of approximately 20 mg represents only about 1/70 000 of the entire liver weight; this finding may also be due to uneven distribution of HCV in liver. Such distribution has been noted in patients with chronic hepatitis C [22]. sup

Several groups [7-1423, 24] have previously reported on clinical and virologic factors that appear to be associated with a greater likelihood of response to interferon, such as the degree of liver histologic results; sex; initial viral levels in serum; and viral subtype. However, other researchers have reported [12] that there were no definite markers. In our study, four factors—including sex, HCV genotype, pretreatment titer of hepatic HCV RNA, and presence of hepatic HCV RNA at the end of therapy—were significant predictors in univariate analysis. Although a relatively small number of patients was studied, the higher proportion of women in the long-term responder group suggests that women have a better response than men. The higher prevalence of genotype II in the short-term responder group may suggest that type II-related HCV is more difficult to eradicate than type III- or type IV-related HCV. Viral load in the liver before treatment and the presence of hepatic HCV RNA after treatment significantly differed between short-term and long-term responders. Multivariate logistic analysis showed that the persistent presence of hepatic HCV RNA at the end of therapy was the most important predictor of a relapse after therapy; patients with detectable HCV RNA in liver at the end of therapy had a significantly higher risk for relapse.

Testing for HCV RNA in serum and liver samples can be helpful in monitoring therapy in patients with chronic hepatitis C, and results from testing show that interferon therapy leads to suppression of viral replication. However, our study and previous studies [6, 7] have shown that the disappearance of serum HCV RNA at the end of therapy cannot be relied on as a marker of long-term response because approximately half of the patients with serum samples negative for HCV RNA at the end of therapy had a relapse after therapy. Approximately 80% of patients in whom hepatic HCV RNA could no longer be detected at the end of therapy had a sustained remission, whereas nearly all patients with persistent presence of hepatic HCV RNA had a relapse after therapy.

The disappearance of hepatic HCV RNA appears to be a more reliable indicator of remission than the disappearance of serum HCV RNA. An explanation for this finding is that HCV replicates in the liver and levels of HCV RNA in the serum reflect this viral replication and that the greater amount of RNA in the liver than in the serum allows for a more sensitive assay in the detection of HCV RNA. Disappearance of hepatic HCV RNA does appear at least to be a necessary precondition for sustained remission. No long-term response occurred in patients with chronic hepatitis C who were treated with interferon- if hepatic HCV RNA could still be detected. Thus, the loss of detectable hepatic HCV RNA may be a useful marker for identifying clearance of the virus and may be useful for evaluating the appropriate timing for discontinuation of interferon- therapy.