These authors contributed equally to this work.
Article first published online: 23 DEC 2008
Copyright © 2009 American Association for the Study of Liver Diseases
Volume 49, Issue 5, pages 1442–1448, May 2009
How to Cite
Basso, M., Giannini, E. G., Torre, F., Blanchi, S., Savarino, V. and Picciotto, A. (2009), Elevations in alanine aminotransferase levels late in the course of antiviral therapy in hepatitis C virus RNA–negative patients are associated with virological relapse. Hepatology, 49: 1442–1448. doi: 10.1002/hep.22810
Preliminary results of this study were accepted for presentation at the Annual Scientific Meeting of the American College of Gastroenterology (Orlando, FL. October 3–8, 2008), and were published in abstract form in the American Journal of Gastroenterology 2008;103(Suppl. 1):S122–S123.
Potential conflict of interest: Nothing to report.
- Issue published online: 27 APR 2009
- Article first published online: 23 DEC 2008
- Accepted manuscript online: 23 DEC 2008 12:00AM EST
- Manuscript Accepted: 16 DEC 2008
- Manuscript Received: 24 OCT 2008
The incidence and clinical meaning of elevated alanine aminotransferase (ALT) in chronic hepatitis C patients who are hepatitis C virus (HCV)-RNA negative during pegylated interferon (PEG-IFN) and ribavirin therapy have not been completely characterized. In this study our aim was to assess the incidence, pattern, predictive factors, and clinical meaning of elevated ALT in a cohort of 173 chronic hepatitis C patients who obtained viral clearance during either PEG-IFNα2a or α2b and weight-based ribavirin therapy. Patients were defined sustained viral responders (SVRs) or relapser responders (RRs) on the basis of a serum HCV-RNA result at 24-week follow-up. SVR and RR were obtained in 141 (58%) and 32 patients (13%), respectively. Among the 173 study patients, 57 patients (33%) had undetectable serum HCV-RNA and elevated ALT in at least one evaluation (weeks 2, 4, 12, 24 in all genotypes, and week 48 in HCV genotype 1 and 4 alone), and this phenomenon was not differently distributed between SVRs and RRs. No pretreatment demographic (age, gender), clinical (ALT levels, histological grade and stage, body mass index) and viral (load, genotype) parameter was associated with this phenomenon. The incidence of elevated ALT levels was not associated with type of PEG-IFN and ribavirin dose. Elevated ALT levels showed a different longitudinal pattern, occurring more frequently between week 12 and the end of treatment in RR as compared to SVR patients (90% versus 9%, P = 0.000001). Conclusion: The occurrence of elevated ALT levels in HCV-RNA-negative patients during PEG-IFN and ribavirin therapy is a fairly frequent and unpredictable phenomenon. Although ALT elevation per se is not associated with a greater risk of relapse, its occurrence in the later phases of therapy is more common in relapsing patients. (HEPATOLOGY 2009.)
The current standard of care for patients with chronic hepatitis C virus (HCV) infection is pegylated interferon alpha (PEG-IFNα) in combination with oral ribavirin for 24 or 48 weeks according to viral genotype. Pegylated formulations have been developed for both interferon α2a and interferon α2b: the two compounds are different in terms of pharmacologic formulation and pharmacokinetics, although they have comparable safety profiles, and patients treated with either drug show similar sustained viral response (SVR) rates.1
During antiviral treatment of HCV infection, early viral response (EVR)—defined as a 2.0 log drop of viral load or HCV-RNA clearance at week 12—is a good predictor of SVR and has a negative predictive value above 95% for nonresponse (NR) to treatment.2 Furthermore, a rapid decrease of alanine aminotransferase (ALT) and its normalization during treatment can be considered an indicator of favorable response to treatment, although this biochemical feature has no role in predicting SVR in patients treated with standard interferon.4 On the other hand, viral response is not always associated with biochemical response, as previously observed in patients treated with standard interferon and ribavirin, and with PEG-IFN α2a monotherapy.5, 6 In fact, Hung et al.5 observed that 13% of chronic hepatitis C patients who obtained SVR to standard interferon and ribavirin treatment showed persistently elevated ALT during treatment, whereas in the study by Zeuzem et al.6 41% of patients did not achieve ALT normalization until the end of therapy. These findings seem to suggest that lack of ALT normalization is not necessarily associated with a decreased efficacy of treatment. However, this phenomenon has not been characterized in a systematic fashion, and little is known about its incidence, clinical characteristics, longitudinal pattern, and clinical relevance in chronic hepatitis C patients treated with PEG-IFNα and ribavirin combination therapy.
In this study we analyzed a cohort of chronic hepatitis C patients who became HCV-RNA-negative during treatment with PEG-IFNα and ribavirin. We compared the pretreatment demographic, clinical, and viral characteristics of patients who showed elevated ALT levels with those of patients who obtained ALT normalization so as to identify parameters associated with this phenomenon. Furthermore, we evaluated whether discordance between viral (i.e., negative HCV-RNA) and biochemical (i.e., elevated ALT levels) responses were associated with response relapse (RR). Lastly, we assessed whether elevated ALT levels showed a peculiar longitudinal pattern in SVR and RR patients.
Patients and Methods
In all, 244 consecutive patients with a diagnosis of chronic hepatitis C who underwent PEG-IFN and ribavirin combination antiviral treatment at a single institution were considered for this study. All patients were positive for serum HCV-RNA detected by polymerase chain reaction (PCR) and had elevated serum ALT levels (normal value ≤40 IU/L) for at least 6 months before starting therapy. Other criteria for inclusion in the study were: hemoglobin >12 g/dL, leukocyte count >3,000/mm3, platelet count >100,000/mm3, and bilirubin and creatinine serum levels within normal limits. Patients were excluded if they tested positive for human immunodeficiency virus infection or had other causes of liver disease, previous organ transplantation, preexisting psychiatric disease, seizure disorders, cardiovascular disease, hemoglobinopathies, hemophilia, poorly controlled diabetes, autoimmune diseases, intravenous drug use and/or alcohol abuse, and if they were unable to use contraception. Details of all patients undergoing treatment were collected prospectively in a database that was retrospectively analyzed. The local Review Board approved the study and informed consent was obtained from all patients.
Patients were treated with either PEG-IFNα2a (180 μg/week subcutaneously, 93 patients) or PEG-IFNα2b (1.5 μg per kg/week subcutaneously, 151 patients) plus oral ribavirin in two separate doses (total dose 800 mg for patients weighing <65 kg, 1,000 mg for patients with a body weight ranging from 65 to 85 kg, and 1,200 mg for patients weighing 85 kg or more). This was not a randomized study, and selection of PEG-IFN was at the discretion of the study physicians in charge of the patients. Treatment duration was 24 weeks for genotype 2 and 3 patients, and 48 weeks for genotype 1 and 4 patients. At week 12, treatment was discontinued in patients who did not clear HCV-RNA or who had a reduction in viral load lower than 2.0 log as compared to baseline.2 At week 24, treatment was discontinued in genotype 1 and 4 patients who were still HCV-RNA-positive.2
Clinical and biochemical evaluations were performed at week 2, week 4, and then monthly thereafter while on treatment, and at weeks 12 and 24 during follow-up. Qualitative HCV-RNA was determined at week 2, week 4, week 12, week 24 in all patients, at week 48 in genotype 1 and 4 patients alone, and at 24 week of follow-up in all patients. Quantitative serum HCV-RNA was determined at baseline and at week 12. Serum HCV-RNA was measured by Amplicor HCV Monitor (Roche, Milan, Italy; cutoff limits, quantitative test: 600 IU/mL; qualitative test: 50 IU/mL). HCV genotype was determined before treatment in all patients with the INNO-LiPA HCV II kit (Bayer Diagnostics, Emeryville, CA). Both end of treatment response (ETR) and SVR were assessed in all patients. ETR without SVR was defined as RR, and NR was defined as lack of HCV-RNA clearance during treatment and at follow-up.2, 7 Patients were considered RR and SVR on the basis of a serum HCV-RNA result at 24 weeks of follow-up even if treatment was discontinued before the assigned schedule due to side effects or noncompliance to therapy. Liver biopsy was performed according to the Menghini technique, the biopsy specimen was formalin-fixed and paraffin-embedded, and histological grading and staging were scored according to Ishak et al.'s classification.8 Liver biopsy was carried out in a proportion of genotype 1 and 4 patients and in none of genotype 2 and 3 patients. Before treatment, body mass index (BMI; weight in kilograms divided by the square of the height in meters) was recorded in all patients.
In this study we included patients who, during therapy, contemporaneously had undetectable HCV-RNA and elevated ALT levels (i.e., >40 IU/L) in at least one of the programmed evaluations (i.e., week 2, week 4, week 12, week 24 in all patients, and at week 48 in genotype 1 and 4 patients alone). In particular, we compared the baseline demographic (age, gender, BMI), clinical (serum ALT, histological grade and stage), and viral (HCV genotype, quantitative HCV-RNA) characteristics of patients who showed elevated ALT levels with those of patients with normal ALT levels. Evaluation of ALT modification during treatment was carried out considering the difference between baseline and on-treatment ALT values of every patient when discordance between viral and biochemical response (i.e., negative HCV-RNA and elevated ALT) was observed, expressed as absolute value. In patients with elevated ALT levels the main putative causes for elevated ALT (i.e., hepatitis A and B infection, cytomegalovirus and Epstein-Barr virus infection, autoimmunity, thyroid dysfunction, metabolic derangement, iron overload, drugs, alcohol abuse) were evaluated by means of thorough clinical investigation and appropriate testing. Finally, we assessed whether the presence of elevated ALT levels had a peculiar longitudinal pattern in SVR and RR patients.
Continuous data are expressed as median and 95% confidence interval (95% CI). Discrete variables are expressed as absolute number and percentage. The Mann-Whitney U-test was used to compare continuous variables and the χ2-test (with Yates' correction for continuity) and Fisher exact test were used to compare categorical variables. A P value of 0.05 or less in a two-tailed test was considered statistically significant.
The main baseline demographic, clinical, and viral characteristics of the 244 patients who underwent PEG-IFNα and ribavirin antiviral therapy are shown in Table 1. Overall, SVR was obtained in 141 patients (58% of the study population: 83/151 treated with PEG-IFNα2b, 58/93 treated with PEG-IFNα2a), whereas RR and NR were observed in 32 (13%) and 71 (29%) patients, respectively, without a statistically significant difference in treatment response between the two PEG-IFNs (P = 0.340). SVR was obtained in 102 genotype 2 and 3 patients (80%) and in 39 genotype 1 and 4 patients (34%). Neither the baseline (1,000 mg/day versus 1,000 mg/day) and on-treatment (992 mg/day and 13.8 mg/kg/day versus 1,000 mg/day and 13.3 mg/kg/day) median ribavirin doses were significantly different between RR and SVR patients.
|Age||years||median (95%CI)||46 (44–49)|
|Gender||males||% (n)||66% (160)|
|Body mass index||kg/m2||median (95%CI)||24 (23–24)|
|ALT||IU/L||median (95% CI)||92 (84–100)|
|HCV genotype 1–4||% (n)||52% (128)|
|HCV genotype 2–3||% (n)||48% (116)|
|HCV-RNA||IU/mL||median (95%CI)||676,500 (508,106–835,966)|
Incidence of Elevated ALT Levels and Baseline Patient Characteristics.
Overall, among the 173 patients who had undetectable serum HCV-RNA during treatment, 57 patients (33%) showed elevated ALT levels (i.e., >40 IU/L) in at least one programmed evaluation. The incidence of elevated ALT levels was not significantly different between SVR (47/141 patients, 33% of all SVRs) and RR patients (10/32 patients, 31% of all RRs) (P = 0.987). Table 2 summarizes the main baseline characteristics of the 173 patients who had undetectable serum HCV-RNA during therapy, subdivided according to the presence of normal and elevated ALT levels. As far as age, gender, BMI, baseline ALT levels, and HCV genotype and load are concerned no significant difference was observed between the two subgroups of patients. A similar proportion of patients in the two subgroups had HCV-RNA serum levels above 400,000 IU/mL (70% versus 71%, P = 1.0). Significant histological fibrosis (score >2) had a similar prevalence in patients with normal and elevated ALT (34% versus 47%, P = 0.547), and median histological grading was not significantly different in the two subgroups of patients (score 6 versus score 6, P = 0.163).
|Variable||Unit||Normal ALT (n = 116)||Elevated ALT (n = 57)||P|
|Age||years||median (95%CI)||45 (40–49)||44 (40–50)||0.730|
|Gender||males||% (n)||60% (70)||68% (39)||0.388|
|Body mass index||kg/m2||median (95%CI)||23.0 (23.0–24.0)||24.0 (23.0–25.0)||0.115|
|ALT||IU/L||median (95% CI)||90 (80–103)||92 (79–124)||0.365|
|HCV genotype 1–4||% (n)||36% (42)||32% (18)|
|HCV genotype 2–3||% (n)||64% (74)||68% (39)||0.670|
|HCV-RNA||IU/mL||median (95%CI)||583,000 (482,409–834,770)||711,500 (298,946–964,882)||0.956|
Influence of Treatment.
The baseline (1,000 mg/day versus 1,000 mg/day) and on-treatment (967 mg/day and 13.0 mg/kg/day versus 1,000 mg/day and 13.5 mg/kg/day) median ribavirin dose was not significantly different between patients with normal and elevated ALT levels, and during treatment daily ribavirin dose had to be reduced in a similar proportion of patients in the two groups (normal ALT = 21%, elevated ALT = 16%, P = 0.539). The occurrence of elevated ALT levels showed a trend toward a greater incidence in patients treated with PEG-IFNα2a (29/71, 44%) as compared to patients treated with PEG-IFNα2b (28/102, 27%), although this difference was not statistically significant (P = 0.095). The same analyses carried out further subdividing patients into SVRs and RRs showed no statistically significant difference (data not shown).
Pattern of Elevated ALT Levels.
Among SVR patients who showed elevated ALT levels, the majority (37/47 patients, 79%) showed a constant decrease of ALT levels as compared to baseline at all the programmed evaluations, whereas six patients (13%) showed an increasing ALT pattern, and four patients (8%) showed a fluctuating ALT pattern. In SVR patients who showed a constant decrease in ALT levels, the median time from HCV-RNA clearance to ALT normalization was 10 weeks. A reduction trend in ALT levels was observed also in the majority (8/10 patients, 80%) of RRs with elevated ALT, whereas two patients (20%) showed an increasing ALT pattern, and none had a fluctuating pattern. Overall, a median serum ALT increase as compared to baseline was 11 IU/L (95% CI, 3–57 IU/L) and 30 IU/L (95% CI, 17–55 IU/L) in SVR and RR patients, respectively.
Longitudinal Evaluation of Elevated ALT.
The presence of elevated ALT levels in HCV-RNA-negative patients showed a different chronological pattern: this phenomenon occurred between weeks 2 and 12 in 91% of SVR patients (43/47 patients), whereas it took place between week 12 and the end of treatment in 90% of RRs patients (9/10 patients, P = 0.000001, Fig. 1). Moreover, it was not an isolated finding: in approximately half of the patients it was observed at more than one timepoint with no different distribution between SVRs and RRs. Figure 2 shows the percentage of patients with elevated ALT in each group (RR and SVR) at the various timepoints of the study. Finally, at the time of identification of biochemical abnormality we performed a thorough investigation of the putative causes associated with elevated ALT levels in each patient (i.e., hepatitis A and B infection, cytomegalovirus and Epstein-Barr virus infection, autoimmunity, thyroid dysfunction, metabolic derangement, iron overload drugs, alcohol abuse), and all these tests resulted completely negative in all patients.
At 24-week follow-up, 44 of the 47 SVR patients who had negative HCV-RNA and who showed elevated ALT levels during treatment had normal ALT (94%), whereas three patients (6%) still had elevated ALT levels. In this latter group of patients, one patient had gained weight (his BMI increased from 28.0 kg/m2 to 30.0 kg/m2), and the remaining two patients had started drinking more than 30 g alcohol per day. In all RR patients the reappearance of serum HCV-RNA was associated with elevated ALT levels.
Elevated aminotransferase levels are considered a sign of hepatocellular damage due to their high hepatic concentration, and ALT is more specific for liver damage than aspartate aminotransferase.9 In patients with chronic HCV infection, persistence of elevated ALT levels despite viral clearance during PEG-IFN and ribavirin therapy is a phenomenon that may be encountered in clinical practice, although its incidence (and its association with patients and viral characteristics) and treatment schedule have not been fully elucidated. Furthermore, the possible association between the presence of elevated ALT levels and viral relapse has not been tested.
In this study, carried out in a large cohort of anti-HCV patients in clinical practice, we observed that ALT levels were elevated in approximately one-third of patients who were HCV-RNA-negative during PEG-IFN and ribavirin therapy. The lower incidence of elevated ALT levels that we observed in this study (i.e., 33%) as compared to the incidence observed in the few previous studies that reported this phenomenon may be due to several reasons. First, we studied patients who were HCV-RNA-negative alone, and second, we treated patients with PEG-IFN and ribavirin10, 11; thus, a possible positive action exerted by the nucleoside analog ribavirin on biochemical liver damage is likely.12 Whereas the majority of our patients (79%) showed a constant decrease of ALT levels as compared to baseline, a subgroup of patients showed either an increase in ALT values (13%) or a fluctuating ALT pattern (8%). From the safety point of view, we observed that the degree of ALT elevation was generally mild, and none of the patients showed an on-treatment ALT level greater than twice the pretreatment level. Noteworthy, this phenomenon disappeared at 24-week follow-up in 94% of patients, and in the remaining 6% there was a clear explanation for persistence of the elevated ALT.
Although in our study we observed that the presence of elevated ALT levels in patients who are HCV-RNA-negative during antiviral treatment seems to be a rather frequent and generally mild phenomenon, we were not able to identify a predictive factor for its occurrence. As a matter of fact, all the pretreatment demographic, clinical, and viral parameters that we evaluated were not significantly different between patients with elevated and normal ALT levels. As far as the treatment schedule is concerned, we found that the type of PEG-IFN and the dose of ribavirin (both at baseline and on-treatment) were not associated with elevated ALT. Despite similarities in the study framework, our results cannot be compared with those recently obtained in a study carried out in a smaller cohort of chronic hepatitis C patients in the United Kingdom.10 In fact, in this study patients were almost exclusively treated with PEG-IFNα2b (i.e., 93% of the study cohort), and patients who obtained SVR and RR were analyzed together with NRs.10
When we analyzed whether the incidence of elevated ALT levels had any prognostic value, we observed that this phenomenon was similarly distributed between HCV-RNA-negative patients who showed SVR and RR, and therefore, as such, cannot be considered a parameter associated with a greater risk of relapse. However, a peculiar finding of our study was that longitudinal evaluation of elevated ALT identified a significantly different pattern of ALT elevation in RR as compared to SVR patients. Indeed, in the majority of RR patients, elevated ALT levels appeared during the late phase of treatment and showed no fluctuating pattern, whereas in SVR patients this phenomenon was mainly observed during the early phase of treatment. A subanalysis performed, lowering the threshold for elevated ALT to 30 IU/L, identified a greater number of patients with elevated ALT levels (52%), as expected, although it did not modify the results of the study.13, 14 In fact, using this “new ALT definition,” 75/141 SVR and 15/32 RR patients had elevated ALT (P = 0.56), and the longitudinal pattern of ALT elevation reflected the one observed with a higher ALT cutoff (data not shown). Once again, these results cannot be compared with those previously obtained by Thurairajah et al.,10 because in their study the longitudinal pattern of elevated ALT was analyzed considering ETR and not SVR.
What can the clinical relevance of our study results be? We observed that elevated ALT levels in the absence of detectable serum HCV-RNA occurred in 33% of patients who attained SVR, and a thorough evaluation of these patients was not able to identify any cause responsible for enzyme alteration. Lack of ALT normalization may discourage patients from continuing therapy and may worry the clinician about treatment outcome, raising the suspicion of the presence of concomitant causes of liver damage. To translate this observation into clinical practice means that the decision of discontinuing treatment solely on the basis of elevated ALT levels in patients who are HCV-RNA-negative would result in a substantial proportion of patients not being cured of infection. A large study reported that biochemical abnormalities, including liver enzymes alteration, led to discontinuation of treatment in 12% of chronic HCV patients treated with PEG-IFNα2a,15 whereas another study found an ALT doubling with respect to baseline values in 2% to 3% of patients treated with PEG-IFNα2b and ribavirin.16 Neither of these studies, however, indicated if serum HCV-RNA was detectable when ALT levels were higher than normal.15, 16
In this study we identified no parameter, either of the patient or of the virus, that was able to predict liver enzyme alteration despite viral clearance. In particular, viral genotype and load and advanced fibrosis seemed to play no role in ALT elevation. Although elevated ALT levels were not associated with treatment schedule, this study cannot prove/disprove the possibility that in some patients ALT elevation was possibly due to a mild PEG-IFN-induced liver toxicity. Finally, although we cannot rule out a specific role played by other histological characteristics that we did not take into account (e.g., pretreatment steatosis), we observed that both BMI (i.e., a metabolic determinant of steatosis) and HCV genotype 3 (i.e., a viral determinant of steatosis) were not significantly associated with elevated ALT levels, therefore indirectly ruling out steatosis as a possible culprit for the observed phenomenon.
Indeed, the most important finding of our study was that elevated ALT showed a peculiar pattern in SVR and RR patients. This finding, if confirmed in further studies, may suggest that patients with elevated ALT levels during the late phase of treatment may hypothetically be in need of a prolonged course of treatment so as to avoid relapse.17 However, what are the bases for this suggestion? In order to explain this phenomenon we formulated two hypotheses. Patients who obtained SVR more frequently showed elevated ALT levels during the first phase of treatment; it is likely that in these patients elevated ALT levels can be the result of an early and stronger immunomodulatory effect exerted by both (PEG)-IFN and the modified cytokines milieu, eventually determining clearance of infected hepatocytes.18–20 On the other hand, elevated ALT levels observed during the late phase of treatment despite HCV-RNA negativity in RR patients might be due to a subclinical viral activity that emerges at the end of therapy, once pharmacological pressure is withdrawn. Indeed, although the HCV-RNA assay we used in this study is commonly used in clinical practice (lower limit of detection 50 IU/mL), we cannot exclude that in RR patients the use of a more sensitive technique such as transcription-mediated amplification (TMA) would have been able to detect the presence of minimal residual viremia before the clinical diagnosis of relapse.21 Unfortunately, the recent studies that assessed minimal residual viremia by means of TMA in patients who relapsed after PEG-IFN and ribavirin therapy did not report the biochemical status of patients who were HCV-RNA-negative/TMA-positive during treatment.22–24
In conclusion, our study showed that mildly elevated ALT levels can be observed rather frequently in chronic hepatitis C patients who clear HCV-RNA during PEG-IFN and ribavirin therapy, although there are no pretreatment clinical predictors of this phenomenon. Noteworthy, if viral clearance occurs the presence of elevated ALT is not a poor prognostic factor per se, and although its appearance during the late phase of therapy may suggest a higher risk of relapse, its occurrence should not be a reason for treatment discontinuation. As far as the mechanism(s) underlying ALT elevation are concerned, an immunomodulatory effect exerted by IFN and minimal residual viremia may play a role in patients who obtain SVR and RR, respectively, although further studies are needed so as to better clarify these aspects.
- 5Is delayed normalization of alanine aminotransferase a poor prognostic predictor in chronic hepatitis C patients treated with a combined interferon and ribavirin therapy? J Gastroenterol Hepatol 2002; 17: 1307–1311., , , , , , et al.
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.