| |
Peginterferon alfa-2b and ribavirin in patients with hepatitis C virus and decompensated cirrhosis: A controlled study
|
| |
| |
Journal of Hepatology Feb 2007
Angelo Iacobellis1, Massimo Siciliano2, Francesco Perri1, Brigida E.
Annicchiarico2, Gioacchino Leandro3, Nazario Caruso1, Laura Accadia1, Giuseppe Bombardieri2, Angelo Andriulli1
1 Division of Gastroenterology, "Casa Sollievo della Sofferenza" Hospital, IRCCS, San Giovanni Rotondo, Italy
2 Institute of Patologia e Semeiotica Medica, Catholic University, Rome, Italy
3 Division of Gastroenterology, "De Bellis" Hospital, IRCCS, Castellana Grotte, Italy
"....in decompensated cirrhotics HCV clearance after antiviral therapy may be life-saving, improves hepatic function, and would limit further clinical decompensation. Treatment should be encouraged in CTP classes A and B, and especially in patients infected by genotype 2. The benefit of treating patients with genotype 1 remains unproven.......As cirrhosis progresses, complications occur and liver transplantation becomes the only option to improve patient survival.....We report a controlled trial of therapy with PEG-IFN alfa-2b and RBV in decompensated cirrhotic patients. Our principal aim was to investigate the impact of therapy on the natural history of these patients with liver failure, based on the expectation of a survival gain in those patients clearing the virus after treatment..... During follow-up, there were 15 deaths in controls, and nine in the non-responders; the number of patients who underwent liver transplantation was four and one, respectively. Liver failure and/or hepatocellular carcinoma was the cause of death or transplantation in 18 controls and nine non-responders. All 13 patients who experienced SVR survived, with no need for transplantation. The annualized incidence of death from any cause was 2.34 per 1000 patient-years in controls, 1.91in non-responders, and 0 in patients with SVR.... the number of decompensated events was significantly higher in controls and non-responders as compared with patients who achieved an SVR, in whom improvement of both CTP and MELD scores was apparent at the end of follow-up.....Deaths and/or liver transplantation occurred in 19 controls and 10 non-responders, but in no patient with a SVR..."
ABSTRACT
Background/Aim: To evaluate long-term outcomes in decompensated HCV-related cirrhotic patients treated with antiviral therapy.
Methods: Of 129 eligible patients, 66 received peginterferon alfa-2b and ribavirin for 24 weeks, and 63 were controls. Survival and recurrence of liver failure events after therapy were main outcomes.
Results:
Therapy was tolerated by 27 patients, dose reduced in 26 for toxicity, and discontinued in 13 for intolerance.
End-of-therapy and sustained virological response (SVR) rates were 82.6% and 43.5% for HCV 2/3 patients, and 30.2% and 7.0% for HCV 1/4 patients.
During therapy, odds ratios for severe infections or deaths due to infection were 2.95 (95% C.I. 0.93-9.3) and 1.97 (95% C.I. 0.40-9.51) in treated patients as compared with controls.
During a follow-up of 30 months off-therapy, decompensated events occurred in 52, 33, and 3 of controls, non-responders, and SVR patients.
Odds ratios for ascites, encephalopathy, and oesophageal bleeding in treated patients significantly decreased as compared with controls.
Annualized incidence of death was 2.34, 1.91, and 0 per 1000 patient-years, respectively, in controls, non-responders, and SVR patients. Survival curves showed early separation of SVR patients from both non-responders and controls at approximately 6 months.
Conclusions: In decompensated cirrhotics, HCV clearance by therapy is life-saving and reduces disease progression.
Discussion
Recent studies have validated the feasibility of IFN-based therapies for decompensated cirrhotic patients, but have provided no data on the impact of therapy in disease progression, avoidance of transplantation, and improvement of life expectancy [6], [7], [8], [9], [10]. These hard clinical end-points are particularly applicable to patients with advanced disease, when liver function is more likely to deteriorate within a few years [13]. We report a significant improvement in overall and event-free survival as well as in clinical status and laboratory profile of patients who eventually cleared HCV after treatment with Peg-IFN alfa-2b and ribavirin administered for 24 weeks. During the follow-up, the number of decompensated events was significantly higher in controls and non-responders as compared with patients who achieved an SVR, in whom improvement of both CTP and MELD scores was apparent at the end of follow-up. Deaths and/or liver transplantation occurred in 19 controls and 10 non-responders, but in no patient with a SVR. Our findings support the use of antiviral combination therapy for patients with decompensated liver disease and might be of particular relevance for patients who will never be listed for transplantation, such as those with contraindications to surgery or those older than 65 years (the current age-limit to enlist in Italy). Indeed, six of 13 patients with SVR were in this age group. In long-term follow-up studies of patients with HCV infection treated with IFN-based therapy, death from liver failure was reduced in comparison to untreated controls [3], [4], [14], but this advantage waned after matching patients for baseline signs of liver function [4]. In the present study, patients chose whether they would receive therapy or not, but all were eligible to therapy and presented with equivalent parameters of liver function at baseline (Table 1), the only difference being their willingness to be or not to be treated. Consequently, the observed differences in their outcome are likely to reflect the real benefit of therapy.
In comparison with the small, uncontrolled case series of decompensated cirrhosis so far reported with IFN-based therapies, the present trial had unique peculiarities, as it included a group of untreated patients, enrolled exclusively patients with decompensated events, and reported survival data. In keeping with previous trials [6], [7], [8], [9], [10], therapy has been successful only in a minority of our patients (19.7%). However, data were particularly encouraging for HCV 2-infected patients in whom EOT and SVR were 82% and 41%. In contrast, in HCV-1, both EOT (30%) and SVR (7%) rates were disappointingly low. In the present trial, a high relapse rate was observed both in genotype 2 (41.6%) and genotype 1 (23.2%) patients, and might be related to either the suboptimal dosages of both Peg-IFN and RBV or the short treatment course. At the time present study was planned, decompensated cirrhotics were not considered candidate to antiviral therapy and, therefore, we purposely decided to use a cautious regimen, although a different one from currently recommended. Nevertheless, it should be considered that the likely benefit of prolonging treatment in these patients or increasing drug dosage needs to be counterbalanced by the poor tolerability of therapy.
In view of the myelosuppressive actions of IFN [15], current guidelines do not advise antiviral therapy in presence of platelets <70,000cells/μL or neutrophils <1500cells/μL, and suggest dose reduction and possible treatment discontinuation if values fall below 50,000 and 1500cells/μL [12]. Nineteen patients in the present series presented with cytopenia at enrollment and would have been excluded from therapy. Twenty-two individuals developed cytopenia during treatment, which was successfully managed by administration of erythropoietin, blood transfusion, and/or granulocyte-growth factor without dose modification in most of them. These findings warrant for a change in current guidelines.
The regimen of antiviral therapy administered to our patients led to an increased risk of severe infections that resulted in death of some of them. As compared with control patients, in treated patients the odds ratios were 2.95 for severe infection and 1.97 for dying from infection. As a precaution against potential harm of treatment, therapy was started one month after the patients recovered from the decompensated event, and growth factors were given whenever neutrophil count dropped <750cells/μL. We need to appreciate the factors predisposing decompensated cirrhotics to infections during antiviral therapy. Spontaneous episodes of septicaemia are to be expected in these patients, due to intra- and extra-hepatic shunting of bacteria delivered to the liver via the portal vein [16], and we estimated an incidence rate of 0.17 per 1000 patient-months of severe infection in our controls. During Peg-IFN therapies, the risk of infections increases independently from neutropenia [17] which would suggest preserved neutrophil function to be more relevant than absolute counts. Conversely, in previous therapeutic trials neutropenia was not associated with infection [18], [19] as long as >750μL neutrophils were maintained. At multivariate analysis, independent predictors of infections in our patients were CTP class C and neutrophil count <900μL during treatment. Therefore, it would be tempting to speculate that by either using growth factors more generously to keep neutrophil counts above these thresholds and denying treatment to CTP class C patients will constitute the safest way to go with decompensated cirrhotics.
In conclusion, in decompensated cirrhotics HCV clearance after antiviral therapy may be life-saving, improves hepatic function, and would limit further clinical decompensation. Treatment should be encouraged in CTP classes A and B, and especially in patients infected by genotype 2. The benefit of treating patients with genotype 1 remains unproven.
Introduction
In patients with compensated HCV-related cirrhosis, antiviral treatment has been investigated in several trials [1], [2], [3], [4]. The highest response rate has been reported from a subgroup analysis of cirrhotics included in a multinational trial, in which peginterferon (PEG-IFN) alfa-2a plus ribavirin (RBV) achieved a sustained virological response (SVR) of 50% [5]. Whether the benefit of viral clearance will, in turn, lead to a significant survival advantage remains to be determined.
As cirrhosis progresses, complications occur and liver transplantation becomes the only option to improve patient survival. In these circumstances, antiviral therapy is deferred due to concerns over treatment-induced side effects with potential to further deteriorate hepatic function. However, in contrast, antiviral therapy has demonstrated encouraging SVR (up to 33%) in decompensated cirrhotic patients; moreover, HCV re-infection of the liver graft may also be prevented in patients who cleared the infection [6], [7], [8], [9], [10]. After reviewing these reports, the International Liver Transplantation Society recommended interferon treatment in decompensated cirrhosis, and issued guidelines for selection of candidates to treatment [11].
Previous evidence of the benefits of therapy with PEG-IFN and ribavirin is comprised of small, uncontrolled case series examining cirrhotics treated with non-PEG-IFN, often at suboptimal dosages and duration of therapy. In addition, no study on current standard of treatment - the combination of Peg-IFN with RBV - is yet available. Furthermore, previous studies have only assessed efficacy and tolerability of therapy, whereas SVR is a surrogate endpoint in this setting, as the primary goal of therapy should be preventing disease progression.
We report a controlled trial of therapy with PEG-IFN alfa-2b and RBV in decompensated cirrhotic patients. Our principal aim was to investigate the impact of therapy on the natural history of these patients with liver failure, based on the expectation of a survival gain in those patients clearing the virus after treatment.
Methods
Patients' selection
From January 2002 to May 2003, all patients with HCV-related cirrhosis, consecutively admitted to hospital for ascites, variceal bleeding, and hepatic encephalopathy, were screened. Eligible patients had to be naive to conventional IFN and RBV therapy, whilst those previously treated with IFN monotherapy could be enrolled 6 months after treatment was discontinued. Exclusion criteria included rapid deterioration of liver and/or renal function (such as deepening jaundice or hepatorenal syndrome), detection of hepatocellular carcinoma, infection with immunodeficiency or hepatitis B viruses, current alcohol abuse, chronic invalidating disease (such as unstable cardiovascular disease or severe chronic obstructive lung disease), bacterial infections, platelets <35,000cells/μL, neutrophils <1000cells/μL, haemoglobin <10g/dL, total bilirubin >3mg/dL, and serum creatinine >2.0mg/dL. Of the 337 cirrhotics admitted, 76 were excluded for hepatocellular carcinoma, 21 for worsening of liver function during the index hospitalization, 53 for non-HCV-related disease (29 alcoholic, 15 HBV, five patients with haemochromatosis, and four biliary cirrhosis), 21 for being older than 75 years, 18 for previous combination antiviral therapy, and 19 for platelet, neutrophil or haemoglobin levels below stipulated. The remaining 129 patients were offered therapy after explanation of expected benefits and potential risks: 66 of them agreed to be treated and signed an informed consent form, whereas the remaining 63 did not and therefore served as controls. The study was approved by the Ethics Committee and conducted according to guidelines of the International Conference on Harmonization for Good Clinical Practice. The trial was designed, conducted, and results analyzed without any financial support from producers of drugs used in the study.
Treatment schedules
Treatment started 1 month after the decompensated event had been treated with appropriate therapy. Data on safety of PEG-IFN and RBV therapy in decompensated patients were not available at the time the study was started. However, since lethal infections had been reported in decompensated cirrhotics treated with non-pegylated IFN [9], [10], we decided to administer antiviral drugs at lower doses and for shorter time than currently recommended [12]. Consenting patients received 1.0μg/kg body weight of PEG-IFN alfa-2b (PEG-Intron, Schering-Plough) weekly in combination with oral RBV (Rebetol, Schering-Plough) at a daily dosage of 800 or 1000mg for body weight < or > 75kg, for 24 weeks.
Virological and safety evaluation
During the first 6 months of the trial, clinical visits and laboratory tests were scheduled monthly for treated patients, and every 12 weeks for controls. Qualitative serum HCV-RNA was determined at baseline and at weeks 4, 12, and 24 during therapy, and at week 24 of follow-up (Amplicor HCV test, Roche). At baseline, quantitative serum HCV-RNA (Cobas Monitor 2.0, Roche) and HCV genotyping (Innolipa HCV, Innogenetics) were also performed. Whenever neutrophil count dropped <750cells/μL or haemoglobin <10g/dL, granulocyte colony-stimulating factor (Neupogen, Amgen Inc., California, 300μg) or erythropoietin analog (Eprex, Janssen-Cilag, 10,000 units) was administered twice weekly; blood was transfused in the event of haemoglobin level <8.5g/dL. If there was no improvement in haematological parameters within 2 weeks, antiviral therapy was discontinued.
Adverse events were graded as mild (not requiring dose modification), moderate (requiring dose modification), or severe (requiring discontinuation of therapy), according to recommendations of the World Health Organization. Episodes of infection were recorded if infection was either suspected or confirmed (by bacteriologic cultures or positive radiograph), and treated with oral or parenteral antibiotics. Infections were categorized as severe if requiring hospital admission and/or treatment discontinuation.
After 6 months, treated and control patients entered a follow-up phase comprising scheduled outpatient visits 12 weeks apart, when they were monitored for reappearance of ascites or encephalopathy, oesophageal bleeding, hepatocellular carcinoma, liver transplantation, and death.
Results
Baseline characteristics
Baseline characteristics of treated and untreated patients (Table 1) did not differ significantly, with the exception of haemoglobin level. Fifty-three patients were older than 65 years, 22 treated and 31 controls. Most patients were naive to therapy, infected by HCV 1, in CTP class B, and presented with ascites.
Outcome of therapy
By intention-to-treat analysis, 32 patients were HCV-RNA negative at the end of treatment (EOT), and 13 patients (19.7%; 95% CI, 10-29%) at the end of follow-up. EOT and eradication rates were 82.6% and 43.5% for HCV 2/3 patients, and 30.2% and 7.0% for HCV 1/4 patients, respectively. At the univariate analysis of baseline features, only HCV genotype was significantly related with viral clearance, whereas viral load, type of presenting decompensated event, CTP class, MELD score, and previous IFN-monotherapy were not.
Tolerance of therapy and haematological effects
Full dosage and duration of treatment were tolerated by 27 (41%) patients whilst 13 patients discontinued for flu-like symptoms and depression (n=4), severe infections (n=7), and low-haematologic counts despite growth factors' administration (<20,000cells/μL or neutrophils <500 cells/μL; n=2). In the 26 remaining patients, dosages of both drugs were reduced for haemoglobin <10g/dL (n=7), neutrophils <750cells/μL (n=8), concurrent decrease of haemoglobin and neutrophils (n=5), and moderate infections (n=6); erythropoietin and granulocyte growth factor were administered, respectively, in seven and five individuals. During therapy, a drop in the three haematopoietic lineages of haemoglobin, neutrophil and platelet counts was observed. For the former two lineages, the Nadir value (1182±588cells/μL for neutrophils, and 74,472±36,697cells/μL for platelets) was reached during the first month of therapy, and persisted throughout the treatment period. Haemoglobin levels decreased smoothly as long as therapy was continued.
Morbidity during the initial 6 months of the study
When analyzed according to the time of the first event, there were 42 decompensated events in controls and 45 in treated patients (Table 2). By analyzing the individual events, the odds associated with therapy decreased when ascites, encephalopathy, oesophageal bleeding and hepatocellular carcinoma were considered. In contrast, in treated patients the odds were 2.95 (95% CI, 0.9-9.3) for severe infection and 1.97 (95% CI, 0.40-9.51) for dying from infection. The incidence of severe infection was 0.17 and 0.45 per 1000 patient-months in controls and treated patients, respectively. The majority of infections (70%) had origin from the urinary or respiratory tracts, and were more frequent in CTP class C, in treated patients, in cases with neutrophil counts at baseline ?2100mm3, with average counts ?1250uL, or single count ?900uL during treatment. After multivariate analysis, independent predictors included CTP class C (OR 3.8, 95% CI, 1.4-10.7), and neutrophil count ? 900μL during treatment (OR 9.2, 95% CI, 4.6-18.3). During the first 6 months of the trial, there were four deaths in the control group, and five in the treated group; deaths were related to sepsis in two and four patients, respectively.
Long-term outcome
Median follow-up period off therapy was 30 months (range 4-38), similar between the two groups. When analyzed according to the time of the first event, decompensated events during the follow-up occurred in 52 controls, 33 non-responders, and three patients with SVR (Table 4). As compared with controls, the odds decreased significantly for ascites, encephalopathy, and oesophageal variceal bleeding in treated patients, independently from viral clearance. As compared with baseline value, the CTP score significantly improved at the end of follow-up in SVR patients (7.8±1.1 vs 6.4±0.7; p=0.001) and worsened both in non-responders (8.0±1.3 vs 8.7±1.5; p=0.02), and controls (8.3±1.3 vs 9.4±1.6; p<0.001). At the end of follow-up the MELD score significantly improved in SVRs (14.1±2.9 vs 10.5±2.3; p=0.0005), and worsened both in non-responders (14.3±3.8 vs 15.8±4.7; p=0.4), and controls (14.5±3.4 vs 16.7±3.2; p<0.0001).
During follow-up, there were 15 deaths in controls, and nine in the non-responders; the number of patients who underwent liver transplantation was four and one, respectively. Liver failure and/or hepatocellular carcinoma was the cause of death or transplantation in 18 controls and nine non-responders. All 13 patients who experienced SVR survived, with no need for transplantation. The annualized incidence of death from any cause was 2.34 per 1000 patient-years in controls, 1.91in non-responders, and 0 in patients with SVR. The Kaplan-Meier curves for the end-point of death showed early separation of patients with SVR from both non-responders and controls at approximately 6 months of the post-treatment follow-up (Fig. 1).
|
|
| |
| |
|
|
|
